AOA Forums AOA Forums AOA Forums Folding For Team 45 AOA Files Home Front Page Become an AOA Subscriber! UserCP Calendar Memberlist FAQ Search Forum Home


Go Back   AOA Forums > Hardware > AMD Motherboards & CPUs

AMD Motherboards & CPUs Questions or comments on AMD products?


Reply
 
LinkBack Thread Tools Rating: Thread Rating: 8 votes, 4.50 average.
  #1 (permalink)  
Old 11th February, 2006, 06:35 PM
Super Nade's Avatar
Member
 
Join Date: December 2005
Location: Indianapolis, USA
Posts: 157
Send a message via Yahoo to Super Nade

ASUS A8V Overclocking Guide

:: Latest Issues and News ::
{Updated April 11, 2005, latest issue first}

Slipstream SATA Drivers! You don't need a Floppy Drive!
Many thanks to crosis.23 for writing this up.
Quote:
Originally Posted by crosis.23
You should go HERE first and follow the steps to create a Windows CD that is streamlined with SP2.The latest Windows Deployment Tools should be obtained from microsoft.com. Using setupmgr.exe, follow the steps again outlined in msfn.org to create your winnt.sif file.

Then usng the "drivers via winnt.sif" method you want to edit txtsetup.sif contained within your i386 folder (of your XP cd backup).This is the bit I used to incorporate the Promise 378 controller.You must add these 4 lines to the txtsetup.sif file in the listed sections:

[SourceDisksFiles]
ulsata.sys= 1,,,,,,4_,4,1,,,1,4

[HardwareIdsDatabase]
PCI\VEN_105A&DEV_3373 = "ulsata"

[SCSI.Load]
ulsata = ulsata.sys,4

[SCSI]
ulsata = "WinXP Promise SATA378 (tm) Controller"

In the $OEM$\$1\DP\M\P folder you only need the files from the XP folder of the driver pack.You must create the cab with the name ulsata.sy_

DIMM Configurations and operation:
Apart from the "recommended" configs in the Manual, the following config's work:-
  • Only both black slots populated, enables you to run Dual channel. I have a booster in the A1 (first blue slot) and nothing in the B1 (2nd Blue slot).
  • Both B1 (2nd Blue) and A2 (2nd Black) slots when populated enable you to run in single channel mode.
  • It is also possible to run in single channel with any three slots populated providing you use a fake RAM, its just a PCB with no chips. You can pull them of an old Dell.


OCZ DDR Booster Hits 3.1V stable with this board (scroll down for a comment from the OCZ Rep posted earlier):

Guys, you won't believe it, but contrary to reports, the Booster hits 3.0 (already known) V and even 3.1 V with this board! Guess what? It is perfectly stable. I'm at 3.1V now . The OVP feature is actually a voltage regulator issue. I'm guessing that a certain VR pulls the voltage down to 3.0V or 3.1V (the exact number is unpredictable). I turned the pot almost fully across but there was no change. I'm pretty confident about the VR thing because I clearly see the booster jump to 3.5 and even 3.7V, but in about 1 second, the VR takes over and drags the voltage down to 3.0 or 3.1 V. There is a sweet spot at which the voltage remains at 3.1V (a slight turn by say 60 degrees does it).Anything more or less keeps it at 3.0V. Now it remains to be seen if I can isolate the offending VR and possibly bypass it with a switch. I'll look into existing Volt Mods and see if it will work. Its going to take a while as this requires a lot of spare time.

ATM I've ditched the TCCD for some Muskin Level 2 Black. A quick run of Memtest #5 did well so I'm quite optimistic. Before this, I used to get gazillion errors on test #5 (at 99% completion). More testing and optimization needs to be done.


GSkill Rep says TCCD doesn't OC well with this MoBo!!! :
I recently upgraded to PDP/XBL TCCD modules. I RMA'd one set because I had errors at stock and I'm testing the current set. After hearing many good things about GSKill, I decided to go to the horse's mouth for an answer (GSKill Rep). To my utter disappointment he clearly said that TCCD doesn't like this MoBo (and possibly VIA Chipsets). Here is the unedited version of our conversation. SOURCE

Me:
Hello,
This is my system:
Mandrake 10.1/XP Pro.
ASUS A8V (Rev 2.00) BIOS 1009.006, A64 3500 NC @ 2.58,

Fortron 530W, 80 Gb Hitachi SATA, SB Live 5.1, 60Gb Maxtor IDE,
Radeon 9600,Vantec Tornado 92mm, Thermalright SLK948U.

Will the GSKill PC4400 work for me? I would like to run my RAM at 279MHz, same as FSB(or HTT).I mean if CPU = 279 X 9.5 MHz then Memory = 279MHz.

I heard that you guys guarentee 275MHz on the PC4400 (Samsung TCCD). Is this true?

Jack (GSkill Rep)
If Your motherboard is asus a8v,I suggset that you do not buy the TCCD memory.If you can not reach DDR550 or above,You will be unhappy.
You can look the what chip it is.

-TCCD is easy DDR550 2.5-4-4-8 or ddr600 above.
-Hynix-d5 is ddr500 ~580 in the 3-4-4-8. hynix-(bt-d43 or dt-d43)it is DDR400 CHIP,But it has more probability to DDR500 3-4-4-8.
TCC5 has less probability DDR400 2-2-2-5 , more probability ddr500 above.
TCCC has more probability ddr500 .(3-4-4-8)vdimm=2.85v.256mbx2 has good performcnce than 512mbx2.
Micron-5b-mt...(very Special chip .it can ddr500 2.5-2-2-5).
winbond BH-5 easy ddr400 2-2-2-5.ddr500~550 2-2-2-5.But it need high voltage.
winbond ch-5 or ch-6(DDR400 2-3-2-5 OR ..)ddr500~550 2-3-2-5.But it need high voltage. This is more special IC in the market.
The hUNIX-"BT"-D43 is cheaper memory than others. It can easy reach ddr500 or above.
You can search www.google.com
key in "Hynix d43 ddr500" ,I think that it will appear many testing reports.
If you still have any problem,please contact with me as fast as possible.
or e-mail to me or msn me.
best regards
jack

Me:
Jack,
Thanks for the reply.
So, there is no chance of reaching DDR500+ with TCCD on this motherboard.
That is very sad because I wanted to get the PC4400 LE's sad.gif
What other GSkill RAM can you suggest? I want to run my RAM at 250MHz FSB.

Another question:
What is the recommended voltage for TCCD chips?
I would like to run 3-4-4-8-1T DDR550.

Jack:
Dear sUPER Nade
If you still use the asus a8v,you just about ~ddr500.
This speed is finished by HYNIX-D5 or BT-D43 or TCCC(2.85v).
These ic can do DDR500 3-4-4-8.
If you want to buy the 4400LE,I have a opinion that strong recommand DFI_NF4 Series or MSI_sERIES(NF4).
It is easy DDR600 1T~DDR700~DDR740(XS DFI_Release ddr740)
msi_nf4 and DFI_NF4 is easy DDR600 1T Command.
and DFI is easy DDR700~740.
If you use the "LE" spec,you will realize the powerful overclocking ability.
NEO2~US100~120
Gigabyte K8NSNXP~US 100~120.
DFI OR MSI(NF4 Lower price is~137~150)
http://www.gskill.com/forum/index.php?showtopic=239
This is my privated-opinion.
Help your friend to setup computer.
And you have DFI_NFI(Ultra-d).
DFI_NFI(Ultra-d).+ ddr4400LE,I think that it is easy ddr600 1t ~ddr700~above.

To help others
TCCD VOLTAGE~2.65~2.85V
0431 0425 has more porbability DDR500 CAS 2-3-3-6 (HIGH VOLTAGE) in th xs.
0434 0437 0440 ,I have no looked the testing report .
This is reference information

From preliminary testing by rdrash, a colleague at ocforums, the TCCD hating seems to be true. You can find his results HERE


Faulty BIOS Chips!! :
I am seeing a lot of complaints about faulty BIOS chips. All of you struggling with BIOS issues may possess a MoBo with a faulty BIOS. This problem seems to have afflicted users who purchased thair Motherboards in the last week of Dec and stll persists.
My sources are (many more on the AMD Forums):-
Source 1
Source 2
Source 3
Source 4


No-boot/ severe startup problems:
Recently, quite a few new members have reported No-POST problems. The symptoms ranging from loss of power immediately after turning on and no video. After the usual round of basic troubleshooting measures viz., taking out the Motherboard to eliminate shorting with the case, reseating components etc...., these problems persisted. In one case reseating the CPU worked and in another case, changinng the Keyboard jumper settings worked momentarily. Both members have since RMA'd their boards. The relevent threads can be found HERE, and HERE.


