• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

ASUS A8N-SLI :: Work Arounds:Expensive Bugs:Tweaking:Info

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

OC-Master

Member
Joined
Jul 14, 2001
Location
Edmonton, Alberta
Here is my A8N-SLI faq;

LAST UPDATED: 3/23/06

Users of the A8N32-SLI Deluxe should check the ASUS website for the latest bios updates and driver fixes. the ASUS A8N-SLI Deluxe and A8N32-SLI Deluxe are two completely different boards which means bios'es are incompatible with each other.

Faq Updates / Beta Bios 1017.004
DOWNLOAD
I've updated a few links that were no longer working. Bios 1017.004 is also available for those of you who wish to update to it. It contains a few updates one which is support for the FX60.

Driver Update To v6.69-6.70 Reminder
v6.69 WindowsXP x64 v6.70 WindowsXP Pro
Just a reminder that everyone should update there chipset drivers to the latest nVidia ones to ensure the best compatibility and bug fix updates for best performance and peace of mind.

Bios 1016 Final Released
DOWNLOAD
With this update, the A8N-SLI Deluxe appearently supports all Socket 939 CPUs to date including Sempron, Opteron, Athlon64, Athlon64-FX and Athlon64 X2 CPUs. ASUS has also finally removed the useless floppy 2.88 option.

Raid Service U.S. English Resources Crashes
It seems that I'm not the only one with this bug. After installing the latest nForce drivers, the RAID utility that is installed seems to always crash on me for some unknown reason. Currently, my only fix around this problem is to manually remove the file to avoid this error. My system is working great without this program loaded and I'm not quite sure what exact task it possesses. UPDATE: Seems that Bios 1012-1013 with Drivers v6.66 or newer completely fixes this issue for those of you who could not get the raid service drivers to function correctly.

nForce4 Driver v6.66+ Fixes Reboot Issue
DOWNLOAD
The reboot issue affects bios'es 1012 beta and newer. Thanks to all who have reported the fix to me. It seems with the startling news ASUS launched the other day has gotten people to update there chipset drivers and have found the reboot issue to have dissappeared.

Some Breaking News
For those of you who have suffered stability issues with your board with bios 1010 or newer, ASUS has finally spoken up today on whats up..
""A8N-SLI Bios version 1010
Please do read the description before updating BIOS
To avoid damaging system, please do update the chipset driver to below version prior to this Bios.
For WinXP 32bit system, please download and update chipset drivers V6.65 or later
For WinXP 64bit system, please download and update chipset drivers V6.65 or later
For Win2k/2003 32bit system, please download and update chipset drivers V6.66 or later
For Win2003 64bit system, please download and update chipset drivers V6.66 or later ""

So there you have it. If you have the 1010~1013 series, give the newer chipset drivers a shot before doing any overclocking or even any system use for that matter.
For those of you who are curious, I found this bit of news on ASUS's download page..
LINK

Lastly, this bug also effects the ASUS A8N-SLI Premium with bios 1006 or newer so if you've got the premium boards, get those new drivers now. For ASUS to say that damaging the board in a news break, it must be serious stuff.

Get Your Bios'es Here:
The Rebels Haven
This website is definately the best place to find odd bios'es, betas or bios'es that have just seemed to have dissappeared. It also has the famous 1010.001 bios for those of you looking for it.

Gain More Voltage From Your A8N/E/SLI/DELUXE/PREMIUM:
SCREENSHOT
I've performed a vcore mod the other night and it was a success! Like I was hoping for, my vcore went from 1.58v with the beta bios 1012 and 1010 to 1.68v after the vmod was performed. I would like to thank the people over at VR-Zone for sharing their ideas with us. You can find out how to perform the vcore mod by clicking HERE . Dont forget to ground all your 4 pins properly and use a meter gauge while performing this operation. I used the fan header nearby for ground.

Vdimm Mod:
The Vdimm mod from here actually works well. I gained a bit more stability with my memory from the mod but unfortunatly due to OVP (Over voltage protection) my ram will only allow up to 3.08v to go into it making the 3.3v not very effective. However, I know many of you out there do have ram without OVP that would love to juice up their memory.

