Asus A8N32-SLI initial impression....WOW!

[TheMooN]

Lets hope so MP

Yes , From reading around the forums I am pretty impressed so far with this mobo especially considering this is a rev 1 board .

I suppose one way around this little problem would cost me the price of another Sata Caddy ~ Running them both off the same channel ?

[Mickey Padge]

Yeah that would boot the OS OK, but you may need the same HDD too. It depends on how the BIOS picks up your caddy.

If it sees just the caddy in the BIOS and not the HDD firmware then you are OK, but if it sees the HDD I think the BOOT config would get mucked up if the HDD were different models.

I hope that makes sense :D

[TheMooN]

Yeah that would boot the OS OK, but you may need the same HDD too. It depends on how the BIOS picks up your caddy.

If it sees just the caddy in the BIOS and not the HDD firmware then you are OK, but if it sees the HDD I think the BOOT config would get mucked up if the HDD were different models.

I hope that makes sense :D

Yep. makes perfect sense m8 , Unfortunately :( ,

Thanks for your advice tho :)

[arrarro]

You gotta flash the Firmware of the drives to BANC1B70 or newer, and use the SATA drivers from version 6.39 forceware. A friend of mine has the same drive firmware but he can use newer drivers without a hitch, so it might just be a WinXP issue.


The problem is: HOW TO FIND THE UPDATED FIRMWARE for those MAXTOR HDD??? Can you send me that BANC1B70 firmware you are speaking of? is it for 300Gb 16mb cache Diamond Max10 drives? Thanx!

[smurfmods]

Is this a HD thread or a Motherboard thread? Did I click the wrong thing?? :hehe:

Just kidding... I'm actually going to try raid 0 SATA for the first time tonight. Still not so sure about it for data backup (movies, games, music, downloads ect) Worried about a HD fail on one drive could kill all my data. I will try it for WinXP though. My winxp install has always been on a seperate drive. I format and reinstall every 3 months because my Windows install gets that "Not So Fresh Feeling." :wth: No matter how carefull you are windows collects to much junk from downloading and installing it gets sluggish over time.

Is raid pretty easy to set up on this board? I don't have the CD or software that come with the board. Can I still setup raid?

[Zeus...]

Is raid pretty easy to set up on this board? I don't have the CD or software that come with the board. Can I still setup raid?

The raid driver is on the CD u dont have, But should be on ASUS site (will mail u a copy if u cant get it) the raid0 was easy to setup when installing XP, not sure how to do it myself know u have already installed XP but I would of thought it was pretty straight forward just format the raid on boot and windows should recognize it and promt u for the driver - I have had NO issue with the raid0 what so ever - So far:cool:

[Mickey Padge]

I remember that there were problems with programs like Ghost and creating backups with Nvraid. Ghost had problems with thre nvidia drivers.

I copied this information from another forum and can confirm it DOES allow you to create backups when using Nvraid.

In order to use Ghost in a NVRAID volume, the user must:
1. Disable the Ghost Direct Disk Access
2. Force it to rely on Extended INT13 to access the disk

Ghost Version 7.5
To set these options for Ghost Version 7.5, do the following:
1. Start Ghost from the DOS prompt. (Not the Windows Command
2. Prompt session)
3. Select the “Options” (ALT+O) menu
4. Scroll to the “HDD Access” tab
5. Select the “Use Extended Interrupt 13h disk access” (ALT+E)
6. Select the “Disable direct IDE access support” (ALT+B)
7. Select the “Disable direct ASPI/SCSI access support” (ALT+B)
8. Press (ALT+A) to activate the “Accept” button to use the new settings
9. Proceed to run Ghost as normal

Ghost Version 8.0
To set the options for Ghost Version 8.0, do the following:
1. Use the command line option to disable direct disk access and enable INT13
2. In DOS prompt, start Ghost with this command line: ghost.exe -ffx -fni
Note:
ffx - Prefers the use of Extended Interrupt 13h disk access for hard disk
operations.
fni - Disable direct IDE access support for IDE hard disk operations.

Sorry if this is common knowledge already, but it may prove helpful to somebody. Also Ghost version 10 may have fixed the problem now.

Cheers :D

[smurfmods]

I remember that there were problems with programs like Ghost and creating backups with Nvraid. Ghost had problems with thre nvidia drivers.

I copied this information from another forum and can confirm it DOES allow you to create backups when using Nvraid.

In order to use Ghost in a NVRAID volume, the user must:
1. Disable the Ghost Direct Disk Access
2. Force it to rely on Extended INT13 to access the disk

Ghost Version 7.5
To set these options for Ghost Version 7.5, do the following:
1. Start Ghost from the DOS prompt. (Not the Windows Command
2. Prompt session)
3. Select the “Options” (ALT+O) menu
4. Scroll to the “HDD Access” tab
5. Select the “Use Extended Interrupt 13h disk access” (ALT+E)
6. Select the “Disable direct IDE access support” (ALT+B)
7. Select the “Disable direct ASPI/SCSI access support” (ALT+B)
8. Press (ALT+A) to activate the “Accept” button to use the new settings
9. Proceed to run Ghost as normal

Ghost Version 8.0
To set the options for Ghost Version 8.0, do the following:
1. Use the command line option to disable direct disk access and enable INT13
2. In DOS prompt, start Ghost with this command line: ghost.exe -ffx -fni
Note:
ffx - Prefers the use of Extended Interrupt 13h disk access for hard disk
operations.
fni - Disable direct IDE access support for IDE hard disk operations.

Sorry if this is common knowledge already, but it may prove helpful to somebody. Also Ghost version 10 may have fixed the problem now.

Cheers :D

May be common knowledge to some but it's news to me! I have wanted to use a ghost image to set things up. It sucks having to reinstall everything everytime. I would love to just set it up and have a good starting point everytime. It would save a great deal of time.

Is this still a problem? Have newer nvidia drivers fixed this or is it Ghosts problem? I may end up using a different program for more simplicity (if there is one.) I would hate to be 3 to 6 months down the road trying to remember "
What did Mickey Padge say?? Was it ghost.exe +ffy +fnq or something else?" ;)

Great advice though, thanks! Any other recommendations for ghost type programs? Man.. this thread is getting a bit off topic. Sorry if i'm boring anybody. :yawn2:

I do have a more MOBO related question though. What temps are you guys getting for the MOBO temps. I'm not running a fan on the fins of the heatpipe and I'm getting 115 to 120 F (46C) Is that to hot? What kind of a difference can I expect with a fan?

[Mickey Padge]

My temps are at arround 40c with that little fan added to the heatsink. Although I don't think the temp readings are that accurate because everything FEELS just warm to the touch after many hours of use.

As for using Ghost 8 your right, you just have to add the -ffx -fni to the floppy disk config boot command, easy really ;)

[Xyriu]

I can't belive...

I set in the BIOS 250x10, RAM 1:1 etc.
The system pass all tests, SPI32MB, 3DMarks, Prime, Games... etc. Really RS!

WHEN ...I reboot, the system won't boot!!!!!!!! I need to shut down it or reset!!

If I set 240x10...he reboot perfectly every time...


Why? :confused: :mad:

[stone_cold_Jimi]

Just want to round up my Z503 Ballistix info in this mobo in one spot, might be useful to someone else:

From Ben805:

"the TREF only has 3 options so I didn't have much to try there, from experience TREF play a major role in overclocking these Ballistix so it would be helpful if I could get more values to play with. I'd also noticed if the timing and overclock speed was instable then the BIOS would either automatically relax the TREF, TWR, MAL and READ Preamble, or it would not work at all."

" set 11ns for Async Latency, 9.5ns for Read Preamble, and 24clk for TRFC,TRC in Auto, once you dial in those values then leave them for now, there is only three settings for TREF so play with it one at a time and see what happen."

Also, use 1T always.

OC on a Venice 3000+ E6, sweet at this frequency, note async latency and read preamble which need Ben805's settings to OC the CPU more, also Tref:

http://www.0wned.org.uk/pics/shots/venice2g4.jpg

Don't know if this has any great meaning. Data rates are 6500MB/s according to Sandra:

http://www.0wned.org.uk/pics/shots/everest_lat.jpg

It's a nice box to use set up like this. CoD 2 runs very nice indeed.

[tommrussell]

smurfmod try acronis true image 9.0 it works better than ghost tried em both and been using acronis for a couple of years now and its been very good to me oh ya i also have this board and set up in raid 0 not a problem with setup but did have a prob with a couple of prog that broke the raid other than that its been rock solid and no temp issues
tom

[bobov]

I do have a more MOBO related question though. What temps are you guys getting for the MOBO temps. I'm not running a fan on the fins of the heatpipe and I'm getting 115 to 120 F (46C) Is that to hot? What kind of a difference can I expect with a fan?

My MB temp is 30C underload without the fan on the fins of heatpipe. It's in an Antec p180 case.

[smurfmods]

My MB temp is 30C underload without the fan on the fins of heatpipe. It's in an Antec p180 case.

You case must sound like a wind tunnel to have a temp that low without a fan directly on the heatpipe.

[qdemn7]

This is the smoothest board I've ever own. I've got my 3000+ Winchester running @ 260 x 9 on stock voltage. Idles at 37C, full load 48C using an XP-120. The board actually runs about 4C hotter than the CPU. Only glitch I've noticed is my -12v rail is showing -1.65v, and my +12 is showing +11.65v. This is with an Anetc Neopower. I know Antecs sometimes have flaky 12v probems, so I pulled out my 'meter. The -12v rail is spot on, and the +12v is showing +11.75. Weird. :confused:

[bobov]

You case must sound like a wind tunnel to have a temp that low without a fan directly on the heatpipe.

No, the two 120mm yate loon case fan running ~1150rpm, and panaflo 120M1BX running ~ 1450 on SI-120. No near sound like a wind tunnel. My opteron 148 running at 3GHz 1.45v prime large fft 24h, cpu temp ~45-46C, MB temp ~ 30-31C, room temp @ 18-19C.

