OC'ing help on Q6600 & Striker Extreme

coolamasta

Member
Jan 8, 2008
57
0
0
Hi all, another noob here in the world of overclocking haha

I've been a AMD man for years and only ever done mild overclocks but have just built a new system using a Q6600 and have heard about all the OC potential with these chips and so have decided I want to take it from 2.4 to around the 3Ghz mark

Machine spec:

  • Intel C2Q Q6600 Retail CPU
    Zalman CNPS9700-LED CPU Cooler
    Asus Striker Extreme Mobo
    Crucial Ballistix Tracer 2x1GB DDR2 PC2-8500C5 1066MHz Dual Channel Kit
    Gigabyte ODIN GT 800W ATX2.2 Quad-SLi Compliant Modular PSU
    2 x 512MB BFG Nvidia 7900GTX's in Sli (for now)
    Seagate 500MB Sata II HDD with 32MB Cache


Now im thinking about running the FSB at 333 as opposed to the stock 266 but need to know what I need to do with voltages, mem timing etc as ive never done this before!

Any help greatly appreciated :D
 

undeclared

Senior member
Oct 24, 2005
498
0
86
I'm just mostly gonna repeat what most guides say but..

Step 0 (which is better than everything I said) - See Sticky thread :p

1) Set ram voltage to what it is (but I assume since it's 1066, it's already set to that)
2) Lock PCI-Express to 100mhz
3) Lock PCI Synchronization to 33mhz
4) See Sticky thread for what CPU options to turn off
5) FSB to 333
6) Up the voltage (see C2D ocing thread for good voltages). Reboot. See if it boots o.k.
7) If it doesn't, you will have to reset the BIOS, two ways: 1) Hold Backspace or DEL on reboot (varies based on motherboard), or 2) There is a jumper on the motherboard (see manual and/or have this information ready if this happens), you need to switch it to the second position, for like a minute, then back to the other. This will effectively reset 6) Once a stable voltage has been found, do all the standard tests you can find to make sure it's stable etc..

I don't know what else to say. I am no expert, and the C2D ocing thread is way better than this info.. But it's a start.

 

Idontcare

Elite Member
Oct 10, 1999
21,110
59
91
@coolamasta - I have a striker extreme with an mulitplier unlocked QX6700 and my striker does not like having the FSB increased much at all.

Not sure if it is the same i680 chipset issue with high-FSB that is also the cause of these boards being incompatible with the 1333MHz FSB 45nm Yorkfield chips, but I can't get my QX6700 to run stable and reliable at 1333MHz on my striker extreme.

Since my chip is unlocked I just run it at 14x267 (3.73GHz).

Good luck!
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
Gentlemen, we need to put our heads together.

I had reasonably good luck with my Striker Extreme. I bought the board only after there were BIOS revisions with signals from forums that "stuff had been fixed." So I started with BIOS version 1004.

I'm compulsive-obsessive about BIOS upgrades. The Striker went through a few more upgrades since the Feb/Mar 07 version 1004 -- I think there was an "1101," then a "1301," and in July, 2007, ASUS posted version 1303.

I think 1301 fixed some anomalies with the fan-control features, and the earlier revisions fixed discrepancies between "set" and "monitored" voltages.

I stuck with the 1303 BIOS, although a funny thing happened: In August, a month after the release of the 1303 version, they released version 1305. People were having trouble flashing 1305 into their system. And I noticed that as soon as 1305 was released, ASUS departed from standard procedure and completely removed version 1303 from their web-site.

So I had some spare change, decided I didn't want to risk flashing, and ordered a new PLCC chip with the version 1305 from BIOSMan.com.

I just put the 1305 version in the system today.

I'm soooo p***ed -- I don't know what to say. I had over-clock settings for 3Ghz/1333, 3.15 Ghz/1400, and 3.2 Ghz/1424FSB. And all those settings would run with really tight latencies on my Crucials -- with BIOS version 1303.

NOW, with version 1305, I can't seem to make it work at even the "easy" 3.0 Ghz, 1334/667DDR setting. Also, there is a voltage discrepancy AGAIN -- A-[freakin]-GAIN -- between set values and monitored values.

I have a funny feeling about this, and a funny feeling about why ASUS would go through 18 months with this board, iron out all the bugs, with the board reasonably over-clockable for Q6600 processors, and then -- AND THEN!! -- release their final BIOS version so as to leave the board a complete dud -- and just about the time they were bringing out the Maximus and some other boards. Oh. Yeah. There's now a Striker II Formula and Striker II Ultimate -- with the new upgraded nVidia chipset.

 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
Coolamasta --

What BIOS version are you running with the STriker?

If it is 1303 or earlier, try this:

1.3188V <= VCORE <= 1.32+V
FSB = 1334
DDR = 667

If not completely stable:

1.2V_HT = 1.30V

You can also boost the NB and CPU_VTT to either 1.30 or 1.35V.

And the (MCH) 1.2V_HT can go up to 1.35V if you still have trouble.

These are relatively mild over-volts for the NB core, VTT and V_HT settings, and just a notch or two above minimum. The "Auto" settings yield voltages that range between 1.26 and 1.32V for these items, as monitored in BIOS.
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
PS

Start with memory Voltage (VDIMM) as follows:

2.0V <= VDIMM <= 2.125V

DO NOT . . . . DO NOT set the VDIMM at 2.2V, unless . . . FOR SOME REASON . . . your Striker Extreme does not show a positive discrepancy between the "Set" voltage and the monitored voltage. For instance, in both 1303 and 1305 BIOS versions, my board shows a steady +0.035V = Monitored - Set. So If I set the voltage to 2.2V, the board is telling me that it is really 2.24V.

I burned out a module set to 2.2V in a Ballistix DDR2-1000 kit.
 

coolamasta

Member
Jan 8, 2008
57
0
0
Im not sure mate, will check later on :)

Thanks very much for the info, will give that a try if im on the 1303 bios, if im not and on an earlier one and you say Asus have removed the 1303 rom from their site, where can I get it from do you know?

Thats a bummer with your new bios chip, seems weird why their final release has messed it all up again :(

Thanks for other peeps replys as well!
 

coolamasta

Member
Jan 8, 2008
57
0
0
Ah cool so set the ram voltage to 2.125v, can i leave the timing 5-5-5-15?

What are these settings mate out of interest?

1.2V_HT = 1.30V

You can also boost the NB and CPU_VTT to either 1.30 or 1.35V.

And the (MCH) 1.2V_HT can go up to 1.35V if you still have trouble.

Cheers :D
 

Idontcare

Elite Member
Oct 10, 1999
21,110
59
91
@BonzaiDuck - In the year that I have owned this motherboard I have yet to have a day go by where I don't regret buying it. It has always been a total finicky POS.

I bought it when it first came out in mid Dec 2006. I always had trouble getting the overclocks to settle in and be stable, but when the weeks of waiting for someone to post a decent review of the mobo on the web began to turn into months of waiting it pretty much confirmed the board really was a big expensive elephant in the room that no one wanted to talk about.

I get better FSB overclocks on my $89 gigabyte DS3L. i680 FTL.

Having said that, I am using the ole crappy B3 stepping of Kentsfield (my QX6700). So I may literally be just too outdated on a number of fronts - both the i680 combined with a B3 quad.

It is vapor-phase cooled, so temps are definitely not my problem.

@coolamasta - sorry that your first experiences with Intel and quads is going to be with a Striker Extreme. I truly hope you have a good experience and that my experiences are due to me being a dud overclocker and not due to a systematic issue with the boards in general.
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
[First, answering IdontCare generally, then coolamasta]

Well, it was more of risk in purchase than I've taken with other motherboards.

I watched forum posts at i4memory, asus and other places, and I DID find some "good reviews." One of them took the board through its paces at different multipliers with the E6600 C2D. When it looked like BIOS revisions were clearing up some oddities with the board, I took the plunge.

Also, I originally configured the board with an E6600 C2D. With Crucial Ballistix, I was able to push the memory clock in 1:1 with the CPU to 390 Mhz, although I had to drop the multiplier to 7. I was able to get to about 375 with multiplier 9, and the only limit I had with that was self-imposed -- I didn't want to push the VCORE higher. Keep in mind this was all with air-cooling.

Also, with the E6600, I had the latencies at these settings tightened down to 4,3,4,9, with some settings at a 1T command-rate.

When I switched over to the Q6600 (B3), there was an entire new regime of FSB settings and latency thresholds needed to keep the VDIMM at 2.175 or below, but I got the B3 to 3.2 Ghz, as I may have said. I just made it a rule not to go beyond VCORE 1.42V. At 3.0 Ghz / FSB = 1,334 or CPU_FSB = 333 Mhz, 1:1 -- I could set the latencies to 3,3,3,8,1T with the VDIMM at 2.125V.

Suddenly, all this seems to change, going from BIOS 1303 to 1305.

Looking at the timing of the BIOS releases from ASUS and the processor releases from Intel, it looks too coincidental: BIOS 1305 was released right after the buying-frenzy began for the G0 stepping of the Q6600. In fact, it may be the case that after a year raking in revenue on this motherboard "Cadillac," ASUS didn't want the customer-base to have any more success with it. Who can tell?

[coolamasta]

If you check Anandtech's article on OC'ing the QX9650, 1.2V_HT is the voltage to the equivalent of the "hypertransport" link used in AMD boards. It is referred to as MCH for "media communications [hypertransport?]" -- someone else can correct me or elaborate, but the article explains that boosting this voltage may clear up residual problems with finding stable memory settings.

The NB core voltage is the voltage to the Northbridge core -- the nVidia chipset. CPU_VTT is the FSB termination voltage of the processor, and boosting it (carefully) should help with higher FSB settings.

I've pushed these voltages as far as 1.45V, although the reviews I mentioned above are more daring in pushing configurations to a higher limit. Personally, I think 1.45 is the limit of prudence. At the same time, the minimum voltage for the Southbridge core is 1.5V, and I haven't seen recommendations that go higher than 1.6. So I either use "Auto" or 1.55V.

A lot of people would agree that you shouldn't need to use other than "Auto" settings for a 3.0 Ghz over-clock, but this may have been speaking of 680i boards generally. I think that I did indeed have a perfectly stable 3.0 Ghz setting with these voltages on "Auto." But I wanted fixed settings in my system, and I discovered that 1.2VHT doesn't even change from its monitored value until you set it to 1.30V. The "Auto" settings for NB and CPU_VTT seem biassed upward slightly, so they seem to fall between 1.26 and 1.32V. You'll see the upward bias when you fix the settings.

 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
coolamasta --

Check your BIOS post screen to determine which version you have, then re-check the ASUS web-site. I swear that I must have visited there several times, but couldn't find that revision, but my eyes are getting old with the rest of me.

If, as I said, they made it "disappear," I saved my download, and I can upload it to my personal web-pages so you can download it. Of course, I take no responsibility for your results, but I wouldn't deliberately contribute to your misery, either.
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
Also, if you have version 1101 through 1301, you could stick with those. I don't think I had any specific problem with version 1301. None that I remember.

I had more of a problem when I tried to flash it. Sooo: ---

1) Download your BIOS, unzip it and save to floppy. DO NOT use the Windows ASUS Update program.

2) Shutdown the system, turn off the power supply, unplug the power supply -- THEN --
3) Set the jumper on the motherboard to make the "CLR_CMOS" pushbutton switch active.
4) Push the CLR_CMOS button once.
5) Push it again.

[Your choice whether to leave the switch active, or put the jumper back on the original pins where you found it. Make sure, though, that you put the jumper back on the board, either way. ]

6) Reconnect the PSU power, switch on -- make sure the pushbutton under the PS/2 I/O-panel plugs has activated the red and green LEDs on the mobo and turned off the blue utility LEDs, then power up the system.

7a) Hold down the delete key, and enter the BIOS. Use the EZ-Flash menu option to flash the BIOS from the BIOS screens, OR --

7b) Follow ASUS instructions for using the DOS/floppy utility to flash the BIOS --

[in either case, from a floppy.]
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
Now I took the time to check the ASUS download pages.

Confirmed -- again -- v.1303 vanished as soon as they posted 1305 in August.

Also -- I suggest you stick with version 1301.

If you ever change over from Conroe 65nm processors to Wolfdale 45nm, that may be the time to experiment with 1305 BIOS version.

I'm still trying to sort this out. Preserving the PLCC version 1305 BIOS chips I ordered, I think I'll re-socket a v.1102 chip and flash it with 1301.