Zalman Cooler Makes A Good Replacement:
I've just completed a huge mod which envolves the NV Silencer 6 series GPU coolers and the Zalman passive northbridge cooler to work in parrellel on the A8N-SLI Deluxe motherboard. This cooler definately destroys the stock northbridge cooling performance wise and with zero noise level. My PCs noise level has been cut in half since removing the northbridge fan.

Venice Handicaps The A8N-SLI Deluxe:
I've had a couple of hours to play around with the Venice 3200+ CPU in an A8N-SLI Deluxe and it appears that all memory dividers when used kill memory stability and severly handicap your memorys clock speeds. Testing with OCZ Premier PC3200 which can easily do 450MHz,,, the memory begins to lock up at around 366MHz and loosening timings or using 2T does nothing for getting the memory to clock higher. As soon as I switched back from 6/4 divider to 1:1, the system would even boot at 10 X 240 (2.40GHz) with memory running at 480MHz CL2.5 3-3-5-1T. This is a terrible limitation which currently is limiting overclocking on the Venice CPU platforms.

nForce4 Hardware Firewall:
Turn this off! Untill further notice, the hardware firewall seems to eventually BSOD all nForce4 based PCs and the bug is nasty. From what I've heard, the problem is global and all nForce4 manufacturers seem to be facing this problem in one way or another. Hopefully, a bios update / driver update will fix this so that we can sit back with peace of mind. BSOD has to be one of the most fustrating things we all face especially when octimizing a PC.

Lack Of 1T Timings At High Speed Is Becoming A Pain:
This is a terrible bug that has been hindering this board since its launch over four months ago. ASUS must address this if they want to keep customers and build new customers for that matter. The fact is, this board does not allow 1T timings with FSB/Mem above 240 without loosing complete stability. Some people have gotten lucky and got to 245 and 250 but thats it! If anyone has a reliable source on this and can confirm if a fix is on the way, PM me ASAP,, this bug needs to be fixed.

PATA Still A No Go Even With the PCI Lock:
7/7/05 UPDATE: Seems as of bios 1010 or newer, the PATA bug no longer exists. I've tested several drives in PATA and I havnt seen any curruption as of over a week of operation. Should be an option now to people who have PATA drives.

Original Bug Dated April 2005;
With the HTT set and PCI lock on, PATA hard drives still seem to currupt after a few weeks of use no matter what bios I use or method of configuring. Of course, my solution was to install my SATA converters which completely eliminated this nasty bug. Hard Drive curruption has to be one of the most fustrating bugs you could run into. For those of you running PATA drives on the A8N-SLI, go out and spend a few bucks on a SATA converter and avoid this nasty bug all together.

HTT 3X Reverting Bug:
UPDATE: 7/7/05
Seems that random bios's do not suffer from this glitch yet all the latest bios'es all definately do. Very strange bug but never the less not to nasty.

This has got to be one of the most annoying bugs yet. Not that it really is a bad thing but the A8N-SLI when equiped with any bios between 1003~1012? will randomly revert your HTT to 3X even if it is set to 4X or 5X in the bios. Most users can identify this issue simply by loading up nTune and monitoring their HTT access. The current only work around for this issue is to manually enable 4X or 5X through nTune everytime your load your OS.

PCI Lock Bug:
EDIT: FIXED! See post #24 below
To lock the PCI lock on your board, all you have to do is make sure you have bios 1003 final or newer and then set your FSB in the bios to 201 or higher. Your lock will then be set. Another alternative since bios 1007 is to set the PCI to 33.3MHz in the frequency menu and NOT to CPU.

Bios Revisions:

Bios 1016.001 Beta
DOWNLOAD
Bios 1016 beta is now available for those of you who are adventurous enough to give it a shot.

Bios 1015 Final
DOWNLOAD
Bios 1015 final is now available and can be downloaded from ASUS's website. Fixes include,
1. Support new CPUs. Please refer to our website at: http://support.asus.com/cpusupport/cpusupport.aspx
2. Fix unstable under 400MHz-Single channel issue.
3. Modify DDR max speed from 400MHz to 333Hz when full loading.

Bios 1015-005 Beta
DOWNLOAD
So bios 1015-005 is now ready for download. I was hoping ASUS would release something soon since bios 1015-001 was giving me huge memory instability issues. It seems that ASUS is tampering with the memory settings of the board.
Long story short, with bios 1015-001, I've been getting nothing but instability with my overclock. It will be interesting to see if bios 1015-005 fixes this wreck!

