Asus A8N-SLI w/ X2 VCore Bug Confirmed. . .

H

Hurin

2[H]4U
Joined
Oct 8, 2003
Messages
2,410
Hi,

Okay, I've been going around saying that my Asus A8N-SLI Premium motherboard didn't suffer from the vcore problem where (with an X2 processor) the vcore settings basically had no effect except to lock the vcore into one of two settings.

Well, it turns out, I actually do have the vcore problem. Though it doesn't seem to fit that description exactly.

What follows is a chart. First is the vcore setting in the BIOS. Next is the vcore reported by the hardware monitor in the BIOS after a restart. The next data entry is which reading is favored where the reading fluctuates (or, in other words, if the reading fluctuates between 1.40 and 1.42, which one is more prevalent). Finally, any comments or notes from me.

Here goes:

ASUS A8N-SLI Premium v1.02 and BIOS 1005

1.3625 = 1.37 Solid
1.3750 = 1.39 Solid
1.3875 = 1.39 Solid
1.4000 = 1.39-1.40 1.40 heavily favored
1.4125 = 1.40-1.42 1.40 mildly favored
1.4250 = 1.39-1.40 1.40 heavily favored Note that this is a move downward.
1.4375 = 1.40-1.42 Evenly split
1.4500 = 1.40 Solid
1.4625 = 1.40-1.42 Evenly Split
1.4750 = 1.50-1.52 1.5 heavily favored JUMP FROM 1.42 to 1.5!
1.4875 = 1.52 Solid
1.5000 = 1.52-1.53 152 mildly favored.

As you can see, there is a huge jump there in reported vcore. And a vcore setting between 1.42 and 1.49 seems impossible. Which is too bad, because that's what I need.

That jump directly to 1.5 costs me about 5-6c in load temps, which is not acceptable.

So, take these numbers as you will.

Best Regards,

Hurin
 
If your load temps really are only 47C - you've got TONS of room to go, so why would 5-6C matter?

On air, 60C is a pretty workable max (I mean, the ACTUAL max is something insane like 98C or so). I regularly ran AthlonXPs around 70C, etc.

As long as it's stable, ya know?
 
Doing some further research on this, there is some evidence that the vcore is being set properly, but the current vcore is not being reported properly. At this point, I'm not sure either way.

H
 
dderidex said:
If your load temps really are only 47C - you've got TONS of room to go, so why would 5-6C matter?

On air, 60C is a pretty workable max (I mean, the ACTUAL max is something insane like 98C or so). I regularly ran AthlonXPs around 70C, etc.

As long as it's stable, ya know?
Click to expand...
Perhaps, in winter. . . but check out my other thread. I don't have air conditioning and my computer room has been getting up into the high 80s every day this week. I don't want to have to watch the thermometer and adjust my OC accordingly.

I read around here that 65c was the danger area for X2 CPUs. With the vcore reporting 1.5 (set at 1.475), the load temps reached 60c. That's too much, and it makes me slightly bitter that I just need another .02 but am instead being forced up .1. :(

H
 
my best advice for you is try the new beta bios. If that doesnt help you then accept the fact that mb's have their quirks. If you are still unhappy than invest in watercooling and you wont need to wory about your temps.:)
 
Well, I don't think that this is a vcore reading bug altogether. However, I do think that more vcore is being put out with each increment and that they aren't necessarily going down or fluctuating as much as reported in my numbers above as the vcore setting is increased.

What does seem apparent to me, however, is that huge gap where the vcore suddenly shoots up. CPU temps raising 5-6 degrees at full load indicates that there is a big jump there. . . And the only thing that can really account for it in my mind is an odd jump in voltage far beyond the actual settings change.

I'm currently running at 1.4625. It is the last one available until the big jump to higher vcores. According to my numbers, it really shouldn't give me any more stability than 1.4375 since they both report as 1.40-1.42v. However, under load, 1.4625 appears to hold at 1.41 more often than does 1.4375 which dips to 1.39 a bit more often. So, I think there may be some stability gain there (and, more importantly, a measurable actual vcore change).

I left the computer running 24-hour dual-prime95 tests. This failed yesterday at 1.4375. Hopefully, it'll make it the 24 hours at 1.4625. I'll know when I get home tonight in about 5 hours.

If this, too, fails, I'll start trying newer BIOSes. However, word is that so far, none of the BETA or release BIOSes have had any affect on this problem.