Another note: I've been flashing BIOS's for 15 years, maybe 20+ computers, and some machines three times or more. Last summer was THE FIRST mishap I'd ever experienced -- but I got careless, didn't clear the CMOS, and used the Windows ASUS Update utility. I had started using the latter utility a few years back with a Springdale ASUS motherboard, and got too comfortable with it.

So again -- First -- flash the BIOS with traditional means or through the BIOS menu. And second -- I highly recommend, but for the loss of $25, to use an online company like www.BIOSMan.com to buy spare PLCC chips. They give you a six-month subscription for free "flashing" if you recycle your spare PLCC chip. Or maybe it's one free update over six months -- check their web-site.

BIOSMan is a firm in Fremont, CA. I'm suspecting that the OP here is from "down under" [Right, mate?] But they always amaze me: I think I ordered the 1305 chips from them sometime last week, and they were in my mailbox sent as US Mail within three days.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
59
91
That may be part of my problem, I gave up very early in the game to try and squeeze out high FSB on my striker. Meaning I am still using Bios 401.

Do you know of any reported issues flahing a bios from such an early rev to the latest and greatest?

Any "ZOMG do NOT flash your striker directly from 400 to 1200, it should never go past 11!" type threads?
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
[First, responding to Idontcare; then second, some progress on one overclock configuration to recapture the "easy" 3.0 Ghz setting . . . ]

Idontcare --

I don't (ha!) know of any such threads, never heard of it.

Both my Strikers are mobo revision 1.00g. I bought one in March 07, and the second in late July (didn't I say that I had a "panic" over the bad BIOS flash? And only came to my senses to order from BIOSMan after I chose to buy rather than wait for RMA from ASUS? -- I have plans for #2, anyway.)

But there shouldn't be any difference if -- there are indeed -- additional hardware revisions, since the only BIOS downloads they post are not specific to hardware revision.

My best understanding of it is this. The BIOS information is stored in the PLCC chip. It's the same PLCC chip for whatever BIOS version came shipped with your board.

You should be OK if you follow my advice above. You also might want to try version 1301, which never "disappeared" for the ASUS web-site. Apparently, they had cleared up the "fan-control code" issues with 1301. And when I was using it, I was able to do . . . . what I said above I did with the board.

The only other question -- "What memory are you using with it?" OCZ apparently works; Crucials work GREAT; G.SKILL apparently OK. There were some issues with Corsair with the earlier BIOS revisions. But I think that even those things have been cleared up.

If I don't find info about memory from customer-reviews and forums that is specific to Striker Extreme, I look for comments per the 680i boards in general. . . .

[MOVING TOWARD 3.0 GHZ AGAIN . . . ]

I started experimenting with memory "multipliers/dividers" other than 1:1 before I started with this accursed BIOS 1305. Just for the heck of it, I tried one setting at 2.88 Ghz and a 4:5 divider so that I could run the memory at the DDR2-800 spec and tighter latencies.

That proved OK with this BIOS:

VCORE = 1.3188V (same as needed for 3.0 Ghz)
FSB = 1280 (CPU_FSB = 320)
DDR = 800 (RAM FSB = 400)
VDIMM = 2.125 to 2.15V (depending on latencies -- I'm still tweaking it . . . )
1.2VHT = 1.35V
NB = 1.35V
CPU_VTT = 1.35V

The 4:5 ratio is, according to Anandtech and my own experience per synthetic benchies, the next best thing to 1:1.

Soooo --

I decided to bump up the frequencies a hair:

VCORE = 1.3188 (same)
FSB = 1312 (CPU_FSB = 328)
DDR = 820 (RAM FSB = 410)
VDIMM = 2.15 (not necessarily lowest -- still tweaking, but same latencies as above)
Other voltages the same.

So far, passing PRIME95 after 30 minutes. [30 minutes stable is a lot better than freezing during POST before the memory-test odometer -- requiring attention to clearing CMOS.]

3.0 Ghz should just be a few Mhz away with some tweaks to VCORE and memory timings. My guess is that running the memory at 4,3,4,9 now, once I push the FSB to 1,334, I can get them to run at 4,4,4,10.

