tag:blogger.com,1999:blog-25509522801585955422024-03-13T12:06:50.434-07:00Computer hardware softwareYou can get some information about computer hardware and softwareUnknownnoreply@blogger.comBlogger228125tag:blogger.com,1999:blog-2550952280158595542.post-52088346578936166952014-01-12T02:21:00.001-08:002014-01-12T02:21:49.492-08:00Hard Disk Data Recovery Services<div align="justify">
Read on to know more about the various causes for Hard Disk Failure and how to go about Hard Disk Data Recovery for desktops or laptops… </div>
<div align="justify">
</div>
<div align="justify">
Hard Disks of computers – either laptops or desktops – are susceptible to failure where the data that you have stored on it either disappears / cannot be found. The good news is that this data can often be recovered. There are various ways of doing so. </div>
<div align="justify">
Before you venture into hard disk data recovery territory, it is essential to know the cause of the hard disk failure. The cause can be either <strong>logical, or electrical or mechanical</strong>.</div>
<div align="justify">
1.<strong>Logical</strong> causes include problems with the boot sector of the hard drive which leads to non-booting of Windows. In this case, you would be able to you’re your computer from another disk. You can then investigate if you have problems with the boot sector or you have bad sectors on the disk itself. Hard disk data recovery in this case can be possible via Data Recovery Software.</div>
<div align="justify">
2.<strong>Electrical</strong> causes include problems caused due to voltage problems which may have led to the burning of the circuit board. For this you may have to open your desktop’s ATX box or your laptop’s bottom cover. Ensure that you have shut off the electrical supply to your machine and are working in a static-free environment. You also need to wear an ESD wrist strap. Under no circumstances must you touch the circuit board as this might damage the disk head. Just try to see if you can see any signs of burnt components or any signs of burns on the circuit board. In this case, you can contact your hardware supplier or a service agent to figure out the next step.</div>
<div align="justify">
3.<strong>Mechanical</strong> problems are caused when the hard drive and / or its parts have been physically damaged due to a number of causes. You may hear clicking sounds from the hard disk itself. This kind of failure can also be caused by dropping the hard disk or exposing it to water or humid atmospheres. Under no circumstances must you try to shake the hard disk or remove the hard disk cover. Most times, a physical repair of hard disks or their parts is not possible. You can find a data recovery service provider who may be able to help you with this problem. Look for service providers who have a ‘No Data, No Pay’ agreement.</div>
<div align="justify">
</div>
<div align="justify">
Thus, based on the causes of hard disk failure you can find different methods of hard disk data recovery. </div>
<div align="justify">
</div>
<div align="justify">
Here are some tips for general hard disk data recovery tips suggested by professionals in this field:<br />
<strong>1</strong>.Always use firewalls and good virus protection programs for your computer.<br />
<strong>2</strong>.Don’t run the disk defragmenter if you want to recover a file that you have ended up deleting. <strong>3</strong>.Use a Data Recovery Software first. Doing so will increase your chances of recovering the lost data.<br />
<strong>4</strong>.Programs such as scandisk, chkdsk or Norton Disk Doctor will not help you with data recovery. <strong>5</strong>.In fact, they may often end up hinder the process of hard disk data recovery.<br />
<strong>6</strong>.If your computer is not booting, you can try booting it in safe mode and try to disable or uninstall any recently installed programs which may have caused the problem. You can also try and see if running System Restore can solve the problem effectively.<br />
<strong>7</strong>.Before you buy data recovery software, ensure that you know exactly what kind of data it recovers. Instead of trying out a variety of softwares, first know what kind of data you want to recover and then look for that data recovery software which will meet your needs.<br />
<strong>8</strong>.Know what kind of file system your computer’s hard drive is using. This will help in locating the right kind of data recovery software for your computer data recovery process. Similarly, knowing the file system used by your laptop will help in the laptop data recovery process.<br />
<strong>9</strong>.The best tip ever is to always keep a backup of your important data.<br />
Sometimes it may not be possible to constantly keep backing up data, especially with recent files. If your hard disk crashes while you are working, don’t try to save the file as you may also end up losing data that has been saved earlier.<br />
<strong>10</strong>.Before you go to the hard disk data recovery experts make sure that you know the relevant details such as the Operating System, the details of what happened before your hard disk failed, and if your computer still recognizes the drive or not.</div>
<div align="justify">
</div>
<div align="justify">
These tips will help you identify the problem with your hard disk and even help you with hard disk data recovery on your own. However, if you are unsure, then it is best that you find a data recovery service provider who offers a ‘No Data, No Pay’ agreement. </div>
<div align="justify">
In fact, you can also negotiate well with such a data recovery service provider better if you have an idea of the cause of the hard disk failure.</div>
<div align="justify">
By Madhavi Ghare</div>
Unknownnoreply@blogger.com1tag:blogger.com,1999:blog-2550952280158595542.post-14445616146487456052009-08-19T18:25:00.000-07:002009-08-19T18:27:30.328-07:00Devices and Drivers.<h2><a name="HDR1">5 Devices and Drivers.</a></h2> <h3><a name="HDR2">5.1 What are they?</a></h3> <ul><li>Devices are presented as <i>special files</i> in /dev<br /></li><li>Devices are either <i>block</i> or <i>character</i> special files<br /></li><li>Any peripheral is seen by Unix as a device<br /></li><li>Even <i>memory</i> is a device (/dev/kmem and /dev/mem)<br /></li><li>Devices are created with mknod just as directories are created with mkdir<br /></li><li>Devices have <i>major</i> and <i>minor</i> number<br /></li><li>The <i>major</i> number represents the <i>device driver</i><br /></li><li>The <i>minor</i> number represents the <i>instance of a device</i> of the type specified by the major device number<br /></li><li>Devices are created by /dev/MAKEDEV script (SunOS 5.x creates devices at boot time; use boot -r)<br /></li></ul> Here is an <a href="http://www.washington.edu/R870/examples/dev">example</a> of the contents of the /dev directory.<p> </p><h3><a name="HDR3">5.2 Description of SCSI.</a></h3> The Small Computer System Interface (pronounced "scuzzy") specification was developed to allow up to 7 devices to be intelligently controlled by each interface rather than the computer. SCSI adapters are common on most workstations today as part of the motherboard.<p> </p><ul><li>Uses 50 pin "Centronics" style connectors<br /></li><li>Devices daisy-chained to form a "bus" topology<br /></li><li>Only <i>one</i> SCSI device should be terminated - the final one (the bus is also terminated internally)<br /></li><li>SCSI controller occupies address 7 - all other devices can use addresses 0 through 6 (addresses do not have to be contiguous, e.g. can use 1, 2 and 4 for devices)<br /></li><li>Addresses <i>must</i> be unique<br /></li></ul> <h3><a name="HDR4">5.3 Disk Special File Names (standard /usr partition)</a></h3> <pre>Table 3: <b>Disk Special File Names</b><br />--------------------------------------------------------------------------------<br /> <b>BSD</b> <b>System V.4</b> <b>SunOS 4.1.3</b> <b>IRIX</b> <br />--------------------------------------------------------------------------------<br />File Name /dev/rdisk0g /dev/rdsk/c1d0s2 /dev/rsd0g /dev/rdsk/dks0d0s2<br />Raw access mode /dev/<b>r</b>disk0g /dev/<b>r</b>dsk/c1d0s2 /dev/<b>r</b>sd0g /dev/<b>r</b>dsk/dks0d0s2<br />Device Type /dev/r<b>disk</b>0g /dev/r<b>dsk</b>/c1d0s2 /dev/r<b>sd</b>0g /dev/r<b>dsk</b>/<b>dk</b>s0d0s2<br />Drive # /dev/rdisk<b>0</b>g /dev/rdsk/c1<b>d0</b>s2 /dev/rsd<b>0</b>g /dev/rdsk/dks0<b>d0</b>s2<br />Disk Partition /dev/rdisk0<b>g</b> /dev/rdsk/c1d0<b>s2</b> /dev/rsd0<b>g</b> /dev/rdsk/dks0d0<b>s2</b><br />Controller # /dev/rdsk/<b>c1</b>d0s2 /dev/rdsk/dk<b>s0</b>d0s2<br />--------------------------------------------------------------------------------<br /></pre> OSF/1 (Digital Unix) and Digital Ultrix look similar to BSD style, while SunOS 5.x uses System V.4 style device files but also provides BSD style device names via use of symbolic links (for backward compatibility with SunOS 4.x).<p> </p><h4><a name="HDR5">5.3.1 Particular naming schemes for SCSI discs</a></h4> <ul><li>SunOS 4.x - /dev/sd<b><i>AP</i></b><br /></li><li>SunOS 5.x - /dev/[r]dsk/c<b><i>C</i></b>t<b><i>A</i></b>d0s<b><i>S</i></b><br /></li><li>Digital Unix - /dev/[r]rz<b><i>NP</i></b><br /></li><li>Linux - /dev/sd<b><i>LK</i></b><br />Where: A = SCSI address, C = Controller number, K = DOS primary/extended partition, L = drive letter "a" through "h", N = 8 * controller # + SCSI address, P = BSD-style partition, S = System V-style slice<br />[Note: Linux device names and portioning is quite different from other flavors of Unix. Partition 0, or no partition number, refers to the whole drive. Partitions 1-4 are DOS primary partitions, while partitions 5-8 are extended DOS partitions.]<br /></li></ul> <h3><a name="HDR6">5.4 Common "tricks" with devices.</a></h3> <ul><li><a href="http://www.washington.edu/R870/examples/tape-links">Example</a> of association between /dev/*tape and tape device<br /></li><li><a href="http://www.washington.edu/R870/examples/floppy-links">Example</a> of association between /dev/floppy and high-density floppy device<br /></li></ul> <h3><a name="HDR7">5.5 Example: Adding a tape drive to a Digital Unix system</a></h3> <ul><li>Determine the "unit number" of the device(assume first SCSI bus, SCSI ID of tape drive is 5)<br /></li></ul> <pre>N = (8 * SCSI bus #) + SCSI target #<br />= (8 * 0) + 5<br />= 5<br /></pre> <ul><li>Create device entries<br /></li></ul> <pre># cd /dev<br /># ./MAKEDEV tz5<br />MAKEDEV: special file(s) for tz5:<br />rmt0l<br />rmt0h<br />rmt0m<br />rmt0a<br />nrmt0l<br />nrmt0h<br />nrmt0m<br />nrmt0a<br /></pre> <h3><a name="HDR8">5.6 Example Problems</a></h3> <ul><li>Determine the number of inodes on the root partition (See "man df"). How many more files can you create on that partition?<br /></li><li>Determine the device names of all devices on your system<br /></li><li>If you have a tape drive, how do you specify use of non-rewind mode? How do you specify the high-density mode?<br /></li><li>If you were to add a new CD-ROM drive or SCSI tape device (assuming you don't have either yet), what would its device name be? (Check in /dev/MAKEDEV to see how its device files are created.)</li></ul>www.washington.edUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-44150751639690546862009-08-19T18:24:00.000-07:002009-08-19T18:25:13.700-07:00Printing.<h2><a name="HDR1">17 Printing.</a></h2> Steps to set up a local printer on your Unix workstation.<p> </p><ul><li>Physically connect the printer to the computer (for serial ports, you may need a null modem cable)<br /> </li><li>For serial line printers, create or modify an entry in the terminal line configuration file<br /> </li><li>Configure the proper files on the system to communicate correctly with the proper port<br /> </li><li>Create a spooling directory and accounting files for this print queue<br /> </li><li>Set up proper filter programs (if necessary)<br /> </li><li>Configure the printer system to allow printing to your new device<br /> </li><li>Enable printing on the new device<br /> </li><li>Send a print job to test the printer<br /></li></ul> <h3><a name="HDR2">17.1 BSD spooling.</a></h3> The lpr system uses a single printer capabilities file to control printing; which device, what filter to use, baud rates, form feeds, remote host and remote device name, etc. The lpr system can work over a network.<p> </p><ul><li>lpr - queues print jobs<br /> </li><li>lprm - removes print jobs form queue<br /> </li><li>lpq - shows jobs in print queue<br /> </li><li>lpc - printer control program<br /> </li><li>lpd - printer daemon<br /> </li><li>/etc/printcap - printer configuration file describing all printers on this host (and possibly other remote hosts)<br /> </li><li>Put entry in /etc/ttys (/etc/ttytab under SunOS 4.1.3) for serial port with status off, type unknown and keyword none.<br /></li></ul> <h4><a name="HDR3">17.1.1 BSD printcap file.</a></h4> Here is an <a href="http://www.washington.edu/R870/examples/etc-printcap">example </a>/etc/printcap file Note that this is for a Postscript printer. It needs special filters to convert ordinary text into Postscript output. The software to do this, Transcript<b>, </b>is available free to campus systems. The Transcript install program made this printcap entry.<p> </p><h3><a name="HDR4">17.2 System V printing.</a></h3> The lp system uses scripts that control the printer devices. (System V.4 adds remote printing and PostScript capabilities)<p> </p><ul><li>lp - initiates print requests<br /> </li><li>cancel - cancels pending print requests<br /> </li><li>lpstat - to list queue contents and show status<br /> </li><li>accept - allows lp to accept requests<br /> </li><li>reject - prevents lp from accepting requests<br /> </li><li>enable - activates named printer<br /> </li><li>disable - disables named printer<br /> </li><li>lpadmin - configures lp spooling system<br /> </li><li>lpsched - line printer scheduling daemon<br /> </li><li>/usr/spool/lp/model - directory where model files are kept<br /></li></ul> <h3><a name="HDR5">17.3 AIX queueing.</a></h3> AIX supports the BSD and System V printer commands and adds extra functionality by integrating printing with its general queueing system.<br /><br />www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-32347184827619641042009-08-19T18:23:00.000-07:002009-08-19T18:24:09.537-07:00Installing Terminals & Modems.<h2><a name="HDR1">16 Installing Terminals & Modems.