Is There a Fix for AN7 VTT Tracking Issue?

[hovo73]

Hi, is there a fix for VTT Tracking issue on AN7?

I have good cooling on the CPU, NB, SB, VDIMM, Mosefets and good case airflow, it still doesnt' work kright.

The pic below was taken while I was running P95. VDIM was set at 2.7V in BIOS, and for as long as I was looking at it the VTT was tracking at ~1.2V. And the VDIMM is not even higher than 2.9V.

And I am having great difficulty in going higher than 210.

http://www.xtremesystems.org/forums/attachment.php?attachmentid=30944

[hovo73]

bump ... ?

[cetoole]

Tell you the truth, I am not sure there is a real problem, but the problem might be that MBM5 doesn't read the vtt correctly. I have done some testing with my DMM (with Hellfire's guidance), and havn't been able to find a readpoint that corrisponded with the MBM5 values. Now, when I put an 80mm fan over my mosfet heatsinks, MBM5 read vtt as much higher than before, about where it should be.

[hovo73]

Hi cetoole, thanks for your input. I believe we have exchanged a couple of messages regarding this issue on EOCF a while back.

I installed really nice heatsinks on the 6 mosefts near CPU socket. I made them myself from a stock AMD HSF. They are really beeffy and high and I have attached them using Arctic Ceramenique (or was it Alumina, dont' remember exactly) adhasive. The 92mm Panaflo H that is on CPU HSF (SLK-947U) is pusing considerable air over them, and they are really cool to the touch.

True, you can't trust MBM5 too much, but I am having real problems going over 210 FSB and this seems to be the only good explanation, given the fact that many of these boards have that problem.

What do you think?

[cetoole]

Well, you have 2x512, so you probably should be using cpc2t, and I believe that utt isn't great with NF2 boards. Have you tried higher vdimm? Also, it is worth a shot, try the L12 mod. What bios are you using?

[hovo73]

Yes, I have tried with VDIMM up to the max 3.2V, and I have have done the L12 mod.

[cetoole]

Yes, I have tried with VDIMM up to the max 3.2V, and I have have done the L12 mod.
What modded bios are you using, and is it cpc1t or 2t?

[hovo73]

It is CPC On (1T). To be exact it is this BIOS: an718de81t.zip

I got it a while ago from users from another forum with good recomendations. What is interesting I haven't seen it on one of Tictac's sites. I have seen these:

ABIT-AN7-N24AAC18-R2-1T.zip (along with 2T)
ABIT-AN7-N24AA131-R2-1T.zip (along with 2T)
ABIT-AN7-N24AAC18-RevC-1T.zip (along with 2T)

None of the match what I have (an718de81t.zip). I am also not sure what is the chronological order of the 3 BIOSes I listed (which one is the latest and the best).

Could you recommend any one of them?

[cetoole]

an7de8 is not a tictac bios, it is modded by Enduracell. I suggest trying 18de8cpc2t. If you cant find it, send me a message, I have it backed up here, but it might be a day or two before I can dig it out of that computer, because I am just switching it from the AN7 to a DFI UI. The difference between the bios you have now and the one I am recommending is just CPC setting, but since you have 2x512 ram, you should at least try the 2t bios. Most people have the best luck with the AN7DE8 derivatives, but there are some that find their highest OC using the tictac bioses you listed. I believe the bios you are using now, and the one I recommended, which is really the same thing but cpc2t are the latest. I forget which AN7 bios was modded for the three others though.

[hovo73]

Thanks cetoole.

I have these two BIOSes too:

an7de8.zip
an718de82t.zip

I think the second one is what you were referring to, right?

[Micutzu]

Guys, the hardware monitor on the AN7 has a bug (uGuru sux), the reported VTT voltage decreases as the PWM temp increases; in reality the VTT tracks perfectly at 1/2 VDIMM (tested up to 3.5V) and is very stable. Did i say that uGuru sux ?

[cetoole]

Guys, the hardware monitor on the AN7 has a bug (uGuru sux), the reported VTT voltage decreases as the PWM temp increases; in reality the VTT tracks perfectly at 1/2 VDIMM (tested up to 3.5V) and is very stable. Did i say that uGuru sux ?
Everything I have found, from the voltage readings I took on the mobo matching almost perfectly at every vdimm value to the mysterious increase in vtt readings via mobo sensors when cooling the mosfets with a fan, so this sounds like the truth to me. I also found that it is possible to have the mosfets cool enough that the reported vtt is well over 1/2 vdimm with nothing more than a low speed fan and some aluminum heatsinks. Hovo73, the second bios you list is the one I am suggesting. I havnt found anything that the uguru did for me other than provide headaches. Good luck with this board, it is a real PITA. Micutzu, :toast: