• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

SR-2 troubles

DooKey

DooKey

[H]F Junkie
2FA
Joined
Apr 25, 2001
Messages
15,722
Ok, just got some extra spicy e5649s in the board. Won't boot, I get a CF code. Cleared CMOS already. Any ideas?
 
Might need an updated bios to see the chips.

That's my guess, anyone else?
 
Ok, it's running A50. Guess I need to talk to Loco about modding a bios for me?
 
Bummer DooKey, mine are running just fine on the A50 bios. I just started overclocking it, 4GHz was easy. Time to push [H]arder.

1zbx8jk.jpg
 
Hmmmmm, guess I need to find out what the CF code means then. Not sure what to do from here. I guess I'll put the E5530s back in a see what happens.
 
Last edited:
Tobit said:
Bummer DooKey, mine are running just fine on the A50 bios. I just started overclocking it, 4GHz was easy. Time to push [H]arder.

1zbx8jk.jpg
Click to expand...

does it get any sexier than a linux command-line cpu temperature gauge? I dont think so

linux pr0n ftw!
 
Ok, making some progress now. Got the system to boot and it's running at default settings just fine. For some reason the CMOS won't always clear when I push the button. I tried similiar settings to my E5530s, but no go on boot. Where do you recommend I start? Slight cpu voltage increase and bclk increase? That's what I'm going to try.
 
DooKey said:
Where do you recommend I start? Slight cpu voltage increase and bclk increase? That's what I'm going to try.
Click to expand...
Dunno, I am having a bitch of a time getting anything FAH stable. I'm scratching my head at the moment.
 
Yeah the SR-2 is sooo tempermental OCing especially w/o all the necessary features in the BIOS to do real OCing.
 
I'm stable at 2.66ghz (20x x 133) running bigadv. I just can't figure out how to make this bitch go above 133 bclk and boot.
 
Have you backed the multiplier down to 12x to see what you can get out of the bclk? Memory is also tempermental in these things - try DDR-1066 with everything else memory-related at Auto. Check you actual Vtt and memory voltages - none of my boards have been that accurate on settings versus actual voltages. I would manually set IOH and both Vtts at 1.35, set the DRAM voltages to whatever gives you the correct voltage for your memory, multiplier at 12x, and start cranking up the bclk.

Just some thoughts.
 
DooKey, I am finding these E5649's need a lot of vcore. I am currently at 4 GHz (200x20) and DDR3-1600 (CAS 8) but I had to crank the vcore up to 1.425 and vtt to 1.375.
I can actually boot into Linux at 208 BCLK but I haven't been able to fold an SMP unit at that BCLK yet.

iqj66a.jpg
 
Last edited:
I had this board running at 200 blck with the E5530s no problem. It might be that I have to crank some vcore to the cpus like Tobit. Gonna try that.
 
musky said:
Have you backed the multiplier down to 12x to see what you can get out of the bclk? Memory is also tempermental in these things - try DDR-1066 with everything else memory-related at Auto. Check you actual Vtt and memory voltages - none of my boards have been that accurate on settings versus actual voltages. I would manually set IOH and both Vtts at 1.35, set the DRAM voltages to whatever gives you the correct voltage for your memory, multiplier at 12x, and start cranking up the bclk.

Just some thoughts.
Click to expand...

I backed the multi down and I can't raise bclk. CPU isn't the prob I guess. I wonder if my RAM is a problem?
 
I've had my SR2 in a place where it seems like nothing is working right. Won't always boot, won't OC, etc.

Then, pulling the cmos battery out and letting it sit for about 12 hours to get its head right seems to fix the problem. Come back the next day and it's working like it should. Something must get fouled up in the BIOS or something from time to time. Seemed to happen a lot when I was switching CPUs regularly.
 
Any more luck today DooKey? My goal of bigadv stable @ 4 GHz with our new E5649's is proving very difficult to reach. I think I am now bigadv stable at 3.6 GHz. If I can get at least 3.8 GHz, I'd be happy at this point. Just hope I don't have to settle for 3.6 GHz.
 