Possible VIA AGP Driver Problems: (Thanks to Beanster)
Quote:
I was experiencing some crashes to desktop and an occasional total crash while gaming (no freezing though) and after I eliminated hardware failure by testing with prime and memtest I turned my focus to drivers/software. I found THIS thread in the Via arena forums. It is about the via agp driver being the cause of a lot of our symptoms. I removed the via AGP driver as per the instructions and installed the one recommended from the list, all was well, no more crashing. today just because I am a glutton for punishment I formated and reinstalled and did not use the agp drivers that come with the latest via's, (4.55 at this time) and left it ride until the windows sp2 installed theirs. same result, no more crashing and problem free all day while gaming.I followed the guideline for windows/driver installation HERE

SB Audigy screeching lock-up problem:
BioVader, Gashman and the rest of the crew have bought this problem to my attention.
SB Audigy users have reported random lockups preceeded by a high pitched wail. Sounds spooky, huh? Apparently, this is a well known issue and Creative have done nothing to correct it! Not A8V specific but this would definitely cause system insability. This is an EAX issue when playing certain games. Also Dirctx 9.0c and the Creative driver incompatibilities have been cited (unconfirmed yet). Please look HERE, HERE and HERE for more info.

Massive BIOS problems:
ASUS are having a hard time releasing a stable BIOS. All ! current Winchester supporting BIOSes have problems.The common problems are HTT instability and freezing. The members here have reported extreme instability (no-boot, random lockups) despite maintaining good tempratures and fairly constant VCore. A few threads detailing these problems can be found HERE, HERE and HERE.

Caution advised if Volt Modding
According to SteveOCZ, a member here ( and an OCZ rep according to many members), the Mobo blew up (as they ramped up VDimm beyond 3.0V) when they tested their product (OCZ DDR Booster). Steve says its a problem with the Mobo. I don't know what version they used or what the test conditions were. As you read later in this thread, some guys at on the xtremesystems forums have reported success with the booster. So this issue is not clear and this note is intended to be a caution sign only.
Here's what SteveOCZ had to say:
Quote:
Originally Posted by SteveOCZ
Overvoltage Protection on the A8V kicks in at 3.01V and it will shut the Mobo down.The ddr booster will work if you keep it under
3.01V.

Prime95 round off errors and stability problems with the 1008 BIOS: (Thanks to Impulses)
There have been reports of system instabilities caused when using the 1008 BIOS. More specifically, people have reported Prime95 round off errors. Although there is NO conclusive consensus among the users, (I browsed through various forums) there may be a glitch with the "for Winchester" BIOS's. Why would ASUS come up with 2 releases (3 betas) within 3 weeks? Please remember that there is insufficient data to point a finger at the 1008 BIOS yet. I'm switching to the 1009 BIOS and retesting @ 258 x 10, 1.65VCore to see if there's a difference.Discussions can be found HERE, and HERE.


Only for Winchester users (with older A8V's):
(If you have bought an A8V this month i.e Nov 2004 you probably have a newer BIOS chip and this may not apply to you.)
All of you who plan on getting the newer 90nm Winchesters i.e 3400, 3200, 3000,4000 and FX-55 please make sure your BIOS is flashed to version 1007.xx or later. Your processor will not be recognized by an older BIOS version. You must contact your seller and ask them to flash it for you before putting your rig together. You cannot do anything to make it work if you get an older BIOS and a Winchester unless you have access to a Newcastle (3500,3800,FX 53), whereby you boot with the newcastle, download the new BIOS and update.ASUS is charging people $25 to flash it for you or ship a new BIOS.Also, if you have an older BIOS and have successfully flashed to a new BIOS, make a copy on a Floppy or CD of the latest BIOS. If you reset the CMOS by removing the battery, you will fall back to the older 1006.02 version.
Q. Oh no! I dont have a spare CPU and I don't want to pay an extra $25 am I stuck now?
A. Fortunately there is a way around this:
Quote:
Originally Posted by Godsmack74@ Anandtech
You can use a 90nm socket 939 chip to get it booted and flashed. You just have to do two things.

1. Reset the bios cmos. (Reset the CMOS by switching the jumper only when your rig is powered off or else your BIOS will become dysfunctional). This is to
ensure that the board boots from a floppy first.

2. Most importantly make sure you have only one stick of ram in mobo and make sure that stick is in B1 slot. Third from the left I believe. Check the
motherboard manual for a graphic layout of the motherboard and it will show which slot is B1.

-Start the computer, boot from a bootable floppy, then use another floppy with the new bios and the AFUdos.exe program on it, not ezflash.

-Then type at the A prompt: afudos.exe /ia8v1007.ami

-It will erase, then write your bios.

-Restart and go to bios and configure as you like. You may now put in additional sticks of ram to get dual channel.

Works here just fine. This is the procedure for the Asus A8V I do not know if this will work in other motherboards, but I have heard similar stories of using
one stick to get it up. My guess would be the only thing that needs to change for a different board would be the flash program and bios ofcourse.

Complaints about HSF obstructions:
People have complained that the Thermalright XP-120 touches the memory module on slot A1 and this is said to be clearly apparent if the RAM sticks have heatspreaders on them. Read THIS Anandtech thread for more info.
For proof in pictures, look at Boomslang's pictures. You can clearly see what a tight fit it is. Clearly, installation of RAM with heatspreaders can be a bit tricky.

--------------------------------------------------------------------------


:: General Questions and Known Issues ::

0. The Overclocking Failed! message and other startup troubles:
It is irritating to hear a beep with the above message when you start up. This is a common error message and after a while it sounds familiar . I'm not sure why this happens but what I've noticed (can be taken as a rule) is that if you hear/see the message "OC Failed press F1..." etc twice repeatedly, then you need to lower your OC.Read the manual first and make sure that the the CLR RTC jumper isn't in the CLEAR position . Look at the Beep/Error code section for more details. Make sure that your processor is seated correctly in the socket! Sometimes reseating the CPU solves the no boot problem! SEE HERE and HERE'S a older article discussing these issues.


1. My Noob Q's during my first ever build last month:
SEE HERE


2 .New Bios update: {Updated Feb 6, 2005}

3. Massive sticky on A64 in general:
SEE HERE


4. Cool & Quiet Technology explained: (Thanks to J3lly)
Quote:
Originally Posted by J3lly
'Cool and quiet" works by down clocking/voltage thorogh the "unlocked downward dynamic multipliers" according to workload and how fast the cpu needs to run,try it and enable it,run tests and benchies,athlon64 was design for this and it knows what it supposed to do it'll add overall OC stability,since the cpu run at full speed only when it needs to, thus allowing cpu to cool down and consumes less power during idle or while doing LOW intensive stuff like web browsing,email,etc desktop stuff. You would need to install the C&Q drivers from the CD.
Additional Info from The Coolest
Quote:
PowerNow!/Cool'n'Quiet (both are essentially the same thing) is a feature in AMD CPUs, starting with AthlonXPs that let the processor change the CPU multiplier internally, using software. This is what mobile AthlonXP CPUs use when they are actually used in a laptop computer, when the CPU is under no load, the CPU slows down to conserve power, by lowering the multiplier, and then speeding back to full speed when its under heavy load, by upping the multiplier. PowerNow! needs two things to function: 1) A startup multiplier, which is basically the multiplier at which the computer first boots when you turn it on. 2) A maximum multiplier, which will set the max speed that the CPU will reach in powernow mode. (Usually the mobile CPUs have a startup multiplier of x5 or x6, and max multiplier will depend on the CPU speed rating).To enable and use PowerNow! on AthlonXPs you had to have the 3rd L5 bridge connected that would enable the feature, and then a motherboard (more like chipset) that supports this feature. On the AthlonXP the startup multiplier and max multiplier was set by bridges, the startup was set by L3 bridges (like the default multiplier on a desktop CPU), and the max was set by L6 bridges. There was also a mod, that removed the limitation set by L6 bridges, by connecting the 2nd L5 bridge, you would get access to x24 multiplier using PowerNow!
Petr anwers a few Basic Questions of mine:
Q1. I have a question about Cool n Quiet. You said that Cool n quiet is essentially the same as Power Now!. When running at stock my 3500 is at 200 x 11. Since it is top locked, that is the maximum multiplier, right? When OC'd, I'm @ 258 x 10. I enable Cool n Quiet when OC'd and my system refuses to boot.Why should this happen if CnQ is actually lowering my OC when idle? From what you have said, I gather that CnQ will adjust the freq with an upper bound of 258 x 10. I don't see why I can't boot ? Is there more to CnQ which fundamentally limits an OC?
A1. If you can't load OS, then answer is simple - CnC uses multiplier / voltage combinations that are loaded to RAM by BIOS. But in this case it loads 11x as a max. (table is predefined for a given CPU). So when you are trying to load OS, driver detects you run at 10x and not maximum, switches to 11x and it crashes.


