nForce 790i SLI/Ultra chipset mobos - Reviews/OC/Guides

[tekjunkie]

Argh!!! It gets worse.Now windows hangs instantly after boot. Has anybody come acroos this!!!!








It's quite possible I might have gotten a dud in the first place. Also, am suspecting that my mobo ate my RAM(much like the earlier striker). When I first tested it with memtest @ default 333 FSB(1333 QDR) unliked with the RAM @ the rated 1600 7-7-7-20 1T i got 3 good passes. Now,memtest is giving errors @ the same setting.

As a last ditch effort, I will have to try the 0508 BIOS and see what happens. I already got the RMA # for the RAM from newegg.

[tekjunkie]

0508 Is a no go on my board. It complains about the id not being compatible with the existing bios or something like that. I guess it's RMA time. Never dealt with NCIX for an RMA. Actually SIIE was my first purchase from them.Hope the RMA goes smooth for a replacement from NICX.

[saaya]

damn, the S2E is sounding worse every day i check this thread :S
now im actually glad i didnt buy one...

tekjunkie, did you oc the fsb unlinked from the mem clocks or linked?
what divider ie memspeed did you reach max with what timings and voltage?
do you think the memory could have been unstable and that caused your corruption issues?

[webwilli]

currently "playing" with my 8500 on S2E using 9,5x multiplier

when setting FSB to 425 in NVCP,system froze several times

w/o changing anything else,i set cpu@1800(aka 450FSB) and system boots/works fine

RAM linked/unlinked?
Voltages?

Help me to get my system stable

0508 Is a no go on my board. It complains about the id not being compatible with the existing bios or something like that. I guess it's RMA time. Never dealt with NCIX for an RMA. Actually SIIE was my first purchase from them.Hope the RMA goes smooth for a replacement from NICX.

i get the same massage. then i use dos flash and it works.

[tekjunkie]

damn, the S2E is sounding worse every day i check this thread :S
now im actually glad i didnt buy one...

tekjunkie, did you oc the fsb unlinked from the mem clocks or linked?
what divider ie memspeed did you reach max with what timings and voltage?
do you think the memory could have been unstable and that caused your corruption issues?

This is what I did

First up QX9650 with OCZ16002GK, everything @ default -> Memtest good for 3 passes, SPI 1M,32M stable and 12 hours+ prime stable.

QX9650 with OCZ16002GK, everything @ default except for vcore @ 1.15V, SPI 1M,32M stable and 12 hours+ prime stable.

QX9650 @ 1333 QDR(333 FSB) and RAM unlinked @ 1600 7-7-7-20 1T, 1.9V, 3 passes of memtest good,SPI 1M,32M stable and 12 hours+ prime stable.I had to manually enter the timinigs as the default timings were way too loose. I think it was like 9-8-8-24 or something like that. This is when I found out the RAM is being overvolted a bit.(~1.93 with 1.9 set in BIOS). Everything else @ auto


Qx9650 @ 1600 linked and synced 7-7-7-20 1T 1.9, all on auto except memory. didn't bother to memtest, SPI 1M no go. Tried upping the vcore only to find out the vcore is also being overvolted, went all the way upto 1.35vcore with loadline calibration. Nope, still no go. Sometimes even BSODed.

Brought the FSB down to 350(1500 QDR) with all on auto. 6+ hours prime stable. Tried 375 FSB from there. No go. This where all the trouble began.
The board sometimes got stuck on "Code Init" and refused to post, had to pull the power cord off the PSU to get it to post.

Swapped my QX9650 with my E3110, went straight to 375 FSB, windows hung on me while running 32M SPI(all @ auto).

Did a memtest @ 1333 FSB with RAM @ 1600 and the rated timings @ volts and let it run overnight. A lot of errors came up then.I had ordered the 4GB OCZ EB 1600 (2x2) from mwave and thought I would return the old RAM. It came in yesterday, popped it in and ran memtest @ 400 FSB linked and synced with my E3110, memtest hung on me.Tried the white slots instead if the blue ones. Still memtest was hanging and the board sometimes would hang on Memory init. on This where it even got worse.I kept trying different things like manually setting tRFC and stuff like that and also did a clear RTC. the board refused to even start up when I pushed the power button.The stupid LCD poster thingy kept flashing "CPU INIT", got it to POST and everything was on default coz of the CMOS reset. Booted into XP, it hung on me. Rebooted, that when I got the message about the NB, CPU and SB error message. It gets better, I kept trying different things, lol @ one time all the lights on the board came on(the LCD for the high, low , crazy on the board itself). I panicked and switched off the PSU and when I rebooted, none of the lights came up and I kept getiing the error message.


Sorry about the long post, but I thought I would share my experience. I think the BIOS got corrupted or the CMOS itself chip is F***ed up someway.The reason am thinking that is when I tried to flash it to 0508, it said the ID does not match and I downloaded the BIOS file from ASUS's FTP site.

I guess it's a good thing that I paid for the Advanced RMA from NCIX when i bought the board. Also, it's quite possible I got a dud since I was one of the early people who got the board probably from a very early batch(They had just 2 of 'em when I ordered).

[saaya]

hummmm, would you mind copy pasting this post into this thread i started about corruption issues on 790?
http://www.xtremesystems.org/forums/...d.php?t=183119

did you try a dos flash of the bios?
since its an early board you might have a very early bios which could be causing the problems...

did you increase the nb and sb voltages etc to get 1600fsb or higher?

