With your FSB at 250, and memory 1:1, and a 2X HT multi, you're getting DDR500 and an HT of 500. Just a guess, but could there be a benefit to having the HT match the memory speeds?
Printable View
With your FSB at 250, and memory 1:1, and a 2X HT multi, you're getting DDR500 and an HT of 500. Just a guess, but could there be a benefit to having the HT match the memory speeds?
Program HT x1 HT x 2 HT x 4
3D2k01 21.871 22.138 22.207
3D2k03 7098 7130 7140
PC2002
cpu 7131 7121 7135
mem. 8840 8939 8996
Aquamark N/A 48,564 48,884
PC2004 4286 4363 4396
well i have tested with epoX KDA3+ (nforce3250Gb) with 4.27 latest nforce drivers i think...
amd athlon 64 3200+ c0 clawhammer 2202Mhz
Corsair BH-6 @ 2-3-3-7 200.4 Mhz
cat. 4.8 by Igor Pavlov (best driver i ve tested)
MSI RX9800 @ 438/378
i cant agree with dagalidis but since i saw his thread i thought i must bench too :D seems to do not as much on the epoX board i wouldnt say that its with all nforce boards so... (more reviews :D plz)
least but not last i want to say i only benched the 3dmarks once and i think the aquamark score on HTx1 would be little bit less than with HTx2
seEn andL
Nice Post andL64.....
I will try again with my new ABIT mobo 1.1 Rev and post some new findings.... ;)
Quote:
Originally posted by Dagalidis
Nice Post andL64.....
I will try again with my new ABIT mobo 1.1 Rev and post some new findings.... ;)
so plz do so!
we need more reviews i want to know if its on alle nforce3 like that...and also k8t800(pro)
and dagalidis i wanted to do this since a few times but i didnt recognize some real difference so i didnt looked at it exatcly but since i saw ur thread i MUST do it :D
i ll bench other benches like sisoft i think memory get s also a little bit better but why it didnt use HTT or does it?
we ll see and also i ll bench with the new "hell on earth" board if i can get one :D
soon i ll also get a pre tested cpu then i ll give ya some more...
seEn andL
PS: lets see how high ur HTT could get !?! i ran 280x4 without problems
...and how the hell i could post a chart , list ?
Was able to begin booting at 350 x 6 @ LTD X 3 but Crassed on Windows Loading SCreen.....
Damn it was to close but my mobo need Aircondition because i have 39 C Case temp :( and dont help for high FSB OC..... :D
Dagalidis
Hmm.. can you maybe bench same resulting HT speed achieved by different methods? For example 5x200, 4x250, 3x333 :) I wonder if that will make any difference @ all.
Last night was trying to find my max 1:1 free of errors using MEMTEST 1.5 as mem diagnostic tool....
I was able to hit the amazing 262 FSB using ONLY 2.62 Vddr and I was really socked….
I the most of my OC I used 2.8 Vddr as default option because I thought this was the only way to get my system stable….
I was wrong… my ADATA VITESTA DDR500 using Hynix –D5 chips just love low vddr …
I conclude that CPU mem controller is heating or don’t like more that 263 MHZ because no matter what I do this was the max acceptable OC for 1:1 FSB/MEM….
Just check my steps at the XLS table and make your judgment……
MEMTEST = 10 loops of 5th test
3D = AQUAMARK03 Bench
SPI = Super PI 2MB
It took me a lot of time to have long Tests....
http://www.xtremesystems.org/forums/...chmentid=15801
I will Run Tweeker to see what changes make my ABIT when change LTD when return home.....
ADATA is a very good perfomer MEM and those Hynix D5 chips are clocking well and with low VDDR as you see at my EXCEL..... :slobber:
wow those are amazing results! extremely interesting!
Yes.... MEM Controller sufer from heat and water is needed for higher 1:1 clocking .....
But ADATA is JUST GREAT :D
So you use #18? I mean the FSB/mem config. Very nice with 1T at those speeds with the volts! Is it not possible to tighten to 8-3-3, at least?
eh...
nice nice nice i heard of it that the a data likes low voltages ver very much and thell ll do not even mor with xtreme or higher voltage so...
but i got bh-6 how far can i go ? i think 3,2V shouldnt be a problem... i ll only get little bit nervous cause of the mem controller in side the amd64 :D
bu no risk no fun!
soOn awaiting the dfi board :D
seEn andL
Not only you.....Quote:
Originally posted by andL64
soOn awaiting the dfi board :D
seEn andL
Believe me .... lol
Not sure how you ran the test as far as order. Meaning did you run LDT 2, then 3, then 4, then 5 and rebooted before each run. If that's the case then reversing the order may give different results
Dagalidis great findings.
btw I think you mean LDT. ;)
Correct ........ Sorry for Mistake....Quote:
Originally posted by iceman2g
Dagalidis great findings.
btw I think you mean LDT. ;)
No....Quote:
Not sure how you ran the test as far as order. Meaning did you run LDT 2, then 3, then 4, then 5 and rebooted before each run. If that's the case then reversing the order may give different results
I was starting from LDT 5 and when not passed MEMTEST or Windows Benches then searching the best BIOS option to make system stable again...
So if needed LDT decreased back to X4 - X3 as you see at the EXCEL....
CodeRed Sorry for this late reply…..
I have screens of A64Tweeker of LDT X 5 & X4 but nothing changed…
I’ve also tested with LDT X3 – X2 – X1 but nothing new appears on Settings…
This is LDT X 5 Screen…
http://www.xtremesystems.org/forums/...chmentid=15939
This is LDT X 4 Screen....
and as you see ... nothing changed.....
http://www.xtremesystems.org/forums/...chmentid=15940
Here is Unbuffered Score of SANDRA MEM @ 260 1:1 @ LDT X 4 @ 2.6 VDDR :D
http://www.xtremesystems.org/forums/...chmentid=15941
Oh well it was worth a try :)Quote:
Originally posted by Dagalidis
CodeRed Sorry for this late reply…..
I have screens of A64Tweeker of LDT X 5 & X4 but nothing changed…
I’ve also tested with LDT X3 – X2 – X1 but nothing new appears on Settings…
This is LDT X 5 Screen…
There are a few more LDT registers, maybe I need to make another version ;) Or easier to just dump the values with WPCREDIT (bus 0, device 24, function 0).
Thanks for doing the testing.
Maybe I missed something I was refering to the 3d runs at the start of the threadQuote:
Originally posted by Dagalidis
No....
I was starting from LDT 5 and when not passed MEMTEST or Windows Benches then searching the best BIOS option to make system stable again...
So if needed LDT decreased back to X4 - X3 as you see at the EXCEL....
Just a small note.
Ive played with many Adata sticks in the past, and found that they have hated any more than 2.7v.
Try using 2.7v as your top limit; might help you reach 270-280 ;)
Hmmm...Quote:
Originally posted by Kinsy
Just a small note.
Ive played with many Adata sticks in the past, and found that they have hated any more than 2.7v.
Try using 2.7v as your top limit; might help you reach 270-280 ;)
Yes i know what you mean...
I found that 2.6 VDDR @ 260 1:1 is the best combination right now....
BTW i hope VTT mod will help to get more than 262 1:1 when watercooling be ready and push VDDR up to 3.2 Volts...
Nothing new about ABIT @ Via Chipset to POST....
But this weekend i will try my new DFI board and run same TESTs again to compare with old (Via Chipset)....
Wish me luck with DFI..... :D
dam it i think have DT43 chips and i need 3.2v to get to 270 1T.
that cant be stable? i only get 2200/2270 unbuffered :(