Many thanks ... I'll give the modified BIOS (1301) a shot later. I'm through messing around today :up:
Printable View
Many thanks ... I'll give the modified BIOS (1301) a shot later. I'm through messing around today :up:
Just as an update, I'm still running the Mr.Toad modified BIOS for the R3 Gene and running strong after the last few months of heavy usage. Looks like Asus still isn't interested in updating their BIOS how it really needs to be fixed though :(
Played with my 980x a bit this weekend. Looks like I was wrong about the IMC. It will only do 3.4GHz vs the 3.6GHz of my 970.
The change in GFlops is minor.
The core of the chip is pretty good. It's the first chip I've owned that can do 4.5GHz with HT on, and the first chip I feel comfortable with the temps to run it at 4.3GHz 24/7.
4.33GHz, new 24/7 clock:
http://i144.photobucket.com/albums/r...-83gflp-ht.jpg
4.5GHz HT on:
http://i144.photobucket.com/albums/r...86GFlop-HT.png
4.5GHz HT off:
http://i144.photobucket.com/albums/r...6-968GFlop.jpg
You can see I did 4.5GHz with UCLK at 3.2GHz. I got 96.8GFlops with this 980x with 3.2GHz UCLK, and got 97GFlops with my 970 at 3.6GHZ UCLK. The difference is very minor.
Zoson, mate, that chip is just bad. :(
Can you sell it and find another?
Yeah, the IMC is a bit weak. But the core is still better than any other chip I've owned... :(Quote:
Originally Posted by donmarkoni
My best 970 does 4.5ghz HT off at 1.525v BIOS... This 980x does 4.5ghz HT off with 1.4875v BIOS... With my 970 I was VERY close to getting a stable 4740 run... I'll probably be able to do that with this 980x, I just haven't had the time yet...
I was hoping for better, so I'll do a 4.6GHz+ run on the 980x, post it to the 4.5GHz club, then try to sell it and get another.
Edit:
Also forgot to mention that the 4.33GHz linx was done with my AC off and a room temp of 31C. So it really can do 4.33GHz 24/7 without worrying about overheating. I did the 4.5GHz runs with my AC on, which was able to keep my room temp between 24C and 25C. On a cool night it can bring my room temp down to 19C, and that's when I shoot for the big clocks.
Edit2:
Maybe I'm the only one with a working Marvell controller... I'm using two Intel X25-M G2's in Raid0 on it with no issues at all... Still running whatever driver windows installed from the get go.
so i can consider my 980x a "good" one.... 4.5ghz ht on with 1.43v ??
I have not seen any linx resutls for your 980x... Just for your old 970... If you can do 20 LinX stable at 4.5GHz HT on with only 1.43v... Yeah that's a nice CPU. You should clock it and submit an entry to 4.5GHz+ club!Quote:
Originally Posted by p3gaz_001
http://www.xtremesystems.org/forums/...nX-Stable-Club
Very few chips are that good though, and you may need more than 1.43v at 4.5GHz to get 20LinX stable.
On another note, even with 32C ambient and 45C coolant, my 980x can run at 4.4GHz HT on stable. Max core temp gets up to 97C in LinX, but still below 101C Tj Max. Never goes above 80C for regular computing. So now I'm running 29x152(comes out 29*152.5 in windows for 4422MHz) with 1.475v BIOS. This is also with all 18GB of my memory in, running at 1830MHz CAS 8-9-8-24-110-1T on 3.2GHz UCLK.
well i'd like to run linx, but since i've changed to this 980x, linx wount work.... it stay there thinking thinking an thinking, cpu usage never goes more than 1% ... don't know what's wrong with it.......Quote:
Originally Posted by zoson
Seems pretty decent but without being able to stress it with LINX, it's hard to say its stable. Prime95 is about as good a test as you can get which I prefer for long runs in addition to LINX testing, so does it pass a few hours of that? Tried re-downloading and using new copy of LINX?
Quote:
Originally Posted by p3gaz_001
Get the latest linx 0.6.4 engine and then download the latest linpack binary:
http://software.intel.com/en-us/arti...pack-download/
Replace, and retry.
You can't say it's stable without LinX... My 980x will do 4.5 HT on with 1.43v, and *seem* stable, but it's not really. It's very temperature related... For example it'll run fine for 3 days, then ambient will increase from 25C to 32C and it'll crash.
i'll try out.
