Jump to content
Sign in to follow this  
raju

975 oc'ing mini guide

Recommended Posts

The second issue that is know cropping up is that anyone who has tried 830 bios and has had to flash back to 0731, is having cold boot issues even though 731 worked fine for them in the past. I did suggest that the newer version of Winflash be used, but that did not help.. So they now have to buy new bios chips, or hope dfi will help out with a replacement ... As well as this board runs for some, it's proving to be a real dud for others. I am not trying to be negative here, there's a lot that needs looking at before DFI issue another grossly non universal bios.

 

:rolleyes: Cheers m8, nice to know your on my side. I've contacted Dfi europe bout getting new bios chip (be calling on you guys ere to help me out if they dont). I'm 100% stable, but cold boots are wrecking my head.

I was trying bios 0823 before 0830 and that is where i first got the cold boot issue, maybe its 0823 that causes the problems.

Share this post


Link to post
Share on other sites

nice work 'maan', :) you are once again my master...

 

 

 

MY 385 first screen is ready, I'm back at work today so I had to leave PI running while I was away, 13 hours no less.

 

ocdbentry2001mv6.th.jpg

 

 

rest will follow in the db this evening..

 

regards

Raja

Share this post


Link to post
Share on other sites

happy games, well done bro'. I did not even think of leaving the memory timings on auto...

 

in any case thanks to your input priming at 390fsb on 4:5... getting closer to the illustrious ddr 1000.....

 

please amend the mini guide with your findings, or write up a new one....

 

 

regards

Raja

Share this post


Link to post
Share on other sites
happy games, well done bro'. I did not even think of leaving the memory timings on auto...

 

in any case thanks to your input priming at 390fsb on 4:5... getting closer to the illustrious ddr 1000.....

 

please amend the mini guide with your findings, or write up a new one....

 

 

regards

Raja

lol hell no dude i got way too much to do on my own (doing the comparisons vs AMD at 1024, 1280, 1600 resolutions, singe and Crossfire, etc)

 

that's going to keep me busy for the next 5 days, as well as I have a computer to build for a customer...

 

remember, I used YOUR guide to get where I and with the 400x9...so it should be you who does it lol

 

 

also note...that soon enough, others will add to it like a wiki, and soon enough after that we'll have the Intel's version of the Overclocking Guide, just like the giant, conglomerate of info in teh AMD version that is linked far and wide across the internet ;)

 

(it has basically become THE BIBLE for overclocking AMD64 cpu's)

Share this post


Link to post
Share on other sites

Just to keep you lot up-to-date. I've already posted this info on xs forums. I've received this response from Dfi about my cold start issue.

I have done a bit of checking and have determined that when updating to the 0830 version certain areas of the DMI are overwritten, when flashing back to the 0731 they are not. Try reflashing the BIOS in DOS mode (not thru windows or a windows DOS session) with the 0731 version. This time however type in the following at the A: "awdflash 97TID731.bin /f/cc/nbl/qi" (Without quotation marks of course). The extra switches will then "format" the chip and allow the DMI to be rewritten in accordance with the 0731 BIOS. Of course I have not had any problems with "cold boot" either way so I can't say if this will help for sure, but it should get rid of any residual traces of 0830 and is worth a try. Let me know if it helps.

 

I've tried it and it seems to completely re-write the chip, not like awdflash which skipped the last segment. It's 2 early to say if it worked, but i'll let you all know.

Share this post


Link to post
Share on other sites

I have to admit, I was pretty dissapointed when I first got this board. After playing around with it a little I'm a little happier with it. I was having a lot of the boot problems with bios that was on the board when I got it and flashing to 830 didn't help at all. I reflashed to the 913 (hope that's the correct number) yesterday and was able to boot at 380fsb consistantly with my 6700es @ 1.5vcore, this is with the ram at 1:1. I was able to boot at 385 once, but system shut down opening cpuz, either needs more vcore or the memory timings were a little too tight. I haven't really had time to tweak everything, but have managed a 13.5 sec 1M Spi with almost nothing tweaked. I'm planning on pushing it a little more on air, then bolt on a tube and do some dice testing. My cpu an nb temps seem to be fine, I'm running the board on the bench and have a 120mm Delta blowing on it :)

Share this post


Link to post
Share on other sites
Speaking of C1E you didn't mention disabling that or the other two things you mentioned disabling in another thread (virtualization and execute disable bit).