UPDATE: Seems my overclock is back with bios 1015-005. Whatever bios 1015-001 did, it completely destroyed my overclock no matter which setting I tried due to something with the memory settings. Bios 1015-005 seems to be OK for download. If I experience more problems, I'll be posting here asap.

Bios 1014 Final
1014 DOWNLOAD | 1015-001 DOWNLOAD
This bios updates the CPU code to recognize the Athlon64 X2 5000+ / 5200+. This bios can also properly identify 64-bit Semprons to allow for 64-bits functionility. Tested and seems to be on par with 1013. Thankyou netlaw1 for contacting me about this release.

I also should mention that Bios 1015.001 beta has been realeased as well but I am not sure of the updates it possesses. If anyone has some experience with this bios, please feel free to contact me with your comments.

Bios 1013 Final
DOWNLOAD
Bios 1013 Final is out and is ready for download. A BIG thanks to Pyro for pointing that one out to me.
Bios 1013;
1. Support new CPUs. Please refer to our website at: http://support.asus.com/cpusupport/cpusupport.aspx
2. Update nVidia onboard Lan PXE ROM to V215.0503
3. Fixed system will hang in memory test 2.09GB if Quick Boot disabled.
4. Change class code of SATA to fit new Driver (WHQL) if RAID is enabled.
5. Fixed system cannot detect ARC 12xx Serial ATA RAID Host adapter.
6. Fixed K8 Vcore will set wrong value under AI-Booster.

I've finally had a chance to test out the 1013 bios and it seems to be a very well balanced bios. With the latest v6.66 nForce drivers, I'd have to now recommend this one to people who are thinking of updating there bios'es from earlier revisions.

Bios 1013.002 Beta
DOWNLOAD
After testing the bios for a few minutes, I can honestly say that the reboot and shutdown issue still isnt fixed. After checking over the entire bios from end to end I cannot see anything new to tinker with. People with X2 CPUs should download this bios to experiment because I have a feeling this is an X2 CPU compatibility improvement. The only new option to me that wasnt working before is the ability to adjust the CPU Vcore even after the mod. I can now set my vcore to whatever I want in the bios even with the mod installed. Vcore between 0.8~1.68v is now fully selectable.

Also now getting more feedback with the bios tampering with vcore threshhold on various CPUs. X2s seem to be taking the biggest hit loosing vcore from 1.55 down to 1.40v or less.

Bios 1012.003 Beta
DOWNLOAD
This bios still has all the issues the .002 previous had with appearently some memory tweaks yet again. However, if you havnt already downloaded this bios, I recommend you stay away from it. Far too many people have had booting and shutting down issues after upgrading to this beta. Back to 1010.001 I go untill further notice. The later 1013.002 also suffers from this strange bug so I've done some experiments and have determined it to be a temperature monitoring issue the bios reads before allowing a proper shutdown.

Bios 1012.002 Beta
DOWNLOAD
This bios is horrible! Do not use it. After installing this update on several boards, each computer had multiple problems rebooting and sometimes even had problems with memory settings in some cases.

Bios 1012.001 Beta
DOWNLOAD
Isnt really anything too noticable from 1011 with the 1012 bios except a few bios memory timings have been changed.

Bios 1011-001 Beta
DOWNLOAD
So far I've noticed that the option for the 1.565v~1.65v vcore adjustment has again been removed. It seems that the final release of this 1011 bios explains that all vcore above 1.55v was removed from the bios on CPUs that sport 1.4V and 1.35Vcore default voltages.

Bios 1010.001 Beta
The 1010 series welcomes back all the memory timings which were hidden in previous bios'es due to some incompatibility issues which are now fixed. Some stability as the 1009.002 beta so if you have the 1009 beta, its recommended you step up to the 1010 beta.

Bios 1009.002 Beta
DOWNLOAD
The 1009 series is here and it has some major memory tweaks and support for the Athlon64 FX 57 processor. I'm currently testing out this bios as we speak. UPDATE: It seems the bios is very stable aside from the whole PC Probe bug that has now been introduced :( . Its a solid beta to test out if your looking to venture behond 1007 final and try something new.