I did get 38-40C MB temp when I first use it. I remember that after I cleared CMOS and load BIOS default some days ago, I get 30-31C mb temp. I also drop the vdimm from 2.75v to 2.60v. Maybe that helps the mb temp. :)

[grizzlybear]

ello all

at default install ai booster shows
cpu frequency 200
multiplier x10
fsb 800

tweak the cpu frequency and the fsb will raise at x4 till you reach 220
cpu frequency 210
multiplier x10
fsb 840

cpu frequency 220
multiplier x10
fsb 880

but at anything over 220 (fid set to auto will drop the multi to x9) report the fsb at 1:1
cpu freqency 221
multi x9 or x10 (set) k8-nb-f 800 or 1000 (set or auto)
fsb 221

see aibooster pic a few post back

i thought the fsb multi was the chipset k8-northbridge frequency
1000=x5
800=x4
600=x3
400=x2

tried @default settings
sb-nb overclock 210 = no change to fsb
k8-northbridge frequency 600 = no change to fsb (individually and combined)
nb-sb frequency 600 = no change to fsb (individually and combined)

any ideas ???????
is it just ai booster not reporting it correctly ??
can anything else confirm the hypertransport speed ????

ram divider (memlock) is set to auto and does drop the memory mhz as cpu frequency is increased
cpu frequency 200x10 mem 200mhz
cpu frequency 295x9 mem 133mhz

my spec q4 stable ;)
cpu frequency 295 (if 800=x4 then my htt is 1180 )
multiplier x9
k8-northbridge frequency 800 (1000 will make it crash )
ai booser reports fsb at 295
3dmark05 5799

a64 3200 venice @2654
twinx2048c2pt 2-3-3-7 t1
xfx 6800gs
bios 703
stock h&f & volts

default settings wall @ 240 (ish)
dropped multi to x9 wall @ 270 (ish)
dropped k8-northbridge frequency to 800 wall @300 (ish)
will boot into windows at 380 multi x6 (not tested tho)

[smurfmods]

This is the smoothest board I've ever own. I've got my 3000+ Winchester running @ 260 x 9 on stock voltage. Idles at 37C, full load 48C using an XP-120. The board actually runs about 4C hotter than the CPU. Only glitch I've noticed is my -12v rail is showing -1.65v, and my +12 is showing +11.65v. This is with an Anetc Neopower. I know Antecs sometimes have flaky 12v probems, so I pulled out my 'meter. The -12v rail is spot on, and the +12v is showing +11.75. Weird. :confused:


I've had very few problems with this board also. Very smooth like you said. No random reboots. Extremely stable. Only thing is I don't like my MOBO temps. Will have to find a little fan to put on there. I don't have any of the stuff that come with the board.

[Titus]

Greetings

I am in the midst of building a new system with the A8N32 board...does the optional fan that came with the motherboard need to be installed if not using a water-cooling setup?...and yes I know the manual states that you should not use the fan unless you are using passive cooling BUT I heard it makes a difference in temperatures by attaching it to the copper heatsink even when using air-cooling?...I will have great airflow withing my case so I was wondering if using it would be worth it or not...

what's this about problems with Maxtor 16 MB cache hard drives??...I bought a Maxtor DiamondMax 10 300GB SATA 150 16 MB cache/with NCQ (Model #6B300S0) drive?...what problems have there been with this hard drive/mobo combination and where can I find the latest firmware?

Thanks

[Yegg]

Just ordered the following

ASUS A8N32-SLI Deluxe
AMD Opteron 165 (god knows when these will be in stock again over here)
G.Skill 2GBZX
TT Big Typhoon

Hope I'll be pleased with the money I spent :/

[smurfmods]

Just ordered the following

ASUS A8N32-SLI Deluxe
AMD Opteron 165 (god knows when these will be in stock again over here)
G.Skill 2GBZX
TT Big Typhoon

Hope I'll be pleased with the money I spent :/

Should be happy. Even happier if you are overclocking. Youd be crazy :hitself: not to. What will you use for sound? If you can afford it the X-FI is pretty slick!

[Yegg]

Should be happy. Even happier if you are overclocking. Youd be crazy :hitself: not to. What will you use for sound? If you can afford it the X-FI is pretty slick!
Of course I'll be overclocking! I just hope the memory will work ok.

I'm planning on getting an X-Fi at a later date, can't really afford one at the moment, I'm $1000 over my budget for the computer :thumbsup:

Does anyone know if this motherboard and the Big Typhoon will fit in an Antec Sonata II case?

[ben805]

Greetings

I am in the midst of building a new system with the A8N32 board...does the optional fan that came with the motherboard need to be installed if not using a water-cooling setup?...and yes I know the manual states that you should not use the fan unless you are using passive cooling BUT I heard it makes a difference in temperatures by attaching it to the copper heatsink even when using air-cooling?...I will have great airflow withing my case so I was wondering if using it would be worth it or not...


Thanks

Asus do not recommend the optional heatpipe fan unless you're on WC or Phase change, I do not think you can fit it on if you're using big cpu heatsink like XP120 or Big Typhoon anyway. As long as your case has sufficient air-flow you shouldn't need it ;)

[qdemn7]

Asus do not recommend the optional heatpipe fan unless you're on WC or Phase change, I do not think you can fit it on if you're using big cpu heatsink like XP120 or Big Typhoon anyway. As long as your case has sufficient air-flow you shouldn't need it ;)On my board the backside of the XP-120 overhangs the heatsink next to the I/O ports.

[mion]

What bios is good for that board and dual core.

[Titus]

Asus do not recommend the optional heatpipe fan unless you're on WC or Phase change, I do not think you can fit it on if you're using big cpu heatsink like XP120 or Big Typhoon anyway. As long as your case has sufficient air-flow you shouldn't need it ;)

thanks for the reply!!

actually the Zalman CNPS 9500 does fit alongside the optional ASUS fan fine but I'm still not going to install the extra fan...I was tempted to install it after hearing about it supposedly lowering CPU/mobo temps...

here are a few pics of the Zalman cooler + ASUS fan...

*the pics are for reference only and not from my system

http://img246.imageshack.us/img246/442/fan13ti.jpg

http://img246.imageshack.us/img246/1129/fan29xo.jpg

[ben805]

Well...if you can get it to fit, then maybe it's worth leaving it on? have you noticed any improvement with the optional fan? check the temperature with and without it....

[situman]

where's Craig with his phase results? He received the board i sold him yesterday. He said he was putting on nail polish and stuff. Wonder how it went. I hope everything is alright.

[safan80]

On my board the backside of the XP-120 overhangs the heatsink next to the I/O ports.

can you take a picture of the xp-120 on your board by any chance?

[HKPolice]

This board is crap. Warm boot issues @ 250Mhz with GSkill ZX sticks, Infineon CE-6 chips. Tried both 703 and 805 bios.

Even with 166 Divider and auto memory timings. I don't know what all the hype is about, this board has serious issues.

[craig588]

Chilly1 has my cascade, so there arn't many results from me. Right now I'w working on maxing out the memory, I'm probably going to do a 3.3V == Vdimm mod, 3.2V isn't nearly enough. I hope the VTT tracks.

There's only one VTT regulator, but it's not the same one used on the DFI so it might be fine. I need to solder some wires to various places to begin modding this mobo.

My board has a 10/27 bios on it, I think I should dump it.

[Blue078]

This board is crap. Warm boot issues @ 250Mhz with GSkill ZX sticks, Infineon CE-6 chips. Tried both 703 and 805 bios.

Even with 166 Divider and auto memory timings. I don't know what all the hype is about, this board has serious issues.

Did you double check to make sure EEC is disabled? That was the source of the warm boot issues I had. The board defaults to EEC enabled, after every optimized default reset you have to go and disable it.

[HKPolice]

Did you double check to make sure EEC is disabled? That was the source of the warm boot issues I had. The board defaults to EEC enabled, after every optimized default reset you have to go and disable it.

Yes, I made sure it was disabled everytime, and that the APM page was not empty.

On a cold boot, it'll run 250x11 no probs, but warm boots = no POST. After that happens, it won't even post after I save changes in BIOS, I have to reset setup defaults before it posts again, and sometimes even that won't work and I'll have to reset CMOS.

This BIOS is pure CRAP.

[qdemn7]

can you take a picture of the xp-120 on your board by any chance?Here you go. http://img526.imageshack.us/img526/1779/img0306large8mt.th.jpg (http://img526.imageshack.us/my.php?image=img0306large8mt.jpg)

[arrarro]

Yes, I made sure it was disabled everytime, and that the APM page was not empty.

On a cold boot, it'll run 250x11 no probs, but warm boots = no POST. After that happens, it won't even post after I save changes in BIOS, I have to reset setup defaults before it posts again, and sometimes even that won't work and I'll have to reset CMOS.

This BIOS is pure CRAP.

you know what? I just ordered an ASUS after having this problem with my DFI SLI-DR and I tried avery single damned bios out there, official, beta or modded. It seems bios corrupts and needs to be cleared...

[Blue078]

Yes, I made sure it was disabled everytime, and that the APM page was not empty.

On a cold boot, it'll run 250x11 no probs, but warm boots = no POST. After that happens, it won't even post after I save changes in BIOS, I have to reset setup defaults before it posts again, and sometimes even that won't work and I'll have to reset CMOS.

This BIOS is pure CRAP.

Oh well I had hoped that might help you too. I wont argue with you on the BIOS :) Asus needs to get on the stick and get a 1000 Bios out with tons of fixes.

[jinu117]

Well... got optie 175 and dual priming at stock volt (hmmm 1.42 is stock volt on this? interesting) 2.6ghz so far on stock air... trying to find unstable point so I can burn in this cpu properly. Only problem I have right now is my pair of ballistix is not even stable @ 200mhz... Thing is, this is the stick that was used in DFI that died and took my X2 3800 with it.... so could be joepardized... had OCZ plat 2gb set I was going to RMA as it didn't work with divider at all on DFI which works fine on this board for some strange reason... so I think I can use this for my backup for time being. But before I send ballistix for RMA I want to double check... what did you use on mobo settings for ballistix Ben? I do suspect the memory has been degraded a lot, but want to give it final go before I consider... RIP.

[bobov]

Yes, I made sure it was disabled everytime, and that the APM page was not empty.

On a cold boot, it'll run 250x11 no probs, but warm boots = no POST. After that happens, it won't even post after I save changes in BIOS, I have to reset setup defaults before it posts again, and sometimes even that won't work and I'll have to reset CMOS.

This BIOS is pure CRAP.

No warm boot issue here. Just done a test 255X11=2805MHz 1.35V, mem 1:1, no problem at all. My memory is Mushkin 2X1GB redline xp-4000, ce-6 chip too. BIOS revision 0703.

[HKPolice]

No warm boot issue here. Just done a test 255X11=2805MHz 1.35V, mem 1:1, no problem at all. My memory is Mushkin 2X1GB redline xp-4000, ce-6 chip too. BIOS revision 0703.

The mushins use a different model of BP PCB, and you're on a single core chip. What settings are you using for memory and chipset, etc etc?