</a></h2> A few facts about (ASCII or "dumb") terminals and modems:<p> </p><ul><li>Terminals and modems are <i>asynchronous</i> devices<br /> </li><li>Most use <i>RS-232</i> standards (such as they are)<br /> </li><li>Terminals and modems used for logins must deal with getty<br /> </li><li>Modems used for both dial-in and dial-out have special needs<br /> </li><li>Asynchronous data transfer is usually much slower than ethernet<br /> </li><li>Official limit for RS-232 lines is 50 feet<br /> </li><li>Most workstations and desktop Unix systems only have two async ports<br /></li></ul> <h3><a name="HDR2">16.1 Pins used for RS-232 serial connections.</a></h3> <pre><b>Table 8: <b>RS-232 pin assignments</b></b><br />-----------------------------------------------------------------------------------<br /><b>Pin</b> <b>Name</b> <b>Function</b> <b>Pin</b> <b>Name</b> <b>Function</b> <br />-----------------------------------------------------------------------------------<br />1 FG Field ground 14 STD Secondary transmitted data <br />2 TD Transmitted data 15 TC Transmitted clock <br />3 RD Received data 16 SRD Secondary received data <br />4 RTS Request to send 17 RC Received clock <br />5 CTS Clear to send 18 (Not assigned) <br />6 DR Data set ready 19 SRTS Secondary request to send <br />7 SG Signal ground 20 DTR Data terminal ready <br />8 DCD Data carrier detect 21 SQ Signal quality detector <br />9 Positive test voltage 22 RI Ring indicator <br />10 Negative test voltage 23 DRS Data rate selector <br />11 (Not assigned) 24 SCTE Serial clock transmit external <br />12 SDCD Secondary data carrier detect 25 BUSY Busy <br />13 SCTS Secondary clear to send <br />-----------------------------------------------------------------------------------<br /></pre> Most of these pins have no practical use, except:<p> </p><ul><li>Ground pins 1 (FG) and/or 7 (SG) are used<br /> </li><li>If only one ground is used, it should be 7<br /> </li><li>Pins 2 (TD), 3 (RD) and one ground pin <i>must</i> be used<br /> </li><li>Pins 4 (RTS) and 5 (CTS) may be used for HW handshaking<br /> </li><li>Modems commonly use pins 6 (DSR) and 20 (DTR)<br /></li></ul> When installing cables in walls/conduit, it is best to use straight-through cables using all 25 pins and use smaller, specialized (e.g., null modem) cables or converters between the ends of the routed cable and the devices.<p> </p><h3><a name="HDR3">16.2 Equipment types.</a></h3> <ul><li>DTE - Data Terminal Equipment (computers are <i>usually</i> DTE)<br /> </li><li>DCE - Data Communications Equipment (modems are <i>always</i> DCE)<br /></li></ul> Connecting a <a href="http://www.washington.edu/R870/img/DTE-DCE.gif">modem to a computer</a> is different than <a href="http://www.washington.edu/R870/img/DTE-DTE.gif">connecting two computers</a> together.<p> </p><h3><a name="HDR4">16.3 Hardware handshaking (RTS/CTS).</a></h3> <ul><li>DTR - "Ready to communicate"<br /> </li><li>Tells modem that computer is ready/able to communicate - if this signal drops, modem should hang up.<br /> </li><li>DSR - "Ready to receive data"<br /></li></ul> Tells computer that modem can accept data (ignored by many U.S Unix systems which use DCD instead).<p> </p><ul><li>DCD - "Have remote carrier"<br /></li></ul> Dropped signal means that modem lost connection to other system.<p> </p><ul><li>RTS (Request to Send) and CTS (Clear to Send) lines used to "throttle" transmission (lines not shown)<br /></li></ul> <h3><a name="HDR5">16.4 Software Handshaking (XON/XOFF).</a></h3> <ul><li>XOFF (ASCII Control-S)<br />Used to tell other side to stop transmitting data.<br /> </li><li>XON (ASCII Control-Q)<br />Used to tell other side that it is OK to continue transmitting data.<br /></li></ul> Since hardware handshaking is implemented in the serial port hardware, it is far more efficient and reliable. This is because:<p> </p><ul><li>the time delay between transmission of XOFF character and actual halting of transmission can cause buffer overflow and lost data (especially at very high baud rates)<br /> </li><li>transmitted data may contain XON/XOFF characters which fool the hardware<br /></li></ul> <h3><a name="HDR6">16.5 Steps in installing terminals/modems</a></h3> <ul><li>Physically connect the terminal/modem to the computer (DTE<->DTE or DTE<->DCE)<br /> </li><li>Determine the special file in /dev that communicates with the terminal/modem<br /> </li><li>With terminals, make sure a termcap or terminfo entry exists for this terminal (you may need to create one)<br /> </li><li>Add/modify an entry in the proper configuration files<br /> </li><li>Force init to reread the terminal configuration information (kill -HUP 1 or init -q)<br /> </li><li>Dial-in/dial-out modems are a special case(May need to send modem initialization codes and save to non-volatile memory using tip or cu)<br /></li></ul> <h3><a name="HDR7">16.6 /etc/ttys file</a></h3> Here is an <a href="http://www.washington.edu/R870/examples/etc-ttys">example </a>/etc/ttys file from a BSD-style system (a NeXT). The uncommented lines with "on" in the fifth column are active and will have associated getty processes. The second field (between the " ") points to entries in the /etc/gettytab file.<br /><br />www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-44103322509010273552009-08-19T18:10:00.001-07:002009-08-19T18:10:38.836-07:00Configuring the kernel.<h2><a name="HDR1">14 Configuring the kernel.</a></h2> <h3><a name="HDR2">14.1 Why?</a></h3> Not necessary with most modern UNIX systems, except when you:<p> </p><ul><li>add or remove a type of hardware, or<br /> </li><li>add/remove some subsystems (e.g., NFS, ISO9660, SLIP/PPP)<br /></li></ul> <h3><a name="HDR3">14.2 Basic Steps</a></h3> <ul><li>Make sure you understand the hardware on your system (e.g., disc/tape drives, network interface, mice)<br /> </li><li>Create/edit the configuration file in the conf directory (usually /sys/conf or /usr/sys/conf).<br /> </li><li>Run config in conf directory.<br /> </li><li>Use make depend and make <i>kernel-name</i> to create the new kernel.<br /> </li><li>Rename the old kernel and copy the new one into / to test it out. If it fails, reboot with the old kernel.<br /></li></ul> <h3><a name="HDR4">14.3 Examples of generating an Ultrix kernel</a></h3> <ul><li><a href="http://www.washington.edu/R870/examples/usr-sys-conf-mips-SOFTY">Example configuration file</a><br /> </li><li><a href="http://www.washington.edu/R870/examples/doconfig.output">Example of </a>doconfig command output</li></ul>www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-8601935352180544232009-08-19T18:09:00.001-07:002009-08-19T18:09:48.401-07:00Unix System Security.<h2><a name="HDR1">13 Unix System Security.</a></h2> Securing any computer system involves many steps, including:<p> </p><ul><li>Knowing how the system may be attacked<br /> </li><li>Knowing how to protect against these attacks<br /> </li><li>Monitoring your system<br /> </li><li>Teaching users how to be "safe" and getting them to work with you<br /> </li><li>Knowing what to do when you detect you've been hit<br /></li></ul> <h3><a name="HDR2">13.1 Knowing how the system may be attacked</a></h3> The best place to start is by reading <a href="file://jhunix.hcf.jhu.edu/pub/miscellaneous_security_papers/">articles</a>, such as:<p> </p><ul><li><a href="http://stimpy.cac.washington.edu/%7Edittrich/R870/cracking_guide.html">Guide to Cracking Unix</a> article<br /> </li><li><a href="http://www.cis.ohio-state.edu/hypertext/faq/usenet/computer-security/">Usenet computer-security FAQ files</a><br /> </li><li><a href="http://www.cis.ohio-state.edu/hypertext/faq/usenet/computer-security/compromise-faq/faq.html">Compromise FAQ</a><br /></li></ul> <h4><a name="HDR3">13.1.1 Detecting an attack</a></h4> Common methods of attack include:<p> </p><ul><li>System crashes<br /> <ul><li>Repeat crashes <i>usually</i> hardware related<br /> </li><li>Read log files/console for panic reasons<br /> </li><li>Any patterns (e.g., time of day) should raise suspicion<br /></li></ul> </li><li>New accounts<br /> <ul><li>Especially with UID 0<br /></li></ul> </li><li>New files<br /> <ul><li>With setuid/setgid bits or root ownership<br /> </li><li>Hidden directories (e.g., "..." or with control characters in the name, such as backspace)<br /></li></ul> </li><li>Modified file<br /> <ul><li>System programs (e.g., login, sh, csh, ps)<br /> </li><li>Configuration files (e.g., /etc/rc, /etc/aliases, /etc/hosts.equiv)<br /> </li><li>User files (e.g., .login, .cshrc, .history, .rhosts)<br /></li></ul> </li><li>Poor Performance<br /> <ul><li>May indicate password cracking or packet sniffing programs<br /> </li><li>Programs to use: ps, iostat, vmstat, pstat, sar, netstat<br /></li></ul> </li><li>Denial of service<br /> <ul><li>Exhaustion of finite resources<br /> <ul><li>File space on partitions (e.g., filling /tmp)<br /> </li><li>inode table<br /> </li><li>process table<br /> </li><li>open files<br /> </li><li>virutal memory<br /> </li><li>network services<br /></li></ul> </li><li>Destruction of data<br /> <ul><li>Deleting user files/system configuration files/system programs<br /></li></ul> </li></ul> </li></ul> <h3><a name="HDR4">13.2 Knowing how to protect against these attacks</a></h3> Make sure you've applied all recommended security patches to your system.<p> </p><ul><li><a href="http://www.cis.ohio-state.edu/hypertext/faq/usenet/computer-security/security-patches/faq.html">Security Patches FAQ</a><br /></li></ul> Learn about, and use, tools and papers from sources such as these:<p> </p><ul><li>The <a href="http://www.cs.purdue.edu/coast/coast.html">COAST</a> project at Purdue University<br /> </li><li>The National Institute of Standards and Technology's (NIST)<a href="http://www.ncsl.nist.gov/cs.html">Computer Security Division</a> (CSD)<br /> </li><li><a href="http://www.alw.nih.gov/Security/security.html">Unix Security Information</a> from the <a href="http://www.alw.nih.gov/">Advanced Laboratory Workstation Project</a> at the National Institute of Health (NIH)<br /> </li><li>The <a href="file://cert.org/pub/">Computer Emergency Response Team</a> (CERT)<br /> </li><li><a href="http://www.tansu.com.au/Info/security.html">Computer and Network Security Reference Index</a> (Note: this site is in Australia)<br /></li></ul> Only provide as much access as is necessary:<p> </p><ul><li>Put access controls on IP services with tcp_wrapper<br /> </li><li>Remove IP services you don't really need, such as fingerd, sendmail, and tftp<br /></li></ul> <h3><a name="HDR5">13.3 Monitoring your system</a></h3> <ul><li>At the very least, log access with something like <a href="http://ciac.llnl.gov/ciac/ToolsUnixNetSec.html#Tcpwrappers">tcp_wrappe</a>r - add to effectiveness by logging to another host (see "man syslogd")<br /> </li><li><b>Read your log files</b>, or use a tool like <a href="http://ciac.llnl.gov/ciac/ToolsUnixSysMon.html#Swatch">swatch</a><br /> </li><li>Detect modified programs with <a href="http://ciac.llnl.gov/ciac/ToolsUnixSysMon.html#Tripwire">tripwire</a><br /></li></ul> <h3><a name="HDR6">13.4 Teaching users how to be "safe" and getting them to work with you</a></h3> <ul><li><a href="http://www.alw.nih.gov/Security/Docs/passwd.html">Select Good Passwords</a> (and change them regularly)<br /> </li><li>Take care with .rhosts files<br /> </li><li>Don't share accounts/passwords (one person, one account)<br /> </li><li>Understand how to <a href="http://stimpy.cac.washington.edu/%7Edittrich/R870/x-security.html">safely use the X Window System</a><br /></li></ul> <h3><a name="HDR7">13.5 Knowing what to do when you detect you've been hit</a></h3> <ul><li>Have security policies in place. COAST has <a href="ftp://coast.cs.purdue.edu/pub/doc/policy">example security policies</a>, including examples from <a href="ftp://coast.cs.purdue.edu/pub/doc/policy/University-Policies/">several Universities</a>.<br /> </li><li>There are some good guidelines in <a href="file://nic.ddn.mil/rfc/rfc1244.txt">RFC1244</a> (chapters 5 and 6) on post-incident response, but it also provides much information on developing policies and procedures<br /> </li><li>Send email to help@cac.washington.edu to report the incident<br /> </li><li>Gather and preserve evidence - it may be needed by law enforcement investigators<br /></li></ul> <h3><a name="HDR8">13.6 Unix System Security Checklist</a></h3> This <a href="http://stimpy.cac.washington.edu/%7Edittrich/R870/security-checklist.html">Unix System Security Checklist</a> can serve as a tool to help you with improving and auditing the security of your systems.<p> </p><h3><a name="HDR9">13.7 Vendor Security/Patch Web Sites</a></h3> <ul><li>Silicon Graphics -- <a href="http://www.sgi.com/Support/security/security.html">http://www.sgi.com/Support/security/security.html</a><br /> </li><li>IBM -- <a href="http://www.ibm.com/Security/html/resources.html">http://www.ibm.com/Security/html/resources.html</a><br /> </li><li>Digital Equipment -- <a href="http://www.service.digital.com/patches/index.html">http://www.service.digital.com/patches/index.html</a><br /> </li><li>Sun -- <a href="http://sunsolve.sun.com/">http://sunsolve.sun.com/</a><br /> </li><li>RedHat (Linux) -- <a href="http://www.redhat.com/support/docs/errata.html">http://www.redhat.com/docs/errata.html</a><br /> </li><li>Microsoft -- <a href="http://www.microsoft.com/security/">http://www.microsoft.com/security/</a><br /></li></ul> <h3><a name="HDR10">13.8 Computer Security related sites on the Internet</a></h3> <ul><li><a href="http://crimelab.com/bugtraq/bugtraq.html">http://crimelab.com/bugtraq/bugtraq.html</a> (Bugtraq archive)<br /> </li><li><a href="http://www.cs.purdue.edu/coast/coast.html">http://www.cs.purdue.edu/coast/coast.html</a> (Purdue's COAST home page)<br /> </li><li><a href="http://csrc.ncsl.nist.gov/">http://csrc.ncsl.nist.gov/</a> (NIST Clearning Security Resource Clearinghouse)<br /> </li><li><a href="http://www.openmarket.com/info/intindex/">http://www.openmarket.com/info/intindex/</a> (Internet Facts)<br /> </li><li><a href="http://www.dct.ac.uk/www/books/hacker-crackdown/hacker.html">http://www.dct.ac.uk/www/books/hacker-crackdown/hacker.html</a>(Hacker Crackdown book)<br /> </li><li><a href="http://www-ns.rutgers.edu/www-security/reference.html">http://www-ns.rutgers.edu/www-security/reference.html</a>(Rutgers WWW Security References)<br /> </li><li><a href="http://www.commerce.net:8000/directories/jumpstation/sectools.html">http://www.commerce.net:8000/directories/jumpstation/sectools.html</a>(Electronic Commerce Jumpstation - Security Tools)<br /> </li><li><a href="http://www.underground.org/">http://www.underground.org/</a><br /> </li><li><a href="http://www.alw.nih.gov/Security/security-docs.html">http://www.alw.nih.gov/Security/security-docs.html</a> (NIH web pages)<br /> </li><li><a href="http://www.faqs.org/faqs/sgi/faq/security/index.html">http://www.faqs.org/faqs/sgi/faq/security/</a>l (SGI Security FAQ)<br /> </li><li><a href="http://sites.inka.de/%7EW1012/freefire-l/tools.html">http://sites.inka.de/~W1012/freefire-l/tools.html</a> (Freefire project)<br /></li></ul> <h3><a name="HDR11">13.9 Example Problems and Reading</a></h3> <ul><li>Get a copy of Crack (a password cracker) from <a href="ftp://cert.sei.cmu.edu/pub/tools/crack/">cert.sei.cmu.edu</a> and run it against your /etc/passwd file. See how many passwords it can break.