Tobit said:
Any more luck today DooKey? My goal of bigadv stable @ 4 GHz with our new E5649's is proving very difficult to reach. I think I am now bigadv stable at 3.6 GHz. If I can get at least 3.8 GHz, I'd be happy at this point. Just hope I don't have to settle for 3.6 GHz.
Click to expand...

I'm going to mess with it again after work today. It's really strange what is happening. If I leave it at auto I can boot and fold at stock. Once I increase bclk even 1 mhz I reboot when entering the windows logon screen.
 
Scratch that, only made it through 12 frames of a 2685 @ 3.6 GHz before the machine rebooted itself.

Frustration continues. Boots into the OS fine at speeds approaching 4.4 GHz, runs IBT and Linpack fine at 4 GHz, SMP stable at 3.6 GHz but bigadv is still a problem. Back to the drawing board.
 
jebo_4jc said:
I've had my SR2 in a place where it seems like nothing is working right. Won't always boot, won't OC, etc.

Then, pulling the cmos battery out and letting it sit for about 12 hours to get its head right seems to fix the problem. Come back the next day and it's working like it should. Something must get fouled up in the BIOS or something from time to time. Seemed to happen a lot when I was switching CPUs regularly.
Click to expand...

Interesting. I may try that if I can't get an OC out of the box tonight.
 
Tobit said:
Scratch that, only made it through 12 frames of a 2685 @ 3.6 GHz before the machine rebooted itself.

Frustration continues. Boots into the OS fine at speeds approaching 4.4 GHz, runs IBT and Linpack fine at 4 GHz, SMP stable at 3.6 GHz but bigadv is still a problem. Back to the drawing board.
Click to expand...

I'm wondering if we need a microcode update in the bios for this extra spicy chip?
 
DooKey said:
I'm wondering if we need a microcode update in the bios for this extra spicy chip?
Click to expand...
Shouldn't. E5649's have been a supported chip for several BIOS revisions now. A48 through A50 are all working fine here, having the most luck with A49 as I did before with my ES L5640's.
 
Let me know how it goes. I'll report back when I get home from work in a few hours.
 
Ok, still no luck overclocking at all. If I increase bclk even up to 134 I will either bluescreen or just reboot going into windows. Strangest thing I've ever seen.

I'm going to remove the CMOS battery like Jeb and let it sit for a while. It didn't start doing this till I changed cpus so I know the board can take up to 205 bclk no probs. I'm thinking bios issues due to the cpu change out.

We'll see I guess.
 
@tobit - dumb question, but is it worth reseating your PSU cables, making sure all the extra plugs are in?

Are there stop errors being logged like #124 that could shed some light?

@dookey - what stop error are you getting on the bsod?
 
may i be worthy said:
@tobit - dumb question, but is it worth reseating your PSU cables, making sure all the extra plugs are in?

Are there stop errors being logged like #124 that could shed some light?
Click to expand...
No stop errors, in every case it has been a spontaneous reboot or:

[20:02:17] mdrun returned 255
[20:02:23] Folding@home Core Shutdown: UNSTABLE_MACHINE
[20:02:23] CoreStatus = 7A (122)

but mostly a spontaneous reboot after X amount of frames. All of th extra cables are plugged in.
 
DooKey, are your temps good? Does all the RAM show up in BIOS? Are the HSFs seated correctly and not touching the heatsinks around the socket? Is there any flex in the motherboard, or is it sitting on a hard surface? I had trouble with all these things, including running the mobo sitting on the box (which doesn't count as a hard surface).

Maybe you can post your BIOS settings, or link to a post from where you copied the BIOS settings. There are some buried gotchas in the SR-2 BIOS.
 
may i be worthy said:
@tobit - dumb question, but is it worth reseating your PSU cables, making sure all the extra plugs are in?

Are there stop errors being logged like #124 that could shed some light?

@dookey - what stop error are you getting on the bsod?
Click to expand...

When I get a bluescreen it is usually machine_check_error or uncorrectable machine error if I remember correctly.
 
SazanEyes said:
DooKey, are your temps good? Does all the RAM show up in BIOS? Are the HSFs seated correctly and not touching the heatsinks around the socket? Is there any flex in the motherboard, or is it sitting on a hard surface? I had trouble with all these things, including running the mobo sitting on the box (which doesn't count as a hard surface).

