680i SLI is too expensive? Asus as the solution : P5N32-E SLI PLUS

[Radical_53]

Yes, but when CPU-Z shows you 1,15, Everest is at roughly 1,23 right? That's why it runs stable then. When you set the voltage manually, Everest shows you the voltage you set in bios, CPU-Z is way lower.

[>GSXR<mrbusa]

I didn't explain it right.
I had to set the vcore manually at 1.132.
Everest is reporting at 1.12 and saying the VID is 1.113.
Core temp is reporting the VID at 1.125.
CPUZ is reporting the VID at 1.152.
The mobo bios sets the auto vcore at 1.25, which is certainly not need and is the rated max vcore for the CPU batch I got, q813a478. All the programs are the latest versions. It seems to just be one of those chips that has a wierd diode output that programs and my mobo have problems with, just like with the temps. It decided to get hot where I live so overclocking is out for right now anyhow.

EDIT: I just realized the tjunction for this chip is 105c, so that would explain the high temp reading right? Are all these 45nm CPU's at 105c for the tjunction?

[Radical_53]

Interesting, which bios are you running? I had the same thing with 2 chips on the P5N and the Striker with the most current bios versions that Everest was the only program to report the voltage as high as the bios. CPU-Z and OCCT for example always showed it lower, ~0,08V.

[>GSXR<mrbusa]

bios 1003.
I guess I could try the 1002 but I am scared of it after the headaches it gave me with not booting or saving settings. I think these chips need a newer board in order to be more than just capable of running. This board is holding this chip back.

[Radical_53]

Well, yes it might hold it back because of the FSB. I had my two 8500s at 4.2GHz on both boards. It takes some time, but it works Yet I'm pretty sure that a bios could really make things much easier, less painful and less time consuming.

[>GSXR<mrbusa]

Another weird voltage detection with the 1003 bios.....I originally set the cpuvtt to 1.2, which is the lowest manual setting, but noticed the board was over volting it to 1.2/3 within the bios. When left on auto it was reading 1.15 in the bios hardware monitor. So I set it back to auto and then set the cpu vcore to the lowest reported vid, Everest, of 1.113, and now the cpu is Everest stress, 3dmark06, and game stable with that vid. Temps dropped a few degrees also. These chips are pretty amazing.
Why are there so many different variations of voltage requirements for these new 45nm chips?

[Leeghoofd]

I need to set 1.3 to get about 1.2 under load on the Striker board... Temp wise use real temp or the latest coretemp and adjust that one to Tjcasemax 95°C... Striker runs fine here with a Thermalright set and the E8400 currently at 3.6ghz at 1.2 volts in bios around 1.12 for CPU-Z...

For my E8400 1501 was the better bios... for the Striker, no clue what it is for your board mate

[>GSXR<mrbusa]

I have hardly any vdroop on my board. Vcore reads true in the bios and most programs. The cpu vtt minimum manual setting of 1.2 when the detected auto setting is 1.15 seems to be what was keeping me from running stable at the lowest reported vid for my vcore of 1.113. Auto vtt at 1.15 and vcore at 1.113 is running stable as can be.
There are only two bios revisions, 1002 and 1003, that are capable of running the e8400 so changing bios is out of the question because 1002 was not stable for me. 1003 is my only choice.
Just modding coretemp to report the tjmax at 95 really proves nothing because it is unknown what the correct tjmax is on these 45nm chips. The only way to be for sure is to use a probe to measure the temp.

[Leeghoofd]

What does CPU-Z give you ? only CPU-Z gives beleiveable readings on my Striker board (Speedfan, Everest home, Asus probe all give incorrect reading, I too have no Vdroop whatsoever), do you believe that without any voltmods you set in bios 1.2 volts and your board will give you 1.18 volts or such under load ? must be one hell of a lucky board then you got there...

