So many A64 deaths

[`schr0et]

I'm thinking moreso that it's mainly to do with the NF3 boards. As I've yet to see anyone killing a CPU/MOBO with the K8T800 by running a high 3,3 line or high vdimm.

Correct me if I'm wrong please though

YoupY:

Your idea doesn't seem bad at all, but considering all the excess work to even get that done I would rather just kill the board and RMA it

[`schr0et]

Seems like I got a good stepping from newegg for my A64 3200+,

CAAKC week 37 wish me luck

[wimpie007]

I wish you luck
Those are quite some frightning stories, but there's not so many chance that I'll run my cpu up so high... (haven't picked up an A64 yet, but planning to...)

[xgman]

and another

http://www.xtremesystems.org/forums/...threadid=28312

[saaya]

so its a weak point in the nf3 design? guess it has to do with the vmods as mm suggests...

[Alex08]

I killed my first A64 3200+ on my an50r

i got the stupid blue screen error of death when trying to boot into windows.

I tried two different hd's, had no luck.

i had my chip at 1.66v and my vdimm at 3.6v

my vdd was kept on auto the whole time.

Im thinking its my vdimm that killed the chip.

[BMORIN]

I have 2 questions:

1) Has anyone killed one WITHOUT doing the vdimm mod?

2) Has anyone killed one on a VIA chipset?

[bias_hjorth]

1: I believe Bowman has

2: dont think so yet

:-)

[`schr0et]

MickeyMouse your idea worked perfectly, only reason my CPU died was cause of my stupidity.

I did the mod so perfectly, I was so proud of myself.

Set the VR 1.984K Ohm, booted in @ 3,17 Vdimm stable.

Then windows rebooted for some weird reason so I thought I would "fix" the problem by changing VDDR from "auto" to 2.6

ROFL did not help at all!!

Next thing I know I'm reading the mosfet and its spiking upwards 3,78 and then randomly falling and flying up again in a span of 5 seconds.

I was like "OMG NOOO STOP CPU YOU NOT SPOSE TO DO THAT!!"

But now she won't boot so I got an RMA approved to send her remains back

EDIT: Yes I'm on a K8V Deluxe K8T800 and I have a VMod on my board and I'm pretty sure my CPU is dead dead dead

[BMORIN]

PRanganathan, nice job on reading those voltages while it was happening. I think that's the first real proof we've seen all this time of a possible voltage malfunction causing this.

[Alex08]

so let me get this right.

if we leave it to auto, then it's all fine? because mine was working fine until i changed it from auto to 2.7 which really was 3.6v

i cant decide to keep my shuttle motherboard or not, i just took off the vdimm and vcore mods?

[Tedinde]

But now she won't boot so I got an RMA approved to send her remains back

So you mod the board, kill the hardware, then RMA what you killed. Did you take the mods off or do you send it back as is??

I was reading in a forum a couple days ago, one place didnt accept a guys MB on RMA, they are starting to look for known solder points i guess. They caught it, and asked if he would like to pay for return shipping on his Modded board.

Guess retailers are getting tired of this.

But good luck with your RMA.

[Karnivore]

Originally posted by Tedinde
So you mod the board, kill the hardware, then RMA what you killed. Did you take the mods off or do you send it back as is??

I was reading in a forum a couple days ago, one place didnt accept a guys MB on RMA, they are starting to look for known solder points i guess. They caught it, and asked if he would like to pay for return shipping on his Modded board.

Guess retailers are getting tired of this.

But good luck with your RMA.

Bad thing is all deaths caused my botched mods, and over stressed components, then RMA'd is going to make it damn difficult for legit returns. If ya can't afford the loss you shouldn't be doing it.....

[`schr0et]

First, removing solder = easy.

All you do is apply heat, and use a bulb to suck up the hot solder.

Easy as that.

Second my board still works fine, just the stupid CPU that died.

[`schr0et]

Oh ya I forgot my third point.

Point #3: It seems those of us with Vmods have very stable Power Supplies that don't have fluctuating voltages going thru our lines.

Many of our Power Supplies are of very high quality: PC Power & Cooling for example.

Before I even started the VDimm mod on my board, I checked my 12v, 5v, and 3,3V lines. Because my Asus Probe was reading my 12v and 5v lines very low.

Just to make sure I checked the 12v and 5v thru the molex, both came out rock stable 12v was @ 12.08V, and 5v was @ 5,04v.

My 3,3 line was perfect also coming in @ 3,31V (measured thru mosfet)

When my VDDR was in "auto" my VDimm line never fluctated, always stayed very stable @ 3,17 Vdimm. It was only when I set it to 2,6 Vdimm that my volts started going out of control, upwards 3,78 VDimm which eventually killed the onboard controller on my CPU.