[Shootingblanks]

my results so far....

http://img.photobucket.com/albums/v226/shootingblanks/SP32-20051208-212436.jpg

http://img.photobucket.com/albums/v226/shootingblanks/SP32-20051208-213557.jpg

Vcore is currently set at 1.3875 and Asus Probe is reporting it at 1.42 and CPU-Z is reporting it at 1.375. Had to up the Vcore so it would pass Prime 95. I know neither are really accurate, but my temps seem fine. At idle my processor and mobo are at 39 degrees celcius. Isnt that a little warm for a motherboard?

[situman]

I think the Asus reads a little high. Stock vcore at 2.7ghz with my 170 reads close to 50c with watercooling. Radiator is cold to the touch.

[ben805]

Yes, I made sure it was disabled everytime, and that the APM page was not empty.

On a cold boot, it'll run 250x11 no probs, but warm boots = no POST. After that happens, it won't even post after I save changes in BIOS, I have to reset setup defaults before it posts again, and sometimes even that won't work and I'll have to reset CMOS.

This BIOS is pure CRAP.

It sounds like the memory timing still need some tweak to be done, leave the data strength alone but play with the TREF, MAL and Preamble, use the 703 official bios and take your time to try different combination of things....have you try different ram and see what happen?


Well... got optie 175 and dual priming at stock volt (hmmm 1.42 is stock volt on this? interesting) 2.6ghz so far on stock air... trying to find unstable point so I can burn in this cpu properly. Only problem I have right now is my pair of ballistix is not even stable @ 200mhz... Thing is, this is the stick that was used in DFI that died and took my X2 3800 with it.... so could be joepardized... had OCZ plat 2gb set I was going to RMA as it didn't work with divider at all on DFI which works fine on this board for some strange reason... so I think I can use this for my backup for time being. But before I send ballistix for RMA I want to double check... what did you use on mobo settings for ballistix Ben? I do suspect the memory has been degraded a lot, but want to give it final go before I consider... RIP.

It's very possible that your Ballistix is dying....I had to RMA both pairs and sold all of them as soon as I get the new one back from newegg, try this:

use Bios 703, 250 x10, set vdimm to 2.8v

Limit
200
Manual
3
8
4
4
3
8
18
3
manual
3.9, also try 7.8
1
8
4
256
Disabled
Normal
Disabled
3clk
Disabled
Disabled
Manual
10ns
7ns
Auto
4 beats
Disabled


run superpi 1m, if it passed, shoot for 32m, if it passed.....then you can start tighten up your timing one at a time and slowly increase your memclock and see what happen. if spi failed....use 2T, if spi 1m/32m passed....your Ballistix still has some life left but won't last too long, I'd say couple weeks top. If it failed under 2T....they're hopeless. If I were you I'd still RMA them either way :lol:

[AppaYipYip]

Just installed mine tonight with Corsair 3500PRO LL DDR. I'm totally blown away by the BIOS and so far it's working flawlessly. Running an Opteron 175 at 2.5GHz stable as can be. My A8N SLi Premium that I upgraded from is being RMA'd tomorrow.

Pics of my install:

http://homepage.mac.com/jtrussell7/hobbies/PhotoAlbum22.html

[craig588]

I've switched my R404 cooler to R134, but it's still too cold. I can only reliably run if I boot up the system at least -5C and very quickly put a load on it. I'm feeding it 1.7V just for the extra heat because the mobo won't let me drop the drive stength low enough to compensate.

What are you guys doing to get over the cold bug on this mobo? This is with a Sandy too, not a Venice. I really don't want to put my 4000+ back in, it has a very crappy memory contoller and doesn't even clock much past 3.1GHz on R404. I've seen the Sandy do 3.3 to 3.4GHz on R134 and it has handled 2GB of memory no problem too. The Claw has a hard time with 1GB.

[phadeout]

This board is crap. Warm boot issues @ 250Mhz with GSkill ZX sticks, Infineon CE-6 chips. Tried both 703 and 805 bios.

Even with 166 Divider and auto memory timings. I don't know what all the hype is about, this board has serious issues.


I have no problems with the Mushkin CE-6 [already reported above].

I was running 260HTT @ 3328 1T and 2.8V... everything else auto.

testing a new chip so... gotta play more!

[jasonh1234]

Hi guys. Just setting up my system right now and I have a question that doesn't seem clear enough in the manual.

Do you have to connect the EZ_Plug if you're using 2 GPUs?
Or is it an optional alternative to plugging in a PCI-E ATX cable into each card?
Or do you need both the individual PCI-E ATX cables and the EZ_Plug connected?

I'm using an Enermax Liberty PSU and x2 EVGA 7800 GTX KO's

[Psyche911]

You should plug it in and the 2xPCI-E connectors. Especially with those GPUs....maybe not with like 6600 or 6800GS.

[Ic3man]

Finally tried the 805 bios hoping for better clocks using the alt+f2 method and.....


it sucks. Wouldn't even post using same settings so dropped them back to 2.5...no post....2.4...no post...2.3 posts but wouldn't boot windows.
Went back to 703 revision and all is well again. 2.65 joy :)
The 'update' to 805 bios went very smooth using ezflash but the 'downgrade' to 703 gave me a bit of a fright as after it flashed the pc hung:eek: :eek: :eek:

Had to unplug the psu for 30seconds before it would boot again...oh the relief:D
I'll wait now until Asus release an 'official' bios before upgrading again...someone else can be asus's guinea pig tbfh.

[Ic3man]

For those who are having problems using Asus Update (failed to load NT service..please reboot or BF.dll error) then the problem stems from using the XFI soundcard driver. I have uninstalled the driver but it has made no difference (must some registry entry that isn't reverted). The only solution is to re-install windows and install ASUS update and update the bios BEFORE you install the XFI drivers as it wont work after you have installed them. I will just have to use EZflash or afudos until a fix is released by either asus or creative:(

[jasonh1234]

You should plug it in and the 2xPCI-E connectors. Especially with those GPUs....maybe not with like 6600 or 6800GS.Yeah I'd just really like to avoid it if I can. My cables are messy enough without having to add another modular cable with 4 connectors on it, for just one lone plug.

[bobov]

The mushins use a different model of BP PCB, and you're on a single core chip. What settings are you using for memory and chipset, etc etc?

I'm using the memory setting suggested by charlie in this thread http://www.xtremesystems.org/forums/showthread.php?t=76672&page=7

Mem: 1:1, 255Mhz, 2.6V 3-3-2-8-7-15

[HKPolice]

I think I might of figured out the problem. Believe it or not, it's caused by a wireless PS2 keyboard. I was using the Microsoft natural Multimedia wireless keyboard, and noticed that it'll reboot memtest86 if I press ANY key. But if I pressed "C" when the floppy was loading memtest, I was able to access the menu and it'll only reboot if I pressed Esc, like any other normal keyboard. At first I thought it was a bug with Memtest86 (I'm using v1.65 which is the newest) but I remember this happening on a DFI NF3 setup, using the BIOS's built in memtest86 as well.

So I swapped it out with a normal wired PS2 keyboard and memtest86 works normally now and the warm reboot hang problem SEEMS to be gone..... Still testing.

[smurfmods]

I think I might of figured out the problem. Believe it or not, it's caused by a wireless PS2 keyboard. I was using the Microsoft natural Multimedia wireless keyboard, and noticed that it'll reboot memtest86 if I press ANY key. But if I pressed "C" when the floppy was loading memtest, I was able to access the menu and it'll only reboot if I pressed Esc, like any other normal keyboard. At first I thought it was a bug with Memtest86 (I'm using v1.65 which is the newest) but I remember this happening on a DFI NF3 setup, using the BIOS's built in memtest86 as well.

So I swapped it out with a normal wired PS2 keyboard and memtest86 works normally now and the warm reboot hang problem SEEMS to be gone..... Still testing.


Where is the ANY key? I can't find that one on my logitech keyboard.

:rolleyes:

[AppaYipYip]

Can anyone tell me why Ai Booster reports my FSB as my CPU Frequency when I overclock, yet when I set everything to Auto is reports it properly as 800?

How can I be sure my FSB is running at 1,000 or close to it?

[craig588]

I pushed my CPU speed up to 2.9GHz and now it seems stable. Those extra 700MHz added enough heat, I think. I was trying to get it running fine stock before.

I really like the idea that my system was too slow to be stable.

[Zeus...]

Can anyone tell me why Ai Booster reports my FSB as my CPU Frequency when I overclock, yet when I set everything to Auto is reports it properly as 800?

How can I be sure my FSB is running at 1,000 or close to it?

that Ai Booster sucks mine does the same here - I use 'everest home free ed' to check some off the settings (Motherboard-motherboard gives the FSB & HT) but cant get any CPU/HD temps with everest ATM only GPU temps

[craig588]

It just froze up after idling for a while, much better than it used to do, but still a problem.

[AppaYipYip]

I am really disappointed that nTune Monitor is not supported, I really like that monitor. I could see my frequencies and temps all at a glance. Is there anything around that might work as well?

[craig588]

Now I've unplugged the fans from my condenser. Once again it seems to be working.

Surely someone else has had problems with the cold bug on this mobo, no?

[smurfmods]

I am really disappointed that nTune Monitor is not supported, I really like that monitor. I could see my frequencies and temps all at a glance. Is there anything around that might work as well?

Ntune and a whole host of other utilities are not working yet. Even asus programs fall flat. This board will be pretty popular I imagine it won't take long for most of them to get fixed.

The next time ntune is updated it should include support for the newer X16 sli chipset.

[situman]

Now I've unplugged the fans from my condenser. Once again it seems to be working.

Surely someone else has had problems with the cold bug on this mobo, no?

I think Ben was able to get his working fine with some of his adjustments. Not sure what he did though.

[Eluzion]

It just froze up after idling for a while, much better than it used to do, but still a problem.

Mine did the same thing the other day. Three times actually. Wasn't doing anything really and the computer just froze. I raised my VCore and it seemed to have fixed it. I'm sitting at 250x10.

Guess I could throw my two cents in this thread. ;) I finally got the system put together the other night and it's been running great! Updated to BIOS 805 using EZFlash (Alt-F2 method). System specs are as follows:

AMD 64 X2 3800+ (250x10, 2.5ghz)
Zalman CNPS9500 w/ AS Ceramique
Asus A8N32-SLI Deluxe
mushkin eXtreme Performance 2GB (2x1GB) DDR500
XFX GeForce 7800GTX
Creative SoundBlaster X-Fi XtremeMusic
Samsung SpinPoint P Series 160Gb SATA 3.0Gb/s
OCZ ModStream 450W PS
Cooler Master Centurion 5 Case
Samsung 940b 19" 8ms LCD Monitor

VCore - 1.39v
CPU - 36c (idle), 45c (load, Prime95)
MB - 32c (idle), 36c (load)

Quick question for anyone who might know - In the device manager I have 2 "other devices". Any idea what they are/do and what drivers I need for them (or how to disable them if I don't need them I guess). They are PCI Memory Controller and SM Bus Controller. Thanks!