---
btw .... i've noticed something wrong since i had 2 580 with the same vid. i remembere that i was able to close benchs at 200.x23 with 1.50 .... now is not more enough, is that normal?
i've changed about 5 580.... and this is the first time that i have two card with the same vid... it seem's like cpu doensn't like gpu with the same vid...... what to check? pci-e bumped to 103, ioh voltege bumped to 1.19 ... what else?
@p3gaz_001
Tried more vIOH? ~1.25-1.3V?
sincerely no, i'd read about 1.16v to get stability, i'd pushed to 1.19v and i was thinking that it was to high... then if you are telling me that i can go to 1.25/1.30 i'll do it right now..... thanks as always for the support.
We recently discovered that vIOH is also QPI PLL... Different chips will like different vIOH because of it. Up to 1.35v should be no problem, and I doubt anything higher will help at all, anyway.
this is a great info, but it sound to me a little bit strange, how can vIOH be QPI PLL?? did a read right? btw... this night i'd tried some combos .... from 1.19v IOH to 1.35 it seems to be no differences ... but i'm leaving it at 1.30. and .... by switching off lane 3 and removing power connector to the second 580, 200x23 rolls back to 1.50v , 2 580 200x23 is now 1.518... but, i'd even noticed that, if i try straight 200x23 1.50v when the room temp and liquid temp is low i can succesfully close any bench .. i take for esample the vantage cpu test, once i've finish the test, and i try it to run again it fails, so i'd think that ... this hot days, aren't good for the cooling sys..... and i have to rice up vcore.Quote:
Originally Posted by zoson
Yes, it is very strange... Raja confirmed for us though. the vDC requirements for ioh and qpi pll are the same, so ASUS just tied them together... It's stupid...Quote:
Originally Posted by p3gaz_001
I'm not sure why your GTX580 does that... Maybe it's bad? Did you try that one by itself? Do you have the EZ_PLUGs connected? For GTX580 SLI you should have at least one EZ_PLUG attached.
well ..... !!!! if raja confirmed that .... we should trust him, so according to that... moving vIOH or QPI PLL , should have the same effect? ....Quote:
Originally Posted by zoson
-----
my 580's aren't bad... they both work great .... and i've also connected both of EZ_PLugs ........ :|
There's no option to change QPI PLL voltage separately so basically bumping vIOH you also automatically bumping QPI PLL.Quote:
Originally Posted by p3gaz_001
ok that was the point.... but what i'm triying to understand is.... should i bump only vIOH to get QPI stability o i must bump the both? i'm sorry i had to ask, but ..... this is some how a shoking news.... don't wan't to rma my cpu or mainboard by failing bumping voltages ...Quote:
Originally Posted by McDown
Quote:
Originally Posted by McDown
You cannot increase only vIOH. That's what he's saying. Increasing vIOH automatically increased QPI PLL voltage.. so no choice of increasing only vIOH or both vIOH and QPI PLL. Always both vIOH and QPI PLL increase together.Quote:
Originally Posted by p3gaz_001
I have a question not realty sure here just thinking, don't want to screw things up.
My R3E board failed last week and I received the replacement board today. I plan on doing the rebuild over the weekend.
Since It took me a while to get all the settings in the bios perfect for my 4.0ghz I-7 930 on air I wondering if I could use the bios chips from the defective board (as long as they are not dead too) and pop them in the replacement board because I don’t remember all the previous bios settings. Plus I don’t want to spend all the time in priming etc. is this doable? Thanks in advance for replies.
Hmm, good question, I dont see why not. Are you sure your board did not die because of overly aggresive settings by chance? Also, I would think a new board also means completely new chips, which means completely different results on your settings. What may have been stable on your board does not mean it will be stable on mine and vice-versa. BUT, with that said, YES, your old bios chips will work just fine. I have a chip puller just for that reason alone. lol....Quote:
Originally Posted by dionysus
thx Sk0rpn Thats what i thought, I also image the setting working on the previous MIGHT not work with the replacement board, however it might be a good starting point and I could work from there.
Yup agreed. Im surprised they didnt ask for the old board, or did they? You just going to swap out the chips? Do both chips have the same settings on them?Quote:
Originally Posted by dionysus
If so, you should be able to just boot up, with minimal complaining from your OS, and immediately start some stress testing on your old settings. Anyway, good luck with that. Im curious if it works myself, lol....
Well....remember the ROG boards have an advance replacement policy :) so I have the replacement board, and the one that essentially won't boot.
Yeah I hope it's as easy as swapping out the 2 bios chips which is why I'm reaching out to the forum here for help thanks again.