<br /> </li><li>Get a copy of COPS (a system security checker) from <a href="ftp://cert.sei.cmu.edu/pub/tools/cops/">cert.sei.cmu.edu</a> and run it on your system to see what it finds. You may wish to set this up so it runs under cron to regularly check your system.<br /> </li><li>Here are some interesting <a href="file://jhunix.hcf.jhu.edu/pub/miscellaneous_security_papers/">articles/papers</a> on system security issues.</li></ul>www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-81003001129249671842009-08-19T18:08:00.001-07:002009-08-19T18:09:01.291-07:00Using an NFS server.<h2><a name="HDR1">12 Using an NFS server.</a></h2> <h3><a name="HDR2">12.1 NFS "exporting" (server side).</a></h3> Server host must first be configured to allow your host to mount directories<p> </p><ul><li>/etc/exports defines the directories that are exported and which hosts are allowed to mount them(SunOS 5.x uses /etc/dfs/dfstab in slightly different way)<br /> </li><li>exportfs is used to make directories available for mounting(SunOS 5.x systems use the share and unshare commands)<br /> </li><li>/etc/xtab lists information about currently exported directories<br /> </li><li>mountd processes mount requests, verifying that requesting clients are allowed to mount by reading the /etc/xtab file<br /></li></ul> Here is an <a href="http://www.washington.edu/R870/examples/etc-exports">example</a> of an /etc/exports file.<p> </p><h3><a name="HDR3">12.2 NFS mounting (client side).</a></h3> Your host must then mount the remote directories<p> </p><ul><li>showmount will show mountable file systems on remote machines(Here is an <a href="http://www.washington.edu/R870/examples/showmount-example">example</a> of using showmount)<br /> </li><li>mount command used to mount directories, e.g.<br /></li></ul> <pre>mount host.dept.washington.edu:/usr/man/usr/man<br />mount host.dept.washington.edu:/usr/users/home<br /></pre> <ul><li>/etc/fstab contains list of all file systems mounted by default<br /> </li><li>Can use mount -p to help build /etc/fstab entries<br /><pre>Tabell 7: <b>Commonly used NFS mount flags</b><br />-----------------------------------------------------------------------<br /><b>Flag</b> <b>Description</b> <br />-----------------------------------------------------------------------<br />ro Mount the file system read-only <br />rw Mount the file system read/write (only if server exports as <br /> writable) <br />bg If the server doesn't respond, keep trying the mount in the <br /> background - may also hang startup scripts, or cause <br /> strange failures, if they rely on files in these file systems <br />hard Blocks processes accessing files until server can respond <br /> (may look like your shell has "hung") <br />soft Processes trying to access files get an error message if <br /> server is not responding <br />intr Allows user to abort a blocked access (returns error also) <br />timeo= Timeout period (in tenths of a second) for requests <br />retrans= Specify number of times to retry requests before return <br /> ing an error when soft mount used. <br />-----------------------------------------------------------------------<br /></pre> </li></ul> <h3><a name="HDR4">12.3 Gotchas! with NFS</a></h3> <ul><li>Ownership represented by numeric UID/GID values, which correspond to logins viathe /etc/passwd file on <b><i>both client and server</i></b>!(What if UIDs don't match?)<br /> </li><li>Security is a problem (e.g., can chown be used to give files away?).See <a href="http://stimpy.cac.washington.edu/%7Edittrich/R870/security-checklist.html">Unix System Security Checklist</a> for details.<br /> </li><li>NFS is a "stateless" protocol. What happens if the server goes down?<br /> <ul><li>"NFS: stale file handle" error message means re-mount file systems<br /> </li><li>"Hard mount" is "Energizer Bunny" mode. Client keeps trying, and trying, and trying.... Unless interrupts allowed, client will "hang" until server comes back up<br /> </li><li>"Soft mount" means client gets an error message.<br /></li></ul> </li><li>File locking can be a problem.<br /> </li><li>Easy to "waste" bandwidth by transferring files twice!(e.g., ftp to/from an NFS mounted file system on a client)<br /> </li><li>Forgetting to use -xdev (or -mount, etc)., when using find.<br /> </li><li>Differences in maximum block sizes between client/server file systems can cause programs to crash<br /></li></ul> <h3><a name="HDR5">12.4 Related Articles</a></h3> <ul><li>"How to Develop a Distributed Filesystem Model," Michael S. Hill, <i><a href="http://www.samag.com/">Sys Admin</a></i>, August 1996<br /> </li><li>"Hidden Dangers of NFS Mounting Foreign File Systems," Doug Morris, <i><a href="http://www.samag.com/">Sys Admin</a></i>, August 1996</li></ul>www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-23085789088972041302009-08-19T18:06:00.000-07:002009-08-19T18:08:12.323-07:00Networking.<h2><a name="HDR1">11 Networking.</a></h2> <h3><a name="HDR2">11.1 UW network concepts.</a></h3> <ul><li>The predominant networking protocols for world-wide internetworking are the <a href="file://ftp.merit.edu/introducing.the.internet/intro.to.ip">TCP/IP</a> protocols<br /> </li><li>UW campus <i>backbone</i> only routes <i>IP</i> traffic<br /> </li><li>The <i>Domain Name System</i> (<i>DNS</i>) is used to reference hosts instead of IP addresses (e.g., rtfm.mit.edu)<br /> </li><li>UW campus is <i>subnetted</i> (128.95.subnet.host)<br /> </li><li>Subnets are connected by <i>routers</i> (campus standard address for gateway router is 128.95.subnet.100)<br /> </li><li>Individual network connections within the building are usually served by ethernet multi-port repeaters (<i>empr</i>)<br /> </li><li>The campus backbone is part of <i><a href="http://www.nwnet.net/">NorthWestNet</a></i>, one of many regional network access providers who in turn connect to one of many national networks (NorthWestNet connects to MCInet).<br /> </li><li>All these tens of thousands of smaller <a href="file://ftp.merit.edu/nsfnet/statistics/history.netcount">networks</a> with millions of connected <a href="file://ftp.merit.edu/nsfnet/statistics/history.hosts">hosts</a> make up the <i>Internet</i> (a.k.a. <i>"The Net"</i>)<br /></li></ul> You can find information about the Internet at many sites. Try looking at what is stored at <a href="file://ftp.merit.edu/">ftp.merit.edu</a>. Here are some other <a href="http://www.openmarket.com/diversions/internet-index/index.html">interesting statistics on the Internet</a>.<p> </p><h3><a name="HDR3">11.2 Example of subnetted network.</a></h3> This <a href="http://www.washington.edu/R870/img/Network.gif">diagram</a> shows an example of a subnetted network.<p> </p><h3><a name="HDR4">11.3 Network topologies.</a></h3> Here is a <a href="http://www.washington.edu/R870/examples/Ethernet">description of the technical details of Ethernet</a> networks. The most common topologies on this campus are:<p> </p><ul><li><a href="http://www.washington.edu/R870/img/StarTopology.gif">Star</a> (e.g., twisted pair ethernet, or 10BaseT). This is the type of network technology currently recommended by C&C and what we would install today.<br /> </li><li><a href="http://www.washington.edu/R870/img/BusTopology.gif">Bus</a> (e.g., thin ethernet, or 10 Base 2). The older standard for inexpensive Ethernet networks, but still found in many places on campus.<br /> </li><li>Ring (e.g., Token Ring, FDDI, CDDI)<br /></li></ul> <h3><a name="HDR5">11.4 DNS name service.</a></h3> <ul><li>Host name + Domain = Fully Qualified Domain Name<br /></li></ul> E.g., host stimpy in domain cac.washington.edu has a fully qualified domain name (FQDN) of stimpy.cac.washington.edu<p> </p><ul><li>Host name set in rc files using hostname command<br /></li></ul> <pre># hostname ren<br /></pre> <ul><li>Resolver configuration file is /etc/resolv.conf<br /></li></ul> <pre>domain cac.washington.edu<br />nameserver 128.95.120.1<br />nameserver 128.95.112.1<br />nameserver 128.101.101.101<br /></pre> <ul><li>DNS is <i>not</i> related to Sun's NIS/NIS+, although they can work together.<br /></li></ul> SunOS 4.x: get replacement libraries from <a href="file://ftp.cac.washington.edu/pub/sun-fix/">ftp.cac:/pub/sun-fix</a>. SunOS 5.x users need to modify /etc/nsswitch.conf.<p> </p><h3><a name="HDR6">11.5 ISO vs. TCP/IP protocol stacks.</a></h3> <a href="http://www.washington.edu/R870/Networking.figure.id.1.gif"><img style="width: 465px; height: 352px;" src="http://www.washington.edu/R870/Networking.figure.id.1.gif" /></a><p> </p><h3><a name="HDR7">11.6 Protocols, ports, and servers.</a></h3> <ul><li>Protocol numbers defined in <a href="http://www.washington.edu/R870/examples/etc-protocols">/etc/protocols</a><br /> </li><li>Port numbers for services defined in <a href="http://www.washington.edu/R870/examples/etc-services">/etc/services</a><br /> </li><li>Servers are managed by the Internet daemon inetd<br /> </li><li>Servers are defined in <a href="http://www.washington.edu/R870/examples/etc-inetd.conf">/etc/inetd.conf</a><br /></li></ul> To force inetd to reread its configuration file, send it a -HUP signal, like this:<p> </p><pre># ps -aux | grep inetd<br />root 111 0.0 1.7 1.44M 344K ? S 0:00 (inetd)<br />root 4451 0.0 1.0 1.52M 208K p2 S 0:00 grep inetd<br /># kill -HUP 111 # ...for BSD<br /># killall -HUP inetd # ...for System V<br /></pre> <h3><a name="HDR8">11.7 Configuring the ethernet interface (BSD).</a></h3> <ul><li>/etc/rc or /etc/rc.local<br /><a href="http://www.washington.edu/R870/examples/rc.local">Here</a> is an example of the part of the /etc/rc.local file that configures the ethernet interface.<br /></li></ul> <h3><a name="HDR9">11.8 Configuring the ethernet interface (System V).</a></h3> <ul><li>/etc/init.d/network<br /></li></ul> <h3><a name="HDR10">11.9 Configuring the networking software.</a></h3> <ul><li>Make entry in /etc/hosts for your host and IP address<br /> </li><li>Register your DNS name<br />(Get IP addresses and Domain Names from your departmental network administrator, or from campus network administrators by sending email to netops@cac.washington.edu )<br /> </li><li>Set up the DNS resolver configuration file<br /> </li><li>Make sure the netmask is 255.255.255.0 (0xffffff00)<br /> </li><li>Make sure the broadcast address is128.95.subnet.255 (use 128.95.subnet.0 for Sun computers)<br /></li></ul> <a href="http://www.washington.edu/R870/examples/sun-network">Here</a> is an example of setting up the network interface for a Sun computer.<p> </p><h3><a name="HDR11">11.10 User commands.</a></h3> <ul><li>telnet<br /> </li><li>ftp<br /> </li><li>rlogin<br /> </li><li>rsh<br /> </li><li>rcp<br /> </li><li>rwho<br /> </li><li>ruptime<br /></li></ul> <h3><a name="HDR12">11.11 Public domain utilities/facilities.</a></h3> <ul><li>Anonymous ftp<br />Log in as user name of ftp or anonymous, and using your email address (e.g., name@host.dept.washington.edu) as the password.<br /> </li><li>gopher/xgopher<br /> </li><li>archie/xarchie<br /> </li><li>Xmosaic<br /> </li><li>WAIS<br />These utilities and facilities are described in detail in <i>Whole Internet: User's Guide & Catalog</i>, <i>Internet Companion: a Beginner's Guide to Global Networking</i>, and<i> Internet Passport.</i><br /></li></ul> <h3><a name="HDR13">11.12 Standard diagnostic tools.</a></h3> <ul><li>netstat<br />The netstat utility can be used to <a href="http://www.washington.edu/R870/examples/netstat-i">determine the name of your interface</a> and to <a href="http://www.washington.edu/R870/examples/netstat-r">check the routing table</a> to see if your gateway is accessible.<br /> </li><li>ifconfig<br />The ifconfig utility is used to configure (or <a href="http://www.washington.edu/R870/examples/ifconfig">check the configuration</a> of) your network interface.<br /> </li><li>ping<br /><a href="http://www.washington.edu/R870/examples/ping">Tells if you can send/receive packets</a> with another Internet host.<br /> </li><li>arp<br />Exercises the name resolution mechanism and <a href="http://www.washington.edu/R870/examples/arp">reports IP/ethernet addresses</a>. (<a href="http://www.washington.edu/R870/examples/EthernetCodes">Here</a> are some Ethernet codes, in case you need them).<br /></li></ul> <h3><a name="HDR14">11.13 Public domain diagnostic tools.</a></h3> <ul><li>lsof<br />A useful tool for finding which processes have open file handles, it also helps find which processes have open sockets. This is very handy for identifying processes involved in network connectivity problems. Available from <a href="ftp://vic.cc.purdue.edu/pub/tools/unix/lsof/">ftp://vic.cc.purdue.edu/pub/</a>tools/unix/lsof/<br /> </li><li>dig<br />An alternative to nslookup. It has a method of doing reverse name lookups, which can be handy in determining the name of hosts that are sending packets across your subnet. Available from <a href="file://venera.isi.edu/pub/">ftp://venera.isi.edu/pub/.</a><br /> </li><li>traceroute<br />Traces the route of UDP packets from your host to another host. Can be used to diagnose "no answer" and "network unreachable" error messages from TCP/IP utilities. [<i>Usually</i>] shows all interfaces that packets go through. Available from <a href="file://ftp.ee.lbl.gov/">ftp://ftp.ee.lbl.gov/trac</a>eroute.tar.<br /> </li><li>tcpdump<br />"Network" or "protocol" analyzer that lets you see some/all packets that are traversing your subnet. Allows you to select or filter packets "promiscuously." Available from <a href="ftp://ftp.ee.lbl.gov/tcpdump.tar.Z">ftp://</a>ftp.ee.lbl.gov/tcpdump.tar.Z<br /> </li><li>tcpview<br />Motif version of the tcpdump utility. Produced by C&C for use here on campus. Available from <a href="file://ftp.cac.washington.edu/pub/noc-tools/tcpview/">ftp://ftp.cac.washington.edu/pub/noc-tools/tcpview</a>.<br /> </li><li>etherman/interman/packetman<br />X Windows programs to monitor ethernet, IP, and packet level traffic on your subnet. Available in binary form for Ultrix 4.2/4.3 and SunOS 4.1.3 (sun4c) from <a href="file://ftp.cs.curtin.edu.au/pub/netman/">ftp://tfp.cs.cur</a>tin.edu.au/pub/netman.