[situman]

Install the Nvidia chipset driver, it will go away.

[phadeout]

Quick question for anyone who might know - In the device manager I have 2 "other devices". Any idea what they are/do and what drivers I need for them (or how to disable them if I don't need them I guess). They are PCI Memory Controller and SM Bus Controller. Thanks!

IIRC, it is the nvidia chipset drivers and/or ata drivers.

[Eluzion]

Install the Nvidia chipset driver, it will go away.
Yeah, I installed them but remember unchecking the SM Bus Controller. I just installed that one and it went away. There was nothing for PCI Memory Controller though *shrug*, so it's still there.

[phadeout]

Yeah, I installed them but remember unchecking the SM Bus Controller. I just installed that one and it went away. There was nothing for PCI Memory Controller though *shrug*, so it's still there.

Maybe this is the AMD 1.2.2.2 drivers you need?

[arctic-k20]

you need the smbus drivers - its the ide drivers that cause issues....

everything else is fine - i install everything MINUS the ide drivers and i say no to the nvidia firewall ;)

[Opuntia]

Somebody gave me a 6800gs and I was wondering if that with this MB and the 6800gs installed, would all 3 PCI slots be available? I was looking a somebody's photos and it looks like that with a single-slot PCIe video-card in the "blue" slot, there's very little room for the adjacent PCI slot.

I don't ever plan on going SLI, but I like this board because of its stability, dual LAN and passive cooling.

Thanks in advance for your answers.

[jasonh1234]

Hi guys. Just setting up my system right now and I have a question that doesn't seem clear enough in the manual.

Do you have to connect the EZ_Plug if you're using 2 GPUs?
Or is it an optional alternative to plugging in a PCI-E ATX cable into each card?
Or do you need both the individual PCI-E ATX cables and the EZ_Plug connected?

I'm using an Enermax Liberty PSU and x2 EVGA 7800 GTX KO's

You should plug it in and the 2xPCI-E connectors. Especially with those GPUs....maybe not with like 6600 or 6800GS.

Can anyone else confirm this is correct?

[craig588]

Always plug in as many connectors as possible. Also, throw out that modular PSU, it's not as good as one with direct lines.

[craig588]

Mine did the same thing the other day. Three times actually. Wasn't doing anything really and the computer just froze. I raised my VCore and it seemed to have fixed it. I'm sitting at 250x10.



Well, I already have 1.7V on it and it can handle 100% loads like from cpuburn for hours without any problems.

[stone_cold_Jimi]

W Only problem I have right now is my pair of ballistix is not even stable @ 200mhz... ... what did you use on mobo settings for ballistix Ben? I do suspect the memory has been degraded a lot, but want to give it final go before I consider... RIP.

You should be able to do this:

http://www.xtremesystems.org/forums/showpost.php?p=1172630&postcount=512

which isn't pushing it. At 200MHz, it should run 2-2-2-5 1T and maybe do that up to 210 (mine's about 207). If it doesn't, it's shot imo.

[smurfmods]

Can anyone else confirm this is correct?

You guys talking about the molex plug on the motherboard? I have sli 7800 gts overclocked and I don't use it.

I do get a prompt saying I should plug it in on boot... but it worked ok without it. Runs 24/7 and plays battlefield 2 for hours on end.

I will most likely plug it in just to get rid of the prompt.

[Eluzion]

you need the smbus drivers - its the ide drivers that cause issues....

everything else is fine - i install everything MINUS the ide drivers and i say no to the nvidia firewall ;)

I installed everything except the IDE drivers as well (heard they're not any better than the MS ones and work the same).

I had two Other Devices listed, SM Bus Controller and PCI Memory Controller. SM Bus Controller is gone but PCI Memory Controller is still there. *shrug*

[ben805]

Now I've unplugged the fans from my condenser. Once again it seems to be working.

Surely someone else has had problems with the cold bug on this mobo, no?


Craig....for phase change make sure you use only BIOS 703, none of the beta are stable enough under subzero IMO, I've noticed my memory controller react slightly different and as mentioned by Tony(aka BigToe) you'll need to use weaker drive strength on phase change....not much to play with but you can try to set from Normal to Weak, and up your vdimm one more notch and see what happen, also try to enable your chipsets and LDT overvolt.

[jinu117]

It sounds like the memory timing still need some tweak to be done, leave the data strength alone but play with the TREF, MAL and Preamble, use the 703 official bios and take your time to try different combination of things....have you try different ram and see what happen?




It's very possible that your Ballistix is dying....I had to RMA both pairs and sold all of them as soon as I get the new one back from newegg, try this:

use Bios 703, 250 x10, set vdimm to 2.8v

Limit
200
Manual
3
8
4
4
3
8
18
3
manual
3.9, also try 7.8
1
8
4
256
Disabled
Normal
Disabled
3clk
Disabled
Disabled
Manual
10ns
7ns
Auto
4 beats
Disabled


run superpi 1m, if it passed, shoot for 32m, if it passed.....then you can start tighten up your timing one at a time and slowly increase your memclock and see what happen. if spi failed....use 2T, if spi 1m/32m passed....your Ballistix still has some life left but won't last too long, I'd say couple weeks top. If it failed under 2T....they're hopeless. If I were you I'd still RMA them either way :lol:

Sligthly better... could do dual channel 240... but not even stock speed. It's in route to RMA... wonder if it is the DFI that killed it or ballistix itself dying... bleh. Will know after it sits on this asus for next couple of months. Also have CF asus coming in today but I kind of like this board already and opty I have... we shall see :)

[qdemn7]

wonder if it is the DFI that killed it or ballistix itself dying... bleh. Will know after it sits on this asus for next couple of months. I'm not as technically knowledgeable as some of you people, so pardon a possibly dumb question.

Could the onboard memory controller of the A64 be responsible for killing memory modules? These CPUs are being put under incredible stress. Could the controller be the weak link in the chain, and thus somehow killing modules?

The reason I ask was that my last CPU and board was an AOpen AK89 Max and Newcastle 3000+. That combo killed not one but, TWO sets of quality modules. First a set of 2 x 256MB of Corsair XMS3500, and then a set of 2 x 512MB of OCZ 3700EB. Note that the AOpen wasn't much of an O/C board since the voltage wouldn't go too high and my skills weren't up to volt-modding the thing (hands too shaky for all the delicate work). The best I ever got was 2.35 x 10 out of either memory. And crappy power isn't at fault either. I've always used Antec PSUS, backed up by an APC UPS.

[ben805]

Sligthly better... could do dual channel 240... but not even stock speed. It's in route to RMA... wonder if it is the DFI that killed it or ballistix itself dying... bleh. Will know after it sits on this asus for next couple of months. Also have CF asus coming in today but I kind of like this board already and opty I have... we shall see :)


2 pairs of my Ballistix were dead when I had them on the Ultra-D, maybe it was just coincidence but when you get your new pair back from RMA put them under the Asus and see how they hold up, these ballistix are excellent overclocker and scream like banshee....it's a shame that most of them only last couple weeks and we still don't know for sure if it was the DFI that killed them, but I hope they work on your Asus though.

[jinu117]

I'm not as technically knowledgeable as some of you people, so pardon a possibly dumb question.

Could the onboard memory controller of the A64 be responsible for killing memory modules? These CPUs are being put under incredible stress. Could the controller be the weak link in the chain, and thus somehow killing modules?

The reason I ask was that my last CPU and board was an AOpen AK89 Max and Newcastle 3000+. That combo killed not one but, TWO sets of quality modules. First a set of 2 x 256MB of Corsair XMS3500, and then a set of 2 x 512MB of OCZ 3700EB. Note that the AOpen wasn't much of an O/C board since the voltage wouldn't go too high and my skills weren't up to volt-modding the thing (hands too shaky for all the delicate work). The best I ever got was 2.35 x 10 out of either memory. And crappy power isn't at fault either. I've always used Antec PSUS, backed up by an APC UPS.

Well good PSU helps but since it IS running through mobo voltage regulator, if the voltage regulator is shakey... it could kill memory. There were certain notorious PSUs that were good at killing memory if you do direct 3.3v to memory mod. I primarily think the main reason memory dies are unsteady voltage getting fed in, too much voltage, or god's act :)
One thing in my case is that DFI did kill a lot of memory for mine which never really happened with other mobo... so my primary suspect becomes DFI in this case but I will leave the room open :P

BTW, funny enough my rig is going strong dual priming at stock speed @ 1.15v I setup... bit hard to burn it in when you can't figure out where it starts going not stable... loving it so far. Opty 175 and A8N32. Wierd thing is, whatever v-core i set, it reads .03 more under load and .05 more under idle than what I set at on asus probe thing... Burning in this opty might take a lot more time than I thought I would.

[craig588]

I think I've got the mobo to a point where it's working well. I sinked the memory fets and VTT regulator because they were very hot and warm, respectivly. All of my stability issues seem to have dissapeared.

[ben805]

I think I've got the mobo to a point where it's working well. I sinked the memory fets and VTT regulator because they were very hot and warm, respectivly. All of my stability issues seem to have dissapeared.

Can you please post some pictures as to exactly where you sink them? it might be helpful to eliminate similar problem down the road. :toast:

[situman]

ASus does recommend adding fans for phase and watercooling so I guess they actually did these kind of tests.

[craig588]

VTT and the little chip next to it (I had a big enough sink, so why not?)
http://img376.imageshack.us/img376/358/dscf21487uf.jpg
Memory mosfets
http://img446.imageshack.us/img446/5182/dscf21468yy.jpg

I just remembered I hit all of the chokes around the CPU too. They were all very hot too. I doubt anyone needs a picture to find them though.

[Shootingblanks]

I am currently running my 4400 at a 11X multiplier at 228Htt to get me a 2508Mhz processor. I am running bios 703 with my mobo an have the following advanced chipset settings:
K8 to NB Frequency: [600 MHz]-this gives me a 3x HTT(LDT)multiplier
K8 to NB Link Width: [Auto}
SB to NB Frequency: [1000]
SB to NB Link Width: [16down,16up]

I can play BF2 for 5 hrs straight with no crashes and my temps are 45 degrees celcius for the processor. System has yet to crash. I am not experiencing any problems at all. I am running my memory at 2-3-2-6 at 1T at 2.75 volts. I am having a problem passing Prime 95. Because of this, I have raised my Vcore up to 1.4 in bios to see if processor is the problem. I loosened my timings to 2-3-3-6 and 2.5-3-3-6 and I am still getting a no pass in Prime 95. Therefore my system is not as stable as I would like it. I plan on running memtest with loop test 5-8 all night long so it should be about 200 passes to see if I get an error with the memory. I performed a quick test with memtest 86 and it passed. I got great bench test results with sisoft sandra 2005, pcmark 2005, and aquamark. I average between 85-100 fps in bf2 and have no problems with performance end. I believe my memory is 438 and at 228 I am running it at 456. Any suggestions for prime 95 would greatly be appreciated. thanks.

http://img.photobucket.com/albums/v226/shootingblanks/SP32-20051208-212436.jpg

It also didnt pass prime 95 at these settings dipicted.

