6625wd BSOD x3 without a trace?

So heres the case, 3 BSODs in 24h wich is wierd. Even worse is that Sytem log doesnt say anything else than "The last system shutdown was unexpected". Nothing before that...

So I installed the 1.81 Realtek driver 2 days ago, but I now uninstalled that one, and got a BSOD after installing the 1.80 and restarting the computer. 2 of the BSODs have said something about TCPIP.sys and bad driver, but when the third happened I werent here.

Heres the numbers:

BCCode: d1
BCP1: 00000010
BCP2: 00000002
BCP3: 00000001
BCP4: 8E91BA05

BCCode: d1
BCP1: 00000010
BCP2: 00000002
BCP3: 00000001
BCP4: 8E69BA05

And the third is unrecorded.
So, Ill see if I get any more of the BSODs after I installed the 1.80 and report more later...

 

have you ran memtest yet?

 

Are you running RMClock?

I disabled RMClock's management a few days back and haven't had a BSOD or crash since.

It seems that RMClock's Santa Rosa compatibility isn't as good as I thought (but is now under development) and this has become an issue since Zepto got the Santa Rosa features properly working.

John

 

memory is fine.

About RMCLock, how come I havent got any BSODs before this then? I mean, im not even using the 011 BIOS. It might be that I recently updated to 008 from 004 but that was almost a week ago, and no BSODs last week.

But I havnt got any more BSODs yet

 

It took me a while before I started to think that RMClock would be the cause of my problems because they didn't start immediately on moving to a SR-enabled BIOS. Anyway, I'm now up to over 90 hours run time without a crash. I did get an unexpected dose of a very noisy fan yesterday when an application went berserk.

I recommend you disable RMClock's CPU management and see how you get on. I hope it won't be too long before RMClock's SR compatibility is fixed (the developer has now got some SR hardware).

John

 

Ok, Ill do thet if I get more BSODs. But how will the undervolting go then? Ill get a hot lappy then

 

You will either have to live with the heat or do without the Santa Rosa features (might be able to disable these in the BIOS) until the next version of RMClock is released (I hope a beta version will appear quite soon).

Anyway, it will be useful to determine whether RMClock is also the cause of your problems.

John

 

I don't have any problems with RmClock on Bios-008

I lowered the max volt from 1.250 to 1.025, and temperatures dropped from 85c to 65c.

But with Bios-011, RmClock will not work. Wrong FID/VID, temperature, CPU load etc. Wait for a newer release

 

Hmm, ill try putting the Volt a bit higher and disabling some of the C-states

btw n0elia, do you have the latest version of RMCLock? It has got pretty much more functions than the old one

 

I have ver. 2.30.1, the latest

 

I get BSOD with newest bios for 3415w running RMclock as well.

 

i suppose that when I checked "enable c4 on c3 command" caused the BSODs, after disable, no BSODs. This would also explain why I nevr got a bsod when I was gaming

 

RMClock 2.35 beta 1 is now posted but seems to be a bit buggy.

John

 

Thanks for the RmClock link, been waiting for this!!! Now gotta test the beta

 

Hmm, it seems to be working fine here.
If anyone's interested here are my FID/VID:

6.0x = 0.9375V
7.0x = 0.9500V
8.0x = 0.9625V
9.0x = 0.9750V
10.0x = 1.00V
11.0x = 1.0125V

Index 0 (SLFM) shows as 8.0x <-- What? Shouldn't it be 5 or 6? EDIT: it IS 8.0x, because 8 * 100 = 800MHz, sry

Haven't installed the new BIOS though, so have not tested SLFM. The reason I don't do this is because of the fan noise in the new BIOS

 

Got the beta too, minimum VID dropped from .935 or something like that to .8500

rly nice

 

That's SLFM (Super Low Frequency Mode), so it's not the minimum VID. Your 6.0x wont be stable at that V

You also need the new BIOS to use that VID, or RmClock will crash

 

I have 011 for the 6625, and yes it is stable at that voltage (VID), Iv been off for about 2h leaving it in that voltage and no BSOD so I think its stable (its hard to test as you cant torture test the idle state )

 

I was planning to select one speed / voltage in RMClock's profile and then run a torture test. However, this will need a large range of tests. At the moment I have 6x to 8x all set below the previous 1.0V minimum.

John

PS: I have disabled the enter C4 on C3 command option and haven't had any further crashes.

 

I have now the option enabled again, and no BSODs yet. How come you had 1.0 as minimum voltage? Mine was like 0.935 or something like that

 

Intel has some mysterious way of determining the appropriate voltage range for each individual CPU. My T7300 was set to 1.0V > 1.25V with 0.9V in SLFM.

John

 

right, after a reboot of the whole system (a non-planned one as I had to cut the power because I got a turned off screen when I resumed from hibernation) all my settings had changed in RMClock. Now I have IDA wich I didnt yesterday, my voltages had been scrambled and performance on demand all my PST states had been disabled.
So I redid all the settings and found this little checkbox:
"Use OS based load management"
So I unchecked that, and guess what, my OS load went down to 10-20% usage!
So now my fan is less used too, or it takes longer time for it to travel from 50C to 58C where the fan kicks in.

 

Ok Sprint, I believe you, my 6.0x is also running stable at 0.85V Well, for 5 minutes Gotta go now, hope it's still stable when I return, in an hour