• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

2x256mb Adata PC4200 finally good results

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

tekOC

Member
Joined
Feb 23, 2004
Location
Apt 2
Finally got my 2x256mb Dual Channel Hyperram Adata chips PC-4200.

With my rig, everything at stock I score 4796mb/s Sandra Mem (2.5-3-3-8-8) timings, 75% efficiency.

yay, I guess the 2x512mb sticks just blow ***

Now time to OC the hell outta this 2.4C
 
Ehhh... I can't even boot at 230fsb.....

With 1.6V on the vcore, and 2.75vdimm, 400mhz 1:1 for the ram (2.5 or 3-3-3-8-8) I cant boot at 230fsb... I get all kinds of windows errors and beeps and crap and I have to reboot and change the BIOS settings.

When I tried 250fsb (I could do 250fsb stable with the old 2x512mb sticks) I couldnt even boot past the BIOS - I'd just get a BIOS corruption thing and it would try to reflash my bios, asking for the BIOS floppy.... but all I need to do is reboot and I'm fine.

Wierd... maybe the RAM has to burn in or something... I'm gonna go set it to 2.85V and just let it burn I guess, with a lot of F@H
 
tekOC said:
Ehhh... I can't even boot at 230fsb.....

With 1.6V on the vcore, and 2.75vdimm, 400mhz 1:1 for the ram (2.5 or 3-3-3-8-8) I cant boot at 230fsb... I get all kinds of windows errors and beeps and crap and I have to reboot and change the BIOS settings.

When I tried 250fsb (I could do 250fsb stable with the old 2x512mb sticks) I couldnt even boot past the BIOS - I'd just get a BIOS corruption thing and it would try to reflash my bios, asking for the BIOS floppy.... but all I need to do is reboot and I'm fine.

Wierd... maybe the RAM has to burn in or something... I'm gonna go set it to 2.85V and just let it burn I guess, with a lot of F@H

try 2.5-4-4-7 or 3 -4-4-7 and bump voltage to 2.9 if u can,
as i remember my stuff didnt like 3-3-3-8 after abou 220
if that works then once it quits then go to 3-4-4-8
 
I'm an idiot lol, thanks flapperhead for reminding me...

I forgot that this RAM likes 2.5 or 3-4-4:)

booted @ 250fsb, 1.6vcore, 2.85vdimm, 1.6vagp, PAT enabled, 400mhz 1:1 with 3-4-4-8-8 timings, AGP/PCI on auto

Gonna go run Sandra and I'll update this thread with results

Thanks again flapperhead

---------------------------------------------------------------------------

5792mb/s Int Buffered
5793mb/s Float Buffered
72% efficiency

@ 250fsb x 12 = 3.0Ghz

I'm gonna keep going of course:cool:

---------------------------------------------------------------------------

270fsb x 12 = 3.24Ghz, all other settings same as with 250fsb

Sandra CPU
10004 MIPS
4065/7174 MFLOPS

Sandra Mem
6351mb/s Int Buffered
6304mb/s Float Buffered
74% Efficiency

AIDA32 Read
5999mb/s (1st place, screw the FX-51)

NOW we're talkin :D
 
Last edited:
Now we're in the big numbers:cool:

280fsb x 12 = 3.36Ghz, all other settings same as 250/270fsb

Sandra CPU
10044 MIPS
4258/7439 MFLOPS

Sandra MEM
6508mb/s Int Buffered
6484mb/s Int Float
73% Efficiency

AIDA32 Read
6206mb/s

AIDA32 Write
1871mb/s
 
Sophisticated said:
very nice OC...remember to hit up that memtest and prime

Of course, I just wanted to hit 3.4Ghz - now I'm gonna chill out for a while and do thorough testing for stability.