[ben805]

Any suggestions for prime 95 would greatly be appreciated. thanks.



raise your vcore one more notch and run the prime, if it failed...raise another notch and see what happen, if it still failed...then back it down and play with your memory settings.

[Shootingblanks]

thanks what can i run with air 1.45 right?

[craig588]

Easily.

[smurfmods]

ben805---

I got the Mach II in and installed. I'm getting 6C from pcprobe and a little less from the bios temps. Is that normal on this board? I have a suspicion that the mach head is not making good contact with the cpu. Maybe too much seal string or something. Have you had issues with cpu mating with your phase? Are you using a custom system?

I fell flat on my overclocking though. Only got to 2900mhz. 3gig was a nogo. Tried letting up on the drive strength but didn't help. Took the mem out of the equation by setting it to 133mhz multiplier. I raised the Vcore up to 1.45. Is that enough? I don't want to pump the cpu voltage up any higher than I have to. I'm still working with the 805 bios. Can't find a floppy drive that actually works in my house. Can you boot this board with a USB thumb drive? Anyone tried it?

I'm getting the "NO CPU FAN IS CONNECTED" Press F1 to continue. Any way to kill this message? (keep in mind I don't have the heatpipe fan or another fan handy to plug in there)

I know it's a lot of questions but anybody is welcome to chime in with help.

[Ic3man]

Use bios 703. Imo its much more stable than the 805 bios.

[HKPolice]

This board is a living hell, but I'm slowly working a way through it.

Some bugs:

Overclock too much, no post, shut down, turn on again and it resets to defaults right? Yes, but after that happens, it won't post after I save and exit. I HAVE to reset to setup defaults, save exit, change the settings again then save exit.

For my board + X2 4400+, CPU overvolt + 1.35v = 1.60v
overvolt + 1.36v = 1.61v
overvolt + 1.38v = 1.60v

and the pattern continues, measurements taken from a DMM. The board reading varies from 1.52-1.54v, totally useless.

I am STILL getting warm boot problems above 2.7Ghz. I was running this CPU 24/7 @ 2.75Ghz in the DFI @ 1.57v.

[ben805]

thanks what can i run with air 1.45 right?


dual core can get pretty toasty but you should be able to run up to about 1.55v....it really depends on your underload temperature, if it's getting up in the mid 50s during prime then back down your vcore.


ben805---

I got the Mach II in and installed. I'm getting 6C from pcprobe and a little less from the bios temps. Is that normal on this board? I have a suspicion that the mach head is not making good contact with the cpu. Maybe too much seal string or something. Have you had issues with cpu mating with your phase? Are you using a custom system?

6'C idle or underload? if it's idle then I'd say that's a bit warm for a Mach II unless your vcore is up in the wazoo, check your mounting again. mine is a custom PCICE unit and I don't have any contact problem with the naked core, my new 148 CABNE idle at -26'C and around -12'C during prime.


ben805---

I fell flat on my overclocking though. Only got to 2900mhz. 3gig was a nogo. Tried letting up on the drive strength but didn't help. Took the mem out of the equation by setting it to 133mhz multiplier. I raised the Vcore up to 1.45. Is that enough? I don't want to pump the cpu voltage up any higher than I have to. I'm still working with the 805 bios. Can't find a floppy drive that actually works in my house. Can you boot this board with a USB thumb drive? Anyone tried it?


if the IHS is still on and if you feel comfortable removing it then by all mean do it, take your time and just be careful with the blade, though don't be surprise if you don't see any gain at all as I have had three X2 3800, two 4200 and two 4400 that can't clock worth a crap...their IHS were all removed, and they all hit the wall at 2.90Ghz and won't prime any higher...not even 1mhz more, my last X2 4200 was the only one that could break 3.0Ghz. stick with BIOS 703...it's the most stable BIOS to date especially under phase. And no I haven't try to boot from any USB device :)


ben805---

I'm getting the "NO CPU FAN IS CONNECTED" Press F1 to continue. Any way to kill this message? (keep in mind I don't have the heatpipe fan or another fan handy to plug in there)

I know it's a lot of questions but anybody is welcome to chime in with help.


I believed I'd already pointed out the CPU FAN error message on page 5 of this thread, you'll need to plug in something in there to rid the error otherwise it will keep coming back during POST, use the optional heatpipe fan on the cpu fan header or maybe use one of your case fan, it has to be occupied that's about it.

hope this help :toast:


This board is a living hell, but I'm slowly working a way through it.

Some bugs:

Overclock too much, no post, shut down, turn on again and it resets to defaults right? Yes, but after that happens, it won't post after I save and exit. I HAVE to reset to setup defaults, save exit, change the settings again then save exit.

For my board + X2 4400+, CPU overvolt + 1.35v = 1.60v
overvolt + 1.36v = 1.61v
overvolt + 1.38v = 1.60v

and the pattern continues, measurements taken from a DMM. The board reading varies from 1.52-1.54v, totally useless.

I am STILL getting warm boot problems above 2.7Ghz. I was running this CPU 24/7 @ 2.75Ghz in the DFI @ 1.57v.


have you try different ram? both of my opty needed slightly less vcore to run at the same clock speed on the Asus than DFI....

[HKPolice]

It's not the ram, I tried 166 divider with the 2GB kit and TCCD as well.

[Mickey Padge]

I am at work at the moment so I can't check where in the BIOS it is, but there IS an option to disable the "press F1 on error" message.

I have it turned off myself.

Running a Mach II and can report no issues running at 3300 1.52v at arround -24 idle -12c load. No cold problems with my board at all. I have not needed to play with drive strength etc.... either :) Thats with an opty CABNE 0528 148 CPU.

I am using the 703 BIOS it shipped with and have tried no other.

Cheers :D

[Ic3man]

This board is a living hell, but I'm slowly working a way through it.

Some bugs:

Overclock too much, no post, shut down, turn on again and it resets to defaults right? Yes, but after that happens, it won't post after I save and exit. I HAVE to reset to setup defaults, save exit, change the settings again then save exit.

For my board + X2 4400+, CPU overvolt + 1.35v = 1.60v
overvolt + 1.36v = 1.61v
overvolt + 1.38v = 1.60v

and the pattern continues, measurements taken from a DMM. The board reading varies from 1.52-1.54v, totally useless.

I am STILL getting warm boot problems above 2.7Ghz. I was running this CPU 24/7 @ 2.75Ghz in the DFI @ 1.57v.


It will only reset to default if you press F2. If you press F1 it will load the previous settings you had and enter the bios screen so you can lower the overclock rather than starting from scratch. Just saving the settings you had earlier will not help any as it will be the same unsuccessful config you had before, you need to manually lower the settings you had previous after pressing F1.

Have you tried lowering the divider more? And have you lowered the HT to 3x? This helped my setup and I can boot into windows @2.8ghz - 1.3(+.2) with divider set to 133 1T 2.5-2-3-5 (3500ll - 165 opteron) although at this speed it aint p95 stable.

[da-key]

For the CPU Fan thing Ben says plug a fan in which is correct. Then go into bios and disable the cpu fan speed in the power section. :p:

[AppaYipYip]

My system seems to be running slightly warm overall. My CPU temps for an Opty 175 are idle 38*, load 44*. It never seems to go about 44 though. I had this same CPU on an A8N SLi Premium and it was around 33* idle, 44 ish load. Also, the MB temps seem hotter at idle and load too, and they are almost identical to the CPU temps.

[Perc]

ben895

this bios you recomend for phasechange do you have a link to it? ive tried using that auto update to get the bios and it either doesnt connect or gives me some lame excuse saying there is not proper image file on server!!

thx perc,

[Zeus...]

ben895

this bios you recomend for phasechange do you have a link to it? ive tried using that auto update to get the bios and it either doesnt connect or gives me some lame excuse saying there is not proper image file on server!!

thx perc,
http://support.asus.com/download/download.aspx?SLanguage=en-us select motherboard - socket 939 then a8n32 - bios

[Perc]

thx man that helps. now ben is sayin that bios 703 is the way to go with phasechange? thats what i already have in here. i gues its what came with the mb :shrug: hmmm will maybe ill just give 703 a go before i try flashing over to somthing else...

*edit* hey guys in the bios there is a CPU FSB freq and a SB to NB freq setting wich one is the one you set for htt? do i set both or just the one? they both have a defualt setting of 200 and go up in 1mhz steps so wich one do i adjust? also whats the Peg Link mode option do?

thx perc,

[Zeus...]

thx man that helps. now ben is sayin that bios 703 is the way to go with phasechange? thats what i already have in here. i gues its what came with the mb :shrug: hmmm will maybe ill just give 703 a go before i try flashing over to somthing else...

*edit* hey guys in the bios there is a CPU FSB freq and a SB to NB freq setting wich one is the one you set for htt? do i set both or just the one? they both have a defualt setting of 200 and go up in 1mhz steps so wich one do i adjust? also whats the Peg Link mode option do?

thx perc,


I just set the top one in the chipset config for HT

1000 is
800 is 4
and so on keep it below 1000 allowing for HTT x HT

Jumper free config -AL overclocking to manual then u get the option for the HTT

also whats the Peg Link mode option do? good question:clap:

[Perc]

so your saying that maybe the SB to NB setting is the LDT setting? this bios realy is tricky if your not used to the way asus sets them up. ive seen settings in this bios ive never seen before! going to take me ages to figure this out. im almost tempted to put my DFI ultra-D back in and forget about it but what fun would that be!!

thx perc,

[craig588]

I've been ignoring the SB to NB setting. The CPU FSB setting is your FSB. It hits the memory and the CPU.

This is one of the hottest running boards I've used, it's a really good idea to heatsink everything.

[AppaYipYip]

Peg Link has something to do with your PCI-e slots I think. Some sort of bus speed enhancer I think. I've read in some reviews that it esentially does nothing noticable when you adjust it.