Maybe you can post your BIOS settings, or link to a post from where you copied the BIOS settings. There are some buried gotchas in the SR-2 BIOS.
Click to expand...

Temps are real good. All the RAM shows up in BIOS and Windows. HSFs appear to be seated nicely. I'll check if they are touching HS around socket. No flex in MB, it's in a 700D.

Seems strange it won't OC at all, but runs at stock just fine.
 
Not much progress here last night, still smp stable at 3.5 GHz but bigadv reboots itself (or mdrun 255 error) after a few frames.

Cooking at a BBQ today, heading outside to build a fire in the smoker now so I can't do much today so I am just going to let it crunch smp units all day.
 
Hmmm, double checking time...

Uncore 18x on BIOS A49 hacked or A50? (the only bios to let you change uncore)

Uncore was locked and thus irrelevant on L5640s, but it is a big deal for all my CPUs - x5650/5660/5680 etc. 18x is the sweet spot - tiny performance hit, decent stability boost to go higher. L5640 folks reported older BIOS being better, but on my CPUs A50 or A49 hacked are equally as good.

IOH at least 1.4v?
vtt at least 1.375v?
Signal tweaks to between -80 and -90, and -16

Tobit said:
Not much progress here last night, still smp stable at 3.5 GHz but bigadv reboots itself (or mdrun 255 error) after a few frames.

Cooking at a BBQ today, heading outside to build a fire in the smoker now so I can't do much today so I am just going to let it crunch smp units all day.
Click to expand...
 
may i be worthy said:
Uncore 18x on BIOS A49 hacked or A50? (the only bios to let you change uncore)
Click to expand...
I have it set at 18x in the A49 BIOS but it could very well be locked and the setting not working. I will check it out after this BBQ is over.
 
Tobit said:
I have it set at 18x in the A49 BIOS but it could very well be locked and the setting not working. I will check it out after this BBQ is over.
Click to expand...

Yeah official A49 will do squat. (check uncore / NB frequency using CPU-Z to see if it is actually changing)

Unfortunately I can't find a copy of the hacked A49 right now and the original link is down ... anyway official A50 will show you in short order if that is the problem.

Oh BTW - almost always after changing uncore you will get a FF no post - just reset and the second boot will be fine.
 
Well, A50 testing isn't going well at all. I can't get it to save any BIOS settings at all, not even "optimized defaults" without getting a CMOS Checksum Bad error upon reboot. I have tried A50 in two of the three slots. I don't have this problem at all with A49 or A44. Fucking SR-2.
 
Yeah the SR-2 BIOS pisses me off it sucks. EVGA pisses me off they won't fix it and add the needed OC featuires it misses. I am sticking with ASUS from now on I only bought the SR-2 due to the capability to OC dual processors but since the BIOS sucks so bad it was a terrible puchase. I have always had solid OCing MBs from ASUS. Hopefully they will do a dual OCer for Ivy Bridge with a ton of PCIe bandwidth for GPUs and more
 
I've given up OCing my hexes on the SR-2. I'm just running stock and putting out 105K ppd with good wu's thanks to musky and the Ubuntu thread. On the bright side it is barely pulling any power at all. I'm disappointed I can't get this board to work with the spicy chips, but that's the chance you take. I just wish I had purchased the chips in the $300-400 price range rather than $500 each.

I ordered an ASUS dual proc board and I'm going to put my spicy E5530's in that. When all is said and done I should have about 250-270K ppd from my farm without killing the electric bill.
 
Bummer DooKey. Yesterday, I ran a total of 15 regular SMP units just fine @ 3.6 GHz with the A49 BIOS. I just now got the A50 BIOS working finally and I'm now trying bigadv with the same 3.6 GHz OC.
 
Yeah, I'm debating putting the E5530s back in the SR-2 since they did OC in it. Then I can put the E5649s in the ASUS board and run them stock. That's probably my best option.
 
DooKey said:
Yeah, I'm debating putting the E5530s back in the SR-2 since they did OC in it. Then I can put the E5649s in the ASUS board and run them stock. That's probably my best option.
Click to expand...

Sounds like a better plan - and if its a ZN8A, they are good boards. Pity they don't OC
 
You must log in or register to reply here.
Back
Top