QC6700 / Striker... bios failure after overclocking

[carbonrider]

I am running Asus Striker extreme with 2GB CorsairTwinX XMS2 Dominator(CAS 3-4-3-9) and Quad Core CPU qx6700.

After a period of stable operating and having read as much as I could about overclocking I decided (with a hindsight, prematurely) to dip my toes in overclocking.

I set the FSB to modest 1400mhz, cpu multiplier to 9 (so 350 * 9 = 3.015ghz). FSB : DRAM = 3:2 so dram freq was 933mhz. I also set the ram to 4-4-4-12. What I now realise (after more reading) that by increasing frequency and lowering timing at the same I might have overcooked things(trying to find a solution to my problem I came across considerably more information then before)

Voltages were set as follows (a recommendation from one of the sites I found previously)
# VCore - 1.5V
# Memory Voltage - 2.4V
# 1.2V HT Voltage - 1.25V
# NB Core Voltage - 1.45V
# SB Core Voltage - 1.60V
# CPU VTT Voltage - 1.35V

I started Windows fine, ran CPU-Z which confirmed the various frequencies. I ran the Asus probe. It warned against VTT voltage being 1.41v and MB temp being 35 degrees but that's about it.
I used an Excel macro running an infinite loop of mathematical calculation to push CPU 60 60% on 4 cores. CPU temp was always less then 60 (and much lower at idel - cannot remember exactly how much). The memory has the supplied fan mounted and working and it never rose above 40 or so according the the optional temp sensor stuck into it.

My problem is that after shutting down the i cannot even get BIOS running at all (i.e. nothing nut a blank screen) and the LCD post error is DET DRAM. I have tried reseting the CMOS using the internal push switch but no joy. One thing I am going to try latter is starting with one of the modules.

Any ideas? I mean, if I can the BIOS to start then of course I can correct all my wrong doing and start afresh with a step by step approach... not much that I can do with a blank screen.

[Reign]

that sucks dude. resetting cmos should've done the trick. but that didn't work.

[Tensai]

Did you follow the direction for clearing the CMOS? The Striker's clearing CMOS direction is a bit more complicated the simply pressing that button. You need to move jumpers to enable the Clear CMOS function.

Also, the button isn't like a reset button, it has 2 positions, On/Off.

[scott]

You can also unplug your power supply and pop out the battery when all else fails- if that's the problem.

[RealBeast]

Two things once you get it going again:

In my experience Asus Probe is not very accurate. I would suggest something like Core Temp.

Also, are you using an up to date bios or the original from 2/07?

[Nater]

That memory voltage is nuts dude.

Anyway, I have the poor man's Striker, and this happens all the time. ASUS sucks, and their CLEAR RTC rarely, if even works for me. Pull the CPU, put it back, and it should POST.

[carbonrider]

Thanks for the advices. Been out all day on the racing track so will try all of them tonight.

Did you follow the direction for clearing the CMOS? The Striker's clearing CMOS direction is a bit more complicated the simply pressing that button. You need to move jumpers to enable the Clear CMOS function.

Also, the button isn't like a reset button, it has 2 positions, On/Off.

I probably should read them again. I though that the CLR reset is enabled by default and the jumper can be changed to disable. Am I wrong here?
I did notice it is an on off switch but did not know what to make of it. Should I try rebooting in one state or the other?


I have the latest bios version - 1303

[carbonrider]

Just found it... the CLR switch is disabled by default so will try enabling it tonight

[RealBeast]

. . . ASUS sucks. . . .

I must strongly disagree.

The ASUS 680i might suck, don't have one and never will, but I have used ASUS boards for years and overclocked them all and I've never had any problem like this one.

[Tensai]

Just found it... the CLR switch is disabled by default so will try enabling it tonight

You may want to read the ASUS motherboard forum for different ways to reset the CMOS. Basically, once you set the pins to enable, press the clear CMOS button (and leaving it at on position for 2 hours), press the clear CMOS button again, and then moving the pin back.

Some ppl even had to remove the battery to complete clear their CMOS.

Also, Bios 1303 doesn't play well with some Razer USB mouse . . . you may want to remove them for now while trouble shooting, if you are using them.

[Nater]

I must strongly disagree.

The ASUS 680i might suck, don't have one and never will, but I have used ASUS boards for years and overclocked them all and I've never had any problem like this one.

I've had many different ASUS boards, they've all had problems and ASUS' support is absolutely worthless. I built a rig for a friend using the P5 Digital Home. Total piece of ****. Abit and DFI are the way to go. For enterprice/2S I'd say Supermicro.

[carbonrider]

ASUS' support is absolutely worthless.

Professionaly, I am in charge on a host of test tool s/w. We have invested more then £500,000 by now. And we still get near worthless support from the vendor. So I am expecting very little from Asus or any vendor of consumer level components


I always liked ASUS tbh but as I have no experience with the 680i or overclokling I will refrain from commenting for the time being.

[carbonrider]

By the way...
can someone explain to me (or refer me to the right article / post) what is the contrinuting of memory timing VS freq to actual performance?

I.e would a 3-4-3-9 DRR 6400 (@800mhz) be slower or faster then a 4-4-4-12 DDR 8888 (@1111mhz)? That is, in term of actual computer performance.

I just cannot get my head around the timing / freq trade-off

[Un4given]

By the way...
can someone explain to me (or refer me to the right article / post) what is the contrinuting of memory timing VS freq to actual performance?

I.e would a 3-4-3-9 DRR 6400 (@800mhz) be slower or faster then a 4-4-4-12 DDR 8888 (@1111mhz)? That is, in term of actual computer performance.

I just cannot get my head around the timing / freq trade-off

Unfortunately, there is no direct 1:1 correlation in terms of how much speed you need to offset latencies. Also, you have a number of latencies to consider on which will get you the most performance, like the CAS, RAS, Trcp, etc. The problem is that some applications react more to latency, and other more to speed.

[Nater]

Professionaly, I am in charge on a host of test tool s/w. We have invested more then £500,000 by now. And we still get near worthless support from the vendor. So I am expecting very little from Asus or any vendor of consumer level components


I always liked ASUS tbh but as I have no experience with the 680i or overclokling I will refrain from commenting for the time being.

No experience with the 680i and overclocking? Are you sure?