<br /> </li><li>getethers<br />Builds a table of Ethernet numbers, host names and manufacturer suitable for use as an /etc/ethers file. Available from <a href="ftp://harbor.ecn.purdue.edu/pub/davy/">ftp://harbor.ecn.purdue.edu/pub/davy/</a><br /></li></ul> <h3><a name="HDR15">11.14 Public domain security tools.</a></h3> <ul><li>nmap<br />Network mapping tool<a href="http://www.insecure.org/nmap/index.html">http://www.insecure.org/nmap/index.html</a><br /> </li><li>trinux<br />All the network monitoring tools you need, on a two-floppy, RAM disc based, Linux distribution! (Can even boot on a laptop). Available from <a href="http://www.trinux.org/">http://www.trinux.org/</a><br /> </li><li>tcp wrapper (a.k.a. tcpd)<br />This utility adds access control and logging features to TCP/IP services. It is available in source code form for most popular Unix implementations from <a href="file://cert.org/pub/tools/tcp_wrappers/">ftp://cert.org/pub/tools/</a>tcp_wrappers/.<br /> </li><li>smrsh<br />This program provides a more secure mechanism for shell execution from other programs such as sendmail. It is avaiable as C code with a man page from <a href="ftp://cert.org/pub/tools/smrsh">ftp://cert.org/</a>tools/smrsh.<br /></li></ul> <h3><a name="HDR16">11.15 Useful information</a></h3> <ul><li><a href="ftp://ftp.isi.edu/in-notes/iana/assignments/">Internet Assigned Numbers Authority (IANA) assignments</a><br /> </li><li><a href="ftp://ftp.ripe.net/iso3166-countrycodes">ISO Country codes</a></li></ul>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-89339303668515208092009-08-19T18:05:00.002-07:002009-08-19T18:06:28.804-07:00Adding new users.<h2><a name="HDR1">10 Adding new users.</a></h2> Steps involved (usually automated).<p> </p><ul><li>Determine login name, user ID (UID), group ID (GID) & netgroup (if using NIS or NIS+)<br /> </li><li>Enter data in /etc/passwd file<br /> </li><li>Assign password<br /> </li><li>Set account parameters (e.g., password aging, account expiration, quotas)<br /> </li><li>Create home directory<br /> </li><li>Copy initialization files<br /> </li><li>Set file ownerships and protections (use chown, chgrp, chmod commands)<br /> </li><li>Add user to other facilities (quota system, aliases file)<br /> </li><li>Perform other site-specific initialization tasks<br /> </li><li>Test new account<br /></li></ul> <h3><a name="HDR2">10.1 Fields in the /etc/passwd file.</a></h3> <pre>logname:passwd:uid:gid:user info:home:shell<br /></pre> <h3><a name="HDR3">10.2 Fields in the /etc/group file.</a></h3> <pre>group:passwd:gid:members<br /></pre> <h3><a name="HDR4">10.3 Choosing passwords.</a></h3> One of the best algorithms to use is to take two non-related words, separate them with a punctuation character, mix in some upper-case characters, and perhaps reverse one of the words. E.g.,<p> </p><ul><li>DOg.tenT (mixed upper/lower case)<br /> </li><li>t00l%p0nd (zeros instead of "O")<br /> </li><li>p33l*BOARd (substitute numeric for alpha)<br /> </li><li>tned-Pile (reversed first word)<br /></li></ul> <h3><a name="HDR5">10.4 Shell initialization files.</a></h3> <ul><li>C shell: .cshrc, .login, .logout, /etc/cshrc,/etc/cshrc.std<br /> </li><li>Borne shell: .profile, /etc/profile,/etc/profile.std<br /> </li><li>Skeleton files: in /usr/skel, /etc/skel, or/etc/security directories<br /></li></ul> <h3><a name="HDR6">10.5 Automating account creation.</a></h3> <ul><li>sysadm (System V)<br /> </li><li>passmgmt (System V)<br /> </li><li>useradd (System V.4)<br /> </li><li>adduser, addgroup (Some BSD, OSF/1)<br /> </li><li>SMIT (AIX)<br /> </li><li>nu (NeXTStep)<br /> </li><li>Public domain scripts<br /></li></ul> <h3><a name="HDR7">10.6 Disabling and/or removing user accounts.</a></h3> <ul><li>Remove or modify entry in /etc/passwd<br /> </li><li>Remove entry in NIS/NIS+ maps<br /> </li><li>Remove $HOME/.rhosts files<br /> </li><li>Remove mail spool file<br /> </li><li>Remove from mail aliases file<br /> </li><li>Remove any cron or at jobs<br /> </li><li>Remove directory<br /></li></ul> <h3><a name="HDR8">10.7 Related articles</a></h3> "Developing Common Login Scripts," Dinah McNutt, <i>Unix Review</i>, December 1995, p73. (includes sample <a href="ftp://ftp.zilker.net/pubs/ur/profile.ksh">.profile file</a> for the Korn shell)<br /><br />www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-61046017095102278752009-08-19T18:05:00.001-07:002009-08-19T18:05:45.484-07:00Startup/Shutdown.<h2><a name="HDR1">9 Startup/Shutdown.</a></h2> <h3><a name="HDR2">9.1 Boot Phases.</a></h3> <ul><li>ROM monitor may run some simple hardware diagnostics and finds boot device<br /> </li><li>ROM monitor loads/runs bootstrap loader<br /> </li><li>Bootstrap loader loads/runs kernel (PID 0)<br /> </li><li>Kernel may perform more elaborate diagnostics, then checks root file system<br /> </li><li>Kernel mounts root file system, then starts init(PID 1)<br /> </li><li>Operator may need to switch to multi-user mode<br /> </li><li>init runs initialization scripts (rc scripts)<br /> </li><li>getty/login produces prompts on terminals (or graphics system starts up) and system is ready to go<br /></li></ul> This example of the <a href="http://www.washington.edu/R870/examples/ps-ef.SVR4">output of System V style ps</a> will help identify these processes.<p> </p><h3><a name="HDR3">9.2 Starting a boot.</a></h3> <ul><li>Autoboot from power-on<br /> </li><li>Boot command from ROM monitor<br /></li></ul> <h3><a name="HDR4">9.3 BSD single-user mode vs. multi-user mode.</a></h3> <ul><li>Berkeley UNIX has just two operating states: single-user mode and multi-user mode.<br /></li></ul> <h3><a name="HDR5">9.4 System V run levels.</a></h3> <ul><li>System V has several states, or "run levels"<br /> </li><li>Use who -r to see current run level<br /></li></ul> <pre># who -r<br />. run-level 2 Mar 5 15:32 2 0 S<br /><pre>Table 6: <b>System V Run Levels</b><br />-------------------------------------------------------------------------------------------<br /><b>Run Level</b> <b>Names and Uses</b> <br />-------------------------------------------------------------------------------------------<br />0 Powerdown state; conditions where it is safe to turn the power off. <br />1 Administrative state. <br />s or S Single-user mode. <br />2 Multi-user mode; Normal operating state for isolated, <br /> non-networked systems. <br />3 Remote File Sharing (RFS) state. Alternative state for networking use. (Normal <br /> multi-user state for OSF/1; other Sys V may use state 2). <br />4 User-definable state. (Also not normally used; state 2 used instead). <br />5 Firmware state; Used for some types of maintenance on some systems. <br />6 Shutdown and reboot state; used to reboot the system from some other running <br /> state (s, 2, 3 or 4) to state 0 then immediately back to normal operating state. <br />-------------------------------------------------------------------------------------------<br /></pre><br /></pre> <h3><a name="HDR6">9.5 Initialization scripts.</a></h3> The initialization scripts, usually written in Borne shell (BSD, System V) or Korn shell (AIX) perform the following tasks:<p> </p><ul><li>Sets computer's hostname (BSD and some System V)<br /> </li><li>Sets time zone (System V only)<br /> </li><li>Checks unmounted discs with fsck (only in multi-user mode)<br /> </li><li>Mounts the system's disc partitions<br /> </li><li>Removes files from the /tmp directory<br /> </li><li>Starts up daemons and network services<br /> </li><li>Turns on accounting and quotas (if used)<br /> </li><li>Configures network interface(s)<br /> </li><li>Starts windowing system and window-based login (on graphics workstations)<br /></li></ul> <h3><a name="HDR7">9.6 BSD rc files.</a></h3> <ul><li>/etc/rc (multi-user initialization)<br /> </li><li>/etc/rc.local (site-specific initialization)<br /> </li><li>/etc/rc.boot (NeXTStep single-user initialization and primary SunOS initialization file)<br /> </li><li>/etc/rc.single (under SunOS single-user mode starts shell; under SunOS multi-user mode runs/etc/rc)<br /> </li><li>/etc/rc.net, /etc/rc.tcpip, /etc/rc.pci,/etc/rc.ncs (AIX)<br /></li></ul> Here is an example of a BSD style <a href="http://www.washington.edu/R870/examples/rc.local">/etc/rc.local file</a>.<p> </p><h3><a name="HDR8">9.7 System V initialization files.</a></h3> <ul><li>init reads /etc/inittab to find what to execute<br /> </li><li>Initialization files stored in /etc/init.d<br /> </li><li>Links made into /etc/rc0.d, /etc/rc2.d, etc.<br /></li></ul> Here is an example of a simple initialization file, the <a href="http://www.washington.edu/R870/examples/init.d-lp">/etc/init.d/lp file</a>.<p> </p><h3><a name="HDR9">9.8 OSF/1 initialization files.</a></h3> <ul><li>Same as System V, but...<br /> </li><li>Initialization files stored in /sbin/init.d<br /> </li><li>Links made into /sbin/rc0.d, /sbin/rc2.d, etc.<br /></li></ul> <h3><a name="HDR10">9.9 System V initialization</a></h3> This <a href="http://www.washington.edu/R870/img/Boot.gif">diagram</a> shows the process of initialization under System V style Unix.<p> </p><h3><a name="HDR11">9.10 Shutting down</a></h3> (<b><i>WARNING</i></b>! Never just turn the computer off!)<p> </p><ul><li>Notify users (use wall and/or /etc/motd to announce shutdown, if not done for you)<br /> </li><li>Disable logins under BSD (create /etc/nologin, if not done for you by shutdown)<br /> </li><li>BSD: shutdown <i>time message</i><br /></li></ul> <pre># shutdown +60 "Shutting down for weekly backups"<br /># shutdown now "There's water coming through the ceiling"<br /></pre> <ul><li>System V: shutdown -g<i>n</i> -i<i>level</i> [-y]<br /></li></ul> <pre># shutdown -g60 -iS<br /># shutdown -g0 -i0 -y<br /></pre> <ul><li>reboot command (BSD, some System V, OSF/1)<br /> </li><li>halt command (BSD, some System V, OSF/1)<br /></li></ul> <h3><a name="HDR12">9.11 System crashes.</a></h3> <ul><li>Hardware failures<br /> </li><li>Power failures/surges<br /> </li><li>Environmental problems<br /> </li><li>I/O problems (including filled partitions)<br /> </li><li>Software problems (kernel faults, bugs, programs)<br /></li></ul> <h3><a name="HDR13">9.12 System won't boot.</a></h3> <ul><li>Always wait at least ten seconds when power cycling<br /> </li><li>Same causes as above<br /> </li><li>Errors in initialization scripts<br /> </li><li>Improperly configured kernel</li></ul>www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-48324164184367649462009-08-19T18:04:00.001-07:002009-08-19T18:04:55.535-07:00Backups.<h2><a name="HDR1">8 Backups.</a></h2><h3><a name="HDR2">8.1 Media types</a></h3> Here are some average example prices as of February 1998:<p> </p><pre>Table 4: Backup media comparison<br />----------------------------------------------------------<br /><b>Type</b> <b>Drive price</b> <b>Media price</b> <b>Capacity </b> <b>Price/MB</b> <br /> <b>(MB)</b> <br />----------------------------------------------------------<br />Floppy disc $150 $0.50 2.88 $0.17 <br />QIC $300 $20 250 $0.08 <br />Jaz drive $600 $125 2,000 $0.0625 <br />9 track $? (2) $10 160 $0.06 <br />CD-R (3) $100 $12 660 $0.018 <br />Hard disc $100 $100 (1) 8,200 $0.01 <br />DLT (4) $4,000 $70 30,000 $0.002 <br />8mm [video] $2,200 $9 5,100 $0.0018 <br />4mm [DAT] $1,800 $7 5,500 $0.0013 <br />----------------------------------------------------------<br /></pre> <ul><li>(1) Since media cost equals the drive cost, this option is about the same overall price as any type of tape drive system. Although not as flexible as tape, it is much faster, is random access, and is immediately bootable!<br /> </li><li>(2) Not very common anymore, but still here for comparison<br /> </li><li>(3) While only useful for one write operation per CD, the media lifetime of CD-ROM is on the order of 50 years vs. ~5yrs for magnetic tapes<br /> </li><li>(4) Digital Linear Tape (DLT). With compression, capacity can exceed 70GB/tape. Tapes are designed to take far more reads/writes than 4mm/8mm tapes.<br /></li></ul> <h3><a name="HDR3">8.2 Utilities.</a></h3> <ul><li>dump/restore - backs up file systems, has interactive mode, can do incremental backups, maintains "sparse files", is most commonly used utility<br /> </li><li>cpio - can back up individual files/directories, handles special files, packs data tighter than tar, skips bad spots on media on restore, use with find (some versions of find have -cpio option for this purpose)<br /> </li><li>tar - backs up directory trees, does not back up special files, poor error handling with media errors, does not pack blocks (GNU tar solves some of these problems)<br /> </li><li>dd - copies/converts files, can go from one medium to another, processes whole entity or select blocks, can swap bytes and do ASCII/EBCDIC conversions<br /> </li><li>rmt - used for remote tape operations<br /> </li><li><a href="http://www.washington.edu/R870/examples/BackupPrograms">Here</a> is an email message comparing several backup utilities.<br /></li></ul> <h3><a name="HDR4">8.3 Backup Model.</a></h3> A useful backup model suggested by 3M is the "Grandfather, Father, Son" principle. This scheme uses a total of ten units of backup media (where a unit is sufficient to backup the entire system) to provide up to twelve weeks of coverage.<p> Label ten sets of media with these labels: Monday, Tuesday, Wednesday, Thursday, Friday 1, Friday 2, Friday 3, Month 1, Month 2, Month 3.</p><p> Start the cycle on a Friday. Do a full backup onto the Friday 1 media.</p><p> Every Monday through Thursday, use the appropriately labelled media to do an incremental backup.</p><p> Each Friday, use the next Friday media set to do another full backup, continuing to use the Monday through Thursday incremental scheme.</p><p> Every forth Friday, use one of the monthly media sets (Month 1 through Month 3) to do the full backup.</p><p> After four months you start the sequence over again. Write the date used on the media each time, and watch error rates, to ensure you recycle tapes before they become unreadable.Backup Model (Example).</p><p> </p><pre><b>Table 5: <b>Example Backup Schedule</b></b><br />------------------------------------------------<br /><b>Monday</b> <b>Tuesday</b> <b>Wednesday</b> <b>Thursday</b> <b>Friday</b> <br />------------------------------------------------<br />Monday Tuesday Wednesday Thursday Friday1 <br />Monday... ... Friday2 <br />... ... Friday3 <br /> Month1 <br /> Friday1 <br /> Friday2 <br /> Friday3 <br /> Month2 <br /> <br />------------------------------------------------<br /></pre> <h3><a name="HDR5">8.4 Copying partitions/portions of file systems.