Recommeded setting ENABLED if not OC'ing DISABLED if OC'ing. My rig won't boot at 270 x 9.5 with C&Q enabled. Tried lowering the HTT(upping multi to keep my OC) but won't boot regardless.


5. Random Shutdowns:
I have had absolutely no problems but some people seem to have problems.
Solutions ranged from reinstalling NVidia drivers, running 2T and relaxing memory timings, removing wireless PCI cards.

6. A8V memory:
Good OC 'able Memory known to work from user data:
OCZ PC3700EB, Corsair TwinX1024-3200XL and anything with Samsung TCCD chips on them. Do not get the Corsair VS PC3200 Kit (me) if you seriously want to OC. Its a great budget RAM but not very good if you are looking to run it in sync with the FSB/HTT. Will probably max out at 210 MHz when in sync. Look at the Memory compatability sheet for a list of working modules.
OCZ DDR Booster is not compatible with the A8V according to the official OCZ compatibility sheet but people have modded their boards to get it to work (mixed results)



7. Hypertransport and performance:
SEE HERE


8. AGP/PCI locks: {Added Sept. 30, 2004}
Last night I was messing around with the AGP/PCI lock switching between 66.6:33.3 and the other ratio. I found that that 66.6:33.3 is more stable with tight 2.5-3-3-7 timings. The 7x.xx ratio did not work for me even with loose 2.5-4-4-12 timings.Also, my RAM is CAS 2.5 so I don't know if the higher lock would be better with CAS 2.0 memory.If the 7x.xx ratio is not working for you, scale down the hyper-transport frequency as you are changing the LTD (scroll down to read the important note). If you have other PCI cards, modem, sound etc., and they are misbehaving, this could be the problem. If you are continually experiencing problems with the PCI/AGP lock try bumping up the PCI/AGP Voltage to 2.6 Volts.Also, this may not be a problem with the locks per se, it could be that the PCI/AGP device(s) cannot operate at an enhanced frequency. The only way to find out is to sequentially test each PCI card. In my case, the $5 PCI modem and soundcard are holding me back.
In case you have a Rev 1.02 board use the 1005.002 BIOS to obtain a working lock.

9. More HTT/LTD Questions:

10. Excellent AMD 101 by Gautam:
SEE HERE


11.Must Read part of hitechjb1's super awesome sticky!!!
HERE (Post #8)


12. Very Important:
I was initially extremely confused about these Q's. d]g[ts nailed it down once and for all. Thanks dude!
Q1. Where do you see or in what form do you see the LTD multiplier in the BIOS?
A1. On those boards (AV8 A8V) LDT multi appears as HTT frequency. 1000=5x, 800=4x, 600=3x and so on.

Q2. Difference between the CPU Multiplier and LTD Multiplier?
A2. CPU multiplier is multi to attain CPU speed I.E. CPUmulti X HTT= CPU speed in MHZ. LDT multi is used to attain speed of hypertransport bus. (speed between CPU and Chipset)

Q3. Do you really need lower values of HTT frequency?
A3. Yes , because there is a limit on how fast the hypertransport bus can run. and as HTT is raised, you start to surpass this limit, which causes instability from overloading the bus. to avoid this you drop the LDT multiplier.(Strangely its the reverse in my board. Lowering the HTT causes instability! Unresolved issue.) Another issue which has come to my attention is about the MoBo automatically lowering the HTF to 8x! Is it a good thing? Maybe!

Q4. Difference between HTT Frequency {1000MHz on the A8V} and FSB (max of 300MHz on the A8V)? I thought AMD's didn't have an FSB.
A4. HTT frequency on your board is LDT multiplier. thats just the way your bios is set up. And FSB on your board is the same thing as HTTspeed only worded differently. truthfully there isnt a FSB, because A64 uses the Hypertransport bus to communicate with chipset. It's termed that way to make things simpler for people to understand how to overclock it. (I beg to differ with ASUS on that one )

13. A note about VCore: {Added Oct. 14, 2004}
I saw people get amazing OC's with Bartons and XPM (HERE, and HERE) and was wondering what was the safe VCore. The general consensus was one should not exceed 1.675V on an A64.


14.Clockgen and A8V:
Do not use clockgen. I don't think it was designed for this chipset/Mobo. My box crashed several times at ~2.4.
*Edit* There is a new VIA K8T800 Pro version of Clockgen which works very well with this board. You can download it HERE


15. S939 OC Results FROM A GOOD RESOURCE :
HERE (Post #3)


16.VIA SATA/RAID Controller problems:
A lot of people have been having problems OC'n with their HDD on the VIA controller. Some people report extreme fragmentation and an eventual no-boot. The PROMISE controller is better and more stable. Certainly no problem OC'ing. See for example THIS for problems our members faced with the VIA controller.
Using the PROMISE SATA controller or switching from VIA to PROMISE:
Important Note!
I have had questions about swapping controlers when the OS is on an IDE HDD and I answer them here. Suppose your boot drive is the IDE. Nothing needs to be done. No swapping is possible or necessary between the Red Port and the Blue/Black ones. The Red port is your RAID port and you will keep getting errors if you swap from the Black--> Red ports. Errors will go away only if you set up RAID.

To double check, goto the BIOS and do this:
1. BOOT > HDD (or something similar) and move your IDE to the top of the list.
2. BOOT > BOOT DEVICE PRIORITY and your selected IDE should show up as a boot device. Bump it up to the top of the list.

You are done!

When your OS resides on the SATA being switched:
A lot of people have questions on how to activate the PROMISE SATA controller or switch over from the VIA SATA controller without reformatting their HDD. This is how it's done: (You need to have your XP install CD)
  1. Copy the PROMISE SATA drivers to a floppy.
  2. Turn off your computer and switch your HDD to the PROMISE controller
  3. Turn on goto the BIOS ; Advanced > OnBoard Devices Configuration > Onboard Promise Controller [Enabled] > Operating Mode [IDE]
  4. In the BIOSchange Boot priority to CD and exit BIOS
  5. Pop in your XP CD and press F6 (look carefully for the message at the bottom of the screen)
  6. Pop in your Floppy and let XP copy the PROMISE Drivers.
  7. Exit install when asked for.
  8. We are done!
[*] If the above method doesn't work, follow the same steps only let XP repair the existing installation. You should not lose any data.

When your OS does not reside on the SATA being switched:
Follow the same procedure as above i.e load the SATA drivers but exit the XP installation when the drivers are loaded.

17. ATi Radeon random freeze syndrome and the NO Video problem:
Some members and other people have reported problems with certain ATi Radeons i.e the 9800XT, 9700, 9800Pro,9800 Series (pm me if your card is messing up). A possible solution to resolving this issue would be disable AGP fastwrites and force AGP 4x operation. Don't worry you won't see your performance cut by too much in 4x mode. On windows do this:Start> Run > smartgart and you cas disable fastwrites here. I've seen one complaint about not seeing any video output. Troubleshooting checklist:-
Quote:
Originally Posted by MechBGon @ Anandtech
1) you got the ATX12V cable plugged in, right?

2) memory is in the blue slots? Try slot B1 with a single module if necessary. I'd use 2.7 volts on the RAM at the first opportunity, but that requires reaching the BIOS first... you may want to start with a "tame" PC2100 module that runs on 2.5 volts in order to get into the BIOS, then let it proceed through a whole POST afterwards so it knows to keep the settings.

3) the CLR RTC jumper isn't in the CLEAR position, by any chance?

4) PSU's input-voltage selector is set to the right voltage for your area (115V or 230V)?

5) the CPU fan is plugged into the CPU_FAN header?

6) Keyboard's plugged into the purple PS/2 port, not green?

7) The AGP card is definitely getting down all the way into the second row of contacts in the AGP slot, and its auxiliary power cable (if so equipped) is hooked up?

