A-DATA 533mhz OC's to 1066MHz with tight timings

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

x2 3600 rules sazakky

Senior member
May 11, 2007
410
0
0
Originally posted by: JAG87
so all three of these wonder sticks can do DDR2-1078 at CAS 3 with 1.95v... cause you know, all memory modules must run at the same speed. nice try though, you almost had me.


what are you talking about. they are all running at 1078
 

Scottae

Member
Jan 19, 2008
127
0
0
jag87 has a great point... so mis matched sticks all doing 1078 @ CAS3 really stretching...dunno just fishy
any benchmarks at these speeds?
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Originally posted by: x2 3600 rules sazakky
Originally posted by: Idontcare
Originally posted by: x2 3600 rules sazakky
also, i forgot to mention i have three modules. 2 512mb adata 533mhz modules. and a generic 667mhz modules. therefore it is single channel

I expect two things here to matter. First 512MB DIMMS...yeah these will be tight. Most folks here will assume you are talking minimum of 1GB DIMMS these days.

Second is single-channel. This is far less demanding on the DIMMS. Kind of like 1T vs 2T for command rate, whole other world when you stop demanding dual-channel access timings.

So I recommend pulling a DIMM and testing for dual-channel stability, and even then don't be surprised if most folks aren't impressed with the timings on 512MB DIMMS. We are usually talking about 1GB and 2GB DIMMS these days.

there is a 1gb dimm in it

If you believe that response is appropriate to address the questions raised in my post then you are completely clueless. Thanks for providing proof of such cluelessness, I will now stop wasting my time here.
 

ionoxx

Senior member
Jan 18, 2005
267
0
0
Are the 512MB dimms single sided and the 1GB gouble sided?

CPU-Z could simply be misinterpreting assymetric dual channel.
 

CptCrunch

Golden Member
Jan 31, 2005
1,877
1
0
It is my understanding that doing this with two MATCHED sticks is hard, but having two matched sticks and one 'noname' stick, quite difficult to believe. While I think this would be a great find and great overclock, i'm still finding this hard to believe
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Originally posted by: x2 3600 rules sazakky
http://valid.x86-secret.com/show_oc.php?id=317786

validation page

Are you sure you aren't running HTT bus of 5 x 245 FSB = 1225 effective FSB with memory unlinked? The fact that FSB:DRAM ratio is CPU/5 doesn't seem to indicate to me that you are running a synchronous setup here, which is why it's hard to verify your claim.

However, another way to check if you are is to run memory bandwidth tests in Sisoftware Sandra (ie. unbuffered). I think your ram is actually running at DDR2-539 (but it seems CPU-Z is thrown off due to Single channel mode).
 

x2 3600 rules sazakky

Senior member
May 11, 2007
410
0
0
Originally posted by: RussianSensation
Originally posted by: x2 3600 rules sazakky
http://valid.x86-secret.com/show_oc.php?id=317786

validation page

Are you sure you aren't running HTT bus of 5 x 245 FSB = 1225 effective FSB with memory unlinked? The fact that FSB:DRAM ratio is CPU/5 doesn't seem to indicate to me that you are running a synchronous setup here, which is why it's hard to verify your claim.

However, another way to check if you are is to run memory bandwidth tests in Sisoftware Sandra (ie. unbuffered). I think your ram is actually running at DDR2-539 (but it seems CPU-Z is thrown off due to Single channel mode).

Alright. I will check with Sisandra. But I can also tell the change because general desktop use is about 10x faster than before.
 

x2 3600 rules sazakky

Senior member
May 11, 2007
410
0
0
Originally posted by: Idontcare
Originally posted by: x2 3600 rules sazakky
Originally posted by: Idontcare
Originally posted by: x2 3600 rules sazakky
also, i forgot to mention i have three modules. 2 512mb adata 533mhz modules. and a generic 667mhz modules. therefore it is single channel

I expect two things here to matter. First 512MB DIMMS...yeah these will be tight. Most folks here will assume you are talking minimum of 1GB DIMMS these days.