But 1:1 settings should be achievable. They sure as heck were achievable with the older BIOS . . . . :frown:




 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
MORE OBSERVATIONS ABOUT STRIKER EXTREME BIOS 1305

I'm only speculating about what BIOS revisions might occur to accommodate G0 Kenstfield steppings and Wolfdale Penryns. Only speculating.

But these newer processors run at lower voltage requirements. It appears that my mobo is "behaving about the same" until you get to a range between 1.31 and 1.32V. After that, there is a widening discrepancy between "set" and "monitored" voltage values.

With my version 1303 BIOS over-clocks, there seemed to be near-perfect correspondence between "set" and "monitored" -- as you go up the scale. But now, I need to boost the "set" value to 1.34+V just to get BIOS monitor (upon reboot and re-entry into BIOS Setup) to show 1.32V.

So you wonder if ASUS, or whoever reprograms the BIOS versions, is striving for greater accuracy for "set" versus "monitored" versus "real" in a lower voltage range, and if the revisions or tweaks to make that happen cause less accuracy and more discrepancy in the higher ranges. You'd think, however, that only changes in u-code to accommodate new processors would be made, but that doesn't cover the sieve of possibilities, or even the interaction between one set of improvements and another set of BIOS features.
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
THE GREATER STRIKER EXTREME MYSTERY SOLVED

I was going to post this as another "Part 2" thread to this one, linking them together. But those who have an interest in this will come to this post, and I won't clutter the forum unnecessarily.

SUMMARY

Some people have had trouble over-clocking their Striker Extreme motherboards. Personally, I never did have such trouble -- until last week. The board may have limits, but ASUS avoided the problems with the other reference-board designs from nVidia and other manufacturers last year, as confirmed by nVidia officials themselves -- early last year.

As I said, the board went through a spate of BIOS revisions. These BIOS revisions occurred during the time that (a) the Conroe-C2D processors were released, (b) the Kenstfield B3-stepping was released, and (c) the release of the G0-steppings, E6x50 and other C2D/C2Q pre-Penryn processors.

If you check the ASUS web-site BIOS download pages for the Striker, you will see several -- more than one BIOS revision -- which shows in the list of "fixes" the following item:

" . . . . improved over-clocking capability for newer processors . . . "

With the Striker board, ASUS probably anticipated a long life-cycle that would span pre-Conroe through Penryn CPU releases. And, as I understand it, they were mostly successful, excepting the "Great Yorkfield Promise Robbery" -- something that still needs to be resolved. If it can't be resolved with this board, then -- so be it.

I noted that the last BIOS revision I had found at ASUS in July -- there were two July releases -- was version 1303. Version 1303 was mysteriously removed from the ASUS web-site when the company posted version 1305 in August. Version 1301 was also released in July, but can be found on the BIOS download pages consistent with standard procedures for these companies.

If you look at the "fixes" noted for version 1305, it says:

" . .. . updated to improve over-clocking capability for 1,333-FSB processors . .. . "

And as I discovered to my dismay, and as mentioned above, you cannot over-clock a B3 Q6600 worth a good g--damn under version 1305. How many of these board shipped with version 1305, I have no idea. Version 1305 appears to be the final BIOS revision to the board, although we'd hope ASUS would release at least one more which would miraculously restore the "Yorkfield Promise." Of course, ASUS never made a "Yorkfield Promise," but they did promise "45nm processor compatibility" -- referring to Penryn. Apparently the board has been certified for Wolfdale, and the rest of you Yorkfield hopefuls probably know the current status of things in that regard.

I agonized briefly this morning (at 4AM -- a good time to focus on this . . . . stuff), whether to salvage one of the two 1305 PLCC chips I had installed and replace it with another PLCC chip to reprogram with either BIOS 1301 or 1303. I'd been using 1303 since late July, without too many problems. But that version literally evaporated from ASUS on-line, and you can speculate all over the place as to why, but that and a dime will get you less than an uncertain cup of coffee. Since 1301 solves the problem of "Q-FAN" configuration, and I'd been using 1301 briefly before flashing to 1303, I decided to revert to version 1301. I also figured that I have still one 1305 PLCC chip to use if I ever update to either G0 or Penryn, and then only if I have some disaster flashing back to version 1305.

