Zepto 6625WD BSOD at startup, delivered with faulty memory? And what to do?

Hello all, I have a rather odd issue with my Zepto 6625WD laptop and wonder what the best course of action would be.

Short story:

-Bought zepto6625WD laptop in late december 2007 from N4all. (which means since they dropped all customer support into Zepto.dk's hands, it is pretty much a nightmare)

-Laptop was not delivered with an OS.

-Installed Windows vista 64 bit and after initial installation all was stable and good. HOWEVER, at the next startup after a long period of shutdown (in other words, turning off the laptop for 2 hours or so) the OS would immediately present me a BSOD regarding memmory errors (various types so hard to narrow down). This means that if I reboot the computer, or shut it down and fire up again in 2 minutes, that this wont occur untill the next long shutdown.

I disregarded this error as rather harmless and or part of Vista and let it slip, however everytime it crashed it caused my computer to chunk up and slow down considerably. But since I installed windows 7 the problems continue, so my suspicions arised regarding a laptop that was delivered with a faulty memmory chip (it is a single 2 gb ram at pc5300 667MHZ.

I will be running a memtest tonight to see if it is indeed the memmory. But this arises the question:

-Does this mean that laptops without OS installed wont be even tested if there are errors like faulty memmory?

-I still have a 2 year carry in warranty NL (which meant I bought the extra year of warranty because before N4all cut off with zepto, it was rather logical choice to send it back to a dutch post adress instead all the way to denmark.) It should still be covered, however I use the laptop for projects and cannot afford it to be gone for one month or more for just one faulty memmory (which I clearly hope it is, if not then there is something wrong with other compartments that are not that easily replaced). Are there other suggestions? Not to mention I dread what to happen to the laptop if it has to travel back and forth between netherlands and denmark or at their repairshop: it might become broken or in a worse state during the process

Other thing I would like to know is what happens if your videocard might experience failure as described by nvidia's post regarding faulty chipsets. Its to my understanding that the 6625WD line is discontinued, so replacement could be either very expensive (time or money) or even impossible. What then?

Last thing: are temperature values of 60 c of idle for gpu, 90 at stress
And 50 c of idle of cpu, and 80~ c at stress normal for an zepto?

 

I asume they test the laptop with a standard hd and switch in yours when the test is done.

If it is only the memory, I would just ask them to send you a new set. You could ask if you can hop by zeptoshop.nl to let them check it out so they have some proof that the memory is indeed faulty.

The temperature is high, but not extreme. I had higher on my 6214

 

It sure does sound like a faulty mem-block, looking forward to hear the results of the memtest.

I your case I would do the memtest - if it passes this, a clean install of everything including flashing the bios could be the way ahead.

@Temperatures: I've had major issues with heat on my own 6625WD (Core Duo 2 x T7500 2.2Ghz, 4Gb mem, 8600m GT 512Mb gfx). First of all, in my oppinion your temperatures are excellent! Be sure to have the 015 version of bios on the system - it have quite improved fan control, though a bit more noisy.

The Intel chipset will take use of down throtteling of your CPU if it becomes too hot. For me, this ended up with the need for Undervolting (_the_ guide: The "Undervolting" Guide). I'll be happy to pass my settings for the undervolting issue, if this turn up as a problem to you - that is if your specs are similar.

Atm, I run with 38*C idle CPU's and 66*C GPU with bios v.015 and ForceWare 179.28 driver. At full load (Orthos or Prime95 torture test) I can push my temps to ~91-92*C CPU and 93*C GPU, but I've had very different temps with different drivers!

Let's hear about the memtest

 

Update 2:

I used orthos this time and the values were alarmingly high, 50 celcius idle, but 96~97 celcius at max stressed, yikes! Trying undervolting now and reducing it to 82 Celcius. [at 1.0625 v right now at 11x, any lower and I get warning from orthos]

I wonder what kind of settings you all have with rightmark cpu clocking, and what kind of values you have usually for the temperatures of the gpu/cpu after tweaks and so forth.

And above all, thanks for the input and help, really appreciate it ^^

[regarding the memtest: something came up so planning it doing it tonight]
[regarding bios: updated it too now at the most recent ^^]

 

Okay, just ran a thorough test on the CPU's. 91*C stable temp after 2½ hours Prime95 test. Idle temps at ~40*C.

I've used RMclock to undervolt my system and uses following settings:
(be sure to follow the guide linked in above post!).

Index - Type - FID - VID
0 - SuperLMF - 6x - 0.85V
1 - Norm - 6x - 1.10V
2 - Norm - 7x - 1.10V
3 - Norm - 8x - 1.10V
4 - Norm - 9x - 1.10V
5 - Norm - 10x - 1.10V
6 - Norm - 11x - 1.10V

I've left 7 unchecked. This is a solid and stable build and enables sufficient performance and nice temperatures for me