Second is single-channel. This is far less demanding on the DIMMS. Kind of like 1T vs 2T for command rate, whole other world when you stop demanding dual-channel access timings.

So I recommend pulling a DIMM and testing for dual-channel stability, and even then don't be surprised if most folks aren't impressed with the timings on 512MB DIMMS. We are usually talking about 1GB and 2GB DIMMS these days.

there is a 1gb dimm in it

If you believe that response is appropriate to address the questions raised in my post then you are completely clueless. Thanks for providing proof of such cluelessness, I will now stop wasting my time here.

Alright. Heres the thing. I did test with 2 DIMM"s (the A-data ones) in dual channel and it did
run stable. I ran 3dmark on loop a few time and it was fine. The temp of the ram and cpu were good.
Also, I don't bother changing form 2t to 1t as it doesn't amke much of a difference in real life usage.
 

DerwenArtos12

Diamond Member
Apr 7, 2003
4,278
0
0
Originally posted by: x2 3600 rules sazakky
Originally posted by: superstition
2.7ghz at 1.264v
I can probably hit 2.8 at that voltage (e2140). I had my BIOS set to 1.3V and with droop I ended up with a voltage that low.

my athlon can hit 2.9. i just don't do it. too much voltage

I believe this is possible with an E2140 but, not with an A64 of any revision. I'm on a fairly short list of people to do 3ghz on an A64 and thats with hand picked items and about a year of tweaking the system to get it to that point on air cooling and thats with 1.5v. I can 2.75 all day and night at 1.375(1.4 bios setting) but, 2.7 on 1.264v is just not believable to me.

As for the ram, I agree, CPU-z must be mis-reporting it, I'd believe it's running at DDR2-539 with these settings. And the extra speed you see in normal use is more than likely just from the extra cpu clocks. The performance gain due to the ram running at 539 instead of 533 would be negligable but, it's not a speed loss for that matter either.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Originally posted by: x2 3600 rules sazakky
Also, I don't bother changing form 2t to 1t as it doesn't amke much of a difference in real life usage.

Originally posted by: x2 3600 rules sazakky
Alright. I will check with Sisandra. But I can also tell the change because general desktop use is about 10x faster than before.

:confused: So on one hand you are quite confident a 1T CR makes no improvement over a 2T CR, but on the other hand you are quite positive that you are running DDR2-1066 because the desktop is 10x faster than when you were running DDR-533? :confused:

The shens material just keeps rolling on in. Keep'em coming, keeps me laughing :laugh:
 

nyker96

Diamond Member
Apr 19, 2005
5,630
2
81
I think the chance that one of the 3 sticks can do 1000+ is rare enough say 1/100? the chance ALL three can do DDR-1000+ at under 2v with 3-4-4-11 is I say 1 in 1/1000000 or 1 in a million. This is either the luckiest guy on the planet or there's something else going on here. I hardly even think the best Ballistix can pull this kind of timings much less something that is totally no name or a-data (which isn;t bad brand).

Edit: A quick review of this Adata stuff on newegg, seems that some OCed it to DDR800 with very loose timings. Haven't seen anyone doing DDR1000+ with this stuff at 3-4-4-11 timings.
 

geokilla

Platinum Member
Oct 14, 2006
2,012
3
81
I believe you can run your Athlon at 2.9Ghz 24/7 without any problems. If CPU-Z is detecting the voltages on my M2N-E correctly, then i'm getting horrible vcore. And by horrible, I mean horrible.

Did you run Memtest to verify it that it's stable at those speeds? DDR1000+ at 3-4-4-11 with just 1.95V is hard to believe. And don't forget that the M2N-E is probably the worst AM2 motherboard created by Asus due to the limited vdimm.
 