THE BIG EMPIRICAL SKINNY ABOUT THE BIOS VERSIONS

With version 1301 successfully flashed to the board through the BIOS EZ-FLASH sub-program, I reset my default settings again to 3.0 Ghz 1:1 configuration (FSB 1,334, DDR 667, 1.32V VCORE, 2.15V VDIMM and 3,3,3,8,2T timings).

The ol' Striker Extreme is now going through a PRIME95 BLEND test and just ticking right along as stable as can be. This also means that I can, at whim, re-configure to settings at 2.88, 3.15, 3.17, and 3.2 Ghz, depending on how I want to configure memory, and of course I can choose between 1:1 and 4:5 ratios I've used.

THEREFORE, ALL YE STRIKER REJECTS AND AFICIANADOS, YE CRESTFALLEN, DISCOURAGED, DEMORALIZED GAMER-OVER-CLOCKER-ENTHUSIAST-SPENDTHRIFT-STRIKER-OWNERS -- TAKE HEART!!

YOU MAY BE ABLE TO MAKE THE BOARD WORK FOR YOU WITH Q6600 B3, POSSIBLY WITH Q6600 G0 AND OTHER LATE PRE-PENRYN PROCESSORS, IF YOU REVERT TO BIOS VERSION 1301 for processors in the Conroe/Kentsfield line, and use 1305 (possibly) for G0 steppings and (most probably) for Penryn Woldfdale. If G0 gives lackluster results under version 1301, then flash to version 1305. If you swap processors to Wolfdale, you may want to test the system under version 1301, and then see how it might work under version 1305.

I hope this helps somebody. It's not that I'm delusional or making up something in my head about what appears on my monitor now. I can only say that everything is back to normal with BIOS 1301, and you should look closely at your BIOS revision if you're having trouble with B3 processor steppings and frustration at their "apparent" over-clocking prospects.
 

coolamasta

Member
Jan 8, 2008
57
0
0
Hi BonzaiDuck, thanks greaty for all your input mate, I am going to read through all your replys now :)

As my Striker is only a few days old I have the 1305 BIOS on it already, DOH!

Ive not played with any overclock settings yet but am just doing some screen dumps from CPU-Z with everything standard...

Would you say that I have the G0 chip as I didn't think I did but CPU-Z says Stepping B, Revision G0 (good thing or bad thing?)

Anyway what do you think I should do mate? Try and get 1305 to work of flash back to earlier 1301?

Im going to be online for next few hours anyway, let me know :D

Cheers for all your help mate, much appreciated!!

Screen dumps on way...

 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
Coolamasta --

It's a G0. If you can't get to 3.0 Ghz, 1,334 FSB, 667 DDR with 1:1 ratio and a VCORE of 1.32V, then it suggests to me that they've designed the 1305 BIOS with the Penryn in mind.

Remember what I said about the timing of the releases. Intel released the G0-stepping processors just about the time that ASUS had just released v.1301. But ASUS shows signs of trying to stay a step ahead of Intel in terms of product releases. They no doubt acquire engineering samples of CPUs so they can either tweak their existing product to be ready for the newer processor, or develop a motherboard that will be compatible with it. So there is no certainty that v.1305 was designed with both Penryn and G0-Conroe/Kentstields in mind.

We're trying to second-guess manufacturers, and they're playing a similar game that is both coy and aggressive in the competitive market.

So if you can't get the easy over-clock to 3.0 Ghz with the G0 using BIOS v.1305, drop back to v.1301 and try that.

After my first BIOS disaster in 15 or 20 years of flashing updates, I'm recovering a little from being overly cautious. The Striker Extreme's EZ-Flash bios-subroutine seems to be quite reliable. So you don't have to make a DOS-stub bootable floppy, but only format a floppy and copy the extracted BIOS BIN file to it. For 1301, this would probably be 1301.BIN. You can also use EZ-Flash to back up your 1305 BIOS to another disk, but why bother when you can just download the zip file and save it for later use?

LET'S REVIEW . . . .

You have a G0-Q6600 -- Zalman cooler, even if second-best, is more than adequate for what I'm guessing will be some nice over-clock settings with the Striker -- not stunning, but pretty darn good. (And I imagine you can probably get the Q6600 G0 to 3.4 Ghz later on).