[Perc]

yeah but if you guys look closer you will see that there is a K8toNB setting as well as a SB to NB setting the K8 to NB is the LDT setting but this SB to NB is confuseng me so for now ill just leave it at 200 til it effect my over clock. also notice that cpu-z hasnt buged off of 1.35v even after setting the v-core to 1.425v in the bios though asus probe shows 1.39v... man this mb beter be worth all this effort!!

thx perc,

[Ic3man]

The peg link is indeed connected with the pci-e slots. Read somewhere its a way of overclocking your GPU through the bios.

I think I have posted this before but here goes : My Vcore wouldn't change no matter what setting I used unless I enabled the 0.2mv overvolt. CPU-Z still shows it at its normal setting minus the over volt i.e

Vcore = 1.55 set in bios

Asus probe reads 1.4
CPU-Z reads 1.35

Vcore = 1.35 +0.2 overvolt in bios

Asus probeII reads 1.55 (correct)
CPU-Z reads 1.35

I guess an update with cpu-z will sort out the bug and an updated bios will sort out the asus bug.

[AppaYipYip]

Mine does the exact same thing as everyone elses. I get strange voltage readings and CPU-z always reports 1.350 regardless of my settings. So far my system has been rock solid so I'm not complaining. The *only* issue I've had were two random reboots while playing Guild Wars, but I believe that's an issues with SLi. That game is not happy with SLi setups.

[craig588]

Don't use software to measure voltages. Whip out your multimeter and check them.

[Perc]

oh yeah i know man im just getting this mb all set up and just found it odd that cpu-z wasnt changing as i changed the voltages. just wanted to make sure the infact the voltages are indeed going up! i think i read a few pages back where the contact points were for the v-core readings on the mb.. anyways guys thx for the info so far so good with this mb. now to get back into the bios and figure some more crap out. oh i have a question about the A.I. over clocking feature i notice that there is a few differnt choices ( Manual/Auto/Standard/A.I NOS/ and OC profile) now some are basic but i gues what im wanting to know is does each differnt setting offer differnt setting for the system over clock? Am i making this harder then it realy is?


thx perc,

[craig588]

I've only used manual because Asuses automatic settings never work well on any of their mobos.

[Ic3man]

Use manual unless your are using EVERYTHING Asus recommends for this motherboard, memory, cpu etc etc.

Manual is more fun anyway and no doubt much more successful.

[smurfmods]

6'C idle or underload? if it's idle then I'd say that's a bit warm for a Mach II unless your vcore is up in the wazoo, check your mounting again. mine is a custom PCICE unit and I don't have any contact problem with the naked core, my new 148 CABNE idle at -26'C and around -12'C during prime.

if the IHS is still on and if you feel comfortable removing it then by all mean do it, take your time and just be careful with the blade, though don't be surprise if you don't see any gain at all as I have had three X2 3800, two 4200 and two 4400 that can't clock worth a crap...their IHS were all removed, and they all hit the wall at 2.90Ghz and won't prime any higher...not even 1mhz more, my last X2 4200 was the only one that could break 3.0Ghz. stick with BIOS 703...it's the most stable BIOS to date especially under phase. And no I haven't try to boot from any USB device :)


Yup that's my idle temp. Volt core was a modest 1.425V. Must have something hooked up wrong. I think I'll leave the IHS on. May not be sticking with this cpu long anyways and I may need to sell it. I have another X2 3800 that I will try. (it didn't do as well on air but who knows on phase) Seems like there is no point in putting phase on a cpu if I can get 2.6gig on air and only 2.9gig on phase.

As soon as I get ahold of a floppy I will definatly go to the 703.

Why are these chips hitting the wall at 2.9 to 3 gig?

*update* Very weird. X2 3800 will boot at 2.99gig but not 3gig. 1 more mhz kills it. Runs all day long benching and gaming at 2.99. :gamer: Has to be a engineered lock in some of the chips not to go any higher.

Also I got the USB Thumb drive to boot up an I'm running the 703 bios now. Don't notice a difference in stability but it's there just in case. If anyone wants to know how do make the thumb drive boot up let me know. ( I hate floppies I'll fight with a thumb drive or cd booting all day long before I'll hook up a floppy. :lsfight: )

ben805--Which drive strength do I need to adjust with phase?? What do I set it to?

[jinu117]

The peg link is indeed connected with the pci-e slots. Read somewhere its a way of overclocking your GPU through the bios.

I think I have posted this before but here goes : My Vcore wouldn't change no matter what setting I used unless I enabled the 0.2mv overvolt. CPU-Z still shows it at its normal setting minus the over volt i.e

Vcore = 1.55 set in bios

Asus probe reads 1.4
CPU-Z reads 1.35

Vcore = 1.35 +0.2 overvolt in bios

Asus probeII reads 1.55 (correct)
CPU-Z reads 1.35

I guess an update with cpu-z will sort out the bug and an updated bios will sort out the asus bug.

Interesting... I haven't had this problem but I only tested from 1.4 down to 1.125 so far. Only wierd thing is v-core always reads about .02v higher than what I set at on asus probe thing or AI clocker thing or what not. Is the clockgen version that works with this? My burning in is taking hell lot longer than it should since I am having hard time finding unstable point.

[qdemn7]

Is the clockgen version that works with this? This one does: Here is a beta of the new clockgen. All currently supported PLLs are not all added yet, but it is in progress.
Please let me know your opinion ! :)
Thanks !

Version 1.0.2.3 (http://www.cpuid.com/download/ClockGen.zip)

[HKPolice]

This one does:

Thanks! Is there a way to make it autostart?

Edit: Doesn't help, still warm boot issues over 2.7Ghz

[qdemn7]

Thanks! Is there a way to make it autostart?Uhhh, don't know about that.

[Eluzion]

Anyone getting random freezes when the computer is just idling? It was happening before, but I upped the VCore and it was good for about two days but just earlier I went away from the keyboard for about 5 mins (screensaver was on) and I come back to find the computer was frozen.

I hit the restart button on my computer and it just hangs. Starts rebooting and just does nothing (screen stays black). I have to shut it off completely for a few seconds then power it back up, but when I do that I get a "CMOS Bad Checksum" or something and it asks if I want to run Setup or Restore Defaults.

Not sure what's causing these freeze ups. I was running 250x10 @ 1.39vcore. Ran Prime95 for 15+ hours no problem and temps were 38c idle, 45c load. I'm going to keep it at stock speeds for the next few days I guess to see if it still freezes but if anyone knows a solution, please let me know. ;) Thanks

[jinu117]

This one does:

Hot damn! You just saved me about 5-12 hours there. Thanks a ton :)

[mackanz]

So far so good with the board. OCZ PC4000 EB just loves this board. 270 Prime stable for 24 hours (custom, all memory used) One thing i'm a little buggered with is the amount of assync latency the board uses. 11 and 9.5 is weird numbers. Doesn't like to be changed either. Another thing...with 4 sticks of ram, the board forces DDR200 and can't be changed except for normal HTT rising. You need a lot of HTT/FSB to reach DDR400. 400 htt. Not many boards/cpu's do that with 4 sticks. Don't touch the PEG is my advice..any other than normal gives tremendous instability.

[craig588]

Anyone getting random freezes when the computer is just idling? It was happening before, but I upped the VCore and it was good for about two days but just earlier I went away from the keyboard for about 5 mins (screensaver was on) and I come back to find the computer was frozen.



I was, but after I heatsinked everything the freezes went away. Look at my pictures a few posts ago for information on what to sink. Also hit your chokes around the CPU because they get very hot too.

[ben805]

So far so good with the board. OCZ PC4000 EB just loves this board. 270 Prime stable for 24 hours (custom, all memory used) One thing i'm a little buggered with is the amount of assync latency the board uses. 11 and 9.5 is weird numbers. Doesn't like to be changed either. Another thing...with 4 sticks of ram, the board forces DDR200 and can't be changed except for normal HTT rising. You need a lot of HTT/FSB to reach DDR400. 400 htt. Not many boards/cpu's do that with 4 sticks. Don't touch the PEG is my advice..any other than normal gives tremendous instability.


4 sticks in 1T or 2T?

[craig588]

This mobo also only uses 1VTT regulator when the datasheet even shows there should be two. (Luckily I happen to have a few of them on hand, I'll try installing a second one)

Also, I'm really lost as to a Vdimm mod. I havn't been able to find the regulator chip, or if the Vdimm is driven off of the 3.3 or the 5V rail. (If it's 3.3 I'll just do a Vdimm = 3.3 mod)

Someone help me.


Vcore is setup really weirdly on this board too. The regulator they chose is only capable of handling 4 phase power, where are the other 4 phases coming from? You can probably guess, that extra .2V option. There's an analog switch and 2 little controllers for those extra fets.


Another update: Now I'm stealing some of Shaminos work from another mobo, but apparantly a resistor between pin 6 and pin 3 of the RT9214 will increse the voltage. Start with 50K and move down. I can't find the datasheet on it though. I'll check the resistance on this mobo to make sure something crazy wont happen, but it looks almost ready to go.

Updated again: The resistance is only around 98ohms, so 50K will be fine. Even 10K would be fine.

[HKPolice]

This mobo also only uses 1VTT regulator when the datasheet even shows there should be two. (Luckily I happen to have a few of them on hand, I'll try installing a second one)

Also, I'm really lost as to a Vdimm mod. I havn't been able to find the regulator chip, or if the Vdimm is driven off of the 3.3 or the 5V rail. (If it's 3.3 I'll just do a Vdimm = 3.3 mod)

Someone help me


Vcore is setup really weirdly on this board too. The regulator they chose is only capable of handling 4 phase power, where are the other 4 phases coming from? You can probably guess, that extra .2V option. There's an analog switch and 2 little controllers for those extra fets.

I'm pretty sure it runs off 3.3v else it would support a lot more than 3.2v. Vdimm is rock solid when I measured, so I don't think you need to really worry about it. I only see one missing mosfet soldering pad underneath the dimms, is that the missing VTT circuit?

I think the warm boot issues I've been having is due to vcore. During a normal reboot, the vcore drops down to 1.42v, then back up to 1.61v or whatever I set it to. But above 2.7Ghz, it just stays at 1.42v, no post, no beeps - as if there was no CPU present.

most of the people who have this board are using single core CPUs, so their vcore options aren't as limited.

Argh

How about a vcore mod, craig?
Thanks

[craig588]

The vcore regulator is :banana::banana::banana::banana:ing nasty. There's a 1MB PDF on it, and it's something I'm not looking forward to modding. It's more complex than I like.

I'll send the PDF to anyone who contacts me on AIM or MSN. (There's a 200KB file uploading limit here)

Edit: Just for you I started looking into it. It's more plausible than it first looked. Someone tell me the resistance between pin 8 and ground. My chips legs are covered with nail polish right now. It's the ADP3168 very near the CPU.

