Corsair Twin-X4000 266FSB

[Liquid3D]

Here ya go, fellars;
2.4C (lapped to the copper) 38C all day even at 3.6GHz.
SLK900U (lapped)
Epox 4 PCA3+
2x512MB VDIMM 2.9V (that's all I needed 1 stinking volt.) :D

[Dr. Possum]

Timings at 2.5-4-4-7? Very nice :slobber:




~DP

[CCW]

It wasnt one stick letting you down then? Just needed a bit more voltage. :D

Nice.

Craig

[Shane5578]

very good

[TodB]

I'll have to drop ya a shot of the Hyper-Cheap A-DATA DDR500 memory runnin 280MHz sync ;) .... 4PCA3+ too :)

I'm really happy that some cheap memory can kick all those expensive buts :D

[Liquid3D]

Originally posted by CCW
It wasnt one stick letting you down then? Just needed a bit more voltage. :D

Nice.

Craig This is different, these are benchmarks from 2x512MB sticks Twin-XMS4000. The other post your speaking of, was with 4x256MB Twin-XMS4000 sticks, so that's very possible, I just haven't tested the theory yet. Irregardless, voltage was definately ONE issue. Probably the most critical.

TodB I'd love to try some of the A-DATA or Twinmoss for a comparison, I'll see if I can find some.

EDIT, I'm mistaken those memories are using true 4ns IC's. The same one's everyone else is using. And they are kicking Butt, and taking "names," name-brands that is. Although A-DATA hasn't released their DDR500 yet, at least their site indicates this.

[Shane5578]

Originally posted by TodB
I'll have to drop ya a shot of the Hyper-Cheap A-DATA DDR500 memory runnin 280MHz sync ;) .... 4PCA3+ too :)

I'm really happy that some cheap memory can kick all those expensive buts :D

lol

i'm thinking of getting those cheap a-data pc4000 and whacking 3.3volts through them.

[eva2000]

i just got my Epox 4PCA3+ yesterday as well.. going to try all my memory on it with some more vdimm :D

interesting to see how these compare on 4PCA3+

2x 512MB OCZ PC4000 Gold dual channel
2x 512MB GEIL PC4200 Golden Dragona dual channel
4x 256MB ADATA PC4000

[eva2000]

Originally posted by Liquid3D
Here ya go, fellars;
2.4C (lapped to the copper) 38C all day even at 3.6GHz.
SLK900U (lapped)
Epox 4 PCA3+
2x512MB VDIMM 2.9V (that's all I needed 1 stinking volt.) :D interesting buffered.. i get more with 4x 256MB ADATA PC4000 @ 255mhz at 3-4-4-8 at 2.75v real vdimm on IC7 at 2.8v

and PAT is disabled too!

http://www.fileshosts.com/pentium4/D1/abit/ic7/memory/adata/pc4000/4x256/255-255-3448-auto/mem-buff.jpg

but ADATA PC4000 unbuffered is lower i guess for ADATA PC4000 = 3189/3262

[eva2000]

Liquid3D could you post aid32 read and write and sciencemark benchies @265mhz ? thanks :)

[Henry]

eva2000, what's the tighest timing of your a-data pc4000 at 255 fsb ? And wha't the highest fsb you can get ? Can you post the sandra bandwith at that fsb ?
Last, what's your a-data chip ??

Thanks :)

[FUGGER]

I am pushing my PC3700 to 266FSB with ease, slightly tighter timing too.

Click 2K1 score, 24k with a 5900FX!!! @ 4.24Ghz

[eva2000]

Originally posted by Henry
eva2000, what's the tighest timing of your a-data pc4000 at 255 fsb ? And wha't the highest fsb you can get ? Can you post the sandra bandwith at that fsb ?
Last, what's your a-data chip ??

Thanks :) my 8 sticks are apart of the same 12 sticks of ADATA PC4000 reviewed at OCAU http://www.overclockers.com.au/article.php?id=194595 i asked chainbolt to buy them for me and he did a review for them before sending them to me :)

on IC7 with 2.75-2.78v underload at 2.8v in IC7 15 bios

2 sticks so far
http://www.bleedinedge.com/forum/showthread.php3?s=&threadid=611

4 sticks so far
http://www.bleedinedge.com/forum/showthread.php3?s=&threadid=578

with 4 sticks I found vcore affected the stability testing and benchmarking of the memory even at assumed 2.4C default stable vcore voltages - i'm looking into that

for certain though IC7/IS7 boards aren't going to do justice for ADATA PC4000

don't forget to quote your vdimm used for memory oc's guys :)

[Shane5578]

mm it seems buffered scores are better on the ic7,
but unbuffered scores are better on the 4pca3+

[eva2000]

yeah Clevor confirmed he gets better unbuffered scores on 4PCA3+ than his P4C800

[Liquid3D]

Originally posted by eva2000
Liquid3D could you post aid32 read and write and sciencemark benchies @265mhz ? thanks :) How do you run the [program? I downlaoded it, extracted it, but can't seem to Run it?

[eva2000]

download aid32 3.70 at http://aida32.hu/download/aida32ee_370.zip

then extract it and there's an application .exe file run that - what i do is create shortcut and put it in my winxp quick launch taskbar :)

[Liquid3D]

ScienceMark didn't have my Chipset, nor OS, so I couldn't run it? Is there something I'm missing? Do I have to close all applications to run Aida?

[eva2000]

oh sciencemark .. to run it you go to the mem benchmark menu and run

it will pop up a screen with 2 button choices

Analyse and OK

hit Analyse and let it run until results popup

sciencemark runs fine on my IC7 and PX865PE PRO2 and on Creative's IS7

[Liquid3D]

I ran Science Mark, it came up in DOS, went through all the tests, and just closed? How the Heck am I supposed to save the results?
Here's the Aida READ/WRITE you asked for, but it's at 260MHz 3-7-3-3/2.9V (Corsair Twin-XMS400 2x512MB)

[Liquid3D]

Here's READ same settings

[eva2000]

you sure you gto the same sciencemark app

i uploaded it here http://www.fileshosts.com/files/sciencemark/ScienceMark2.zip

[keaolyen]

I read that review and am pretty certain the cheap AData's in that review are using Hyix IC's and are NOT cheap here in the US or Europe and will most likely have a tariff slapped on them in Taiwan soon also.

[eva2000]

Originally posted by keaolyen
I read that review and am pretty certain the cheap AData's in that review are using Hyix IC's and are NOT cheap here in the US or Europe and will most likely have a tariff slapped on them in Taiwan soon also. ouch that's 30% more expensive then right ?

[keaolyen]

47% in the US I believe... =\

[Liquid3D]

That's right on the money. I read EETimes, and here's a quote from a recent article about this;

"The trade commission's decision, announced last week, upholds a June ruling by Commerce, which imposed a 44.71 percent duty against all DRAMs Hynix exports to the United States from its several manufacturing plants in South Korea. The tariffs do not apply to DRAMs manufactured at Hynix's plant in Eugene, Ore.
The tariffs result from what Commerce said were illegal subsidies that the South Korean government provided to Hynix as part of three separate financial bailouts, totaling $11.7 billion, made from January 2001 to June 2002 through domestic banks owned or controlled by the government."
http://www.eetimes.com/story/OEG20030728S0023