</a></h3> <ul><li>Example using dump<br /></li></ul> <pre># dump 0df 100000 - /oldtree |<br />(cd /newtree; restore xf -)<br /></pre> Specify a very high density parameter (d 100000) to make dump think it's dumping to a very high-capacity "tape." Otherwise, dump may stop in mid-copy and demand that you mount a new "tape."<p> </p><ul><li>Example using tar<br /></li></ul> <pre># (cd /oldroot; tar cvf - ./dir) |<br />(cd /newroot; tar xf -)<br /></pre> <ul><li>Make sure directories have proper ownership and protection masks after transferring (especially when done across systems)<br /></li></ul> <h3><a name="HDR6">8.5 Determining a backup strategy</a></h3> Assumption: The next time you sit down at your system, you find that the hard drive has crashed or a cracker has done an "rm -rf /" while you were gone.<p> </p><ul><li>How many drives and partitions do you have?<br /> </li><li>What does each partition hold? How unique is it?<br /> </li><li>How often do the contents change?<br /> </li><li>How likely is the disc to fail? (E.g., a six year old i386)<br /> </li><li>How much does it cost to replace the contents?<br /> </li><li>How much can your media hold?<br /> </li><li>How do you to recover, since you can't boot from disc anymore?<br /> </li><li>How long does it take to back up/restore? (How much time can you afford?)<br /> </li><li>What if you only need to restore some files?<br /></li></ul> <h3><a name="HDR7">8.6 Example Problems</a></h3> Do a backup to tape/floppy disc using the <a href="http://www.washington.edu/R870/examples/cbackup">cbackup</a> script. After you do this, first make sure it worked by listing the contents using the <a href="http://www.washington.edu/R870/examples/clist">clist</a> script, then restore a set of files to the /tmp directory using the <a href="http://www.washington.edu/R870/examples/crestore">crestore</a> script to make sure you can get them back. (<a href="http://www.washington.edu/R870/examples/README.cbackup">This</a> describes the three scripts).<br /><br />www.washington.edu<br /><h2><a name="HDR1"></a></h2><h2><a name="HDR1"></a></h2>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-44771216674192599272009-08-19T18:02:00.000-07:002009-08-19T18:03:32.057-07:00Adding a disc drive.<h2><a name="HDR1">7 Adding a disc drive.</a></h2><h2>There is never enough disk space. The minute a new disk is added to the system, it is half full, or so it seems. Getting users to clean up their disk space is as difficult as getting a teenage boy to clean up his room. Therefore, the system administrator will occasionally have to install new disk drives.</h2><p> </p><h3><a name="HDR2">7.1 Disc drive terminology.</a></h3> <ul><li>The circular plate coated with magnetic media is called a <i>platter</i> - drives usually have several platters<br /> </li><li>The <i>heads</i> ride on a microscopic film of air on the platter surface - all heads move together<br /> </li><li>The platters spin rapidly and the heads move in and out to position on a <i>track</i> - this action is called <i>seeking</i><br /> </li><li>Each track is made up of a number of <i>sectors</i><br /> </li><li>The collection of tracks under all of the heads is called a <i>cylinder</i><br /> </li><li>When a head touches the surface of a platter, it is called a <i>head crash</i> - this is a Bad Thing (go buy a new disc and get out your last backup!)<br /> </li><li>Discs are logically divided up (by groupings of cylinders or sectors) into <i>partitions</i> -partitions usually overlap<br /> </li><li><i>File systems</i> occupy entire partitions (not all partitions have file systems, e.g. the swap partition)<br /> </li><li>Every file system has a <i>superblock</i> which contains information such as the file system's label, size, and total number of inodes<br /></li></ul> This <a href="http://www.washington.edu/R870/img/Disc.gif">diagram</a> shows an idealized disc drive. Here is an <a href="http://www.washington.edu/R870/img/FileSystem.gif">example</a> of how several file systems on three disc drives might be mounted.<p> </p><h3><a name="HDR3">7.2 Steps in adding a disc drive.</a></h3> <ul><li>Make a <i>device</i> entry in /dev for the new disc (use /dev/MAKEDEV)<br /> </li><li>Make an entry in /etc/disktab (4.2 BSD) or /etc/device.tab (System V.4) for your new disc<br /> </li><li>Set the SCSI address to a legal, unused SCSI id<br /> </li><li>Connect the drive to the SCSI bus<br /> </li><li>Make sure the disc is formatted/partitioned correctly<br /> </li><li>Make files systems using newfs or mkfs<br /> </li><li>Make an entry in /etc/fstab for the new partitions<br /> </li><li>Mount the new<i> </i>partitions<br /> </li><li>Transfer file systems (if necessary)<br /> </li><li>Do site specific configurations (e.g., BSD quotas)<br /></li></ul> <h3><a name="HDR4">7.3 Commands for determining device names/numbers</a></h3> <ul><li>SunOS 4.x/5.x - dmesg<br /> </li><li>IRIX - hinv<br /> </li><li>OSF/1 (Digital Unix), Ultrix - uerf<br /></li></ul> <h3><a name="HDR5">7.4 Commands for partitioning/formatting drives</a></h3> <ul><li>SunOS 4.x/5.x - format<br /> </li><li>IRIX - fx<br /> </li><li>OSF/1 - disklabel (gets partition information from /etc/disktab)<br /> </li><li>Ultrix - chpt, rzdisk<br /> </li><li>Linux - fdisk<br /></li></ul> <h3><a name="HDR6">7.5 Commands for examining partition information</a></h3> <ul><li>SunOS 4.x - dkinfo<br /> </li><li>SunOS 5.x, IRIX - prtvtoc<br /> </li><li>Ultrix - chpt<br /> </li><li>Linux - fdisk<br /></li></ul> <h3><a name="HDR7">7.6 Example BSD partitioning.</a></h3> BSD systems usually use 8 partitions (labelled "a" through "h") <p> </p><pre>Example Eagle disc partitions<br />------------------------------------<br /><b>Partition</b> <b>Cylinders</b> <b>Typical use</b> <br />------------------------------------<br />a 0-15 / <br />b 16-86 Swap <br />c 0-841 Entire disc <br />d 391-407 Alternate Root <br />e 408-727 <br />f 728-841 <br />g 391-841 /usr <br />h 87-390 <br />------------------------------------<br /></pre> There are several possible ways of creating file systems on these partitions. Here is a <a href="http://www.washington.edu/R870/img/DiscPartitions.gif">diagram</a> of three alternatives.<p> </p><h3><a name="HDR8">7.7 Example System V partitioning.</a></h3> System V systems number their partitions 0 through 7. (Example from IRIX system)<p> </p><pre># prtvtoc c0t0d0s2<br />* c0t0d0s2 partition map<br />*<br />* Dimensions:<br />* 512 bytes/sector<br />* 84 sectors/track<br />* 15 tracks/cylinder<br />* 1260 sectors/cylinder<br />* 1658 cylinders<br />* 1656 accessible cylinders<br />*<br />* Flags:<br />* 1: unmountable<br />* 10: read-only<br />*<br />* First Sector Last<br />* Partition Tag Flags Sector Count Sector Mount Directory<br />0 2 00 0 103320 103319 /<br />1 3 01 103320 132300 235619<br />2 5 00 0 2086560 2086559<br />5 6 00 235620 1106280 1341899 /opt<br />6 4 00 1341900 437220 1779119 /usr<br />7 4 00 1779120 307440 2086559 /usr/openwin<br /></pre> <h3><a name="HDR9">7.8 The Berkeley /etc/disktab file.</a></h3> Here is an <a href="http://www.washington.edu/R870/examples/disktab">example of an /etc/disktab</a> file used here in C&C for Ultrix systems. The descriptions of the entries can be found in the man pages for disktab (see "<a href="http://www.washington.edu/R870/examples/disktab.man">man disktab</a>").<p> </p><h3><a name="HDR10">7.9 The SunOS format.dat file.</a></h3> SunOS uses a file called format.dat. <a href="ftp://ra.mcs.anl.gov/sun-managers/format.dat">This example</a> is from the <a href="ftp://ra.mcs.anl.gov/sun-managers/faq">sun-managers</a> email list.<p> </p><h3><a name="HDR11">7.10 /etc/fstab file (System V).</a></h3> <pre>blck-spcl-file mnt-loc [fstype] [options]<br />/dev/root / efs rw,raw=/dev/rroot 0 0<br />/dev/usr /usr efs rw,raw=/dev/rusr 0 0<br /></pre> <h3><a name="HDR12">7.11 /etc/fstab file (4.3 BSD).</a></h3> <pre>blck-spcl-file mnt-loc fstype opts dump-freq passno<br />/dev/sd0a / 4.3 rw,noquota,noauto 0 1<br /></pre> <h3><a name="HDR13">7.12 /etc/fstab file (ULTRIX).</a></h3> <pre>blck-spcl-file:mnt-loc:type:dump-freq:passno:fstype:options<br />/dev/ra0a:/:rw:1:1:ufs::<br /></pre> <h3><a name="HDR14">7.13 Partitioning considerations</a></h3> <ul><li>When a file system approaches 100% full, fragmentation tends to occur in the newly created files. If they stick around, and are used frequently, this can have an adverse effect on disc I/O.<br /> </li><li>When a file system on a partition fills up, you can't create any more files. Period. The glass is full. This means that any programs that need to create files here will start to fail. This means you might want to put your Usenet news spool on a partition other than, say, / (where the operating system usually keeps /tmp) or /usr (where you may have home directories).<br /> </li><li>The inode table is fixed in size. It does not get bigger to accommodate more files. You may have plenty of space left, but with no more inodes available, you can't create any more files. (Same problem as disc filled.)<br /> </li><li>If a partition will have a high percentage of predominately large files, or a high percentage of predominately small files, the default block size may not efficiently allow the files to be organized on disc.<br /></li></ul> <h3><a name="HDR15">7.14 Remaking a disc partition</a></h3> <ul><li>Dismount file system<br /></li></ul> <pre># umount /data<br /></pre> <ul><li>Backup file system (assumes 6250 BPI tape density and length factor of 24000)<br /></li></ul> <pre># dump 0ds 6250 24000 /dev/disk2g<br /></pre> <ul><li>Check to see the backup is good<br /></li></ul> <pre># restore t<br /></pre> <ul><li>Remake the file system (assumes m2694esa 1.1gb 3.5" Fujitsu drive per disktab example)<br /></li></ul> <pre># newfs -b 16384 /dev/rdisk2g M2694ESA<br /></pre> <ul><li>Mount the new file system<br /></li></ul> <pre># mount /data<br /></pre> <ul><li>Restore files<br /></li></ul> <pre># cd /data<br /># restore r<br /></pre> <h3><a name="HDR16">7.15 Example Problems and Reading</a></h3> <ul><li><a href="ftp://sunsite.unc.edu/pub/Linux/docs/HOWTO/mini/Partition">Linux Partition HOWTO</a><br /> </li><li><a href="http://www.nyx.net/%7Esgjoen/disk.html">HOWTO: Multi Disk System Tuning</a><br /> </li><li><a href="http://www.washington.edu/R870/examples/fragmentation">Notes on disk fragmentation</a></li></ul>www.washington.edu<br /><h2><a name="HDR1"></a></h2>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-7671950564874306582009-08-19T18:00:00.000-07:002009-08-19T18:01:37.644-07:00Hardware maintenance tips.<h2><a name="HDR1">6 Hardware maintenance tips.</a></h2><h2><a name="HDR1"></a></h2>Some simple ideas that will help keep your hardware alive and healthy.<p> </p><ul><li>Shutdown, then power off the system before changing any part or plug (exceptions: serial port devices and ethernet interfaces)<br /> </li><li>Use anti-static mat and/or wrist strap<br /> </li><li>Be firm but gentle when seating boards, chips, connectors, etc.<br /> </li><li>Keep it clean, especially air filters & ventilation paths<br /> </li><li>Leave it powered up all the time (exception: video display)<br /> </li><li>Don't drink anything around your machinery<br /> </li><li>Put a power surge suppressor on the power line<br /> </li><li>Avoid vibrations and shocks (e.g., avoid stacking components</li></ul>www.washington.eduUnknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-85236266187183982892009-08-12T19:12:00.000-07:002009-08-12T19:13:43.317-07:00Core 2 Quad Q8200 O/C Performance And Efficiencyby <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-13.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><table class="editorTblTablecenter editorTblSize100 editorTblStyleStyle2"><tbody><tr class="tblEven"><th colspan="3">Intel Quad-Core Test Settings<br /></th></tr><tr class="tblRow"><th><p><strong> </strong></p></th><td><p><strong>Default Settings</strong></p></td><td><p><strong>Overclock Settings</strong></p></td></tr><tr class="tblEven"><th>CPU<br /></th><td rowspan="1"><p><strong>Intel Core 2 Quad Q8200</strong><br />2.33 GHz 4 MB L2 Cache, 1.16V</p></td><td rowspan="1"><p>2.69 GHz, (7x 384 MHz)</p><p>1.29V core, 1.40V FSB</p></td></tr><tr class="tblRow"><th>RAM<br /></th><td><p>DDR3-1333 CAS 9-9-9-24, 1.50V</p></td><td><p>DDR3-1536 CAS 6-6-5-16, 1.65V</p></td></tr><tr class="tblEven"><th>Motherboard<br /></th><td colspan="2"><strong>MSI P45 Diamond</strong> LGA-1366, P45/ICH10R, BIOS 1.5 (10/10/2009)<br /></td></tr><tr class="tblRow"><th>Graphics<br /></th><td colspan="2" rowspan="1"><p><strong>Zotac GeForce GTX260²</strong><br />576MHz GPU, 999 MHz Shader, 896MB GDDR3-2484 </p></td></tr><tr class="tblEven"><th>Hard Drive<br /></th><td colspan="2" rowspan="1"><p><strong>Western Digital VelociRaptor WD30000HLFS</strong><br />300 MB, 10,000 RPM, 16 MB Cache</p></td></tr><tr class="tblRow"><th>Sound<br /></th><td colspan="2"><p>Integrated HD Audio</p></td></tr><tr class="tblEven"><th>Networking<br /></th><td colspan="2"><p>Integrated Gigabit LAN</p></td></tr><tr class="tblRow"><td colspan="3"><p><strong>Software</strong></p></td></tr><tr class="tblEven"><th>Operating System<br /></th><td colspan="2"><p>Microsoft Windows Vista Ultimate x64 SP1</p></td></tr><tr class="tblRow"><th>Graphics<br /></th><td colspan="2"><p>GeForce 182.08 Desktop</p></td></tr></tbody></table> <p><br />A stock FSB clock that was already near the processor’s limit made this the most disappointing overclocking experience we can remember since National Semiconductor’s Cyrix MII. Its measly 15% gain felt like a tremendous achievement, however, considering the great effort required to reach a 2.69 GHz clock frequency.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/6/8/216656/original/image018.png" alt="" title="" /></span></p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/6/9/216657/original/image019.png" alt="" title="" /></span></p> <p>Sandra Arithmetic and Multimedia show CPU performance gains of 14 to 15 percent.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/6/A/216658/original/image020.png" alt="" title="" /></span></p> <p>A memory bandwidth increase of 19% looks much better, but isn’t as noteworthy in a CPU overclocking guide.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/6/B/216659/original/image021.png" alt="" title="" /></span></p> <p>Low core voltage plus high FSB voltage brings a significant penalty in power consumption, even though performance gains were mediocre at best.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/6/3/216651/original/image022.png" alt="" title="" /></span></p> <p>An efficiency loss of 9% results from a ratio of performance to power consumption for the overclocked configuration.