Bios 1008.001 Beta
DOWNLOAD
I'm currently testing it out. It seems to support the AMD Athlon X2 series with it now stating how many CPU cores are present on the AMD A8N-SLI Deluxe. The Pci Express features have also been updated now with the support of PCI Express Apecture Size Auto adjustment. Thanks Kektex for pointing this new bios out to me :D

Bios 1007 Final
DOWNLOAD
It's here and its official, go grab the 1007 bios asap. I've been examining it now for a few days and it seems to be just like the 1007-004 which was a real keeper with the slew of new features. After playing around with the Vcore, it seems the highest setting is actually when setting the vcore to 1.575v in the bios. CPU-Z now reports a nice 1.60v for those of you looking for more juice. Maybe in the near future, we may actually expect to see a full 1.65v or even higher depending on how much these boards can put out. Yo can download the bios directly from this link below,

1007SD-03 Beta
DOWNLOAD
Been testing this bios now for almost two weeks and it seems to be a real keeper for me. I've noticed that this bios gives us a very slight voltage increase at 1.55V (1.565V before and now 1.58V with 1007SD). Some options in the bios have been renamed to avoid confusion in some matters. Also, stability is back from 1004 in all areas so I no longer am faced with disk curruption like I was faced with the 1005 series.

Bios 1004 Final
It essentially is almost identical to the 1004.001 beta asside from more fan monitoring options and very minor adjustments. Some people however are reporting that they are loosing a small portion of their overclock from updating to this revision. However, after a few days testing, this bios seems to be on par with 1004.001 beta for me.

Overclocking:
This is a huge topic and this board is very complicated when it comes to overclocking. While everyone has their own styles in overclocking including the type, brand and quality of hardware choosen, I will give you a general idea on how exactly to configure your board.
Overclocking on the ASUS A8N-SLI is a little tricky especially if your migrating from a socket A platform. Unless otherwise stated, I recommend using 4XHTT multiplier when overclocking. Unless you plan on overclocking past 2.71GHz (3000+) or 3GHz (3200+) more than enough flexibility should be available with the AN8. Some valuable information, the ASUS A8N-SLI on average will cap out at maximum 2472HTT or 309X4 HTT in the bios regardless of setting choosen. This is more than enough HTT to work with in any case. When overclocking this board, I strongly recommend the latest beta 1004.001 bios for greatest stability and overclockability. This board overclocks the easiest with software bar none. The software I used is nForce4 ClockGen which can be found here,
http://www.cpuid.com/clockgen.php
Go get CPU-Z to find the real clock speed of your CPU and memory. Use Prime, SETI, Folding and 3Dmark01~05 to stress test your PC at any given time.
Ai Booster is to me not a very good tool at the moment since most of the time, it seems to lock my system up when I wanna change multipliers or change a huge FSB frequency at a time. Before I completely blow off Ai Booster, it still achieves one important thing,, being it allows your to adjust your ram voltages which can come in very handy sometimes. Ai Booster also has a huge bug being when you change a setting with it, it will lower your Vcore to under 1.5v or 1.35V in the bios which can be a real pain for most. Not to worry, go load up ClockGen and up the VID back to 1.55V to override this glitch without having to restart. Keep repeating this step untill you achieve your desirered frequency. Remember, with the current bios design, no overclock made at this time is ever final. Their is still many bugs to be worked out giving us alot more overclocking potential.

A Little 101 On Q-Fan
Ok, if you really wanna silent your rig while having 3, 4 or even 5 casefans within, hookup and turn on Q-Fan (Requires beta bios 1004.001). Q-Fan can dynamically lower the RPMs of all your fans in your machine cutting noise by more than 50%. Although Q-Fan at the moment has one downside being that the lowest temp we can set it at is 51c for the time being while I think is a little high but I guess ok. To really see the results of Q-Fan, go get a 4pin to 3pin converter and hookup all your case fans to the SF1 connecter located near the bottom right hand corner of the motherboard. By doing this, all your case fans will be driven by Q-Fan meaning all will be temp controlled and help cut noise by insane amounts. If your reading this for the first time, do not pass up this awesome feature, it really does the job. BTW, I was lazzy and just made my own homade moded 4pin to 3pin adapter to get the job done. I simply took a case fan, cut the cord in half and joined the red/black cords to a 3pin hookup from a mod device which I wasnt using. I soldered the two pieces together and then I was good to go.
Qfan is even more efficent when you hook up all your case fans to your CPU FAN HEADER. By doing this, your fans will all now have variable speed termperature control and will spin faster as your CPU temperature rises. Setting your temperature in the bios to 51c insures cool quiet operation while still allowing your case fans to perform when needed. Dont be worried about overloading the 3pin header fore it supports up to 71.5Watts before it reaches its limit. Considering the average LED 80mm fan uses less than 5 watts of energy and the average 120mm fan uses just over 10watts you'll be fine running six or so fans off that single header along with your CPU fan. Go get those 3pin Y splitters from your local shop and start Qfan today!