And now for one that might be less obvious: if you plugged the PSU's fan-RPM-monitoring wire into a 3-pin header on the motherboard, unplug it. Low-RPM fans can bogglize motherboards sometimes, for no good reason at all.

18. MOSFET Cooling :

Made no difference to my OC. I could not increase FSB by even 10 MHz! Still at 257 x 10. But, VCore remains stable (as shown by ASUS Probe) @ 1.648 V. My setting in the BIOS is 1.675 V! I may have hit the wall at 2570 MHz. :cry:
*Edit(s): newest last*
(i) My VCore is fluctuating wildly as shown by Speedfan. Could either be the PSU or the Mobo. Still testing.
(ii) Not a PSU problem. This Motherboard undervolts. VCore fluctuation seems to be a well documented problem endemic to ASUS MoBos (will provide links soon).
SEE HERE
--------------------------------------------------------------------------




:: MEMORY SETTINGS ::

We see a lot of questions about optimizing memory settings and I'll attempt to address them here.


Memory terminology and necessary calculations:-
There is a huge body of detailed information on memory, how it works and various other intricate aspects. Lets look at the relevant one's in brief and put them to use in the context of our A8V.
SETTINGS AS IN THE BIOS: [SETTINGS I USE]
{Explanation of my choice}


MEMCLK MODE:[LIMIT]
This enables us to change the other settings manually i.e more control over other parameters.You can reach ~ 2.45 stable with an AUTO setting. Nothing wrong with it, but not recommended if you want killer Sandra Bandwidth scores .

MEMCLK:CPU [5:3 DDR333]
I find this more stable than the AUTO. As before AUTO works quite O.K. But since the Mem is not forced to operate at 5:3 you may encounter instability in the intermediate multiplier area.Lets see what this means. Remember that the RAM frequency is derived off the CPU frequency (2565MHz in my case). Here I'm setting DDR333 i.e forcing the RAM to operate at a lower frequency. This is where the CPU RAM divider comes in and tries to match our target frequency. The CPU RAM divider has specific programmed values based on the CPU Multiplier. If you have high performance memory (PC3700) or low latency RAM, then I'm sure you can run at 2:1 i.e synchronously with the CPU.

Lets look at this table:

CPU......................DRAM SPEED SETTING/DIVIDER APPLICABLE
MULTIPLIER..200MHz(DDR400).....166MHz(DDR333).....133MHz(DDR266).....100MHz(DDR200)

7.0x.........7.0................9.0................11.0................14.0
7.5x.........7.5................9.0................11.0................15.0
8.0x.........8.0................10.0...............12.0................16.0
8.5x.........8.5................11.0...............13.0................17.0
9.0x.........9.0................11.0...............14.0................18.0
9.5x.........9.5................12.0...............14.0................19.0
10.0x........10.0...............12.0...............15.0................20.0
Let us look at my settings and do a sample calculation.
Target DDR333 i.e 166MHz. My CPU Freq= 270MHz (HTT) x 9.5 (Multiplier)=2565MHz. So the corresponding Divider from the chart is 12.0. So my RAM actually runs at 2565/12= 213.75MHz.
Since this RAM is rated to run at 200MHz (DDR400) I'm actually OC'ing my RAM to run at 213.75 MHz. Suppose I pick a setting of DDR400. The corresponding divider is 9.5 and the actual clock speed of my RAM becomes 2565/9.5 = 270MHz (not surprising since this setting ensures RAM in in sync with the HTT)! Surely my RAM can't handle that and so a crash is inevitable.Gautam helpfully pointed out that the half multiplier settings are useless as the A64 does not support half dividers. I could have used 257 x 10 instead of 270 x 9.5 as the divider is the same number 12. This issue is yet unresolved as my rig is stable at 9.5 and fails P95 in 3 min at 12.
Issue miraculously resolved itself! Stable at 257 x 10. Is this an anomaly or is everybody facing the same "problem"?


DRAM [DISABLED]
I'm not sure about this. Will update.

BANK INTERLEAVING [ENABLED]
I think this is necessary if you are running in Dual channel mode, i.e identical sticks in A1 and B1 slots.

NODE INTERLEAVING: [DISABLED]
Not sure about this. I read somewhere that this is useful in a cluster. Single machine users need not bother.

BURST LENGTH: [8 CLK]
Supposed to offer wider bandwidth than the 16 clk setting. I didn't see a big difference in Sandra scores. (Will update with a fresh test and detailed scores).

CAS [2.5 CLK]
The default Column Access Time latency on my stick. You can expect a crash if you go below the rated CAS. My box crashed when I tried 2.0. Stick with the rated CAS.

TRC [9 CLK]
Row Cycle Time. (TRC= Minimum of TRAS+TRP).

TRAS [7 CLK]
Row Active Time. TRAS= TRCD + CAS + 2 CLK. In my case it is 3+ 2.5 + 2 = 7.5 So choose 7 or 8. {Testing at 6 Clk...works for me}. Some people swear that on Athlon64 systems, tras=10 is the best value. Didn't find any difference in Sandra scores.

TRFC [AUTO]
Leave it on auto. I could not boot at 15 Clk!! Strangely, only the AUTO setting works.

TRCD [3 CLK]
RAS to CAS delay.Default is 3 for my chip(s). Prime95 unstable when lowered to 2.0. Use defaults.

TRP [3 CLK]
Row Active Strobe Precharge. Defaults. P95 unstable if lowered.

TWCL [AUTO]
Makes no difference in Sandra scores.

ASYNC LAT [AUTO]
(Still testing)

2T [DISABLED]:
Better memory bandwidth.
From ACE's hardware:
Quote:
Considering that the FX-53 Socket 940 was able to pull 5.4 GB/s with two DDR400 CAS 2.5 registered modules, something was holding the Socket 939 FX-53 back. I started to scrutinize all the memory BIOS settings, and noticed that the bus turnaround was set to 2 by default. Now that is quite weird, as this is only necessary if you load each channel with two DIMMs.An incredible difference: with a faster bus turnaround, the memory subsystem is able to serve up to 24% more bandwidth, and the latency goes down from 51 (21.25 ns) to 47 cycles (19.6 ns). This results in measurable real world performance gains:
* In 3DS Max 5.1, we gained 3% of performance
* In Medieval War, Comanche we also gained 3%
* In Wolfenstein: Enemy Territory, we gained 5.5%
* In WinRAR and Plasma, the performance advantage was no less than 9%!

Bus turnaround happens because the memory bus is a half duplex bus: it can either write or read. The turnaround is the time required to switch between a read and write cycle (or vice versa), and it is indeed a critical performance factor. With DDR, this is a very important performance parameter (contrary to RDRAM) as the "write" or "read" commands are sent and decoded simultaneously with the addresses. Because of this, you could say that the controller doesn't "know" in advance what's going to happen next. If a write is decoded just before a read has been ordered, the write command will have to be delayed to avoid collisions on the data bus. The data bus can only send signals in one direction. Reading data from and writing data to the DRAM will cause the two signals to collide and corrupt the signals.Turning around the bus takes a number of nanoseconds. During this time, no data traffic is possible as the DRAM does not receive any commands. The higher the clockspeed of the controller, the longer the turnaround will take in terms of clock cycles. In other words, the higher the clock speed, the more critical bus turnaround delays are.

As the Athlon 64 already accesses memory with very low latency, a few cycles saved on bus turnaround can really make a difference. At the same time, it must be said that if you use more than one DIMM per channel (so 4 DIMMs in total, or two DIMMs in the first channel), AMD suggests that you use DDR333 with bus turnaround set to 2. However, we have to confirm that we had no trouble at all running with two DIMMs per channel at DDR400 and "2T" enabled.

To sum it up: if you only limit yourself to one DIMM per channel, you will be able to set this BIOS setting to "1T". If you want to use 4 DIMMs or have bought only one DIMM, you will lose about 2 to 9% performance.
Tested by me and proven true, low Sandra scores if 2T is invoked.