</p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-57225780596770186332009-08-12T19:11:00.000-07:002009-08-12T19:12:36.779-07:00Overclocking Intel's Core 2 Quad Q8200by <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-12.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><div class="KonaBody news-elm"> <p>Intel’s value-priced Core 2 Quad Q8200 uses two of the same processor dice as the Pentium E5200, at a lower clock speed and a higher front side bus clock. The combination of moderate CPU frequency and higher FSB also requires a lower CPU multiplier, and Intel designs these so that the multiplier cannot be increased.</p> <p>Intel typically uses a low FSB on mainstream processors to modulate performance and expand compatibility, so we’re not certain why the company chose FSB-1333 for its cheapest quad-core models. We do, however, know that many overclockers specifically select low-cost processors for the higher multiplier that typically accompanies a lower bus speed, so that its use of FSB-1333 for the Q8000-series has all but prevented its adoption amongst enthusiasts. After looking at the far higher prices of Q9000-series processors, we returned resolute to get big gains from the Q8200, viewing its lower multiplier as a challenge.</p> <p>Unfortunately, the Q8200 would barely budge beyond its original 2.33 GHz frequency, regardless of how much voltage we applied to its core, reaching the same 2.5 GHz overclocked speed at core voltage settings from stock to 1.45 volts. The problem, it seems, is that FSB-1333 is almost the limit for these cores at stock FSB voltage.</p> <p>Dual-die processors of this design use the front side bus for both CPU-to-chipset and die-to-die communication, and increasing the CPU FSB beyond 354 MHz (2.5 GHz CPU clock) would require an increase of the “VTT FSB Voltage” setting seen in the second screenshot below.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc1,0101-216652-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Core2-Quad-Q8200,6-4-216652-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc1,0101-216652-0-2-3-1-png-.html">Zoom</a></span></span></p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc2,0101-216653-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Core2-Quad-Q8200,6-5-216653-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc2,0101-216653-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>Our research showed that CPU FSB voltage had the same practical limit as core voltage: 1.45 volts peak and “something less” under load for continuous long-term use. As with the core voltage of the E5200, we chose 1.40 volts as a target voltage for Q8200’s FSB. We were then able to increase the FSB to 384 MHz, but the resulting 15% overclock is barely worth the risk and effort.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc4,0101-216655-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Core2-Quad-Q8200,6-7-216655-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc4,0101-216655-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>We really wanted to reach at least the next “Intel standard” FSB clock of 400 MHz, or FSB-1600, but getting there required far more “VTT FSB Voltage” than we can safely recommend. Further research into other far-more-successful Q8200 overclocks revealed that those units were actually cream-of-the-crop “Q8200S” models.</p> <p>The CPU cores certainly wouldn’t need a full 1.40 volts at so low an overclock, so we began back-tracking. While “VTT FSB Voltage” remained at 1.40 volts for a stable 384 MHz FSB, we were able to drop the “CPU Voltage” setting to 1.30 volts. Anything less resulted in an eventual crash under <a target="_blank" href="ftp://mersenne.org/gimps/p64v258.zip">Prime95 v25.8 build 4</a>.</p> <p>With the chipset’s maximum memory clock rate of twice the CPU FSB clock, the fastest selectable memory clock of 768 MHz provided a data rate of DDR3-1536. As with the E5200, we then began stability tests using <a target="_blank" href="http://www.memtest.org/">Memtest86+ v1.70</a> at progressively lower DRAM latency settings until the best stable timings of 6-6-5-16 were determined.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc3,0101-216654-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Core2-Quad-Q8200,6-6-216654-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/core-2-quad_q8200_oc3,0101-216654-0-2-3-1-png-.html">Zoom</a></span></span></p> </div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-46656683616413843342009-08-12T19:09:00.001-07:002009-08-12T19:10:32.922-07:00Pentium E5200 O/C Performance And Efficiencyby <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-11.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><table class="editorTblTablecenter editorTblSize100 editorTblStyleStyle2"><tbody><tr class="tblEven"><th colspan="3">Intel Dual-Core Test Settings<br /></th></tr><tr class="tblRow"><th><p><strong> </strong></p></th><td><p><strong>Default Settings</strong></p></td><td><p><strong>Overclock Settings</strong></p></td></tr><tr class="tblEven"><th>CPU<br /></th><td rowspan="1"><p><strong>Intel Pentium E5200</strong><br />2.50 GHz 2 MB L2 Cache, 1.26V</p></td><td rowspan="1"><p>4.11 GHz, (12x 342 MHz), 1.40V</p></td></tr><tr class="tblRow"><th>RAM<br /></th><td><p>DDR3-800 CAS 6-6-6-15, 1.50V</p></td><td><p>DDR3-1368 CAS 5-5-5-12, 1.65V</p></td></tr><tr class="tblEven"><th>Motherboard<br /></th><td colspan="2"><strong>MSI P45 Diamond</strong> LGA-775, P45/ICH10R, BIOS 1.5 (10/10/2009)<br /></td></tr><tr class="tblRow"><th>Graphics<br /></th><td colspan="2" rowspan="1"><p><strong>Zotac GeForce GTX260²</strong><br />576MHz GPU, 999 MHz Shader, 896MB GDDR3-2484 </p></td></tr><tr class="tblEven"><th>Hard Drive<br /></th><td colspan="2" rowspan="1"><p><strong>Western Digital VelociRaptor WD30000HLFS</strong><br />300 MB, 10,000 RPM, 16 MB Cache</p></td></tr><tr class="tblRow"><th>Sound<br /></th><td colspan="2"><p>Integrated HD Audio</p></td></tr><tr class="tblEven"><th>Networking<br /></th><td colspan="2"><p>Integrated Gigabit LAN</p></td></tr><tr class="tblRow"><td colspan="3"><p><strong>Software</strong></p></td></tr><tr class="tblEven"><th>Operating System<br /></th><td colspan="2"><p>Microsoft Windows Vista Ultimate x64 SP1</p></td></tr><tr class="tblRow"><th>Graphics<br /></th><td colspan="2"><p>Forceware 182.08 Desktop</p></td></tr></tbody></table> <p><br />A CPU overclock of 64% yielded Sandra Arithmetic and Multimedia benchmark improvements of 63% and 64%, respectively.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/U/216642/original/image013.png" alt="" title="" /></span></p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/V/216643/original/image014.png" alt="" title="" /></span></p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/W/216644/original/image015.png" alt="" title="" /></span></p> <p>Optimized memory settings resulted in a Sandra Memory Bandwidth improvement of 77% over default values.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/6/E/216662/original/image016.png" alt="" title="" /></span></p> <p>At 1.40 volts, the overclocked Pentium E5200 required an average 34% power increase.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/X/216645/original/image017.png" alt="" title="" /></span></p> <p>A performance increase of 64% with an average power increase of 34% allowed the overclocked E5200 to improve its efficiency by 22% over stock settings.</p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-50183468310112110752009-08-12T19:07:00.000-07:002009-08-12T19:08:37.945-07:00Overclocking Intel's Pentium E5200by <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-10.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><p>As with AMD, Intel’s production technology already has well-known voltage limits affecting the majority of samples. For 45 nm processors based on the Core 2 architecture, CPU core voltage of 1.45 V is generally considered to be the maximum a processor can withstand over a period of many weeks or a few months. We’ve already seen an “office” system that was overclocked using 1.45 volts lose much of its overclocking capability over a period of around three months. This family of processors continues to scale well, even at much higher voltage levels, but cooling and longevity become an issue.</p> <p>Because we want our overclocks to last at least several months (and we keep our fingers crossed for 1-3 years of reliable service), we chose 1.40 volts as a target setting under full CPU load and 1.43 volts peak under no load. Knowing the desired voltage level ahead of time negated the normal practice of increasing voltage in small steps until the system became stable after reaching a clock rate that would have otherwise been unstable.</p> <p>The screenshots below show our final settings: P45 Diamond users must be forewarned that this memory setting required a jumper change which we’ll discuss further down this page.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/pentium_e5200_oc1,0101-216647-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Pentinum-Dual-Core-E5200,5-Z-216647-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/pentium_e5200_oc1,0101-216647-0-2-3-1-png-.html">Zoom</a></span></span></p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/pentium_e5200_oc2,0101-216648-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Pentinum-Dual-Core-E5200,6-0-216648-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/pentium_e5200_oc2,0101-216648-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>Except for its Extreme Edition models, Intel doesn’t allow adjusting its CPU multipliers upward. Using a 12.5x multiplier and 200 MHz FSB clock (FSB-800 via QDR technology), the only way this 2.50 GHz CPU would go faster would be to set a higher FSB. Knowing that the Pentium E5200 should reach at least 3.60 GHz with air cooling, we first tried the next-higher Intel-standard FSB of 266 MHz (FSB-1066). The system booted normally and passed a 40-minute stress test using <a target="_blank" href="ftp://mersenne.org/gimps/p64v258.zip">Prime95 v25.8 build 4</a>. CPU-Z reported core voltage dropping to 1.38V under load however, so we increased the BIOS setting “CPU Voltage (V)” (second screenshot above) to 1.4132 volts. This resulted in 1.424 volts at idle and 1.408 volts at full load.</p> <p>The MSI P45 Diamond supports most FSB clock settings, but we knew that the chipset would be most stable at or near one of Intel’s standard bus speeds. Our next attempt, FSB-1333 (333 MHz clock) would boot inconsistently, resulting in either a black screen or a system reset following the POST screen. After finding success at 320 MHz “most of the time,” we began to believe that our previous memory ratio problem was being caused by a <a target="_blank" href="http://www.tomshardware.com/reviews/ddr3-1333-speed-latency-shootout,1754-17.html">dreaded bootstrap issue</a>.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/p45-diamond_jumpers,0101-216646-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/Pentinum-Dual-Core-E5200,5-Y-216646-13.jpg" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/p45-diamond_jumpers,0101-216646-0-2-3-1-jpg-.html">Zoom</a></span></span></p> <p>The P45 Diamond doesn’t have “bootstrap” settings in BIOS but does have two jumpers for altering detected bus speed. According to the manual, changing both jumpers from pins 1-2 to pins 2-3 would allow a 200 MHz FSB processor to be detected as a 333 MHz FSB version. Following those instructions solved both the memory ratio and boot inconsistency problems.</p> <p>The system now booted at 333 MHz, but extended stability tests proved it wasn’t completely stable. In an effort to keep the system close to the 333 MHz “Intel standard” bus, we dropped the CPU multiplier to 12x.</p> <p>The CPU clock of 12 x 333 MHz proved stable for a longer one-hour Prime 95 test. Jumping ahead slightly, 338 MHz was also stable. We continued increasing FSB and testing stability until it was found that the highest stable CPU speed was 4.1 GHz at 12 x 342 MHz FSB.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/pentium_e5200_oc4,0101-216641-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Pentinum-Dual-Core-E5200,5-T-216641-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/pentium_e5200_oc4,0101-216641-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>Using our memory manufacturer’s rated setting of 1.65 volts, we began chasing good memory performance to match that impressive 64% CPU overclock. Intel’s DRAM multiplier limit of 2 x FSB clock unfortunately meant that our RAM could be set no higher than a 684 MHz clock, corresponding to a data rate of DDR3-1368.</p> <p>We began testing lower memory latency values to improve response time, using a bootable CD version of <a target="_blank" href="http://www.memtest.org/">Memtest86+ v1.70</a> in the same manner that Prime95 was used for CPU stability testing. Since the memory controller is part of the northbridge, we experimented with increased “MCH Voltage” (second screenshot above) until it was found that anything higher than 1.352 volts provided no further improvement.</p> <span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/pentium_e5200_oc3,0101-216649-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Pentinum-Dual-Core-E5200,6-1-216649-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /></span></span>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-62559001643340082682009-08-12T19:04:00.000-07:002009-08-12T19:05:23.177-07:00Phenom II X2 550 O/C Performance And Efficiencyby <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-7.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><table class="editorTblTablecenter editorTblSize100 editorTblStyleStyle2"><tbody><tr class="tblEven"><th colspan="3">AMD Phenom II X2 Test Settings<br /></th></tr><tr class="tblRow"><th><p><strong> </strong></p></th><td><p><strong>Default Settings</strong></p></td><td><p><strong>Overclock Settings</strong></p></td></tr><tr class="tblEven"><th>CPU<br /></th><td rowspan="1"><p><strong>AMD Phenom II X2 550</strong> 3.1 GHz<br />1MB L2+6MB L3 Cache, 1.288V</p></td><td rowspan="1"><p>3.94 GHz (19.5x 202 MHz), 1.50V</p></td></tr><tr class="tblRow"><th>RAM<br /></th><td><p>DDR3-1333 CAS 9-9-9-24, 1.50V</p></td><td><p>DDR3-1616 CAS 6-6-5-18, 1.65V</p></td></tr><tr class="tblEven"><th>Motherboard<br /></th><td colspan="2"><p><strong>MSI 790FX-GD70</strong> Socket AM3, 790FX/SB750, BIOS 1.3 (04/27/2009)</p></td></tr><tr class="tblRow"><th>Graphics<br /></th><td colspan="2" rowspan="1"><p><strong>Zotac GeForce GTX260²</strong><br />576MHz GPU, 999 MHz Shader, 896MB GDDR3-2484 </p></td></tr><tr class="tblEven"><th>Hard Drive<br /></th><td colspan="2" rowspan="1"><p><strong>Western Digital VelociRaptor WD30000HLFS</strong><br />300 MB, 10,000 RPM, 16 MB Cache</p></td></tr><tr class="tblRow"><th>Sound<br /></th><td colspan="2"><p>Integrated HD Audio</p></td></tr><tr class="tblEven"><th>Networking<br /></th><td colspan="2"><p>Integrated Gigabit LAN</p></td></tr><tr class="tblRow"><td colspan="3"><p><strong>Software</strong></p></td></tr><tr class="tblEven"><th>Operating System<br /></th><td colspan="2"><p>Microsoft Windows Vista Ultimate x64 SP1</p></td></tr><tr class="tblRow"><th>Graphics<br /></th><td colspan="2"><p>GeForce 182.08 Desktop</p></td></tr></tbody></table> <p><br />A clock speed increase of 27% won’t surprise many experienced overclockers, but the Phenom II X2 550 started out at a fairly high 3.10 GHz. Its final clock rate of 3.94 GHz is fairly impressive for an AMD processor, even though the percent-gained is not. Does this increase translate directly into CPU performance?