Volt Modding:
If you like living on the dangerous side, you can also do a Vdimm mod, Vcore and even a chipset mod to help you overclock even further. I will simply leave this section to VR-Zone,
http://www.vr-zone.com/?i=1624&s=1
Great article with lots of volt modding and pictures.

Tweaking The PCI-Express Bus:
So far to date, the best results I've found is setting the PCI Bus transfer rate from 100MHz to 120MHz (3030MHZ PCI Express). A gain of 200 points in 3Dmark03 can be noticed from tweaking this variable option.

When connecting a 4pin molex to your EZ-PLUG hookup, make sure you use your Extra 12v connecter if available to maximize your graphics cards stability.


Updated frequently so check back weekly..


OC-Master
 
Last edited:
OC-Master said:
The AN8 series does not lock the PCI bus while PATA hard disks are connected to the motherboard.
Wow! Excellent write-up! Question, though. Do you mean actual hard disk? Or anything PATA? My hard disks are SATA, but I've a PATA DVD drive in my system. Thanks!!
 
Aurora40 said:
Wow! Excellent write-up! Question, though. Do you mean actual hard disk? Or anything PATA? My hard disks are SATA, but I've a PATA DVD drive in my system. Thanks!!

As far as I know, the motherboard only has issues when PATA hard drives are connected. I have one SATA converter and when I use it, my system runs flawlessly with FSB of 283 and higher.

Hopefully this is a bios limitation and eventually will be fixed. I might add that when the hard disks begin to fail, you'll notice your DVD drive acting up as well by making noise and the little light coming on.


OC-Master
 
Last edited:
OC-Master said:
PCI Lock Bug:The AN8 series does not lock the PCI bus while PATA hard disks are connected to the motherboard. I must stress that things really get hectic as soon as you venture above 210MHz FSB.

Really? I was having problems too with 1002 (I was stuck at 214x11), but with 1003 theres a BIOS setting called PCI lock or something which appears to work for me...I'm at 220x11 and I have 2 IDE HDs. No problems so far (I'll run P95 overnight). For me, the problem was the RAM (look below for details).
 
I have 2x 80GB IDE drives setup in a RAID-1 array on the NVIDIA RAID controller (master/slave on secondary controller) and here are my current OC settings:

BIOS=1002 (my pci lock works as documented)
Actually just upgraded to 1003 last night to get the IDE RAID array working properly.
FSB = set to 250/2.2Ghz @ bootup, can max out @ 270/2.43Ghz
CPU Multipier = 9x
HTT = 3x
vCore = 1.50V
DIMM = 2.85V
Memory divider set at 333Mhz
CAS = 2.5
RAS to CAS = 3
Row Precharge = 3
Min RAS cycle = 6
So I guess that's 2.5-3-3-6

Have run benchmarks, game demos, prime95, superpi all without error. System runs about 32C with no load and gets to about 49C running prime95.

RAM also seems to be my weak link. Have to keep the volts up over 2.8V otherwise I can't run memtest86 (test 5 & 6 ) for more than 30 seconds.

Still need to play around with HTT, move it to 4x. Also plan to loosen the memory to something like 3-4-4-10 and see if I can run stable at my current max of 285ish FSB / 2.6Ghz cpu.

3DMark05 = 3654
3DMark03 = 9018

John


-------------------------
AMD 64 3000+ [email protected]
ASUS A8N-SLI Deluxe
2x512MB Corsair XM3200C2PT
Leadtek nVidia 6600GT Extreme
Enermax Noisetaker 495
 
Question, I currently am running 104.001 bios and watercooling my chipset. everytime I start my computer I get a Chipset fan error message. It says you can disable it in the bios but it is not highlighted so you can't select it. I know I could throw a small fan in there but I rather not. Any suggestions?
 
tinytool said:
Question, I currently am running 104.001 bios and watercooling my chipset. everytime I start my computer I get a Chipset fan error message. It says you can disable it in the bios but it is not highlighted so you can't select it. I know I could throw a small fan in there but I rather not. Any suggestions?