READ PREAMBLE [AUTO]
Looks like setting this to higher values is better if your rig is unstable. (Still testing)

HYPERTRANSPORT FREQUENCY [1000 MHz]
Scroll up for the Important note #12.
Logically my system is supposed to be unstable because Max Hypertransport frequency is 1000 MHz where as the corresponding LTD x HTT is 5 x 270 = 1350 MHz! My system is stable (primed for >14 hrs). Makes me think its either a dummy or has a high threshold before failing or I'm a lucky dog with a cool MoBo .
{Added Oct. 21, 2004} I tried lowering my Hypertransport frequency from 1000MHz to 800MHz to test the above premise. Guess what? I got crazy boot-up errors which did have what I assume CPU register dumps and exception error codes. I don't know if it was the BIOS or Windows. This means I'm unstable at lower HT Frequencies!! I had PCI:AGP lock to the 74.xx MHz but I don't think that was the problem because I got the same errors when I used 66.66 MHz. Another possibility could be the 1008.002 beta bios I'm using messing up.
Issue resolved. It was the beta BIOS.

Basic Memory Troubleshooting guide:

These are the basic steps I used to TS my RAM. More detailed information can be found in the Memory section.

1. Update to the latest BIOS.
2. Set Mem timings to AUTO.
3. Get these timings using CPUz or Sandra etc.
4. Now select Manual Mem config.
5. Set VDimm = 2.8 V
6. Keep all timings as in AUTO and tweak one parameter at a time, reverting back to the timings you noted down earlier. I mean for example:- change CL to 2.5 Test and change it back to 2.0 (assuming AUTO seeting is at 2.0) etc..

These steps will tell you what parameter is causing the failure. It may also be a combination of parameters. In such a case select Manual config and change only one parameter at a time, leaving the others in AUTO.
-------------------------------------------------------------------------



:: ASUS Tech Support Troubleshooting List::

Please read this very helpful albeit generic e-mail. Pay particular attention to MS related problems.

Introductory Note:

The problem you are reporting is a stability issue and is usually caused by either the memory, power supply or the CPU running too hot. Please check with the memory maker to see if the module you are using has been tested or certified with the chipset on your motherboard. Also, check that your power supply can supply enough power for your system configuration and that your CPU's heatsink and fan are properly installed. First, please check that your CPU is not overheating. You can check this in the BIOS setup, under Power, then Hardware Monitor. If the CPU is overheating, please check that your heatsink and fan are properly installed and that you are using exactly one heat transfer agent between the CPU and heatsink (ex. Artic Silver). Make sure that the heatsink compound is applied ONLY to the CPU die and that it is NOT shorting any surface mount components on top of the CPU or contacting the CPU's pins. Please check http://www.asus.com/support/cpusupport/cpusupport.aspx to verify that your board will properly support the CPU that you have. This link will list the minimum BIOS and PCB version that you must have to PROPERLY support your CPU.


Memory Related

Please ask the memory manufacturer the following questions:
Q1.Have you done any certification testing with this memory module and our MB (motherboard) or any MB using the same chipset at the density you are trying to use (ex. 3*512MB DIMMS totaling 1.5GB of memory)?

Q2.If the answer is no and they just build to a certain specification/standard, then who's specification/standard do they use and how do they validate their design? If they just build the memory and do not validate it how do they know it will perform?


When we design our MB we follow the chipset makers spec. with regards to memory. We do validate the memory IC maker's reference designs with our MB to make sure they are up to spec. and if we need to add support for Higher density IC. Most memory module makers do not follow the IC maker's reference designs so they can use whom ever IC's on their memory Module. There is nothing wrong with this, but the problem is if they deviate from the IC makers reference designs and or even if they use the IC makers reference they should do some level of validation to insure quality.

We are one of the largest MB makers so one would think that if you were making a product you would test your product with the most popular associated products. Here is a list of memory manufacturers that use our MB's as well as others to do this type of testing and also will post a guarantee that their memory will work in our MB's as well as others. Please check the following websites for compatibility:
www.corsairmicro.com
www.unigen.com
www.smartm.com
www.mushkin.com


Startup TS Tips:

Also, here are some setup tips that may help you to set up the system.

-Please start by clearing the system c-mos memory following the procedure in your manual. Make sure the power AND the battery on the motherboard is removed before shorting the CLR RTC solder points or jumper pins.

-You may need to refer to the "Interrupt Request Table" in your manual and arrange your PCI add-in cards so they do not cause interrupt conflicts.
Some devices just do not get along with each other.

-You may also remove your PCI cards to see if this corrects the problem. If it does, add the cards back one by one to see which card is causing the problem. Anytime you add, move, or remove a PCI card, please enter BIOS setup and set the option "Reset Configuration Data" to "Yes" if your BIOS has this option. This will force the BIOS to reassign resources to your PnP devices.

-Next, enter Bios Setup and load setup defaults. Under Boot, if needed, please set the BOOT order. Then set "Reset Configuration Data" To "YES". Any other changes you need to make for your specific system configuration please do so at this time.Next, exit "Saving Changes" and this should correct the issue.

-Also for Windows 9X or ME, they have some issues with CPU speeds greater
than 2.1 GHz (see MS Q312108) or memory above 512 MB total system memory (see MS Q253912). These O.S. cannot cache the memory properly. If your CPU is greater than 2.1 GHz, you will need to contact Microsoft Product Support Services to obtain a patch to correct this issue.


Microsoft Issues

Here are some links for common memory related problems:

-Err Msg: "Out of Memory" Error Messages with Large Amounts of RAM
Installed


-Err Msg: Windows Protection Error in NDIS with a CPU That Is Faster Than
2.1 GHz


-Err Msg: While Initializing Device NTKERN: Windows Protection...

-Err Msg: Insufficient Memory to Initialize Windows

-Computer May Reboot Continuously with More Than 1.5 GB of RAM


-Computer Speed and Performance May Decrease

- Msg: Not Enough Extended Memory Available to Run Windows

-Troubleshooting Stop 0x24 or NTFS_FILE_SYSTEM Error Messages


-Make sure you have a FRESH install of your OS on THIS motherboard and that the drive was partitioned and formatted on this motherboard. Make sure
that you install the most recent drivers in the following order:
1) Chipset/Motherboard drivers (Example, VIA 4 in 1's for VIA chipsets,
Intel INF and Application Accelerator for Intel chipsets, Etc.) Do this
BEFORE loading any other driver!
2) Latest version of Direct X.
3) Latest Video Card drivers.
4) SCSI/ATA drivers
5) Lan/NIC drivers
6) Modem drivers, then any other drivers
7) Finally, install sound card drivers last.



PSU Recommendations

Also, please check to make sure your power supply can supply enough power to your system.If the CPU is below 700 MHz, the power supply we suggest as follows:

300 Watts with +5 Volt @ 25 amps, +12 Volt @ 10 amps, and +3.3 Volt @ 14 amps.

If the CPU speed is above 700MHz, the power supply we suggest as follows:
350 Watt with +5V @ 30 amps, +12V @ 12 amps, and +3.3V @ 22 amps.

For Dual Intel MB, Min. P.S. should be:
350 Watt with +5V @ 35 amps, +12V @ 15 amps, and +3.3V @ 22 amps.

For Dual AMD MB, Min. P.S. should be:
400 Watt with +5V @ 38 amps, +12V @ 15 amps, and +3.3V @ 20 amps. Please note that these are only suggestions and the actual size of the power supply may be higher than we suggest. Also, depending on the config of your system, you may need more power than listed above. For example, running multiple hard drives, CR-Rom drives or other accessories may increase the power requirements of your system!


Electromagnetic Interference Issues

If that still doesn't work, then please check that the MB is not picking up EMI interference or improperly grounding against the case. Please remove the MB from the case and set it up DIRECTLY on the cardboard box it came in or some other non-static, non-conductive surface like a telephone book. Connect up everything to the MB and see if the problem persists. If this corrects the problem, you will need to electrically isolate the MB from the case. You will need to use electrical tape over the brass stand-offs that support the MB to the case and use the red paper washers between the screw heads and the MB. The red paper washers should have came with your case hardware or they can be purchased at Radio Shack or a computer store. Also double check that a motherboard standoff is NOT placed in a spot where the motherboard is not designed to be grounded, as some cases will have different standoff locations to accommodate different styled of motherboards. Also check that there are no other metal objects that could contact the motherboard or any other electrical device attached to the system, such as a metal burr, loose screw, metal rod, or any other object that could cause a short.


IF ALL ELSE FAILS

Please call 502-995-0883 if you need further assistance.