</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/A/216622/original/image003.png" alt="" title="" /></span></p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/B/216623/original/image004.png" alt="" title="" /></span></p> <p>CPU Arithmetic performance improved by 25%, while Multimedia extensions performance increased by 26%. The small difference between frequency improvement and performance improvement can likely be attributed to our use of a near-stock HT clock, as described on this guide’s previous page.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/C/216624/original/image005.png" alt="" title="" /></span></p> <p>Our efforts to reduce memory timings using the processor’s highest memory ratio resulted in a tiny 8% gain in memory performance.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/D/216625/original/image006.png" alt="" title="" /></span></p> <p>Average power consumption increased by 33%, mostly because of the increased CPU core voltage.</p> <p><span class="imgContent imgCenter"><img src="http://media.bestofmicro.com/5/E/216626/original/image007.png" alt="" title="" /></span></p> <p>An average CPU performance increase of 26% at an average power increase of 33% yields an average efficiency decrease of around 5%. Overclockers looking for improved efficiency can instead choose a lower core voltage, as overclocking at stock voltage, though limited in performance gain, usually increases efficiency.</p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-52542052158596296622009-08-12T19:02:00.000-07:002009-08-12T19:03:15.419-07:00Overclocking AMD's Phenom II X2 550by <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-6.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><p>We follow the overclocking scene fairly closely and have found several overclockers using AMD’s 45 nm processors at voltage levels between 1.50 and 1.56 volts since the Deneb quad-core was first introduced last winter. This level of voltage tolerance is far greater than that of competing Intel models, but to play it safe we chose a maximum setting of 1.50 volts (give or take a few millivolts) under full CPU load, mindful to keep peak unloaded voltage below 1.55 volts.</p> <p>AMD publishes overclocking software under its “AMD OverDrive Utility” name that allows many of the most important settings to be changed inside Windows. While these can prove useful for finding the processor’s operational limits, many users will eventually want to make these adjustments semi-permanent through BIOS settings.</p> <p>The traditional overclocking method is to increase clock speed and test for stability, in small steps, until it’s no longer stable. Then increase voltage slightly to make it stable, and repeat until either a thermal limit (too hot) or clock ceiling (where more voltage doesn’t help) is reached. But a little research on the Phenom II X2 550 showed that most samples continue to scale upwards at voltage levels beyond our desired limit. Because of this, we started with our target voltage and attempted to find the highest stable speed it would run at that voltage. The following BIOS images show the results of our efforts, so let’s discuss how we arrived at each setting.</p> <p>The stock X2 550 clock speed of 3.10 GHz is attained by multiplying the HT clock of 200 MHz by 15.5. MSI's BIOS lists the HT clock as "CPU FSB frequency", though a technical inaccuracy, as AMD insists HT is not an FSB. Since this is a Black Edition processor, most of our overclocking efforts will focus on raising its 15.5x multiplier.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc1,0101-216627-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Phenom2-x2-550,5-F-216627-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc1,0101-216627-0-2-3-1-png-.html">Zoom</a></span></span></p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc2,0101-216628-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Phenom2-x2-550,5-G-216628-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc2,0101-216628-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>In MSI's BIOS, “CPU VDD Voltage” refers to the base voltage at which the processor is supposed to be detected, while the “CPU Voltage” setting acts as a type of fine-tuning for load voltage. We started with “CPU VDD Voltage” set to 1.50 volts and “DRAM Voltage” set to the memory manufacturer’s recommended 1.65 volts. The CPU multiplier, listed in the BIOS as “Adjust CPU Ratio,” was then increased to 16x.</p> <p>We use <a target="_blank" href="http://www.mersenne.org/freesoft/">Prime95</a> for stability testing, and find it handy that <a target="_blank" href="ftp://mersenne.org/gimps/p64v258.zip">v25.8 build 4</a> (64-bit Window version) allows every core to be tested simultaneously from one application launch. A launch menu offers several types of tests. The “Small FFTs” option allows full CPU stress without much DRAM testing.</p> <p>After around 20 minutes of load, we rebooted and increased the CPU multiplier to 16.5x, retested with Prime95, and continued this pattern until the system crashed at 18.5x. Detection program CPU-Z reported that the core voltage was dropping to 1.48 volts, so we went back into the BIOS and increased the “CPU Voltage” setting by 0.20 V (to 1.520 volts) as compensation.</p> <p>Upon rebooting, the 18.5x setting was found to be Prime95-stable, so we continued making 0.5x increases until the system again crashed at the 21x BIOS setting.</p> <p>Since we had already reached our target voltage, we tried dropping “Adjust CPU Ratio” in BIOS to 20.5x and let the stability test run longer. After around 45 minutes, the system again crashed. The same was true at a BIOS setting of 20x.</p> <p>At a BIOS “Adjust CPU Ratio” setting of 19.5x, the system ran stable for several hours. Knowing that we could reach 19.5 x 200 but not 20 x 200, we began increasing the HyperTransport clock, which is the “200” part of 19.5 x 200. Using “Adjust CPU FSB Frequency (MHz)” in the MSI BIOS, we tried an HT clock of 202 MHz with great stability over a one hour test. We then tried 204 MHz and found the system crashed in around 45 minutes. At 203 MHz, the system crashed at around one hour of Prime95 test time, so we reverted back to 202 MHz and again found stability.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc5,0101-216621-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Phenom2-x2-550,5-9-216621-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc5,0101-216621-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>Though it also allows some overclocking adjustments, we used “<a target="_blank" href="http://game.amd.com/us-en/drivers_overdrive.aspx">AMD OverDrive Utility</a>” primarily as a temperature monitor throughout testing. Note that its voltage monitor corresponds to the MSI BIOS' “CPU VDD Voltage” setting, not its “CPU Voltage” setting.</p> <p><strong>Note to non Black Edition CPU owners: </strong>Overclocking an AMD processor that doesn't allow multiplier increases requires raising the HT clock by a far greater amount. The higher data rate will eventually overwhelm the processor's internal HT link, but using the "Adjust CPU-NB Ratio" setting in BIOS to reduce the data multiplier can help. We generally try to keep the HT link data rate (listed below the adjustable setting in MSI BIOS) within 5% of its original speed when overclocking a "locked" AMD processor.</p> <p>Though this isn’t a memory overclocking guide, we did want to optimize our modules for performance. Our Kingston RAM is rated at DDR3-2000, but the highest DRAM external clock rate available from AMD is four-times the CPU's HT clock. With a HT clock at 202 MHz, this corresponds to a DRAM external clock of 808 MHz and a DRAM data rate of 1616 MHz (see “FSB/DRAM Ratio” in the first BIOS screenshot above).</p> <p>While the “DRAM Voltage” in the second screenshot above was set to the manufacturer’s recommended 1.65 volts, added stability can often be found by increasing memory controller voltage (“CPU DDR-PHY Voltage” in the same screenshot). With our DRAM data rate limited to 1,616 MHz, we used this added stability to enable lower latency, or wait time between operations, with our modules.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc3,0101-216629-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Phenom2-x2-550,5-H-216629-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc3,0101-216629-0-2-3-1-png-.html">Zoom</a></span></span></p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc4,0101-216630-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/Phenom2-x2-550,5-I-216630-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/phenom2_x2-550_oc4,0101-216630-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>Starting at our memory’s DDR3-1866-rated CL-tRCD-tRP-tRAS timings of 8-8-8-20, we followed the same method as used for CPU overclocking to decrease each setting until the lowest stable timings were found. A bootable CD version of <a target="_blank" href="http://www.memtest.org/">Memtest86+ v1.70</a> was used following each setting change for stability testing.</p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-22661621996974707822009-08-12T19:00:00.000-07:002009-08-12T19:02:02.279-07:00More Shared Hardwareby <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-5.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><p>Though processor families must be used with specific types of motherboards, other parts, such as the power supply, RAM, and hard drive work across multiple platforms.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/coolermaster_rs-850-embajpg,0101-112279-0-2-3-0-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/Cooler-Master-RS-850,M-V-112279-13.jpg" alt="" title="" /></a><span style="width: 404px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/coolermaster_rs-850-embajpg,0101-112279-0-2-3-0-jpg-.html">Zoom</a></span></span></p> <p>Cooler Master’s RS850-EMBA power supply has far more capacity than needed for today’s guide, but was chosen because it was already on the bench. Its 80 PLUS rating should allow realistic comparisons of power draw between stock and overclocked speeds of each processor.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/jan09_kingston_KHX1600D3T1K3-6GX,0101-176466-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/High-End-RAM-Jan09,5-U-176466-13.jpg" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/jan09_kingston_KHX1600D3T1K3-6GX,0101-176466-0-2-3-1-jpg-.html">Zoom</a></span></span></p> <p>We didn’t need three modules to test this guide’s dual-channel systems, but two of the same parts can be used in dual-channel mode. Kingston’s DDR3-2000 wasn’t just handy; it’s also capable of low latencies at various speeds, available in single-module packages for dual-channel kits, energy efficient, and able to extract peak performance from each processor. Builders should look forward to a cost-conscious comparison of modern dual-channel kits later this month.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/wd_velociraptor_profile,0101-192957-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/Zotac-GTX-285-Infinity,V-X-192957-13.jpg" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/wd_velociraptor_profile,0101-192957-0-2-3-1-jpg-.html">Zoom</a></span></span></p> <p>Western Digital’s VelociRaptor was again chosen for convenience, since its higher-than-average data rate allows quicker load times, but with little to no affect on most benchmark scores. It certainly won’t affect the outcome of today’s overclocks.</p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-39596997128935540062009-08-12T18:58:00.000-07:002009-08-12T18:59:59.981-07:00Keeping It Coolby Thomas Soderstrom<br /><p>Cool processors clock higher and survive longer, but finding an inexpensive cooler in the preferred 120mm tower design able to support both AMD and Intel processors isn’t easy. Rosewill surprised us with a review sample that included an AMD-style clip, since its Fort 120 doesn’t advertise Socket AM2+/AM3 compatibility on the box. Readers should look forward to a review of this unit later this month.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/rosewill_fort-120_kit,0101-216661-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/Rosewill-Fort-120,6-D-216661-13.jpg" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/rosewill_fort-120_kit,0101-216661-0-2-3-1-jpg-.html">Zoom</a></span></span></p> <p>This is the point where some die-hard overclockers might point out that, since we used top-end motherboards, we should also use a top-end liquid cooling system. But while budget overclockers might be able to find less expensive motherboards that replicates our results, the same cannot be said of liquid cooling. We wanted to provide a realistic, yet optimistic target for value-overclockers to use as a goal.</p> <p>One other place we didn’t go cheap was in thermal compound selection. The Fort 120 cooler does not include enough thermal paste for multiple uses, so we instead relied on our established thermal grease choice.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/zalman_zm-stg1,0101-180946-0-2-3-0-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/X58-Comparison-Images,M-A-180946-13.jpg" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/zalman_zm-stg1,0101-180946-0-2-3-0-jpg-.html">Zoom</a></span></span></p> <p>Zalman’s ZM-STG1 was chosen for previous reviews based on its easy application, quick set in time, and upper-range thermal performance. Upon request, the firm supplied enough samples for each U.S. editor to have two bottles.