This pisses me off too, though my problem is with the CPU fan. I've got it connected to my fan controller. A simple workaround is to hookup another fan to that connecter. that should fix it.
 
I've only got the DVD drive hooked up to the PATA, though I am using the Silicone Image SATA connectors (which runs off the PCI bus). I just got 1004-001 and the PCI lock set to 33.33 still does nothing for me. It still moves with the HT bus.

I tried 210MHz (tried 217 for DDR433 and the comp crapped out) for the HT and a 4x mupltiplier, then ran 3DMark03 just to see. My score dropped from 8750 or so to like 5550... So now it's all back to stock. I think I'll just OC my graphics card and leave it at that. I hope the board implements stock SLI very well when I finally do decide to do that. Otherwise I doubt I'll ever buy another Asus motherboard. As it is I miss the ease of use of my Abits (who also include a useful description in the BIOS help for each item)...
 
Another question, I notice MBM5 does not support this board as of yet... Is there any alternative boards you can put down that will monitor it the same? I'm currently using speedfan but I like MBM 5 better.
 
Aurora40 said:
I've only got the DVD drive hooked up to the PATA, though I am using the Silicone Image SATA connectors (which runs off the PCI bus). I just got 1004-001 and the PCI lock set to 33.33 still does nothing for me. It still moves with the HT bus.

I tried 210MHz (tried 217 for DDR433 and the comp crapped out) for the HT and a 4x mupltiplier, then ran 3DMark03 just to see. My score dropped from 8750 or so to like 5550... So now it's all back to stock. I think I'll just OC my graphics card and leave it at that. I hope the board implements stock SLI very well when I finally do decide to do that. Otherwise I doubt I'll ever buy another Asus motherboard. As it is I miss the ease of use of my Abits (who also include a useful description in the BIOS help for each item)...

Did you try the 1003 BIOS?

Is there a prog that will tell you your PCI bus speed?

You can see my sig for my SLI benches...good, but not what I was hoping for. I thought SLI was supposed to get 10k+ in 05...and I heard of people getting my AM3 score with ONE gfx card. could be the drivers, could be the progs. The one good thing is that it'll only get better.
 
LoneWolf121188 said:
Did you try the 1003 BIOS?

Is there a prog that will tell you your PCI bus speed?
Yeah, I tried 1003, then got 1004-001. In both cases, setting it to "33.33" does nothing. You can launch the nTune to see if it stops telling you they are locked. But also what I used was ClockGen mentioned in the first post: http://www.cpuid.com/clockgen.php When I move the HT speed up, the PCI speed goes up with it... :-\ Maybe it's because I'm not using the nVidia SATA controller... I dunno. I'm not a big OC'er, but I think I ought to be able to squeeze out 10MHz or so without a problem. It's not like my big plan was to OC the crap out of my machine. But I find it annoying that the problem is a very expensive mobo, not just that my cpu can't handle the speed or heat or the memory can't handle the timings, etc. Those would be a lot more understandable. That said, if I had it to do over again right now, I'd still probably buy the same board. If the Abit SLI board were out already, though, I would buy it instead.
 
Aurora40 said:
Yeah, I tried 1003, then got 1004-001. In both cases, setting it to "33.33" does nothing. You can launch the nTune to see if it stops telling you they are locked. But also what I used was ClockGen mentioned in the first post: http://www.cpuid.com/clockgen.php When I move the HT speed up, the PCI speed goes up with it... :-\ Maybe it's because I'm not using the nVidia SATA controller... I dunno. I'm not a big OC'er, but I think I ought to be able to squeeze out 10MHz or so without a problem. It's not like my big plan was to OC the crap out of my machine. But I find it annoying that the problem is a very expensive mobo, not just that my cpu can't handle the speed or heat or the memory can't handle the timings, etc. Those would be a lot more understandable. That said, if I had it to do over again right now, I'd still probably buy the same board. If the Abit SLI board were out already, though, I would buy it instead.

You have to use the SATA controller if you want to overclock with your PATA hard drives installed in the machine. I recommend the sata converters to do the job niceless. Funny thing is, even when the PCI lock in the bios is turned on, ClockGen seems to show the PCI bus locked yet when you overclock, all the same stupid stuff happens,, hard drive begins to currupt, video cards begin to act up, and worst off, the on-board ethernet cards begin to fail.