Best Regards,
Technical Support Team
NT001
-------------------------------------------------------------
ASUS COMPUTER INC.
Technical Support
Email: TSD@asus.com << <mailto:tsd@asus.com>>>
Phone: +1-502-995-0883
URL: << <http://usa.asus.com/>>>
<< <http://download.asus.com.tw/>>> (for download page)
<< <ftp://ftp.asus.com/>>>
Customer Service
Email: Rma@asus.com << <mailto:Rma@asus.com>>>
Phone: +1-510-739-3777 EXT. 5105
------------------------------------------------------------------




:: Volt-Mods ::

Caution: Perform at your own risk! I have not done any Volt Mods on my board (yet!). There is no way (AFAIK, yet) to override the overvoltage protection even with a VoltMod. O.K folks, section is going to be updated soon. Meanwhile, you can find more information with the following links:-
-------------------------------------------------------------------------



:: Useful Links and other Resources ::

A good collection of information with active discussion about this board can be found here:-
--------------------------------------------------------------------------



::BIOS Settings for Video/PCI Devices and Troubleshooting ::



All definitions are taken (in verbatim) from Adrian Wong's excellent BIOS Optimization guide.


SETTINGS AS IN THE BIOS: [SETTINGS I USE] {Explanation of my choice}
Advanced> Chipset Configuration

PRIMARY GRAPHICS ADAPTER [AGP]
Sets up the AGP Card for use. Self Explanatory.

SEARCH FOR MDA RESOURCES [YES]
Not sure about this one.

VLINK 8X SUPPORTED [ENABLED]
The VLink 8X Support BIOS feature is used to toggle the V-Link bus mode between the original V-Link and the newer and faster 8X V-Link. If this feature is enabled, the quad-pumped 8-bit V-Link bus will switch to the new 8X V-Link mode, which runs at 133MHz and delivers a bandwidth of 533MB/s.
If this feature is disabled, the V-Link bus will use a clock speed of 66MHz, essentially reverting to the original V-Link standard. It will then deliver a bandwidth of 266MB/s.This BIOS feature was most likely included for troubleshooting purposes. It is highly recommended that you enable this BIOS feature for better performance.

AGP MODE [8X]
This BIOS feature is a toggle for the motherboard's AGP 8X support.
When enabled, it allows the AGP bus to make use of the AGP 8X transfer protocol to boost the AGP bus bandwidth.

If it's disabled, then the AGP bus is only allowed to use the AGP 4X transfer protocol.The AGP 8X protocol must be supported by both the motherboard and graphics card for this feature to work. Of course, this feature will only appear in your BIOS if your motherboard supports the AGP 8X transfer protocol! So, all you need to do is make sure your graphics card supports AGP 8X transfers.

If it does, enable AGP 8X Mode to take advantage of the faster transfer mode. You must disable it if your graphics card doesn't support AGP 8X transfers. The BIOS will then report that the maximum supported transfer mode is AGP 4X.

Set to 4x for troubleshooting purposes.

AGP FASTWRITE [ENABLED]
This BIOS feature controls the AGP bus' Fast Write capability. Fast Write is a feature which accelerates memory write transactions from the chipset to the AGP device.Fast Write allows the AGP device act like a PCI device. This allows it to bypass the main memory and directly access the data which improves AGP read performance. However AGP write performance is not affected. It is recommended that you enable AGP Fast Write for better AGP read performance but disable it if any of your PCI cards start acting funny.


GRAPHICS APERTURE SIZE [64 Mb]
This BIOS feature does two things. It selects the size of the AGP aperture and it determines the size of the GART (Graphics Address Relocation Table).
The aperture is a portion of the PCI memory address range that is dedicated for use as AGP memory address space while the GART is a translation table that translates AGP memory addresses into actual memory addresses which are often fragmented.

The GART allows the graphics card to see the memory region available to it as a contiguous piece of memory range.Host cycles that hit the aperture range are forwarded to the AGP bus without need for translation. The aperture size also determines the maximum amount of system memory that can be allocated to the AGP graphics card for texture storage. Please note that the AGP aperture is merely address space, not actual physical memory in use. Although it is very common to hear people recommending that the AGP aperture size should be half the size of system memory, that is wrong!
The requirement for AGP memory space shrinks as the graphics card's local memory increases in size. This is because the graphics card will have more local memory to dedicate to texture storage. So, if you upgrade to a graphics card with more memory, you shouldn't be "deceived" into thinking that you will need even more AGP memory! On the contrary, a smaller AGP memory space will be required. It is recommended that you keep the AGP aperture around 64MB to 128MB in size, even if your graphics card has a lot of onboard memory. This allows flexibility in the event that you actually need extra memory for texture storage. It will also keep the GART (Graphics Address Relocation Table) within a reasonable size.

AGP 3.0 CALIBERATION CYCLE [DISABLED]
Set to enabled for Troubleshooting purposes.

DBI FOR AGP TRANSFER [ENABLED]
The full name for this BIOS feature is Dynamic Bus Inversion Output for AGP Transmitter. It is an AGP 3.0-specific BIOS feature which will only appear when you install an AGP 3.0-compliant graphics card.When enabled, the AGP controller is allowed to use the Dynamic Bus Inversion scheme to reduce power consumption and signal noise.

When disabled, the AGP controller will not use the Dynamic Bus Inversion scheme to reduce power consumption and signal noise.The AGP bus has 32 data lines divided into two sets. Sometimes, a large number of these data lines may switch together to the same polarity (either 1 or 0) and then switch back to the opposite polarity.

This mass switching to the same polarity is called simultaneous switching outputs and it creates a lot of unwanted electrical noise at the AGP controller and GPU interfaces.To avoid this, the AGP 3.0 specifications introduced a scheme called Dynamic Bus Inversion or DBI. It makes use of two new DBI lines - one for each 16-line set. These DBI lines are only supported by AGP 3.0-compliant graphics cards.

Dynamic Bus Inversion ensures that the data lines are limited to a maximum of 8 simultaneous switchings or transitions per 16-line set. It does so by switching the DBI line instead of the data lines when the number of simultaneous transitions exceeds 8 or 50% of the data lines. This ensures that electrical noise due to simultaneous switching outputs are minimized. In short, DBI improves stability of the AGP interface by reducing signal noises that occur as a result of simultaneous switching outputs. It also reduces the AGP controller's power consumption. Therefore, it is recommended that you enable DBI Output for AGP Trans. to save power as well as reduce signal noise from simultaneous switching outputs.
Toggle when troubleshooting.


Advanced> Southbridge Configuration

MPS REVISION [1.4]
This feature is only applicable to multiprocessor motherboards as it specifies the version of the Multi-Processor Specification (MPS) that the motherboard will use. The MPS is a specification by which PC manufacturers design and build Intel architecture systems with two or more processors.
MPS 1.1 was the original specification. MPS version 1.4 adds extended configuration tables for improved support of multiple PCI bus configurations and greater expandability in the future. In addition, MPS 1.4 introduces support for a secondary PCI bus without requiring a PCI bridge. If your operating system comes with support for MPS 1.4, you should change the setting from the default of 1.1 to 1.4. You also need to enable MPS 1.4 support if you need to make use of the secondary PCI bus on a motherboard that doesn't come with a PCI bridge.

You should only leave it as 1.1 only if you are running an older operating system that only supports MPS 1.1. According to Eugene Tan, Windows NT already supports MPS 1.4. Therefore, newer operating systems like Windows 2000 and Windows XP shouldn't have any problem supporting MPS 1.4.
However, users of the ABIT BP6 motherboard and Windows 2000 should take note of a possible problem with the MPS version set to 1.4.

Dan Isaacs reported that when you set the MPS version to 1.4 in the ABIT BP6, Windows 2000 will not use the second processor. So, if you encounter this problem, set the MPS Revision to 1.1.


PCI DELAYED TRANSACTION [ENABLED]
To meet PCI 2.1 compliance, the PCI maximum target latency rule must be observed. According to this rule, a PCI 2.1-compliant device must service a read request within 16 PCI clock cycles for the initial read and 8 PCI clock cycles for each subsequent read. If it cannot do so, the PCI bus will terminate the transaction so that other PCI devices can access the bus.
But instead of rearbitrating for access (and failing to meet the minimum latency requirement again), the PCI 2.1-compliant device can make use of the PCI Delayed Transaction feature. With PCI Delayed Transaction enabled, the target device can independently continue the read transaction. So, when the master device successfully gains control of the bus and reissues the read command, the target device will have the data ready for immediate delivery.
This ensures that the retried read transaction can be completed within the stipulated latency period. If the delayed transaction is a write, the master device will rearbitrate for bus access while the target device completes writing the data. When the master device regains control of the bus, it reissues the same write request. This time, the target device just sends the completion status to the master device to complete the transaction. One advantage of using PCI Delayed Transaction is that it allows other PCI masters to use the bus while the transaction is being carried out on the target device. Otherwise, the bus will be left idling while the target device completes the transaction.