JAG87

Diamond Member
Jan 3, 2006
3,921
3
76
Originally posted by: x2 3600 rules sazakky
Originally posted by: JAG87
so all three of these wonder sticks can do DDR2-1078 at CAS 3 with 1.95v... cause you know, all memory modules must run at the same speed. nice try though, you almost had me.


what are you talking about. they are all running at 1078


exactly my point einstein. its not just your a-data memory, but even your no-name 1GB memory stick.

its obviously a false reading, dont you think? or you just magically stumbled across the 3 best DDR2 memory modules every produced?
 

john3850

Golden Member
Oct 19, 2002
1,436
21
81
I believe this is possible with an E2140 but, not with an A64 of any revision. I'm on a fairly short list of people to do 3ghz on an A64 and thats with hand picked items and about a year of tweaking the system to get it to that point on air cooling and thats with 1.5v. I can 2.75 all day and night at 1.375(1.4 bios setting) but, 2.7 on 1.264v is just not believable to me.

Have a 939 ut4 148 opty run 10x300 for 6 months with lose tccd ocz on water.

If t1 t2 doesnt make a difference it is because your on a divider and not 1 to 1.
ut4 148 opty runs ut 3 server 24 7 @2700
 
May 30, 2007
1,446
0
0
Had an Opty 144 that did 3ghz on a DFI NF4 Ultra but only 2960mhz on the A8N32-Sli Deluxe I had it on. Oh, that took 1.55v to pull off btw. Best that chip did with 1.3v was like 2300mhz.

Agree with everyone here when I say you have no chance in hell making any ram ever do that, it's quite obvious you know little to nothing about overclocking or you'd smell your own bullshit.

I say nothing short of having one of us come to your house and see your rig with our own eyes will buy you an ounce of credibility. That or give remote access to your PC to Zap so he can solve this once and for all by properly verrifying the results.
 

sskk

Junior Member
Dec 1, 2007
19
0
0
He only OCed a 533 to 1066, he DID NOT OC a 1066 to 2133, which I would call BS or a true 1 in 100000000000 billion chance.

There is a very good chance A-Data had lots of 800 or 1066 lay around and just labeled them 533, or maybe they've been getting lots of good chips lay around and just happened to put them in these "533" Rams.

 
May 30, 2007
1,446
0
0
This is A-Data we're talking about. Not G.skill which does do that. A-Data has no prob selling thier 800mhz kits so I doubt they'd re-label them. Even if they did, A-Data does not use any IC's that are capable of 1000mhz @ cas3 and sure as hell not at that voltage.
 

Scottae

Member
Jan 19, 2008
127
0
0
Originally posted by: sskk
He only OCed a 533 to 1066, he DID NOT OC a 1066 to 2133, which I would call BS or a true 1 in 100000000000 billion chance.

There is a very good chance A-Data had lots of 800 or 1066 lay around and just labeled them 533, or maybe they've been getting lots of good chips lay around and just happened to put them in these "533" Rams.

Say that is true... whats the odds of that third no name 1Gb stick doing the same ?
 

DerwenArtos12

Diamond Member
Apr 7, 2003
4,278
0
0
Originally posted by: john3850
I believe this is possible with an E2140 but, not with an A64 of any revision. I'm on a fairly short list of people to do 3ghz on an A64 and thats with hand picked items and about a year of tweaking the system to get it to that point on air cooling and thats with 1.5v. I can 2.75 all day and night at 1.375(1.4 bios setting) but, 2.7 on 1.264v is just not believable to me.

Have a 939 ut4 148 opty run 10x300 for 6 months with lose tccd ocz on water.

If t1 t2 doesnt make a difference it is because your on a divider and not 1 to 1.
ut4 148 opty runs ut 3 server 24 7 @2700

:thumbsup:
Glad to hear that. what voltage is it running? and keep in mind that it's on water. and what board is this on? I'm not attempting to discredit you at all but, how many AM2 boards are even capable of this overclock let alone on 1.3v?
 

x2 3600 rules sazakky

Senior member
May 11, 2007
410
0
0
Originally posted by: geokilla
I believe you can run your Athlon at 2.9Ghz 24/7 without any problems. If CPU-Z is detecting the voltages on my M2N-E correctly, then i'm getting horrible vcore. And by horrible, I mean horrible.