After testing many different ways, all these problems go away as soon as I hook up my hard drive to the nVidia sata port. I presume the Silicon Image ports would work great too.


OC-Master
 
OC-Master said:
You have to use the SATA controller if you want to overclock with your PATA hard drives installed in the machine. I recommend the sata converters to do the job niceless. Funny thing is, even when the PCI lock in the bios is turned on, ClockGen seems to show the PCI bus locked yet when you overclock, all the same stupid stuff happens,, hard drive begins to currupt, video cards begin to act up, and worst off, the on-board ethernet cards begin to fail.

After testing many different ways, all these problems go away as soon as I hook up my hard drive to the nVidia sata port. I presume the Silicon Image ports would work great too.


OC-Master

I have SATA drives, not PATA. They are hooked to the Silicon Images controller, not the nVidia one. I couldn't get WinXP to install on the nVidia one with RAID0, so I used the Sil Im controller. Perhaps the Sil Im ports don't work great too, or perhaps my board is just weirder. But the lock doesn't lock anything on my comp... You can see my comp in my signature --v
 
Aurora40 said:
I have SATA drives, not PATA. They are hooked to the Silicon Images controller, not the nVidia one. I couldn't get WinXP to install on the nVidia one with RAID0, so I used the Sil Im controller. Perhaps the Sil Im ports don't work great too, or perhaps my board is just weirder. But the lock doesn't lock anything on my comp... You can see my comp in my signature --v

Aurora, this is off topic but I am very curious to your CPU voltage when set at 1.55v in the bios. Does your Vcore still flucuate like mine (bounces between 1.58 and 1.601V or does it stay still at 1.61Vcore?

EDIT: I guess I should ask, are you using the latest 1004.001 bios as well?

My PSU should be here today or Monday and I'm very curious to if I will see a difference in stability with my overclock.


OC-Master
 
Aurora40 said:
I've only got the DVD drive hooked up to the PATA, though I am using the Silicone Image SATA connectors (which runs off the PCI bus). I just got 1004-001 and the PCI lock set to 33.33 still does nothing for me. It still moves with the HT bus.

The only way to lock the PCI bus under 1004.01 is to leave the pci lock at AUTO. It then dosent move :beer:
 
thats interesting, because ClockGen shows my PCI bus at 33.34 at the speed I'm at.

I just thought of something: What Silicon revision and board revision are you guys using? I have board revision 1.02 (found under the 1st PCIE slot and on top of the SLI connecter thing) and silicon revision A03 (esay way to find this is with CPUZ: Under the mainboard tab, theres a box that says "Revision: A2" or "Revision: A3". This info can also be found under the chipset cooler).
 
OC-Master said:
Aurora, this is off topic but I am very curious to your CPU voltage when set at 1.55v in the bios. Does your Vcore still flucuate like mine (bounces between 1.58 and 1.601V or does it stay still at 1.61Vcore?

EDIT: I guess I should ask, are you using the latest 1004.001 bios as well?

My PSU should be here today or Monday and I'm very curious to if I will see a difference in stability with my overclock.


OC-Master

My vcore is set to 1.400 I believe (auto), and it flips between 1.392 and 1.408. It is always the same numbers, never any inbetween, so I suspect it may not actually fluctuate much at all, just the scale of the sensor isn't very fine-grained. For some reason it shows 1.550v in the nTune monitor, though.

Where are you looking to see the setting/actual?

Yes, I have 1004-001 bios.
 
stu_allen said:
The only way to lock the PCI bus under 1004.01 is to leave the pci lock at AUTO. It then dosent move :beer:
But the manual choices are 33.33 and "track CPU" or something. Wouldn't auto just pick between the two, and not pick the secret 3rd option "work properly"?
 
Dunno, but 33.33 dosent work on mine - ntune reports its still tracking. When i set it to auto it works!
 
hmm, maybe I lied. I left my clock at 223x11 and tried to get the memory at 1:1 with higher latency. It was working fine until I tried 2.5-4-4-6 or something, and that gave me an error like "system file x is corrupted". Now I have to reinstall windows...weird. Oh well, no big deal. At least nothing's broken...I hope.
 
Back