So many A64 deaths

Quote:

---

Originally posted by BMORIN
Well, that the first I've heard of a "damaged" mem controller. I've heard of various incompatibility issues and that's all. Shim, you need to try your chip in a different model mobo and/or try a different chip in your current mobo. In my opinio, it's worth buying a piece of hardware then paying a 15% restocking fee just to get it checked.

---

hmm... i see...

maybe i do need to try another board and another chip...

well.. i'm waiting for the dfi so i'll try it on that one for sure... and if i still have the same problems, then i'll get another chip...

but the chip works perfectly fine at stock settings though...

weird

People may be you can send me your broken CPU's ? For hardcore experiments

has anybody noticied if there are more C0 A64 deads than CG? i'm going to mod the vdimm directly to the 3'3 atx pin because I don't want to mod my 3'3 psu rail (my vdimm max is 2'93v).
Although I think my mem controller is already damaged... I can't boot with 2 sticks of memory pluggeds at sync mode (and async freezes at 230 htt), but with 1 stick i can do anything... (350 htt if i want). This problem appeared since 1 month.

I have killed my A64 3000+ NC CG. :( Got a 3200+ last weekend. Yesterday it did not boot into Bios. I had Vmod the Abit KV8 Pro VTT and Direct 3.3v Rail connected to VDimm. It was unstable. The next thing it won't boot. error No. 04 on the mobo.

Little question guys ,How much Vcore can the A64 handle before it breaks down ?
I dont increase Vdimm cos i know its bad for the CPU afther reading this treath...
I am using 1.7Vcore to get 2700Mhz and its running fine ,but for how long with only vcore mod ?

Don't worry about the vcore or the vdimm, it's not so much an issue now as it was with some of the early chips and early mobos. I still wouldn't push vdimm past 3.6v. I have run 3.6v on the vdimm and run about 3.3v 24/7. I keep my vcore at about 1.925v

I've bought ocz ddr booster and i'm running at 3'3v for 6 days... still no problems. (3400+ C0 year 2003).

This issue doesn't seem to be a problem with 250gb, CG stepping and so on.
I have run several hours of testing@3,7v and 24/7 on 3,4v, so I guess it's not a problem anymore.

so you people are saying i shallnt be worried of my a64 dying? ive got a gig of mushkin and a dfi UT on the way and i was thinking of getting a 2800 35w mobile chip. my 3.3rail is at 3.4 ish. should i be worried much about deaths? what vcore is considered "safe" meaning it wont kill the chip in a couple months time?

ok guys i think im going to be joining the dead a64 club im afraid... i was ocing my 2 day old DTR3400 proven 2900mhz cpu last night and i was trying to get up into the high ocs like the last guy was getting with the chip. well i was at 240x11 1.55v at 113% when i seen it wasnt stable so i went back into the bios and jacked it up to 1.55v @ 123% on the v-core and 3.1v on the memory and just for good measure i added 5mhz to the fsb! well i hit exit and save and it started to post it got past the memory count past the HD screen then when it was suposed to load windows it just went blk :shrug: nothing i do helps... soo does this sound like i smoked the memory controler or is there hope? im thinking i may have scrambled the bios but i got this 280 dollar feeling the chip is dead.... anyways what do you guys think?

peace perc,

Quote:

---

Originally Posted by htkm1

I have killed my A64 3000+ NC CG. :( Got a 3200+ last weekend. Yesterday it did not boot into Bios. I had Vmod the Abit KV8 Pro VTT and Direct 3.3v Rail connected to VDimm. It was unstable. The next thing it won't boot. error No. 04 on the mobo.

---

exactly the same happend to me. i had my vtt at 1.65 and vdimm at 3.28, which would also comfirm what saaya said.

perc: so it doesnt boot at all? what about the psu or memory? is evap making good contact? you may need to remount it esp. on a DTR

Quote:

---

Originally Posted by Perc

ok guys i think im going to be joining the dead a64 club im afraid... i was ocing my 2 day old DTR3400 proven 2900mhz cpu last night and i was trying to get up into the high ocs like the last guy was getting with the chip. well i was at 240x11 1.55v at 113% when i seen it wasnt stable so i went back into the bios and jacked it up to 1.55v @ 123% on the v-core and 3.1v on the memory and just for good measure i added 5mhz to the fsb! well i hit exit and save and it started to post it got past the memory count past the HD screen then when it was suposed to load windows it just went blk :shrug: nothing i do helps... soo does this sound like i smoked the memory controler or is there hope? im thinking i may have scrambled the bios but i got this 280 dollar feeling the chip is dead.... anyways what do you guys think?

peace perc,

---

me hopes your wrong perc, I've slammed the 1.33% into the chip and whatever's after that. And no problems. I've not seen the A64 problems like they used to be, Is it still very common place??