It seems that while the Motherboard can handle the volt spikings for a longer period of time being showing signs of wearing down, the CPU cannot because of mem controller being inside the CPU (which of course we all know ). So the only solution to this problem as of right now, as MickeyMouse stated, is to always leave your VDDR @ "auto" NEVER EVER CHANGE IT! If you need to change your Vdimm volts ALWAYS change it thru your Volt Mod and not through your Bios.

Everything else will be peachy!

[BMORIN]

Well, it will be interesting to see if we continue to get deaths after people follow this advice.

[`schr0et]

Originally posted by Tedinde
So you mod the board, kill the hardware, then RMA what you killed. Did you take the mods off or do you send it back as is??

I was reading in a forum a couple days ago, one place didnt accept a guys MB on RMA, they are starting to look for known solder points i guess. They caught it, and asked if he would like to pay for return shipping on his Modded board.

Guess retailers are getting tired of this.

But good luck with your RMA.

Only thing I killed was the CPU, no mods were done to it.

I sent it to them with the TIM still applied on the CPU and fan, just to let them know I used the stock HSF.

I don't see why you guys are against Vmodding boards and sending it for RMA.

First off they get products at retailer prices, they still make a big margin of profits off of selling it to us.

Secondly only a very small percent of customers RMA their product, compared to the percent of people who buy the product.

I know honesty, and being a nice good boy is what most people want to see. But frankly I know that the company can afford to send me a new motherboard because the one I got didn't meet my expectations .

To be completely honest, I think the more they start receiving Vmodded mobo's the more they realize they should just give in to us and let us use higher volts thru bios.

[Overlag]

Originally posted by PRanganathan
Only thing I killed was the CPU, no mods were done to it.

I sent it to them with the TIM still applied on the CPU and fan, just to let them know I used the stock HSF.

I don't see why you guys are against Vmodding boards and sending it for RMA.

First off they get products at retailer prices, they still make a big margin of profits off of selling it to us.

Secondly only a very small percent of customers RMA their product, compared to the percent of people who buy the product.

I know honesty, and being a nice good boy is what most people want to see. But frankly I know that the company can afford to send me a new motherboard because the one I got didn't meet my expectations .

To be completely honest, I think the more they start receiving Vmodded mobo's the more they realize they should just give in to us and let us use higher volts thru bios.

overclocking IS against the warrenty and RMAing something that you've voided the warrenty on it really stealing....

im a overclocker to, but never sent something back that I broke

[`schr0et]

Thats very true, kudos to you on noting that.

But it still doesn't change my perspective on RMA'ing parts that don't meet my expectations.

Buy a car lately? If it doesn't meet your expectations would you keep it? I wouldn't, because when I pay for something I have certain criterias that must be met, if they don't meet them then the manufacturer hasn't fulfilled his end of the deal, as a result I don't fulfill my end of the deal as an end user.

[joe2004]

I wouldn't, because when I pay for something I have certain criterias that must be met, if they don't meet them then the manufacturer hasn't fulfilled his end of the deal, as a result I don't fulfill my end of the deal as an end user.

What if a manufacturer did not come close to your expectation but did not advertise as such to start with? I mean if you have bought Honda Accord expecting to do an engine modification to be able to run 250 miles per hour. And when it doesn't you think they would accept the return of a burned out car if they could prove what you have done???

Would you tell to RMA guy when you are doing RMA that the CPU/mobo died because of what has been done or you just pretend it was all by default?

I find this RMAing to be a fraud much like insurance fraud people do with cars. If this was more of a mainstream activity I think there would be laws on books to penalize it. It might be anyway but the guys that do this are lucky that manufacturers/resellers aren't interested to chase small time crooks for one CPU. In either case it is certainly unethical at least.

[Goldlocke]

this forum is called "Xtremesystems", not "RMA ethics"!
Joe and Overlag go back to http://www.community.tomshardware.com

[joe2004]

I didn't really intend to discuss ethics, just the fact that he claims, for some reason, how he is right to do the return while manufacturer "did not fulfill his expectation." Ridiculous at least.

If one expects somebody else to take the financial responsibility for one's botched hardware experiments that one ought not to expect everybody else to buy his infantile reasoning.

[hotwheelz]

@ wich point ill have to measure the vdimm with a multimeter?

ill set the vdimm in bios to 2,9v then via VR up to 3,6v

[Goldlocke]

exacty hotwheelz! look here @ OC.at

CONFIRMED!
Over @ OC.at we too got a little "A64 killer". Hotwheelz just confirmed PR's findings about highly fluctuating VDimm when set at 2.9V in bios!here he states fluctuation from 3.1V to 3.8V. Already killed a 3200+ and a 3400+

(owned by himself )

[hotwheelz]

So ill have a Shuttle AN50R

In Bios i set the vdimm to 2,9v and then with the VR up to 3,1v

when i measure the vdimm its jumping from 3,1 up to 3,8v


and my two A64 (3200+ / 3400+) wont go higher then 2500mhz with mach2 so ill killed them!