Best Regards,

H
 
Im sure they are aware of this and hopefully a future bios release will rectify this problem. try watercooling its virtually silent and with a minumum investment can do loads to ease ones mind.:)
 
hurin you can buy a good hsf and it can cost you a few bucks. depending on the fan it can be quiet or loud. invest in watercooling and 95% of the time any mb or cpu change is transferrable and the real bonus is you can crank up the vcore without worry.Plus you have your cake and eat it too.
 
Water-cooling isn't an option for me. But I appreciate your advice. (No, I don't want to debate why I don't want water cooling! :D).

I'm hoping that my 24-hour run will succeed today. Then I'll leave it alone until a new BIOS is released.

H
 
Im on your side and we all want our rigs to run right.I hope you will find a happy medium to this solution! peace out my friend. by the way nice rig rig you have.
 
hurin i run an old system. my mobile claw gave me fits (besides my dfi board) lots of frustration. old hardware and plenty of cursing! she wouldn't go past 2.5 and be stable without major vcore. air wouldn't cut it.I watercooled it and temps were real good. went further and pumped more vcore and got her stable 40 hrs prime @2.64 with max load 45degrees at are you ready 1.75 vcore. I run f@h 24/7 7mos now. watercooling saved the day.
 
you have a great board and sometimes they need a little help. invest in watercooling you wont be sorry.I wished I had did it earlier. I'm a believer. real nice and if you do it yourself you get more pride.
 
oh and my temps are a single dtek 120 heatercore that also cools my video card.you cant beat that with a stick.ambient room temp is about 28 degrees. no ac in my home.
 
Well, my Prime95 crapped out at 17 hours this time. So, I just need a nudge of voltage. Unfortunately, I'm at the point where it pops up to 1.5 and raises temps by 6c.

I tried the new BIOS (1007 non-beta). It does not address the problem. Behavior remains identical:

@1.4625 vcore is read as 1.40-.142
Up one notch. . .
@1.4750 vcore is read as 1.50.1.52 and temp increases 5-6c.

Just info for anyone else out there.

H
 
Thanks, I had heard rumors of that. But, unfortunately, I tried it, cleared my CMOS via jumper, but the vcore behavior remains exactly the same.

I'm concerned that this may not be fixable by a BIOS update. Perhaps that is why Asus is so quiet about it.

H
 
I have a deluxe SLI version of this boared and i definetly notice vcore issues. My 4200 X2 has a default of 1.3 volts. I can increase the vcore in the bios up to 1.36 succesfully but it does not increase beyond that. The new 1013 bios limits the max to 1.45. At that voltage the core still reads 1.36. I believe this is correct due to my observations with overclocking, its a no go unless i use AI Booster in windows which allows you to adjust the voltage beyond 1.45. Selecting 1.475 with AI tool results in a vcore reading of 1.48 and success in overclocking is achieved. I've called tech support regarding this and they told me "Asus does not support overclocking". I want to return the board now and perhaps get an MSI. The chip will run an extra 200 mhz and i'd like to get it without having that dumb as AI Booster running at startup, its gay!
 
So will most resellers RMA this board as "defective" due to the vCore issue?

Thanks,
BoB
 
BoB-O TiVo said:
So will most resellers RMA this board as "defective" due to the vCore issue?

Thanks,
BoB
Click to expand...
Well, I'll need to pay 15% restocking when I ship it back next week. I mean, it runs perfectly within spec. It's only a dud for overclocking.

H
 
This is seriously getting on my nerves.

Just swapped froma DFI Ultra-D to the Premium as I needed the extra SATA slots.

Now only will the Silicon RAID not allow a single JBOD, I have now found that I am not able to overclock the X2 4400+ to the previous 2.4ghz that I was running on the DFI as I'm not able to select and run 1.4v that the CPU needs for the overclock.


Not a happy customer at all, worse of all is the fact that on the actual Asus Forum, no one from Asus seems to have any input or infact care.

It seems to me that Asus should fix this problem, or take away their overclocking motherboard claims, at least DFI will acknowledge and do their best to fix any problem that arise.
 
Ch!lledBudwei2er said:
This is seriously getting on my nerves.

Just swapped froma DFI Ultra-D to the Premium as I needed the extra SATA slots.