[arrakys]

I dunno if 511 has been released yet here for striker 2 extreme, i just got it from asus (dated from 04/02/08)

You can dl it here

Just flashed it, but bot tested any things yet.

[Big Myke]

I dunno if 511 has been released yet here for striker 2 extreme, i just got it from asus (dated from 04/02/08)

You can dl it here

Just flashed it, but bot tested any things yet.

Get back to us when you tested 0511, let us know the method you used to flash. awdflash/ezflash

[arrakys]

used bios flash tool.

As my test rig is fixed., I can't test it until saturday maybe later, cause i ve i big article to finish

[Broly]

that p03r2 bios bricked evgas if you use a proper boot cd with sata support (I had a sata dvd drive) and type the awdflash /f/prot command.
boardis going in for repair. cpu hot red light is on doesn't post.

pissed off now.

[slim142]

BIOS 0511 available already, this might be the fix everybody has been waiting for?

Lets see whos the brave one to try it.

[OBR]

I tryed 511, testing in progress, but first results ... FSB 1600 + LInked mems at 1600MHz - 22-5-6-7-2T 2hours OCCCT Mem test stable, Prime95 2 hours too ...

1700MHz + Linked 1700MHz mems at 22-7-7-7 OCCCT and Prime95 fully stable too, no data corruption issue yet ... next step will be 1800MHz with Linked ...

[slim142]

I tryed 511, testing in progress, but first results ... FSB 1600 + LInked mems at 1600MHz - 22-5-6-7-2T 2hours OCCCT Mem test stable, Prime95 2 hours too ...

1700MHz + Linked 1700MHz mems at 22-7-7-7 OCCCT and Prime95 fully stable too, no data corruption issue yet ... next step will be 1800MHz with Linked ...

No corruptions yet, but did you have any before? I think you are one of the first here who had their hands on the SIIE...

BTW, I want to ask, when you use memtest 2.01, what tests and how many times do you run them? I usually run 1 full standard test and 10 times test #5 to make sure they wont BSoD in windows, then run orthos blah blah...

[OBR]

i had no data corruption issue before ...

I testing 2 complete passes memtest 3.2, then 2 hours OCCCT mem test, then 2 hour Prime95 FFT, when every test are ok continues with 3Dmark06, 3 or 4 loops at max settings with AA AF, last is some round in Crysis from 1920x1200 to 800x600 with various AA AF ... when is everything ok, i can said TEST PASS, Stable settings ...

[cptkoulis]

RAM linked/unlinked?
Voltages?
Help me to get my system stable

RAM unlinked (4sticks)
NB@1,6Volts
Quad-SLI on
stock air cooling ATM

so i guess i have an FSB hole around 420+

[noobzed]

@dinos22: why you put 1,7+ on MCP/Southbridge ? ..

[|SiLA|]

now duno which mobo to get...xfx or asus s2e

[webwilli]

after two days full of pain i think i finally beat the striker II extreme beast

There was a FSB wall @ 375. This wall was pretty strong.

Yesterday i would sell the board...
But today?
i find the key



Validationscreen:



Link: http://valid.x86-secret.com/show_oc.php?id=340269

RAM unlinked 900 MHz

the solution for my system is little raise VTT and little raise NB. This extra voltages helps to boot with 375, 450, 500, 550.

I still not testet prime.

to break the 375 barrier is a little victory for me...

some benchmarks with 450 FSB (4050 MHz)

RAM-Timings not optimized
PCIe 110
all Boardvoltages default
VTT 1,28 Volt
NB 1,42 Volt



3DMark2001:

http://service.futuremark.com/compare?2k1=9359240

2003:



http://service.futuremark.com/compare?2k3=5796586

2005:



http://service.futuremark.com/compare?3dm05=4161095

2006:



http://service.futuremark.com/compare?3dm06=6093915

[|SiLA|]

congratz web

[slim142]

Hey web, what vcore? 1.28?

[slim142]

I Think This Might Be Interesting For ALL of US

Sata Bricks

http://www.theinquirer.net/gb/inquir...-driver-bricks

[webwilli]

congratz web

thanks

Hey web, what vcore? 1.28?

for the 4050 MHz Test VCore in Bios was set to 1.3250. Everest say 1.36x

the CPU is new, so i must find the right settings now.

[dinos22]

RAM unlinked 900 MHz

the solution for my system is little raise VTT and little raise NB. This extra voltages helps to boot with 375, 450, 500, 550.

shame ppl don't read my testing thread hehehehe

[webwilli]

shame ppl don't read my testing thread hehehehe

here are the volts set in bios which gave me the highest stable FSB i could benchmark with (595MHz)

Under System Voltages Section:

CPU Core >> set value required for MHz OC
CPU FSB >> 1.45v
Memory >> set to value required for MHz OC (1.9-2v is plenty for 2GHz speeds and decent timings)
Nforce SPP (NB) >> 1.45v (VERY important) going lower or higher reduced stability > higher probably due to possible better cooling required (will have to test that but NB is generally pretty cool
Nforce MCP (SB) >> 1.75v (PUT A FAN ON SB)

[slim142]

Why do we have to put so much v to the MCP? :S I thought it was all about SPP and CPU FSB