I've been pushing 1.92v into this 3000+ newcastle you sent me back perc running It @ 2.8ghz And 3.1v to my Ballistix memory. No problems running 24/7 F@H non stop. I know it's running nice and toasty, No load my Vapo LS will show -57c, loaded right now with F@H running and plenty of volts it's down to -39 Celsius Which means i have really good contact. The 3400+ that's taking a nap on you right now was a little harder to get good contact i could tell.

unloaded temps were the same -57, but 100% cpu it only would pull down to -48 Celsius on the front panel. I had problems getting good contact with it. I used to have to pull the Vapo Hose to the left to get better contact and zip tie it a little..

Again Perc, i hope this chip wakes back up for you in the morning, I've seen many a chip of mine just somehow come back to life!!! And dam that was a good chip. I ran it @ 2.65ghz with a Zalman ALCU!!!! The extra 200 mhz using the Vapo and electric was hardly worth it.

I see what you mean now that im back with the newcastle chip and 512 cache, it does make a big difference. Especially on super Pi up above 2.8ghz!!!!

I guess i'll buy us a couple more, I seem to be lucky buying good chips.

Talk to you tomorrow Perc, good luck.

later

tedidne what do you think of IHS? should it be removed, and will zalman alcu still mount?

@tedinde

yeah i think it will man. i dont think its the cpu cause when this happen to me i only made it out of the bios and past the memory count screen then it went blk. i think it may either be a scrambled bios or a cold boot issue. i got a new bios chip comming my way this week so one i hve that ill be able to start working on it. the thing is right now i have one system and its hard to work on a broken system and not have internet to go and find answers too. i need a psu and a video card before i have enough parts to run a system and work on another. know what im saying? anyways i dont think the chip is dead either. after hearing from you on how much voltage youve given this chip theres no way i killed it with 1.875v and 3.1v to the memory... ill keep you posted....

peace perc,

i want that dead chip :D

Quote:

---

Originally Posted by reject

exactly the same happend to me. i had my vtt at 1.62 and vdimm at 3.28, which would also comfirm what saaya said.

perc: so it doesnt boot at all? what about the psu or memory? is evap making good contact? you may need to remount it esp. on a DTR

---

yeah i think i need to start over with this sytem. im going to rip it all out of the case and remount the damm mounting kit and pray like hell i can get it going. the more i think about this the more im convinced its not the cpu. i think its a cold boot issue cause when i first built this system i wouldnt boot just like its doing now. i think that at auto the bios is setting the v-core at 1.04v wich i think is the cause of all this but im not 100%sure..i have a spare bios chip just need to hot flash it in my intel rig. i hope that maybe a diff bios i may be able to get jump started again... ill keep you guys posted on how it turns out. thx for your replys guys.



peace perc,

Is my A3400+ NC dead too?
Or is it something else?

http://xtremesystems.org/forums/showthread.php?t=46292

I am running my ADA3700AEP5AR 0432 at 3ghz / ~1.8v. With 3.3v @ ~3.7v, 5v @ 5.2v, 12v @ 12.4v on the PSU(OCZ520W), vDimm @ 3.5v, and I haven't seen any problems so far. I have the DFI UT and AFAIK there haven't been any problems with the DFI and these failures, right? Is this a s939 issue only?

Only problems I have had was two dead DFI boards. First because of a 10 minute blackscreen limbo, caused by a 3.15ghz suicide run. It simply froze and I wasn't around to shut it down, 10 minutes later I return, the board never posted again. Second board died because of OCZ Booster, 2.8v read by the booster, 2.8v set inside BIOS, no OC on CPU @ 1.5v - worked for 12hours stable, then booted PC and black screen, board dead. I'll never, ever use the Booster again on a DFI UT. First board RMAed, I bought the second one(didn't even tried to RMA it). Lucky for me, the CPU has survived all of this and even more.

Is it for real, when AMD says no more than 2.6v vDimm is recommended? This can't be true, and the OCZ guys said they have never seen a dead CPU from high vDimm, but I wonder about the 3.3v rails.

I have seen spikes on the vDimm, but that was from reading the BIOS' readings, and I have been told that the DFI UT' voltage readings are attrocious. Another funny thing is, that sometimes vDimm voltage readings don't change in BIOS after a bootup.

It can be at 3.5v and I change it back to 2.8v, but after reboot it still says 3.48-3.5v as current vDimm. Only after another reboot, will the reading be correct, but still fluctuating up to +/-0.2v. I have had this issue with all three DFI boards. Anyone know anything about this?

I don't think I'll ever have a problem with my 3700+, because of the 3.5v vDimm or my 3.7v 3.3v line, but I'll report back if I am wrong :p:

up....is this still an issue??

I think this issue is way over exaggerated... I personally haven't come across any...

Quote:

---

Originally Posted by A_X

I think this issue is way over exaggerated... I personally haven't come across any...

---

You probably weren't around when the Shuttle AN50R was "hot". :stick:

so has this issue been resolved? for Socket 754

personally i think i will stop at 1.65 vcore and 3.1 for vdimm....but thats just me... :)

is this still a problem?

Quote:

---

Originally Posted by Unhooked

is this still a problem?

---

If breaking CPU's with too high voltage, or by some abuse is a problem, then yes it is a problem. :D