Jump to content

NF3 - Beta BIOS 3/31/2005


Recommended Posts

  • Replies 98
  • Created
  • Last Reply

Top Posters In This Topic

unfortunately i have tried all that, and it still hard locks the machine when going into idle mode. i think its because i have two sticks of double sided ram 2x512MB Geil Value PC3200 i understand there was some work done by the aurther of RMClock into the matter and the bios tables needed an update but i have no idea weather that was done.

Share this post


Link to post
Share on other sites

Nice bios DFI. I lost my ethernet connection a month ago and tried reflashing countless times with different ones and could not get it back. Just flashed the 3/31 last night and couldn't believe when I had my onboard network back.

 

Also seems my TwinMos SP utt ch5 sticks plays more friendly with this bios, now up to 260htt at 3.45v. Benchmarks seem to have a slight increase nothing great but nice to have.

Share this post


Link to post
Share on other sites

Please explain? If I could use it what would it do for me?

 

Thanks/Rubidoux High

 

For me, Bank Interleave enabled/disabled is the difference between ~600-700MB/sec of Memory Bandwidth.

 

I've noticed something this bios fixed actually, apart from making Bank Interleave more stable (than 1/28) with my setup.

 

With the 10/15 bios, if i enabled "USB Mouse Support", most of the time i would have to unplug and re-plug my mouse from the USB port once in Windows, or the no movement would register.

 

All good now though :D

Share this post


Link to post
Share on other sites

Guest shaolin95
i tried last night, it is also cant run stable @ 300MHz 1:1 with 2 sticks of TCCD, just the same as the old bios.........

 

I dont think a Bios is going to fix that as thats mem controller related...

Share this post


Link to post
Share on other sites

Well, great job Oskar Wu and the rest of the DFI guys... thanks a lot.

I flashed it last night and for the first time EVER I was able to pass MemTest Test#8 at 257 3.2vDimm all tightest timings :)

 

For me Test#8 is basically either going to work or not going to work... timings will reduce errors at a specific MHz but have never completely removed them (DQS & Drive Strength are useful here though ;) )

 

With this BIOS I jumped right back in with the same EXACT settings as with 1/28 and it passed 16 loops before stopped it... previously an error every 2-3 passes was the norm.

 

On the other hand, I am now not completely stable in Test#5, but at least Test#5 has always been more responsive to timings for me... so hopefully I will get her ironed out and be 100% stable with 257 @tight (uwackme TM)

 

Bout the added bandwidth.... well, yes and no for me.

With EXACT same timings as in sig I get the same bandwidth in MemTest as before.... at 9x257 I get 1965MB/s and with 1/28 I got the same.

BUT, now the effect of Tras has changed...

For some reason with 1/28 Tras=8 had some magic effect for me where I would gain ~3MHz in test#5... Tras=8 was less efficient clock-for-clock, but it was a lot more stable.

Now, however, Tras=7 & Tras=8 have a fair bit more bandwidth than Tras=5 & Tras=6 (1965MB/s vs 1923MB/s @ 9x257) though they don't gain me any MHz either, from what I have seen so far...

 

I'm still in the process of trying to get 257 100% stable, but even if I dont, at least now I won't have those pesky test#8 errors and I'll be able to get 256 error free in ALL tests.

Share this post


Link to post
Share on other sites

Well I got my 3400+ today! I will flash this bios after I run some prime thru her and see what shes stable at (dont want to use an unknown bios to test stability I knew I previously had!). Right now im running 250x10 @ stock volts!! YAY! This chip looks like it has 2.7ghz in it ! 270x10 here I come :)

Share this post


Link to post
Share on other sites

OK... messed around more last night and I am ooohh soo clooose to getting 257 2-2-2-8 3.2vDimm rock solid in ALL tests... this BIOS is the most responsive/predictable in terms of timing adjustments. I haven't seen any wierd quirks yet, Test#7 & Test#8 have ALWAYS been beeeaarrches to get stable, but with this BIOS they are rock solid.

 

Like I said in previous post, Test#5 is the first to throw it's toys out the cot. But, Test#5 is also the easiest test to get stable from past experience.... this test seems to be almost purely memory related, rather than memorycontroller.

Loosening Trc from 7 to 8 and Trfc from 15 to 16 gets me 3 errors per several hours so hopefully one more increment will get it stable. Then I'll start taking Trc or Trfc back down and see what happens... hopefully only one of them is causing the occasional errors and I can tighten the other one right back up.

 

If anyone is running a Mobile ClawHammer with OCZ VX memory on this board I STRONGLY recommend giving this BIOS a go... use the timings in my sig as a starting point and you should be almost at the memory limits... Trc & Trfc may need slight adjustments but that's no biggy.

 

[EDIT] Tony, what's up with your Tras=3? Do you get more speed, more stability, or both?

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×
×
  • Create New...