PCI Delayed Transaction also allows write-posted data to remain in the buffer while the PCI bus initiates a non-postable transaction and yet still adhere to the PCI ordering rules. Without PCI Delayed Transaction, all write-posted data will have to be flushed before another PCI transaction can occur.It is highly recommended that you enable PCI Delay Transaction for better PCI performance and to meet PCI 2.1 specifications. Please note that while many manuals and even earlier versions of the BIOS Optimization Guide have stated that this is an ISA bus-specific BIOS feature which enables a 32-bit write-posted buffer for faster PCI-to-ISA writes, they are incorrect! This BIOS feature is not ISA bus-specific and it does not control any write-posted buffers. It merely allows write-posting to continue while a non-postable PCI transaction is underway.
Disable it only if your PCI cards cannot work properly with this feature enabled or if you are using PCI cards that are not PCI 2.1 compliant.


Advanced> PCI PnP

PLUG AND PLAY OS [YES]
I think this option is necessary to support the PnP features of Windows or hotplug in Linux (not very sure).

PCI LATENCY TIMER [64]
This BIOS feature controls how long a PCI device can hold the PCI bus before another takes over. The longer the latency, the longer the PCI device can retain control of the bus before handing it over to another PCI device.
Normally, the PCI Latency Timer is set to 32 cycles. This means the active PCI device has to complete its transactions within 32 clock cycles or hand it over to the next PCI device.

For better PCI performance, a longer latency should be used. Try increasing it to 64 cycles or even 128 cycles. The optimal value for every system is different. You should benchmark your PCI cards' performance after each change to determine the optimal PCI latency time for your system.

Please note that a longer PCI latency isn't necessarily better. A long latency can also reduce performance as the other PCI devices queuing up may be stalled for too long. This is especially true with systems with many PCI devices or PCI devices that continuously write short bursts of data to the PCI bus. Such systems would work better with shorter PCI latencies as they allow rapid access to the PCI bus.

In addition, some time-critical PCI devices may not agree with a long latency. Such devices require priority access to the PCI bus which may not be possible if the PCI bus is held up by another device for a long period. In such cases, it is recommended that you keep to the default PCI latency of 32 cycles.

ALLOCATE IRQ TO PCI VAA [YES]
No idea what this means!


PALETTE SNOOPING [DISABLED]
This BIOS feature determines if your graphics card should allow VGA palette snooping by a fixed function display card. It is only useful if you use a fixed-function display card that requires a VGA-compatible graphics card to be present (i.e. MPEG decoder card).

Such fixed-function display cards generally do not have their own VGA palette. So, they have to "snoop" VGA palette data from the graphics card to generate the proper colours. Normally, the graphics card's Feature Connector is used for this purpose.

When this feature is enabled, the graphics card will not respond to framebuffer writes. It will forward them to the fixed-function display card via its Feature Connector. The fixed-function display card will then snoop the palette data and generate the proper colours.When this feature is disabled, the graphics card will display all framebuffer writes.

It is recommended that you disable this feature if you do not use any fixed-function display card like a MPEG decoder card.
But if you are using a fixed-function display card that requires palette snooping, enable this feature. Otherwise, the colours displayed may not be accurate and the monitor will blank out once you stop using the fixed-function display card.


Advanced> System Freq Voltage Configuration

AGP/PCI FREQUENCY [66.66/33.33]
This is your AGP/PCI lock. Most devices won't work at the higher 74.xx setting.

AGP VOLTAGE [1.6 V]
Set to 1.6 volts if you have problems.

V-LINK VOLTAGE [2.5 V]
Toggle if you have problems.


Advanced> Power Configuration

I didn't realize that there was so much to ACPI! I picked all the material off HERE.

Advanced Configuration and Power Management:

Q. What is ACPI?
A. Users should be aware that ACPI does not improve performance or reliability. If there is a thing that it can improve that thing is called functionality. ACPI is an abstract hardware interface that provides a standard way to integrate power management in a PC system, including hardware, operating system and application software. The advantages of ACPI are obvious, but may not be required for all users. With ACPI functions you can turn peripherals on and off, power-on the system on external activities like the ring of the telephone, etc. Without any doubt it may be a very useful feature, but does it worth?

The reasons why you should use ACPI are multiple, but the most important are:

* ACPI provides a new advanced Plug and Play environment for today computers
* ACPI uses it's own language so it's supported by other operating systems
* ACPI is not Microsoft property
* ACPI provides control of Power state
* ACPI use the System Control Interrupt (SCI)
* ACPI should make your life easier because it handles the resources problems
* ACPI should allow you to insert / remove devices in the fly


Q. How do I use ACPI in Windows 2000/XP?
A. In order to use ACPI in Windows 2000 you must have it enabled in BIOS at the setup stage. Win 2K use a different HAL (Hardware Abstraction Layer) for ACPI enabled machines! Assuming that you installed Windows 2K with ACPI support and you want to disable ACPI in BIOS you will realize that your computer won't boot. Normally most motherboards provide ACPI enable / disable settings in BIOS but there are several which do not allow the user to disable the ACPI support. When something goes mad apparently you can not get rid of ACPI. Just apparently, because there are solutions.

The computer ACPI support is detected in the setup stage, but there are less frequent cases when ACPI is not detected and the Standard PC HAL is installed. If you want to force the installation of the ACPI HAL:

1. At the blue screen of the Windows 2000 setup when it prompts you to press F6 for additional mass storage devices press F5 and you will have a list of options.
2. Choose the proper HAL for your machine remembering that there are also different HALs for multiprocessor machines.
3. Complete the setup, go to Setup / Control Panel / System / Hardware / Device Manager / Computer. If you have listed the Advanced configuration and Power Management (ACPI) the process was completed successfully.

Warning: You can do the same thing if you don't want to install ACPI and your BIOS doesn't support ACPI disable setting. Just choose the Standard Computer HAL from the list of options.

Q. Any other methods to disable ACPI?
A.Wait for the text-phase of Win2K setup to complete
before W2K Setup Wizard to load, boot into DOS and change ACPIEnable = 0 from ACPIEnable = 1 in file c:\txtsetup.sif


The reason why ACPI is not installed on a computer with known ACPI compatibility may be a Microsoft known subtle incompatibility which can lead to hardware instability. In the installation step Windows 2000/XP checks the BIOS compatibility with ACPI and the BIOS presence on a known bad list. If the BIOS is there ACPI is not installed. Of course than errors can occur so you can force the ACPI installation (not recommended).

SUSPEND MODE [S1 --> AUTO]
Not exactly sure what S1, S3 etc mean. I suspect this is related to Energy Star Compliance.

REPOST VIDEO ON S3 RESUME [NO]
Not too sure about this. Looks to be Energy Star related.

ACPI 2.0 SUPPORT [YES]
Explained previously.

ACPI APIC SUPPORT [ENABLED]
Explained previously.


Advanced> Power Management/APM

QUICK BOOT [ENABLED]
Disable for Troubleshooting purposes.

FULL SCREEN LOGO [DISABLED]
If you don't like the A8V logo or if you are as impatient as I am, disable this!

ADDON ROM DISPLAY MODE [FORCE BIOS]
I don't know about this. No noticable change when I toggle this setting.

BOOTUP NUM LOCK [ON]
Just an indication that your keyboard is ready for use. Your Num Lock LED turns on.

INTERRUPT 19 CAPTURE [DISABLED]
This invokes the BIOS Bootstrap loader. I got the details from HERE. I don't understand all of it myself but I figured it would be helpful to somebody with a system programming background.

Interrupt 19:
This interrupt reboots the system without clearing memory or restoring interrupt vectors. Because interrupt vectors are preserved, this interrupt usually causes a system hang if any TSRs (Memory Resident Programs: Terminate but Stay Resident) have hooked vectors from 00h through 1Ch, particularly INT 08.

Usually, the BIOS will try to read sector 1, head 0, track 0 from drive A: to 0000h:7C00h. If this fails, and a hard disk is installed, the BIOS will lead sector 1, head 0, track 0 of the first hard disk. This sector should contain a master bootstrap loader and a partition table (see #00650). After loading the master boot sector at 0000h:7C00h, the master bootstrap loader is given control (see #00653). It will scan the partition table for an active partition, and will then load the operating system's bootstrap loader (contained in the first sector of the active partition) and give it control.