You have high-end Crucial Ballistix RAM. I started with their DDR2-1000 2GB kit, and need to RMA back to Crucial because a module went south just for my volting it at their warranty-recommended max (but tight timings -- to squeeze blood out of a stone.)

I'm currently using Crucial Tracer DDR2-800s. BEFORE I FORGET -- YOUR BOARD WILL RUN ALL MODULES AT "NATIVE-DDR2-800" MODE, EVEN IF SPD /EPP CONTAINS LATENCY SETTINGS FOR THE HIGHER SPEEDS. That's why your CPU-Z is showing "400 Mhz." The double-data-rate is 800.

So when you start (cautiously) overclocking, you would go into BIOS, to the "Extreme Tweaker" submenu, set the first item to "Manual," enter "System Clocks" and set the first three PCI-E items to 100 Mhz if not already set to that, and set the last item to 200 Mhz. This last item is crucial, because if left at "Auto" it will over-clock your PCI_E bus automatically in proportion to the processor over-clock, and you don't want to damage any hardware. I believe it is similar to the "AGP/PCI ratio" setting on ASUS Pentium 4 Springdale and Canterwood motherboards.

You would then enter "FSB and Memory Config," set FSB/Memory Clock mode to [Unlinked], set FSB [QDR] to [1,334], and set MEM [DDR] to 667. Here, you're running your high-end Crucials way below spec in bus speed, but it's a 1:1 ratio with the processor frequency [333.5 CPU and 333.5 memory, or 4 x 333.5 = 1,334 and 2 x 333.5 = 667 -- with "QDR" = 2 x "DDR"]

At this point, leave the Overclocking menu alone, although you may want to explore it. But the Multiplier should be defaulted to 9, and the Memory Timing should show "Auto" for everything.

Go to "Over Voltage" and (carefully) set the VCore voltage to either 1.31875V or 1.32500V. This should be in excess of what the G0 will need, but it's still at well below the maximum spec -- you have about 0.05V wiggle-room before the setting yields "load" readings that will exceed 1.35V.

I recommend setting the Memory Voltage to 2.0V initially -- and you wouldn't hurt by starting at 2.1V. The Crucials recommended maximum is 2.2V, and that's about three notches above 2.1. Of course, you could leave the memory at "Auto" because the latencies and command-rate are still set to "auto," so the default should probably work.

In the other menus, set all "Spread Spectrum" to disabled, set "Vanderpool" to disabled, set SpeedStep, Thermal Control to disabled, Enhanced C1 to disabled -- leave the other settings in that menu alone. The remainder under "Advanced" is up to you -- you'll want to set Primary Display Adapter to "PCI_E" -- Plug-n-Play for Win XP should probably be "Yes". If you have another audio card, disable "HD Audio."

The other voltages -- you can try them at "Auto," but it won't hurt to bump up 1.2V_HT to 1.30 -- probably the same for NB core and CPU_VTT. At this initial 3.0 Ghz setting, I leave the Southbridge set to "Auto".

With these settings, if the system doesn't even post, you know you want to go back to BIOS 1301. If it posts and runs MEMTEST86+, but shows errors, you might want to fix or bump up the memory voltage. If it boots to Windows, and PRIME95 errors occur, or if it freezes requiring a Reset, -- BIOS version 1301 is probably needed.

If you get into a "lockup" and can't get the system to boot into BIOS again, follow the procedure for enabling and using the CLR_CMOS button. Then on anothr computer or with this one at all "Auto" settings -- download the v.1301 BIOS if you need to and prepare a floppy.

EZ FLASH is on the "Tools" menu.

Once you have a stable setting at 3.0 GHz, 1:1, 1,334 / 667, several hours running PRIME95 BLEND test, you can experiment with the other divider ratios, or you can try for 3.15 Ghz 1:1, 1,400 /700 . . . . . Excelsior! Once you have the voltage requirements for any or all of several 1:1 over-clocks, you can see what happens with 4:5 (or other ratios) to push your memory closer to their full spec. You can also run them under spec and start trimming down the latencies. Those modules should have default values identical to the Tracer DDR2-800's -- so you should be able to run them even tighter (like me) between DDR2-800 and DDR2-875 with timings 4,3,4,9 (800 Mhz) and loosened to 4,4,4,10 above maybe 820. I'd say the possibility of tighter timings with speeds at 900 or above is also promising, but I recommend voltage settings at 2.175 or lower. You can ask around -- running them at 2.2 or higher is risky.