 

I'm just starting to play with OC settings right now and the C1E has got me a little stumped. Specifically, I started doing exactly what Happy did in the "Happy's & Rgone's Core2Duo Thread" -- E6600 @ 1.4v, 334x9. Ram 2.1v, sys mem frequency to 667, and all drive strengths to 3x but otherwise all RAM settings auto.)

 

With C1E set to auto, (so the CPU's @~2.0GHz when idle, 3.0GHz under load), Dual prime test goes happily for 12-13 hours. But going back into BIOS and turning C1E *off* (making no other changes), Prime fails consistently and very quickly (under 10 minutes).

 

When it's under load, the settings seems to be identical per Everest (RAM is reporting as defaulting to the same settings, 5-5-5-15 @ 400MHz, etc) but the behavior sure isn't.

 

And once thie rig is stable, it will be running BOINC as its default screensaver, so the C1E functionality won't do a darn thing for me anyway...

 

What am I missing?

 

-G.

Share this post


Link to post
Share on other sites
With C1E set to auto, (so the CPU's @~2.0GHz when idle, 3.0GHz under load), Dual prime test goes happily for 12-13 hours. But going back into BIOS and turning C1E *off* (making no other changes), Prime fails consistently and very quickly (under 10 minutes).

...-G.

 

Something i noticed is that after attaining the best stable O/C that i could get (360 fsb)going by the 'mini-guide', i was able to reduce my voltages to allmost stock (VID + 25mV, VTT 1.25, NB 1.65) AND enable C1E without detriment. Said goodbye to a few Deg C in the process. :D

Share this post


Link to post
Share on other sites

i gotta say the mini guide is 'AWESOME' been able to get 360 bus speed ram at 5-5-5-18 1:1 and i'm happy with it

 

thanks everyone

Share this post


Link to post
Share on other sites
I'm just starting to play with OC settings right now and the C1E has got me a little stumped. Specifically, I started doing exactly what Happy did in the "Happy's & Rgone's Core2Duo Thread" -- E6600 @ 1.4v, 334x9. Ram 2.1v, sys mem frequency to 667, and all drive strengths to 3x but otherwise all RAM settings auto.)

 

With C1E set to auto, (so the CPU's @~2.0GHz when idle, 3.0GHz under load), Dual prime test goes happily for 12-13 hours. But going back into BIOS and turning C1E *off* (making no other changes), Prime fails consistently and very quickly (under 10 minutes).

 

When it's under load, the settings seems to be identical per Everest (RAM is reporting as defaulting to the same settings, 5-5-5-15 @ 400MHz, etc) but the behavior sure isn't.

 

And once thie rig is stable, it will be running BOINC as its default screensaver, so the C1E functionality won't do a darn thing for me anyway...

 

What am I missing?

 

-G.

make sure you have all that junk in the same place as the C1E disabled (Virtualization, Limit CPUID MaxVal)

 

when you do this, the machine will scare you by shutting completely off...then powering itself right back on.

Share this post


Link to post
Share on other sites

as HG aid it's best to stick to the usual method of turning all those extra functions off.

 

On a sidenote (just for explanation sake), if you find you are passing Orthos with C1E 'ON' and failing it with it 'OFF', run memset in XP with C1E 'OFF' and note the timings all the way down the list. Then turn C1E on and run memset again to see if there is any difference, between the 2 positions (assuming all else is left alone in bios).

 

From your description it appears all your ram timings are on auto (which is fair enough and probably best).

 

The upshot is that with the C1E fuinction 'ON' if the board is booting at a lower divider it may set the secondry ram timings differently..

 

If it turns out that there is a difference in the 2 sets of memset timings, just apply the C1E 'ON' timings to the C1E 'OFF' timings in memset.

 

 

The second theory is that C1E in the 'ON' postion is not really allowing your CPU to run at full speed during Orthos (which would be bizzare and highly unlikley), which is why you pass.

 

 

Other than that on my part, much wasted text, with little to say....

 

 

regards

Raja

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...