And for ya temps believe what ya will, must be zillions of users wrong then....we all thought E8400 was 95&#176;C Tjcasemax

[>GSXR<mrbusa]

The vid cpuz reports is 1.152. This much vcore is not needed so I believe Everest's reported vid of 1.113 to be most accurate because now after setting the cpu vtt to auto I can now run completely stable at 1.113 vcore. I have only had this chip a couple days and I still am reading up on other peoples results.
I performed the pencil vdroop mod and at load my board is very stable.
What I have read so far about the tjmax issue with the 45nm chips is that programmers are using the whitesheet reports from Intel based off of the mobile chips and that the actual tjmax differs between the two chips. In that case, what I meant by changing the tjmax in the cpuz inf so that I get temps based off of a tjmax of 95 would not really mean anything but just to set my mind at ease. No biggy to me whether I see 40c or 3?c in a programs readout as long as I know the 40 is based off a tjmax of 105c

[Radical_53]

I prefer to trust the vcore reading in bios and Everest, they are the highest ones and seem to be the most realistic ones too. Temperatures are definitely not worth a look so far, those sensors do just show "something"
What I still can't understand really is this strange stability. I had this with the P5N32 and I do have it with the Striker also. The system can do like all tests on one day, all run like a charm, no problems whatsoever. The other day, not all but some of these tests will fail within the first minutes.
I'm happy with the switch to the Striker so far as it doesn't have stability problems when it's running for a longer time, my P5N always had trouble running over night. Some of the problems do remain though and I'm beginning to think it has something to do with my RAM...

[>GSXR<mrbusa]

When I was talking about the plus board holding this chip back I was meaning that at the same clock speed when compared to a newer board like 750i/up and x35/up, the performance difference can get to be quite different. That is one thing I surely sux at is calculating chipset bandwidth between all the boards.
I was expecting a little performance increase over the e6600 when compared clock for clock because of the higher fsb. It is just not the case on the plus board like it is on a board truly developed for 45nm chips. On some of the Everest cpu tests the 8400 is slower at 3.0 than the 6600 is. On the photoworxx test it is allot slower.

[Radical_53]

Oh, ok. I haven't tried that so far. I had better results with higher clocks for sure, but I guess I haven't tried an exact 1:1 so far. On the other side, I don't have exact numbers from my old 6600 at standard clocks...
The only thing where I did a 1:1 was between dual and quad cores, the reason why I got myself an E8500 instead of keeping the quad.

[>GSXR<mrbusa]

I got my new ram this morning. G-Skill 2x2gig pq's ddr2 1000.
Ran into another issue trying to get these to run at there rated speed and timings. In the A slots they just would not run at 1000 no matter what. Just when I was ready to give up I decided to switch to the B slots and they booted right up at spec. Weirdest of all is I can run the G-Skill hz d9's at 1150+ with 2.2v 24/7 5-5-4-15 2t in the A dimms. All is up and running now stable as can be for now. As soon as they get a week or so of burn in then I will see what they can do.

[Radical_53]

Interesting A and B wasn't such a big difference for my machine, but the order of the sticks somehow. "Wrong" stick in slot 1? Oh my... terrible clocks, unstable very fast and it needed lots of voltage. Swapping them around was the solution for me then.

[>GSXR<mrbusa]

With the hz 1gig sticks the a and b slots had the exact same overclock results. It kinda upsets me because if for some reason I wanted to run eight gigs then now I know I would have issues for sure. Luckily this boards life expectancy is way shorter than the amount of time it will be for me to need eight gigs.

[Radical_53]

Need? Since when the "need" is important? But you're right about that. Still, it has always been and I guess always will be problematic to fill all RAM slots and still overclock.

[ballin247]

guys i need help with downgrading. i updated to 1003 but its pretty wacked out, and i want to go back to 0801, 0805 and maybe 0901. I am getting that downgrade is not possible, ROM... blah blah message.
problem is i dont have a floppy, and i dont know how to boot to dos/prompt from usb to access awdflash and P5N32-EP.BIN.
during post WHEN you press f5 it gives you choices, and there is safe mode in command prompt, but i need command prompt only.
i am really desperate, i need help...

[>GSXR<mrbusa]

guys i need help with downgrading. i updated to 1003 but its pretty wacked out, and i want to go back to 0801, 0805 and maybe 0901. I am getting that downgrade is not possible, ROM... blah blah message.
problem is i dont have a floppy, and i dont know how to boot to dos/prompt from usb to access awdflash and P5N32-EP.BIN.
during post WHEN you press f5 it gives you choices, and there is safe mode in command prompt, but i need command prompt only.
i am really desperate, i need help...