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/paste-applied,0101-216619-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/AM3-Black-Edition,5-7-216619-13.jpg" alt="" title="" /></a><span style="width: 432px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/paste-applied,0101-216619-0-2-3-1-jpg-.html">Zoom</a></span></span></p> <p>Thermal grease or paste fills small gaps between the processor and heat sink to provide a greater contact area. Many experienced builders swear that too heavy a layer will prevent proper sink contact, citing the lower conductivity of thermal compound compared to the aluminum or copper surface it fills, but most modern thermal materials are thin enough that heat sink pressure will squeeze out any excess. The real problem of applying too much paste is that it can make a mess of the motherboard, and its low-conductivity is still enough to potentially cause signal or voltage problems.</p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-25561575191877462192009-08-12T18:57:00.000-07:002009-08-12T18:58:41.694-07:00Getting Started, The Hardwareby <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-3.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /><p>Continually-falling DDR3 prices are allowing the memory technology to displace DDR2 in mainstream-performance builds. And with the future of DDR2 desktop memory drawing short, we selected two DDR3 motherboards from MSI to support our chosen AMD and Intel processors.</p> <p><span style="clear: both;"><table class="imgTab" align="center"><tbody><tr><td class="imgContent"><a href="http://www.tomshardware.com/gallery/msi_790fx-gd70_profile,0101-206291-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/790FX-AM3-IMAGES,6-B-206291-1.jpg" alt="" title="" /></a><span style="width: 200px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/msi_790fx-gd70_profile,0101-206291-0-2-3-1-jpg-.html">Zoom</a></span></td><td class="imgContent"><a href="http://www.tomshardware.com/gallery/msi_790fx-gd70_top,0101-206292-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/790FX-AM3-IMAGES,6-C-206292-1.jpg" alt="" title="" /></a><span style="width: 181px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/msi_790fx-gd70_top,0101-206292-0-2-3-1-jpg-.html">Zoom</a></span></td></tr></tbody></table></span></p> <p>Picked for its <a href="http://www.tomshardware.com/reviews/790fx-socket-am3,2277-11.html">best-in-class HT clock speed capability</a>, MSI’s 790FX-GD70 should provide optimal results for our Socket AM3 overclocking tests. Choosing the standout motherboard from previous reviews allows us to set a high goal for owners of less-expensive motherboards to attempt using the same processor models.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/955-550,0101-216614-0-2-3-1-png-.html" class="iZoom"><img src="http://media.bestofmicro.com/AM3-Black-Edition,5-2-216614-13.png" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/955-550,0101-216614-0-2-3-1-png-.html">Zoom</a></span></span></p> <p>Our budget limit for dual-core and quad-core processors was $125 and $250, respectively. AMD sent its Phenom II X4 955 Black Edition (<a target="_blank" href="http://stores.tomshardware.com/search_getprod.php/masterid=721279602/search=phenom%20ii%20x4%20955/st=query">$245 retail value</a>) and Phenom II X2 550 Black Edition (<a target="_blank" href="http://stores.tomshardware.com/search_getprod.php/masterid=730710754/search=phenom%20ii%20x2%20550/st=query">$100 retail value</a>) for today’s overclocking guide. Black Edition processors are special from other Athlons and Phenoms in their ability to manipulate the clock multiplier upward, allowing high overclocks to be achieved at or near the processor’s original 200 MHz HyperTransport reference clock.</p> <p>We requested MSI’s top P45-chipset motherboard to maintain fairness between processor brands, and the firm responded with its P45 Diamond.</p> <p><span style="clear: both;"><table class="imgTab" align="center"><tbody><tr><td class="imgContent"><a href="http://www.tomshardware.com/gallery/msi_p45-diamond_profile,0101-216617-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/AM3-Black-Edition,5-5-216617-1.jpg" alt="" title="" /></a><span style="width: 200px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/msi_p45-diamond_profile,0101-216617-0-2-3-1-jpg-.html">Zoom</a></span></td><td class="imgContent"><a href="http://www.tomshardware.com/gallery/msi_p45-diamond_top,0101-216618-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/AM3-Black-Edition,5-6-216618-1.jpg" alt="" title="" /></a><span style="width: 182px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/msi_p45-diamond_top,0101-216618-0-2-3-1-jpg-.html">Zoom</a></span></td></tr></tbody></table></span></p> <p>A higher average price gets buyers fewer graphics card slots. MSI makes up for the value loss with added features, such as a PCIe audio card and a chipset water block with copper line adapter kit.</p> <p>Nobody said we had to spend our entire budget on processors. Focusing on the value segment brought us to the <a target="_blank" href="http://stores.tomshardware.com/search_getprod.php/masterid=88947467/search=pentium%20E5200/st=product/sv=title">$70</a> dual-core Pentium E5200 for its high CPU to front side bus multiplier and good overclocking reputation, and the <a target="_blank" href="http://stores.tomshardware.com/search_getprod.php/masterid=89121766/search=Q8200/st=product/sv=title">$160</a> Core 2 Quad Q8200 for its reasonable cost.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/e5200-q8200,0101-216616-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/AM3-Black-Edition,5-4-216616-13.jpg" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/e5200-q8200,0101-216616-0-2-3-1-jpg-.html">Zoom</a></span></span></p> <p>Intel doesn’t produce a 45 nm desktop quad-core with anything less than FSB-1333, and each model up gets us a slightly higher (0.5x) CPU multiplier at a noticeably higher price. Like AMD’s Black Edition, Intel also offers Extreme Edition processors with CPU multipliers that can be manipulated upwards, but Intel charges so much more for this feature that we couldn’t possibly consider any of these for use in a value-oriented overclocking guide.</p>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-72952546833313874512009-08-12T18:56:00.000-07:002009-08-12T18:57:21.359-07:00Understanding The Lingo<div style="text-align: justify;">by <a href="http://www.tomshardware.com/reviews/overclock-phenom-pentium,2366-2.html#" onclick="$('form_cont').submit()"> </a>Thomas Soderstrom<br /></div><p style="text-align: justify;">Many new overclockers leave our user forums never to return when they ask "How do I overclock?" and receive “Raise the FSB or HT clock” as an answer. But once you're armed with the lingo, the principles are fairly easy to understand. Let's cover a couple of the basics. </p><div style="text-align: justify;"> </div><p style="text-align: justify;"><strong>Frequency</strong></p><div style="text-align: justify;"> </div><p style="text-align: justify;">A processor is made up of a complicated series of microscopic electronic switches (transistors) on a pulsating power circuit. The number of pulses (power cycles) per second is called the circuit's “frequency.” It takes at least one cycle for the transistor to change state between on (1) or off (0), and the ones and zeros become part of a data stream.</p><div style="text-align: justify;"> </div><p style="text-align: justify;">Modern central processors run at thousands of millions (billions) of cycles per second, or gigahertz. This is the same range of frequencies at which microwaves and mobile phones operate, so that a relatively short piece of wire can become a fairly good radio antenna. Preventing cross-communication between circuits, where one circuit acts as a transmitter an the other an unintended receiver, is extremely important.</p><div style="text-align: justify;"> </div><p style="text-align: justify;">The conductors on motherboards, called traces, are much longer than those of an integrated circuit, such as a central processor (CPU) or graphics processor (GPU). In order to reduce noise, signal loss and cross-talk, the pathways that connect various processors must run at slower frequencies.</p><div style="text-align: justify;"> </div><p style="text-align: justify;"><strong>The CPU Multiplier</strong></p><div style="text-align: justify;"> </div><p style="text-align: justify;">As the need for increased data speed outstripped the ability of various busses to support it, companies developed a variety of methods to send more than one bit of data per conductor, per cycle. These methods include double data rate used in memory modules, quad data rate used by Intel’s front side bus (FSB), AMD’s HyperTransport (HT) interconnect, and Intel’s recent QuickPath Interconnect (QPI).</p><div style="text-align: justify;"> </div><p style="text-align: justify;">Because Intel’s most recent FSB uses quad data rate technology, its clock frequency is a quarter of its data frequency. That is to say, the clock rate of FSB-1333 is 333 MHz (megahertz, or millions of cycles per second). The CPU itself relies on an actual electrical frequency (the clock rate) to set its internal speed, so a CPU multiplier of 10x on an FSB clock rate of 333 MHz (FSB-1333) results in a CPU frequency of 3,333 MHz, or 3.33 GHz.</p><div style="text-align: justify;"> </div><p style="text-align: justify;">AMD’s internal HT link uses a 200 MHz clock speed with data rates of five to ten times clock speed, resulting in 1,000 to 2,000 transfers per second. But since HyperTransport supports full bandwidth in both directions at the same time, AMD doubles its name to HT 2,000 (1,000 MHz data rate, 200 MHz clock rate) and HT 4,000 (2,000 MHz data rate, 200 MHz clock rate). The most important thing to remember when overclocking is that both HT 4,000 and HT 2,000 use a clock rate of 200 MHz, so that a CPU multiplier of 10x would provide a CPU clock speed of 2,000 MHz, or 2.0 GHz.</p><div style="text-align: justify;"> </div><p style="text-align: justify;">Though we won’t use an Intel QPI-based system today, users should know that it operates in a similar fashion to AMD’s HT link, but at a slower 133 MHz base clock frequency.</p><div style="text-align: justify;"> </div><p style="text-align: justify;"><strong>Voltage</strong></p><div style="text-align: justify;"> </div><p style="text-align: justify;">Frequent overclockers will discuss BIOS settings such as VCore (voltage of the CPU core), VDIMM (memory voltage), and various data pathway/memory controller voltage settings under a variety of different initializations. Some of these will be discussed in detail as we encounter them in BIOS screen shots.</p><div style="text-align: justify;"><br /></div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-32726259889666287542009-08-12T18:53:00.000-07:002009-08-12T18:56:14.497-07:00Guide: Overclocking AMD And Intel CPUs On A Budgetby Thomas Soderstrom<br /><div class="KonaBody news-elm"> <p>Overclocking is a collection of methods for making components run faster than the manufacturer intended. Once little more than a hobby for die-hard geeks and value-seekers, overclocking has become a way—sometimes the only way—for performance fanatics to get the system performance they really want/need. With graphics and memory technologies forging ahead at a brisk pace, central processors are quickly becoming the second-most restrictive component in many high-end systems.</p> <p>If you feel forced into overclocking just to get a high-performance benchmark from the best parts, mid-budget enthusiasts are certain to find their lower-cost parts mind-numbingly slow. Because most buyers can’t afford the best components, the majority of overclockers come from the mainstream market.</p> <p>There are two groups who overclock out of perceived necessity: those who need more performance than the market provides, and those who need more performance than they can afford to purchase.</p> <p>Tom’s Hardware puts much of its editorial efforts into testing and overclocking the latest high-end parts, but today we’re going to focus on a few processors that most mainstream readers can afford and enjoy: AMD’s Phenom II X2 and X4, and Intel’s Pentium Dual-Core and an entry-level Core 2 Quad.</p> <p><span class="imgContent imgCenter"><a href="http://www.tomshardware.com/gallery/amd-intel-oc,0101-216615-0-2-3-1-jpg-.html" class="iZoom"><img src="http://media.bestofmicro.com/AM3-Black-Edition,5-3-216615-13.jpg" alt="" title="" /></a><span style="width: 450px;" class="spipLegend"><img style="margin-right: 3px;" alt="" src="http://m.bestofmedia.com/i/presencepc/design/loupe.gif" /><a href="http://www.tomshardware.com/gallery/amd-intel-oc,0101-216615-0-2-3-1-jpg-.html">Zoom</a></span></span></p> <p><strong>Mitigating Risks</strong></p> <p>Though we’re obligated to tell everyone that overclocking is a great way to put important data at risk, many Tom’s Hardware editors even employ it on their all-important work PCs. Methods that ensure stability are just as important as those that assure longevity, and any data that can't be replaced should be backed up to at least two devices, regardless of whether or not the primary system is overclocked.</p> <p>All machines wear out, and forcing a component to run beyond its specifications is a sure way to make it wear out faster. In electronics, the biggest source of wear is a phenomenon known as electromigration, whereby ions are slowly transferred from a structure to the adjacent structure under the force of electrical current. Major contributing factors include increased heat and voltage, but the limits of heat and voltage vary with different materials, different production technologies, and expected component lifespan.</p> <p>Increased voltage allows a stronger signal to be carried between various components, reducing signal loss that can occur as the result of overclocking and thereby allowing higher component operational frequency. As we overclock today’s four processor samples, we’ll discuss the voltage and temperature limits we’ve chosen as well as the expected lifespan, testing each part for complete stability.</p><p>www.tomshardware.com<br /></p> </div>Unknownnoreply@blogger.com0tag:blogger.com,1999:blog-2550952280158595542.post-88588772835219471682009-08-12T18:51:00.001-07:002009-08-12T18:51:45.695-07:00Windows XP Trick - Increase Performance of PC, XP Tricks<object width="425" height="344"><param name="movie" value="http://www.youtube.com/v/-9Nycdiq3x8&hl=en&fs=1&"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/v/-9Nycdiq3x8&hl=en&fs=1&" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="425" height="344"></embed></object>Unknownnoreply@blogger.com0