I'd see what you can do with multiplier 9 first. We've tried several lower multipliers on this board, but I'm inclined to stick with the stock value. YOU CAN test the limits of your memory subsystem first by dropping the multiplier all the way and running up the FSB, but since there is likely a BIOS issue here regarding your choice of processors, best to try the 3.0 GHz clocking first with these or similar settings -- get that out of the way.
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
PS Just a sample, but before I even ordered my first board, I was visiting this web-site for info. And I think some of these people are in "your neck of the woods," mate!

EVA2000 on (early) Striker Extreme with E6600

Poke around the i4Memory web-site. EVA2000 and some others have posted more than just this thread on the Striker. They've run through configuring it with several graphics cards and memory modules, also. Remember that this thread linked above was put together when the older BIOS versions showed much worse discrepancies between SET, MONITORED, and ACTUAL (multi-tester-measured) voltages.

The best guideline is that many boards seem to post 1.37V as Q6600 "default" Vcore, although the board will run at otherwise stock settings at well-below 1.30V. The retail-box "Maximum voltage" is 1.35V, but to reach that under load conditions, the set value probably needs to be around 1.37+V. Some articles have said that 1.46V is still "safe" for the B3 Q6600 stepping. I won't let mine go over 1.42. I'd be a bit more cautious with the G0 -- but see what people are saying about what's needed to push the speed to the stratosphere. As I said, 3.4 seems to be reasonably attainable with that stepping, at voltages somewhat lower than the B3.
 

coolamasta

Member
Jan 8, 2008
57
0
0
:| :| :| :|

Thanks for the reply mate but it didn't want to know at all :(

Tried what you said in this order:

  • PCI-E clocks to 100Mhz
    The last item fr PCI-E bus to 200Mhz
    QDR FSB to 1334
    MEM DDR to 667
    Vcore to 1.31875
    All spread spectrums to disabled
    Thermal control to disabled
    Enhanced C1 to disabled

Tried this, wouldn't even post!!! Had to reset Bios which meant I had to take my bottom graphics card out grrrrrrrr lol

Next after a Bios reset I done the same as above and also done:

  • Mem voltage to 2.1v

Tried this, wouldn't post, reset bios then done all above plus:

  • 1.2v_HT to 1.3v
    NB core to 1.3v
    CPU_VTT to 1.3v
    Southbridge to auto

And guess what, still wouldn't post :( :(

Back to auto now so it would boot lol

Where can I get 1303 or 1301 Bios from mate? which one would you say to use out of the 2?

Cheers :)
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
See, I could put my 1303 BIOS file up on my web-site for a day or so for you to download, but I don't think you need to do that.

I've got some questions about the wisdom of using it. Why did they put it up at ASUS web-site for three weeks, then pull it? What's the difference between that one, and v.1301, when they only posted 1301 two weeks before posting 1303? And of the two, why did they pull 1303?

I'm back again "certifying" settings for v.1301, and all the voltage and over-clock settings I used before seem to be stable. Try that one first.

I have to say, though, that if you did all that you summarized above, it sounds like the frustrations I had after feeling comfortable with over-clocks between 3.0 and 3.2 with earlier BIOS versions. It really looks very much as though v.1305 is exclusively for:

a) Conroe/Kentsfield "flagship" processors designed to run at 1,333 FSB
b) Penryn processors designed to run at 1,333 FSB

I think some G0-steppings of C2D -- like the E6850 -- are spec'd at 1,333 FSB, so I'd probably be interested in seeing how they work with this board and BIOS v.1305.

If you can't get your G0 Q6600 to work with the earlier 1301 BIOS, it would leave me totally clueless. Why would I be able to OC my B3 stepping with the earlier versions to 3.2 Ghz? We're using the same board, same board revision, same memory, and only a different stepping of processors designed to run at 1066 FSB.
 

BonzaiDuck

Lifer
Jun 30, 2004
16,053
1,681
126
. . . only difference is that you have two 7900 GFX cards in SLI, and I have one 8800 GTS. But somehow I doubt that has anything to do with failure to get to 3.0 Ghz.