[Perc]

well guys i hate to throw in the towel so easily but this mb isnt for me. im sure this mb is as good as the DFI and all but im just not getting along with this bios. i think im going back to my DFI ultra-d. i can deff see why that one guy traded me for the expert mb.... oh well no biggie im sure ill get my money back outa this thing. well thx guys for the help.

peace perc,

[craig588]

Before you go, measure pin 8!

[craig588]

250K pot between pin 8 and ground will do Vcore on the ADP.

Thanks to HKpolice for getting the resistance measurements for me.


It's almost 1AM, I'm going to sleep. Any more mods will have to wait untill tomorrow. (Or be made by someone else)

[SVTSnake]

Well after several days of being very frustrated by this board, tonight drew the last straw. That being I fried my Opty 175 with given it only 1.6v! ARGHH!!!

The board has its pluses, but there are a few things that really annoy me!
- Give it 1 wrong memory parameter, and poof, clearing of the CMOS is necessary to get it to turn back on. Never could get the recovery function to work.

- The Clear CMOS jumper and battery are in the WORST place for SLI! You have to remove the bottom card to get to the jumper and battery!

- The layout of the slots are really annoying. Asus should have kept to there original A8N-SLI layout!

- The BIOS to me is frustrating. I have used AMI bios' in the past, but why don't they group all of the memory and cpu settings together in the same menu?

These are just a few things. I know they are probably minor, but after taking the computer out of the desk countless times in the past few days to remove a video card just to be able to reset the CMOS, enough is enough! I'll have to contact AMD Monday to set up an RMA. For now, I am back on my NF4 SLI Expert with my trustworthy 4400+.

Good luck with the board!

[HKPolice]

250K to Pin8 doesn't work. Well, I soldered a 220K fixed + 50K VR and turned the VR down all the way, still no effect. Will try 150k fixed + 50K VR next.

[mackanz]

4 sticks in 1T or 2T?

2T.
Won't post with 4 DS sticks and 1T.

There's no need to reset cmos at all on this board. If it doesn't post, kill the power on the psu and wait 2 seconds for the mainboard to drain the power. Flip it back on, hold Insert key while power on as usual. You'll then have to hit F1 to enter setup/bios.

[Zeus...]

Good to see one off the OCZ team have got there hands on the mobo :toast:

I most welcome post:woot:

nothing to do with my E-mail was it 'mackanz' ?

[HKPolice]

50K VR to Pin8 mod works. Thanks to craig588! I turned up the vcore to 1.65v and now the warm boot issues @ 2.75Ghz are gone 50% of the time.. if I press reset, it'll post 100% of the time, whereas before, I had to shutdown then power up again. It feels like there's something else causing the warm boots and upping the vcore only helps SLIGHTLY.

I was priming perfectly fine @ 2.75Ghz 1.61v before the mod, just that warm reboots would never post.

It is NOT caused by memory because I've tried UTT BH5, CE-6, and TCCD chips, all with and without dividers.

My theory is that during a warm reboot, the mobo applies the overclocked FSB before it applies the upped +0.2vcore (at least for X2 cpus). I assume this won't be a problem for cpus that don't need more than 1.55v or so, but any higher and it'll cause problems.

The current bioses are still full of bugs, so only time will tell if it will be fixed....

[mackanz]

Not at all Zeus, not at all.

[Zeus...]

Not at all Zeus, not at all.

U have got your hands on the mobo thats all that counts :)

[bullit]

The current bioses are still full of bugs, so only time will tell if it will be fixed....
I f this thing of bad bios would be like the waiting for the correct bios for the A8N-SLi to fix the bug o 1T...:slapass:

[mackanz]

bullit,

I have a feeling that will never happen. I would ditch that sorry board asap.

[bullit]

Mackanz,

I've dtiched it already,now i'm with a DFI Expert...:woot:

[craig588]

Yeah, I was too tired yesterday and screwed up in my math. I added in an extra zero. A 25K pot will be fine for the Vcore mod.


Someone help me with the Vdimm mod. The voltage goes up a little bit, about .1V, but then the machine shuts off. What's going on? I can't find any information on the chip either.


I really just want to do a 3.3= Vdimm mod, but I don't know if the memory is getting fed from the 5V line or the 3.3 line.


Edit: I'm tired of waiting. I'm going to do the mod and if I fry my regulators it will just be a warning to other people.

[tricknasty]

Well don't know what to think about this....but a guy just mentioned to me that a friend of his that works at Asus said the A8N32 had been recalled because it was frying cpu's with certain types of PSU's and there were other things still unknown. :confused:

Now I'm not saying this is a fact, I just don't know what the hell is the truth regarding this matter any more. :confused:


i know this sounds crazy but my friend actually FRIED a brand new X2 4400 on this mobo when he was installing windows. He doesnt even know what happened the computer reset and then his chip was dead. i tested it in my system and i have a A8n32 as well and the chip is definetely dead. could have been a bad chip i guess.

he was using an older thermaltake psu with a 20-24 pin adapter.

[specofdust]

Hmm, might be the right time to ask, is using a 20 ping ATX connector with a 20>24 ping adaptor going to make much of a difference to this mobo?

Good thread, well done Ben 805 for being so helpfull. But to be honest, the DFI SLI-DR looks more tempting compared to all the problems this ones having.

[Zeus...]

i know this sounds crazy but my friend actually FRIED a brand new X2 4400 on this mobo when he was installing windows.

It happens one Winnie died on me once within a week RMAing sorted it quick enough

[craig588]

The memory regulators are crap on this mobo, just crap. I did the 3.3= vdimm mod and all of my warm and cold boot issues have dissapeared. Normally I need to try a few times to make the thing boot, but it popped up right away, first try after the mod.

It's memtesting now. (I need to make sure it's not gonna pop the fets in a half hour when I'm trying to do something)


Do the 3.3 = Vdimm mod, you'll love it. The VTT it tracking too. (But it's still droopy, I need to add in another regulator to stabilize the VTT circuit)

[cronic]

can some one post a pic of where to measure the vdimm and vcore on the motherboard with a dmm. if it ahs already been posted just point me in the right direction. thanks again for all the great work.

[craig588]

http://img376.imageshack.us/img376/358/dscf21487uf.jpg

The white side of that capacitor spot is Vdimm.

I don't have a handy photo for Vcore, but someone else might.

[cronic]

thanks bud

[craig588]

Be careful with 2GB, the cable I soldered is getting a little warm from all the current it's drawing. I might add another to help with the load.


Edit: There is good news though, after an hour of memtesting nothing on the board has popped so the mod's deemed safe by me. Don't try my other Vdimm mod, the memory regulators shut down after you add about .1V. It's pretty much useless.

We have Vcore and Vdimm, does anyone want any other mods?

[MAXIMILIAN]

Im sorry I read about 5 pages of this thread then I couldnt go on, I am in the process of buying a new mobo and I just wanted to know If you guys had come to a verdict concerning A8N32-sli-vs DFI lanparty.

Thanks
-Maximilian

[craig588]

If you don't mind soldering and modding this is a better mobo. If you want to use it stock get the DFI.

I own both.

[cronic]

Im sorry I read about 5 pages of this thread then I couldnt go on, I am in the process of buying a new mobo and I just wanted to know If you guys had come to a verdict concerning A8N32-sli-vs DFI lanparty.

Thanks
-Maximilian

I am extremely happy with the asus a8n32-sli. I have not done any mods to it and love it. I switched from a dfi sli-dr and it was the best upgrade i have ever done.

[jinu117]

Need to give couple months of playing to give verdict really as my DFI boards usually lasted that long each. But so far things looks fine even without v-modding. I am pretty sure I will v-mod for CPU down the line though... :P

[MAXIMILIAN]

I am currently using an asus mobo(a7n8x delux) and i like the idea of using one again cuz i like the bios, Although i understand you can DL other bios:confused: but if the new asus can really OC better than the dfi then i will go for it. What would i need to solder?

O and can this mobo fit two watercooled 7800gtx 512's?

Thanks
-Maximilian

[craig588]

Yes it can.

You should do a 3.3=Vdimm mod and if you have a dual core CPU you should do a Vcore mod.

[cronic]

wouldn't it be better to say if you have a dual core that likes more than 1.58v you should do the vcore mod. i have a dc 3800x2 with no mod at 1.52v and it is fine.

[bullit]

Cronic ,fortunately you have a good chip because mine(rma) usualy gave 2750 but with 1.6 and fried.

[SVTSnake]

i know this sounds crazy but my friend actually FRIED a brand new X2 4400 on this mobo when he was installing windows. He doesnt even know what happened the computer reset and then his chip was dead. i tested it in my system and i have a A8n32 as well and the chip is definetely dead. could have been a bad chip i guess.

he was using an older thermaltake psu with a 20-24 pin adapter.


I don't think it sounds crazy! Fried a perfectly good Opty 175 last night in this board. I am using an PC Power & Cooling 850 SLI power supply that is 24 pin native. I did notice as Craig the chokes and surrounding mofsets got VERY hot.

I tested the 175 today in my other system and she is dead! Also tried it in my Expert before reverting back to my 4400+ and no go!

[MAXIMILIAN]

well i have been considering either the opteron 165 opteron 275, or the athlon 64 fx 57, the 165 is 400$ and the other two are 1000$ but people have been telling me that with my watercooling loop I could get the 165 to outperform the other two. can anyone verify this for me?

[craig588]

I using some random crap power supply that I've thrown together from 3 other power supplies. (Sure it can put out about 1000 watts, but it's not any real brand name or anything)

This board does run very hot though, people should be sinking everything as soon as they get this board. If you do the 3.3 = Vdimm mod the memory regulators turn off so they stay cold and don't need any sinks.

[HKPolice]

Craig, I don't think the 3.3v rail = vdimm mod is a viable option because most people would want to use 2GB and this board limits 4x512MB UTT to 100Mhz memory divider.

I was still getting warm boot issues with 166Mhz memory divider, forcing the ram to run at 229Mhz which should be able to post even at 2.5v or whatever the regulators are giving it.

To the people who fried their CPUs, I think it's because of overheating mosfets. The first thing I did when I got the board was pop off the heatpipe. There was NO thermal grease on the mosfets, only some crappy grey thermal pad which insulates more than anything. Dab on some AS5 and you'll be good to go. My chokes never get hot.

[craig588]

Yeah, that's another issue. Mine was annoying to get off, someone at Asus had the idea to cover the pins with hot glue and the glue embedded itself on the pin in such a way that it's impossible to squish down the sides of the pin to pop it out. I needed to cut them off and replace them.

[HKPolice]

What 2GB kit were you using with the 3.3v=vdimm mod? I don't know of any that likes more than 2.9v

[MAXIMILIAN]

hmm, when i get my mobo i think il need to pm you to do the mod with thermal grease. :)

Also anyone know the answer to the question i posed in my last post?