Now only will the Silicon RAID not allow a single JBOD, I have now found that I am not able to overclock the X2 4400+ to the previous 2.4ghz that I was running on the DFI as I'm not able to select and run 1.4v that the CPU needs for the overclock.


Not a happy customer at all, worse of all is the fact that on the actual Asus Forum, no one from Asus seems to have any input or infact care.

It seems to me that Asus should fix this problem, or take away their overclocking motherboard claims, at least DFI will acknowledge and do their best to fix any problem that arise.
Click to expand...


You should be able to pick 1.4v. At least I can, by picking 1.40V. At least that voltage seems to work as advertised.

BoB
 
BoB-O TiVo said:
You should be able to pick 1.4v. At least I can, by picking 1.40V. At least that voltage seems to work as advertised.

BoB
Click to expand...

When you select this voltage, does cpuz reflect the change? I too can select voltages up to 1.45 with a dual core installed but the max cpuz, AI booster, etc report is between 1.36-1.38. It never increases beyond that, even with 1.45 vcore selected in the bios. Only AI booster set to 1.475 will give me a voltage of 1.48. The crux of the problem is this; 1.48 volts is too much, 1.38 is not enough.
 
I'm in the same boat. Running my 4400+ at 2.6 and can only get 1.475 to work. This is too much, as it runs 60c under full load.
 
superfire said:
I have a deluxe SLI version of this boared and i definetly notice vcore issues. My 4200 X2 has a default of 1.3 volts. I can increase the vcore in the bios up to 1.36 succesfully but it does not increase beyond that. The new 1013 bios limits the max to 1.45. At that voltage the core still reads 1.36. I believe this is correct due to my observations with overclocking, its a no go unless i use AI Booster in windows which allows you to adjust the voltage beyond 1.45. Selecting 1.475 with AI tool results in a vcore reading of 1.48 and success in overclocking is achieved. I've called tech support regarding this and they told me "Asus does not support overclocking". I want to return the board now and perhaps get an MSI. The chip will run an extra 200 mhz and i'd like to get it without having that dumb as AI Booster running at startup, its gay!
Click to expand...

exact same problem I have. I cant get anything over 1.36 on the vcore no matter what I put in the bios. Called tech support, "we cant help you overclock, sorry". I guess asus+x2= no overclocking. Shame, I really like the board otherwise. Back to newegg you go, vile motherboard.
 
So from what I've read, if I'm using this motherboard with an X2 4400, I should have no problems unless I am doing serious OCing?
 
freeman2k6 said:
So from what I've read, if I'm using this motherboard with an X2 4400, I should have no problems unless I am doing serious OCing?
Click to expand...

X2's run without issue on this board provided you have bios update 1008 or later. But there are voltage nuance's that make it less than optimal to get an extra 200 mhz out of the cpu's. I'm starting to believe there is a conspiracy in place to sell more 4600 & 4800 X2's by preventing convenient OC'ing of the 4200 & 4400's. But i also believe in underware-stealing gnomes...
 
EnderW said:
mine works great :)

Click to expand...

I've got similar results with mine, my question to you is, are you using ai booster to select the 1.48 vcore you're running? Mine runs great too, but to get to 1.48, I use ai booster.
 
White94Cobra said:
I'm in the same boat. Running my 4400+ at 2.6 and can only get 1.475 to work. This is too much, as it runs 60c under full load.
Click to expand...

With two instances of prime95 running after about 8 or so hours, I hit 65c running at 11x227(2500). Using the xp90 with medium panaflo. not sure if it's entirely healthy for the cpu but it runs well. I normally run 11x220(2.4ghz) and all is well.
 
BoB-O TiVo said:
What BIOS and which board (Premium or Deluxe)?

BoB
Click to expand...
BIOS is 1005 I think, final, not Beta.
Premium.



jonesmaster said:
well hey good for you
Click to expand...
Thanks.



superfire said:
I've got similar results with mine, my question to you is, are you using ai booster to select the 1.48 vcore you're running? Mine runs great too, but to get to 1.48, I use ai booster.
Click to expand...
No, I selected 1.5v in the BIOS. I think the "bug" has to do with the reporting, because at 1.475 I was Prime 95 stable for about 4 hours, increased to 1.4875 and it was stable for about 16 hours. Finally I went to 1.5v and as you can see, it was stable for 39 hours.



superfire said:
With two instances of prime95 running after about 8 or so hours, I hit 65c running at 11x227(2500). Using the xp90 with medium panaflo. not sure if it's entirely healthy for the cpu but it runs well. I normally run 11x220(2.4ghz) and all is well.
Click to expand...
That's about 18 C higher than my temps at 2.6Hz.
I'm using a XP-120 with Panaflo M1A.
Something is wrong somewhere.
 
