wow... nice result bingo13.... congrats... :clap:
Printable View
wow... nice result bingo13.... congrats... :clap:
Nice results Bingo, that's some nice ram you got there. Cas 3 @ 449mhz...
abit is officially back in the game.
lets see what bingo gets with a 4mb cache cpu..
Switching over now to the E6600 and then on to the Asus P5W-DH with a bios that we assisted on this week, had dual prime stable for about three hours and then ran a couple of the 3dMarks with the 7950GX2, not bad scores for the combination. Figure 7x450 is the limit on our board with the E6300, air, and no mods. One strange issue is that booting at 7 x 450FSB with this bios will either result in 449.8 or 451FSB readings depending upon the instant you take a screenshot. I loaded uGuru (it has some issues with this bios) and the most I could extend the FSB was maybe one or two before the system would crash. Still buggy in my opinion, better to boot with optimized bios settings at this time and leave uGuru on the CD. ;)Quote:
Originally Posted by raju
http://img89.imageshack.us/img89/574...0x73d06dz9.png
http://img181.imageshack.us/img181/2...0x73d01oy7.png
that's still nice for a 975 board on the 1066 strap..
This is not so nice, still playing with it but I was able to do 9x433 with the previous bios with memory timings of 4-4-3-10. The board would do 9x422 at 3-4-3-9 with previous bios also. I could post at 9x449 and do CPU-Z screenshots at 9x444. With this bios I am able to post at 9x441 and do screenshots at 9x434 now, nothing is real stable until 9x421 and that is with the memory at 4-4-4-12 but so far 9x415 with 3-4-3-9 has been quicker to some degree. Will update again shortly and test with a X6800 later on.Quote:
Originally Posted by raju
http://img170.imageshack.us/img170/8...5x7sp1mgh5.png
Thanks for taking the time to do that, I think the bios still needs work if this board is really gonna put ABit back in the frame..
I'll let you know how I get on with the E6600 next week hopefully...
regards
Raja
If I increase the FSB the memory timings go to hell, 9x416 requires 4-3-3-9 setting, 9x422 requires 4-4-4-12 for stability. I topped out at 9x427 and all that did was get me a single SPI32M run and five minutes in Quake4. I think this beta is a step backwards for the 4MB chips. I am sticking with 9x415 for 24/7 type operation at 1:1 for our CPU and board sample. Anything above 9x422 and Orthos Beta would crash within a minute. At 9x415, it ran for four hours before I stopped it. Also, my ram usually does 3-3-2-8 at this speed, 3-3-3-9 would crash at 9x415.:rolleyes:Quote:
Originally Posted by raju
http://img171.imageshack.us/img171/1...x7sp32mqd8.png
Time to pull out the X1950 CrossFire setup, moving on now to the 4:5 testing and will test some high end Elpidia modules. Those are separate runs of SPI 32M, testing the thermals on each core. Will post up true dual runs shortly. :)
http://img87.imageshack.us/img87/171...5x93d06bw5.png
Wow, thanks for your efforts bingo :toast:
Obviously your Elpidas run a bit better than mine :D
What kind of mem timings and vdimm did you have for the RAM at 520Mhz? (4:5 ratio = FSB 416 : 520 RAM ) :)Quote:
Originally Posted by bingo13
Some update with SEMI-Moded mobo and Corsair C3 now.......1:1 action........Vdimm = 2.75Volts via DDR Maximizer......:)
http://www.thelab.gr/images/Hipro5/C...1_1_C3_png.png
I'll be back of course......;)
looking good, hipro & bingo! Glad to see that ABIT's board is doing pretty well!
Nice hipro :toast:
What was your RAM again?
he had 2x512 fatbodies and now hes on 2x1gb corsair 6400 c3.Quote:
Originally Posted by Sicness
Nice results hipro!!!
Now show us pictures of your board!! :) (just kidding mate, im sure you will when you are ready. ;))
Revv, old grouper mate ;)Quote:
Originally Posted by Revv23
I'm rather new to DDR2, all I know is that Micron D9 are supposed to be the best arround. But what exactly are D9 fatbodies? Thanks.
He is using the Corsair 6400C3 (what all of my tests were based on at this time). It appears at this time that the high end Corsair - 8500C5, 6400C4, 6400C3, and Dominator are the only modules that Abit has tuned for 1:1 operation in a reliable manner up top. My testing with a variety of PC8500 modules from other suppliers indicates this pattern currently. I am still working on it. ;) While I can certainly post at higher FSB settings and run basic benchmarks I am finding with the E6600/6700 chips that 24/7 capability is quite a bit lower as indicated in previous posts, especially on a board that has not been modded. As usual HiPro5 works his magic in extremely innovative ways. :woot:Quote:
Originally Posted by Sicness
I bet hipro could even make EDO RAM working on this board :DQuote:
Originally Posted by bingo13
I'm looking for a new 2GB kit, currently looking at OCZ 6400 Plats Rev2, Cellshock (some new german brand) 6400s with D9 and TeamGroup 6400s. My budget is rather limited :(
i need to chime in with my usual remarks cuz im a bit put off by what im seeing.
to be honest. i see nothing too different than my $145 DS3 when using a 9x Mutli..
i boot 24/7 at 412 x 9. why that ? cuz its a nice even 3.7ghz... it only needs 1.43v vcore, and i dont have to increase anything else. no vmch voltage adjust ments needed or anything.
now i had hopes for this baord aside from the fact that its price is way too high, i would have paid for it had i seen better results.
but i do realise this is still early in the game. but since the day i got this DS3 i have been 100% stable at 412 x 9
now i can boot at higher FSB but it requires more volts.. and even with water.. this chip gets hot, not insanely but once i start hitting 60c i back down a little.
anyways, for me,, im pretty disappointed thus far but am holding off final judgement for now.
and yes i appreciate all the time you guys are taking to post this stuff, it takes alot of time and work to do all the screen shots and upload them and write up your findings.. not to mention the time spent doing god knows how many 32m Pi runs lol..
so keep up the good work, but when faced with $145 965 board or a $300 975 board.. of which hasnt shown me that it can do any better at 9x multi and definately hasnt done better at a lower multi. (the DS3 still can do 500 fsb on a lower multi) i honestly can't see a reason why someone would switch.
switch for 2 seconds difference on Pi ?? come now,, seems a bit silly so do that.
anywho keep up the good work guys, and i hope abit is able to make the board act better with 9x multi.
We figured out some of the bios optimizations and now have OCZ PC-8000 VX2 working at 1:1 on this board with improved timings. :D
http://img209.imageshack.us/img209/4...9d9x416bc2.png
this is why i got badaxe, badaxe is a pretty mature board in terms of OCing
i think in 2 months this board should be able to reach 450-ish fsb,
bingo, did you change any other settings besides timings to get the VX2 to work? i cant get teh vx2 or soe pieces to run 1:1.
I'm starting to see fewer & fewer reasons to keep this board over my P5W DH when it arrives. Keep the benchmarks coming guys. They are some very impressive results:)
http://img151.imageshack.us/img151/3...nchmarkkv8.jpg
We had to change the SPD to match up with some internal bios timings. I am testing a final SPD release from OCZ currently, will post up results shortly. I also spoke with Abit about the issues tonight so maybe it will be resolved in a new bios quickly.Quote:
Originally Posted by Dubz