Thanks gents
-Maximilian

[craig588]

I wasn't I'm using 2x512MB of BH5. I'm going to continue using BH5 untill DDR2. I'm convinced there isn't anything that can beat it in the life DDR1 has left.

I meant that anyone that's using 2GB of memory might need more wires because mine was already getting slightly warm.

[specofdust]

This is just a stupid question from a newby, but, couldn't you just use a better/higher capacity wire?

[craig588]

I don't feel like doing that, but yeah that's the easier way if it was done in the first place. I'm using the stock 16-14 gauge wire that's coming out of the PSU.

[jinu117]

I don't feel like doing that, but yeah that's the easier way if it was done in the first place. I'm using the stock 16-14 gauge wire that's coming out of the PSU.

Wow... that 16-14 gauge wire shouldn't really get that warm... But than not sure what you are really running there for memory :P
Awesome job Craig on finding out this v-mods. I really appreciate this pioneering type of activity and your willingness to share knowledge :) :toast:

[craig588]

I think I just killed my board. I did the VTT in parallel mod and now it doesn't boot up. That's not what I think killed it though, while I was working I moved my hand into the heatpipe cooler and it got tilted. After it didn't boot up I checked under it and the NB is chipped.

Time for something else, I guess.

[jinu117]

I think I just killed my board. I did the VTT in parallel mod and now it doesn't boot up. That's not what I think killed it though, while I was working I moved my hand into the heatpipe cooler and it got tilted. After it didn't boot up I checked under it and the NB is chipped.

Time for something else, I guess.


Awww and you were flying with it... -_-; How does Asus CF mobo sound to you? have one here now but no spare CPU really to test until I am done with my opty burning in on A8N32. My son stepped on brand new 3200+ venice yesterday and it is irreparable really... -_- (that was spare cpu I got 2 days ago to play with while burning in opty...-_-)

[craig588]

Send me photos and I'll ask you for numbers on the chips. I'll ask you for resistance values and I'll send you the mods.

I don't have the money to buy another board right now so I'd have to mod it remotely like that.

[MAXIMILIAN]

ok so just to clarify, the new asus sli mobos(with modding) are better in almost all aspects including oc'ing and can fit dual 7800gtx 512's with waterblocks?

Thanks again
-Maximilian

[situman]

Craig, did you even have the board for a week yet? Chipping the NB is a tough way to lose a board like that.

[craig588]

Nope, I didn't.

The crappy heatpipe is worthless. It just makes it really easy to dump alot of torque onto something. They definatly were not transferring heat, the little blocks were hot and the pipes were only mildly warm. I'm still opposed to the idea of them and have been from the beginning.

[MAXIMILIAN]

Is this refering to the asus A8N 32? and what exactally happened?

[specofdust]

It is, and he chipped his northbridge by tilting his heatpipe cooler accidentally while doing a voltmod. Don't really get that much more exact then that :|

[MAXIMILIAN]

sheety thaqts no good :stick:
I noticed that he started a new threa asking about a new sli mobo and the one recomendd is not out yet, Is it worth waiting or should i get DFi or asus

[Jailer]

I'm with craig588 concerning the heatpipe. Take it off and look at the blocks where the pipes are. They are squashed down and crammed in with some thermal past around it. Very poor design. Good idea, poor execution.

I'll post picks as soon as my coolers get here and I take the heatpipe back off so you can see what I mean. It's going in a Lian Li V2000 so I had no choice but to replace it. Unfortunately the spacing of the mounting holes and proximity of the video card limits your choice in a cooling solution.

[MAXIMILIAN]

oh i am going for the exact same case, wil this mobo fit?

[HKPolice]

Sorry to hear about your loss craig. You said that you had to replace all the mounting clips for the heatpipe? I think that was the mistake, because the default clips are really tightly sprung so they apply a lot more pressure which would make it really hard to apply any damaging amount of torque.

Quite franky I don't think the chip caused it to die because I've seen tons of chipped mobos/video cards/cpus and they were all functional, but of course it would depend on how serious the damage was.

What exactly did you do for the VTT mod, and does it help with the warm boot issue? I'd like to try it if it's not too complicated.

[jasonh1234]

I am using a watercooler for my CPU cooler.
a 120mm fan is blowing through the radiator from the back of my Lian-Li V1000.
How essential is that little optional fan in my situation?
http://www.asus.com/999/images/products/744/optionfan.jpg
From Asus' site...
Optional Fan (for Water-Cooling or Passive-Cooling only)
Maximize performance and minimize noise!
The optional fan is specifically designed to provide sufficient airflow over the CPU power modules and chipset area when water-cooling or passive-cooling is utilized, ensuring effective heat dissipation for the entire system.

[HKPolice]

Software hardware monitors are CRAP.

Plugged my CPU back into the DFI to take some DMM readings for vcore. Results:

BIOS setting 1.475*110% = Theoretical 1.6225v.
Actual reading: 1.64v
Bios/CPU-Z/HMonitor reading: 1.57v

BIOS setting 1.55v*110% = Theoretical 1.705v
Actual reading: 1.74v
Bios/CPU-Z/HMonitor reading: 1.69v

Now this makes sense - the warm reboot issues were caused by the lower actual vcore I was running (1.61v vs 1.64v) on the Asus.

The DFI software reading was way off so I assumed that 1.61v Actual reading off the Asus was enough to ensure stability.

I don't think my board is the only DFI that reports way lower vcore - I think it was intentionally designed this way. Which would explain why the new Expert and DFI CF boards 'require more voltage' to run stable than the old DFI NF4 SLI

[stone_cold_Jimi]

FYI:

On my board, vdimm is:

BIOS: 2.80v
AiBooster: 2.85v
Metered: 2.813v

So the BIOS setting is closer to the truth than I'd thought

[jasonh1234]

Finally got my computer finished last night.

Got windows installed, motherboard drivers, videocard drivers, directX 9, X-Fi drivers.
Then installed FEAR. Configured FEAR to run at highest possible settings.
It froze my machine trying to run the game with SLI enabled.
With SLI disabled I was able to load the game and look around for a bit but eventually froze. CPU Temp was between 45-50 in the bios at boot.
Don't know the room temp but it's generally the coldest room in the house.
All bios settings were at default.
Went to bed at 4:44am

Not sure what's the issue here. Maybe a patch is needed for FEAR? I guess I'll try another game when I get home from work.

Hardware involved:
A8N32-SLI (obviously)
FX-57
Lian-Li V1000
Enermax Liberty 620W
x2 EVGA Geforce 7800 GTX KO's 256 MB
74 GB SATA Raptor

[jinu117]

Software hardware monitors are CRAP.

Plugged my CPU back into the DFI to take some DMM readings for vcore. Results:

BIOS setting 1.475*110% = Theoretical 1.6225v.
Actual reading: 1.64v
Bios/CPU-Z/HMonitor reading: 1.57v

BIOS setting 1.55v*110% = Theoretical 1.705v
Actual reading: 1.74v
Bios/CPU-Z/HMonitor reading: 1.69v

Now this makes sense - the warm reboot issues were caused by the lower actual vcore I was running (1.61v vs 1.64v) on the Asus.

The DFI software reading was way off so I assumed that 1.61v Actual reading off the Asus was enough to ensure stability.

I don't think my board is the only DFI that reports way lower vcore - I think it was intentionally designed this way. Which would explain why the new Expert and DFI CF boards 'require more voltage' to run stable than the old DFI NF4 SLI


So what does 1.55*113% mean :P Probably 1.8v or so.... I guess time to check v-core matching on Asus board before I get to have too much fun.

[wydeboy]

just been given 901 bios by asus, not sure how it performs yet, if anyone needs it give me your email

[specofdust]

Is anyone else having trouble figuring out which of the two seem more problematic, the A8N32-SLI or the DFI-NF4 range? Whats a boy to pick :(:confused:

[jinu117]

For me DFI was nothing but trouble. So far A8N32 hasn't. Thing is, I haven't pushed it far enough yet or used long enough to say something about this.

[LowRun]

CPU Temp was between 45-50 in the bios at boot.

If this is somewhat accurate then i would say this is your problem.

[smurfmods]

just been given 901 bios by asus, not sure how it performs yet, if anyone needs it give me your email


Any way you can post it here?? If not... you got PM!

Ideas on what may have been changed or fixed?

[MrBean]

Not sure what's the issue here. Maybe a patch is needed for FEAR? I guess I'll try another game when I get home from work.


@ Jasonh1234: Make sure you have latest FEAR patch, then in config switch off Soft Shadows, and FSAA. I had exactly the same problem with Abit AN8 Fatality, 4800+ and BFG 7800GTX OC's in SLI.

Now game is much better, everything set to high, except 2 mentioned above - oh yeah, dump 81.85's, and load 81.95 GFX Drivers.

Soz for off-topic guys :)

If this is somewhat accurate then i would say this is your problem.

This temp is fine as read within BIOS, when your system is in Bios, cpu is loaded around 90%. YOu will see this drop by a good 5-8 deg C when in OS. Nothing wrong with 45-50 Deg C temp on that cpu.

[stone_cold_Jimi]

Tomorrow my 146 arrives, all things being equal. I need to know 2 things to proceed:

1/ Anyone know of a testpoint on the board for vcore? Or is it a case of poke around and see?

2/ Seal string. What the bally heck is it called in Blighty? Any brand names?

tia

[specofdust]

Has anyone found anywhere in the UK that has the A8N32-SLI in stock/has declared a date for getting them in?

[stone_cold_Jimi]

Has anyone found anywhere in the UK that has these in stock/has declared a date for getting them in?

Mine's coming from CPUCity - they just got a batch and I had one on order. They seem to think there's another batch due, because they said if I didn't like the stepping, I could wait for the next ones.

Or did u mean the mobo? I got mine from the same place.

[MAXIMILIAN]

Is anyone else having trouble figuring out which of the two seem more problematic, the A8N32-SLI or the DFI-NF4 range? Whats a boy to pick :(:confused:

http://i20.photobucket.com/albums/b206/190vincent/zzwithstupid.gif

[Jailer]

jasonh1234 you are going to have to ditch that heatpipe cooler if you have it mounted in a Lian-Li V1000. It won't work with the board upside down and is very likely the cause of your problems.

[Marios]

L1 and L2 Cache ECC is dissabled by default.
I think this should be enabled.
The question is what are the defaault L1 and L2 values for my AMD Athlon64 x2 4400+?

[wydeboy]

just killed my 146 optie on this board, i think tightened block on naked die too hard...nice burnt smell. mobo survived tho