To EnderW in response to

"That's about 18 C higher than my temps at 2.6Hz.
I'm using a XP-120 with Panaflo M1A.
Something is wrong somewhere"

I'm using the deluxe version of this board, latest non beta bios is 1013 and limits voltage selection in the bios to a max of 1.45. Older revisions of the bios would allow selection of higher voltages via the bios. That's why I'm stuck using ai booster to select higher voltages. I don't want to roll back, the newer bios eliminates "unknown hardware" in Win 64 (testing done in in xp-32) As far as my temps go, perhaps the HSF is not seated properly, i'll investigate when i get home. The temperature delta between yours and mine does seem high, despite the fact your using an 120mm fan.
 
superfire said:
With two instances of prime95 running after about 8 or so hours, I hit 65c running at 11x227(2500). Using the xp90 with medium panaflo. not sure if it's entirely healthy for the cpu but it runs well. I normally run 11x220(2.4ghz) and all is well.
Click to expand...

Yeah, that's really strange. I'm using an XP90 with SLOW speed Nexus 92mm. I just tickle 60C running 10.5x250(2625). I'm going to switch to the low speed Panaflo which moves swice the air for 7dB increase. If that keeps me in the 50's I'll call it good.

You might want to check your thermal paste application.

BoB
 
superfire said:
To EnderW in response to

"That's about 18 C higher than my temps at 2.6Hz.
I'm using a XP-120 with Panaflo M1A.
Something is wrong somewhere"

I'm using the deluxe version of this board, latest non beta bios is 1013 and limits voltage selection in the bios to a max of 1.45. Older revisions of the bios would allow selection of higher voltages via the bios. That's why I'm stuck using ai booster to select higher voltages. I don't want to roll back, the newer bios eliminates "unknown hardware" in Win 64 (testing done in in xp-32) As far as my temps go, perhaps the HSF is not seated properly, i'll investigate when i get home. The temperature delta between yours and mine does seem high, despite the fact your using an 120mm fan.
Click to expand...

I removed my HSF (XP90) and found too much arctic silver. I removed it all from the cpu and HS and cleaned with isopropal alcohol. Re-appled a thin coat of AS with a razor blade to the top of the cpu only. I also remembered the HS bracket i installed may have had too much thread protrusion and was bottoming out under the mobo, so i replaced the mounting screws with shorter, more suitable ones. After re-installation of the HSF I seem to be topping out dual instances of prime95 at 55c, an eight to ten Celcius drop. I would have done it sooner but wasn't convinced that it may be a problem until you guys shared your experiences, thanks! Still hoping for a bios update for the vcore issue though, my extra heat has nothing to do with what I've observed.
 
something you can try to lower temps on this mobo if your running an XP-120 is screw 2 120mm fans together as the clip only catches the bottom lip of the fan ...have them both blowing into heatsink as well

..dropped my idle temps from 38c to 27c ..both fans are just generic 120's too... :)
 
Just to add my experience with this board...

I am able to get four different Vcore values as follows:

Set BIOS= Auto, Vcore = 1.3V
Set BIOS<1.475V, Vcore = 1.36V
Set BIOS>=1.475V, Vcore = 1.475V

If I use 1.475V and suspend to RAM and then come out of sleep mode, the Vcore is 1.425. :(

So it seems the Vcore problem manifests itself differently for different folks.

Only BIOS 1005 and 1006 even offer Vcore BIOS settings above 1.45V correct?

I have been able to get a stable 2.6GHz overclock on my 4400+ at 1.475V with peak temps under dual prime at 55-degrees(case temp around 35-degrees). Idle is case temp +4 degrees.

Even if I had a wider range of Vcore values, I'm not sure I would use them. Perhaps I could squeak another 100MHz out of my CPU but it's not going to make enough difference given the added heat, etc.
 
People need to check out the new 1008.3, although the vcore isnt rock solid at what you set it at, it's at least now possible to get roughly the vcore you were after!
 
You must log in or register to reply here.
Back
Top