True IBM PCs and most clones issue an INT 18 if neither floppy nor hard disk have a valid boot sector to accomplish a warm boot equivalent to Ctrl-Alt-Del, store 1234h in 0040h:0072h and jump to FFFFh:0000h. For a cold boot equivalent to a reset, store 0000h at 0040h:0072h before jumping. VDISK.SYS hooks this interrupt to allow applications to find out how much extended memory has been used by VDISKs (see #00649). DOS 3.3+ PRINT hooks INT 19 but does not set up a correct VDISK header block at the beginning of its INT 19 handler segment, thus causing some programs to overwrite extended memory which is already in use.

The default handler is at F000h:E6F2h for 100% compatible BIOSes MS-DOS 3.2+ hangs on booting (even from floppy) if the hard disk contains extended partitions which point at each other in a loop,since it will never find the end of the linked list of extended partitions under Windows Real and Enhanced modes, calling INT 19 will hang the system in the same was as under bare DOS; under Windows Standard mode, INT 19 will successfully perform a cold reboot as it appears to have been redirected to a MOV AL,0FEh/OUT 64h,AL sequence when loading the remainder of the DOS system files fails, various versions of IBMBIO.COM/IO.SYS incorrectly restore INT 1E before calling INT 19, assuming that the boot sector had stored the contents of INT 1E at DS:SI instead of on the stack as it actually does.


-------------------------------------------------------------------------



:: AMI BIOS Beep/Error Codes & OCF Troubleshooting guides::

This is a very generic AMI BIOS code list. Look at the A8V Manual if there are any conflicts with this. I have never gotten any of these errors except the 1 beep!


1 beep : Refresh failure

2 beeps : Parity error

3 beeps : Base 64K memory failure

4 beeps : Timer not operational

5 beeps : Processor error

6 beeps : 8042 - gate A20 failure

7 beeps : Processor exception interrupt error

8 beeps : Display memory read/write failure

9 beeps : ROM checksum error

10 beeps : CMOS shutdown register read/write error

11 beeps : Cache memory bad



-For persistant 1 beep, 2 beeps, or 3 beeps try reseating the memory first. If the error still occurs, replace the memory with known good chips.

-For 4 beeps, 5 beeps, 7 beeps, or 10 beeps the system board must be sent in for repair.

-For 6 beeps try reseating the keyboard controller chip. If the error still occurs, replace the keyboard chip. If the error persists, check parts of the system relating to the keyboard, e.g. try another keyboard, check to see if the system has a keyboard fuse.

-8 beeps indicates a memory error on the video adapter. Replace the video card or the memory on the video card.

-9 beeps indicates faulty BIOS chip(s). It is not likely that this error can be corrected by reseating the chips. Consult the motherboard supplier or an AMI product distributor for replacement part(s).

-If no beeps are heard and no display is on the screen, The first thing to check is the power supply. Connect a LED to the POWER LED connection on the motherboard (The LED could go bad!). If this LED lights and the drive(s) spin up then the power supply will usually be good. Next, inspect the motherboard for loose components. A loose or missing CPU, BIOS chip, Crystal Oscillator, or Chipset chip will cause the motherboard not to function.Next, eliminate the possibility of interference by a bad or improperly set up I/O card by removing all cards except the video adapter. The system should at least power up and wait for a drive time-out. Insert the cards back into the system one at a time until the problem happens again. When the system does nothing, the problem will be with the last expansion card that was put in.


OCF Troubleshooting tips can be found
HERE, and HERE
__________________
R.P Feynman : "If I can't create it...it doesn't exist "
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #2 (permalink)  
Old 11th February, 2006, 07:52 PM
cloasters's Avatar
Asst. BBS Administrator
 
Join Date: September 2001
Location: Location, Location
Posts: 21,947

Holy crudomatics, this is fantastic information. Thank you very much, Super Nade!
__________________
When the world will be better.
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #3 (permalink)  
Old 11th February, 2006, 08:10 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Acknowledge your sources! Otherwise, we risk having to remove it due to copyright infringement.
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #4 (permalink)  
Old 11th February, 2006, 08:46 PM
Super Nade's Avatar
Member
 
Join Date: December 2005
Location: Indianapolis, USA
Posts: 157
Send a message via Yahoo to Super Nade

All of them have been acknowledged. If you find anything missing, please let me know. I had a horrendous time editing all of this from here, and hence, may have missed something:-

http://www.ocforums.com/showthread.php?t=332591
__________________
R.P Feynman : "If I can't create it...it doesn't exist "
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #5 (permalink)  
Old 11th February, 2006, 09:43 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

For example, some of the text from the DBI section appears to have been lifted from Adrian's Rojak Pot.
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #6 (permalink)  
Old 12th February, 2006, 12:22 AM
Super Nade's Avatar
Member
 
Join Date: December 2005
Location: Indianapolis, USA
Posts: 157
Send a message via Yahoo to Super Nade

Maybe you missed it, but its been mentioned in the very first line, in bold, that I used Adrian Wong's material in verbatim

Quote:
All definitions are taken (in verbatim) from Adrian Wong's excellent BIOS Optimization guide.
__________________
R.P Feynman : "If I can't create it...it doesn't exist "
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #7 (permalink)  
Old 12th February, 2006, 12:33 AM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

Indeed you did - I had missed it because it's half way through the post!
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #8 (permalink)  
Old 12th February, 2006, 07:13 AM
Super Nade's Avatar
Member
 
Join Date: December 2005
Location: Indianapolis, USA
Posts: 157
Send a message via Yahoo to Super Nade

Well, originally, it fit 3 seperate pages, but I didn't want someone to post as I was writing it up, so I mashed it into one big post.
__________________
R.P Feynman : "If I can't create it...it doesn't exist "
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #9 (permalink)  
Old 12th February, 2006, 11:19 AM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

BTW, what customisations did you go for on a Stage 1 install?
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #10 (permalink)  
Old 13th February, 2006, 02:25 AM
Super Nade's Avatar
Member
 
Join Date: December 2005
Location: Indianapolis, USA
Posts: 157
Send a message via Yahoo to Super Nade

Here are the C-Flags I used:
Code:
CFLAGS="-O3 -march=athlon64 -ffast-math -funroll-loops -funit-at-a-time -fpeel-loops -ftracer -funswitch-loops -fomit-frame-pointer -pipe"
The customizations I made when compiling the kernel was to include the sk98lin networking module in my kernel. Other than that, i didn't do anything else. Still working my way through the various options and customizations.
__________________
R.P Feynman : "If I can't create it...it doesn't exist "
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #11 (permalink)  
Old 13th February, 2006, 10:18 AM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

I just wondered why you started with a Stage 1 install rather than a Stage 3. Setting the CFLAGS can be done in stage 3. It's only if you need to customise specific options in the bootstrap do you need to bother with stage 1.
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #12 (permalink)  
Old 13th February, 2006, 03:10 PM
Super Nade's Avatar
Member
 
Join Date: December 2005
Location: Indianapolis, USA
Posts: 157
Send a message via Yahoo to Super Nade

I'm not sure if that is correct. You need to set correct CFLAGS if you are compiling everything from scratch? Stage 3 is basically for Genkernel users, right?
__________________
R.P Feynman : "If I can't create it...it doesn't exist "
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #13 (permalink)  
Old 13th February, 2006, 04:52 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

CFLAGS are set for optimisations. Whilst it is true that setting the wrong CFLAGS can cause your system to not work at all, not setting any CFLAGS will result in a working system.

Stage 3 is for those users who just want to set CFLAGS and install the system. You can use Genkernel to build your kernel whatever stage you used for installation - it's just a tool to make building the kernel much easier.
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
Reply



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Overclocking guide for dummies map9078 Random Nonsense! 3 17th July, 2006 04:40 AM
Overclocking your Nvidia 6800/6800LE DDR graphics card: Overclocking/Unlocking Guide. MONKEYMAN AOA FAQ 0 16th July, 2005 10:09 PM
Guide to overclocking for me plz. y2butonz General Hardware Discussion 8 6th July, 2005 01:55 PM
Overclocking Guide deeppow EPoX MotherBoards 2 5th April, 2003 07:54 PM
Overclocking Guide Uncle Bob EPoX MotherBoards 10 24th March, 2003 09:40 PM


All times are GMT +1. The time now is 08:59 PM.


Copyright ©2001 - 2010, AOA Forums
Don't Click Here Don't Click Here Either

Search Engine Friendly URLs by vBSEO 3.3.0