I pmed you the only two ways I researched when I had this problem with 1002 bios and downgrading. My suggestion to everybody that wants to downgrade from the latest two bios is to follow my method a couple pages back. This board doesn't natively support booting from a usb drive and my low skillz in this area could not get it to at all. Get a floppy drive and it will take for sure and there is only the standard failure rate to worry about instead of all the other headaches. Even with a floppy you have to force the downgrade from these two bios revisions because asus did something different to them. So follow my floppy method to the letter and I guarantee it will work. good luck.
If anybody with coding experience that knows how to modify the files layed out by the HP usb boot program and the bartpe usb boot batch file, then maybe you can help these guys out without a floppy drive more than I can.
http://www.bootdisk.com/pendrive.htm
http://www.nu2.nu/pebuilder/

[Snyxxx]

I pmed you the only two ways I researched when I had this problem with 1002 bios and downgrading. My suggestion to everybody that wants to downgrade from the latest two bios is to follow my method a couple pages back. This board doesn't natively support booting from a usb drive and my low skillz in this area could not get it to at all. Get a floppy drive and it will take for sure and there is only the standard failure rate to worry about instead of all the other headaches. Even with a floppy you have to force the downgrade from these two bios revisions because asus did something different to them. So follow my floppy method to the letter and I guarantee it will work. good luck.
If anybody with coding experience that knows how to modify the files layed out by the HP usb boot program and the bartpe usb boot batch file, then maybe you can help these guys out without a floppy drive more than I can.
http://www.bootdisk.com/pendrive.htm
http://www.nu2.nu/pebuilder/

I tried these many times and it worked for me. I flashed from my USB stick using these.

http://www.weethet.nl/english/hardwa...omusbstick.php

and this

http://gocoding.com/page.php?al=bootflashdos

[>GSXR<mrbusa]

Those are basically the same as using the hp and bartpe instructions. The problem is the 1002 and 1003 bios has a problem downgrading to the 800 and 900 series revisions. It comes up with the same error about being unable to downgrade. Have you downgraded from the new bios to an old bios using this method?

[ballin247]

Those are basically the same as using the hp and bartpe instructions. The problem is the 1002 and 1003 bios has a problem downgrading to the 800 and 900 series revisions. It comes up with the same error about being unable to downgrade. Have you downgraded from the new bios to an old bios using this method?

You guys are blooooddy brilliant lol

at first i used bootflashdos tool to create bootable usb (the one that snyxxx suggested). It took like 5 seconds, then i placed awdflash.exe and bios bin in it. Then booted from usb and tried to downgrade. It gave me message that it can not be downgraded. Then i loaded xp back, and created autoexec like musa suggested (page 46 in this thread i think), booted back from usb and it worked fine. I downgraded to 901 successfully.

I am about to read all the posts in this thread... because i had troubles to get 400fsb stable even with vcore of 1.6. I have e6600 (revision/stepping B2, antec 900 set to max, 120 tower tuniq set to max) running @ 3.282 at stock voltages for 6 months and 0801 with no problems, just struggling with 400 and such.

Now testing 0901 revision @ 3.33 (370 fsb) stock voltages. 45 mins already passed on orthos stress test (core 1: 49-51c, core2: 47-49c). I think my mobo likes 0901 better than 0801.

Thanks once more

edit: omg, finished all 52 pages, my brain is done, good nite

[Snyxxx]

Those are basically the same as using the hp and bartpe instructions. The problem is the 1002 and 1003 bios has a problem downgrading to the 800 and 900 series revisions. It comes up with the same error about being unable to downgrade. Have you downgraded from the new bios to an old bios using this method?

No, I am still at 801 BIOS and will not be upgrading.

[ballin247]

guys, i always kept everything auto except vcore and memory. Now i tried to change nb to 1.4 and cpu vtt to 1.4. I dont have water cooling or extra fans over nb. Just 120 tuniq tower set to max and my case is antec 900 set to max. Is it possible that i fried mobo or somethin? also i tried multiplier of 8x and fsb of 450 (i dont think thats a case), i am worried if nb of 1.4 and cpu vtt of 1.4 is to high ? pc simply wont post. Everything is on, but monitor light keeps blinking and nothing happens. Anyone? i have sli plus.

edit: of course i tried to take the battery out. Also i thought maybe one of the ram sticks died, still nothin. All lights are on, fans blowing, but getting no post, monitor light keeps blinking.

[>GSXR<mrbusa]

Glad to here ya got it working.
Now I know how to use a usb stick also.
When you say you created the autoexe, you mean how I said earlier with the force switch commands based off the floppy autoexe.? If so, I could not get it to work with usb that way. I must of done something wrong. The bios I was downgrading from though was 1002, it might have bigger issues flashing this way than the 1003 version. grats though