285fsb x 12 = 3.42Ghz, all other settings same as 250/270/280fsb

Sandra CPU
10367 MIPS
4335/7569 MFLOPS
PR4554 (estimated)

Sandra MEM
6696mb/s Int Buffered
6638mb/s Int Float
73% Efficinecy

AIDA32 Read
6243mb/s

Now for stability testing, including 3dmark01, 3dmark03, PCmark04, Prime, F@H, and memtest, and whatever else I find

P.S. Is this RAM supposed to be single-sided? It only has RAM chips on one side of the PCB
 
Last edited:
this is really gonna get u going.. last nite i temporarily, upped my vtt(abit boards have a vtt prob and did a temp fix) i got 300+ on one of the stix.... b4 the temp circuit heated up too much. gonna do a permanent fix this week.. cant wait heheheheheh....
 
real weird, at first the memory quit at 295 1-1. so i went to 5/4 and it quit at 295. then iwent to 3/2 and it quit. i knew this cpu was capable of 3.7+ so i was confused as to what the prob was. i put in my bh6 at 3/2 and it ran up to 318 till the cpu died. so i concentrated on the vtt issue and once i got it temporarily higher the stick i was testing actually went around 305-310 then quit. so it seems that the adata needs correct vtt voltage when running extreme fsb..
 
flapperhead said:
real weird, at first the memory quit at 295 1-1. so i went to 5/4 and it quit at 295. then iwent to 3/2 and it quit. i knew this cpu was capable of 3.7+ so i was confused as to what the prob was. i put in my bh6 at 3/2 and it ran up to 318 till the cpu died. so i concentrated on the vtt issue and once i got it temporarily higher the stick i was testing actually went around 305-310 then quit. so it seems that the adata needs correct vtt voltage when running extreme fsb..

Sorry for the n00b question.... but what is VTT voltage? Chipset voltage?

P.S. Memtest stable 8x, 3dmark01 and 3dmark03 stable, very hot in my room so I'm getting like 41-43C fulll load temps:mad:

I score 19755 3dmark01 and 6036 3dmark03 with my 9800pro at STOCK clocks and Omega Drivers, with the 285fsb settings

:) Now I'm gunna go run 3dmark01 and 03, and Aquamark with my 9800pro OCED
 
vtt voltage is a termination voltage used for ur memory. it needs to be set at 1/2 of the ddr voltage or the memory wont run right at extreme speeds or quit.its an automatic thing with the mb and the asus vtt works correctly. unfortunately most of the abit boards have a vtt prob. for instance my vtt voltage to my ddr at 2.8 is 1.25 not 1.4v. so i have to find a vtt mod that works. the first onei did doesnt work right. secondly chipset voltage means voltage to ur northbridge, the northbridge is what controls the fsb among other things.. so u need higher chipset or nb voltage at extreme speeds for stability. again the asus does a nice job, cause its chipset circuitry is built to follow the cpu voltage.. Pretty smart eh? so when u need more voltage to the cpu, they know u are ocing the fsb and give u more chipset voltage too.
as for temps at full load, heck thats low in my opinion.
 
Thanks man.

I dont think its too hot, it would be great temps for an OCed AMD, but I'm just used to lower loads on this P4 - maybe when the AS5 breaks in I'll see some better temps.
 
Hmm.... @ 285fsb and 1.60vcore Prime had an error after only 50 minutes, bumped the voltage up to 1.625.... maybe it helps, if not then 1.65v - I dont think I'm gonna go higher, would probably need the damn droop mod

:( trying to have the board unmodded
 
looking forward to push my 2.4C D1 beyond 3300 with my new A-Data PC4000 and my new AI7...
my rig runs 2400 @ 3240 at stock voltages.

3408 works, but i soon get memory errors :(

39°C sandra burn-in temp @ 3120°C.
32°C idle

yeeha!

going to burn in my actual twisters PC4000 a little (max out @ 262 :( )
 
I'm still at 285fsb x 12

Full load is at 42-43C with F@H and CPUBurn4 realtime priority running, along with AIM, IE, and everything else.

Ran CPUBURN4 overnight @ 1.625v last night and it passed.

Now running it @ 1.600v cause this board droops like crazy at anything more than 1.6v, and I wanna try to stay with 1.6v - but Prime gave me an error after 50 minutes yesterday when I was at 1.6v, I changed up something else in the BIOS and stayed at 1.6vcore so now I'm testing the CPU all day with CPUBURN4 and all my other crap

speedfan.JPG


1.55v at full load with 1.6vcore in BIOS, but with 1.625vcore in BIOS it droops down to 1.48v at full load:eek:
 
Last edited:
Back