Did you run Memtest to verify it that it's stable at those speeds? DDR1000+ at 3-4-4-11 with just 1.95V is hard to believe. And don't forget that the M2N-E is probably the worst AM2 motherboard created by Asus due to the limited vdimm.

all of you's. i'm damn confused about whats happening
 

DerwenArtos12

Diamond Member
Apr 7, 2003
4,278
0
0
Originally posted by: x2 3600 rules sazakky
Originally posted by: geokilla
I believe you can run your Athlon at 2.9Ghz 24/7 without any problems. If CPU-Z is detecting the voltages on my M2N-E correctly, then i'm getting horrible vcore. And by horrible, I mean horrible.

Did you run Memtest to verify it that it's stable at those speeds? DDR1000+ at 3-4-4-11 with just 1.95V is hard to believe. And don't forget that the M2N-E is probably the worst AM2 motherboard created by Asus due to the limited vdimm.

all of you's. i'm damn confused about whats happening

First and foremost, don't get offended, we're not trying to attack you, just trying to weed out false information so the youglings here don't get confused too. At this point the best thing you can do is to download a memory testing software like sisoft sandra or everest and benchmark the memory. We can then compare that against other, similar systems to figure out what speeds the ram is actually running. We'll go from there, yeah?
 

x2 3600 rules sazakky

Senior member
May 11, 2007
410
0
0
Originally posted by: DerwenArtos12
Originally posted by: x2 3600 rules sazakky
Originally posted by: geokilla
I believe you can run your Athlon at 2.9Ghz 24/7 without any problems. If CPU-Z is detecting the voltages on my M2N-E correctly, then i'm getting horrible vcore. And by horrible, I mean horrible.

Did you run Memtest to verify it that it's stable at those speeds? DDR1000+ at 3-4-4-11 with just 1.95V is hard to believe. And don't forget that the M2N-E is probably the worst AM2 motherboard created by Asus due to the limited vdimm.

all of you's. i'm damn confused about whats happening

First and foremost, don't get offended, we're not trying to attack you, just trying to weed out false information so the youglings here don't get confused too. At this point the best thing you can do is to download a memory testing software like sisoft sandra or everest and benchmark the memory. We can then compare that against other, similar systems to figure out what speeds the ram is actually running. We'll go from there, yeah?

i'm not that pro at oc'ing, especially he weird amd hypertransport technology which is confusing. also my mobo is known to be bad for oc'ing. max voltage for the dimm's is 1.95 volts
 

DerwenArtos12

Diamond Member
Apr 7, 2003
4,278
0
0
Originally posted by: x2 3600 rules sazakky
Originally posted by: DerwenArtos12
Originally posted by: x2 3600 rules sazakky
Originally posted by: geokilla
I believe you can run your Athlon at 2.9Ghz 24/7 without any problems. If CPU-Z is detecting the voltages on my M2N-E correctly, then i'm getting horrible vcore. And by horrible, I mean horrible.

Did you run Memtest to verify it that it's stable at those speeds? DDR1000+ at 3-4-4-11 with just 1.95V is hard to believe. And don't forget that the M2N-E is probably the worst AM2 motherboard created by Asus due to the limited vdimm.

all of you's. i'm damn confused about whats happening

First and foremost, don't get offended, we're not trying to attack you, just trying to weed out false information so the youglings here don't get confused too. At this point the best thing you can do is to download a memory testing software like sisoft sandra or everest and benchmark the memory. We can then compare that against other, similar systems to figure out what speeds the ram is actually running. We'll go from there, yeah?

i'm not that pro at oc'ing, especially he weird amd hypertransport technology which is confusing. also my mobo is known to be bad for oc'ing. max voltage for the dimm's is 1.95 volts

Nobody here is going to argue that with you but, without more info this thread is notihng more than a big 'ol NEF at best and BS at worst. I know I for one would like to figure it out as it's just odd but, with nothing here besides some screeshots that most of the really knowledgable guys here think to be an erronious reading, to no fault of yours, I blame no-one for believing the latter option about this thread. You were willing to post CPU-Z SSs, why not sandra or everest? I for one at this point am attempting to help you by solving this with you instead of just mocking you and moving on, that could very well backfire and make me look stupid and you do not want to make me look stupid for trying to help you.