Corsair TwinX 3700... hmmm :)

[mechBgon]

Spotted this at XBit Labs. They say this stuff is oriented towards the i865/i875 boards. A snip:

Corsair recommends using its TWINX-3700 memory modules with i865PE and i875P and Intel Pentium 4-based systems when overclocking. The timings of the new memory modules are relaxed to 3-4-4-8 (CAS Latency - RAS Precharge - RAS-to-CAS Delay - RAS Active to Precharge) and the modules should work at higher frequencies (e.g. 466MHz) in order to bring advantages in speed.

 

[DAPUNISHER]

Overclocked 3500? Anyways, I don't put much stock in faster timings unless they can achieve the same clockspeed as with looser timings so I say sweet!

 

[Chumpman]

nice

 

[gramboh]

3-4-4-8? That is garbage, you could probably get the XMS 3500 or OCZ 3700 to do 466 at 3-4-4-8 with ease. Isn't the OCZ 3700 rated for 2-3-3-7 (possibly single channel though). Disappointing considering the cost they will charge.

 

[Peter D]

Originally posted by: gramboh
3-4-4-8? That is garbage, you could probably get the XMS 3500 or OCZ 3700 to do 466 at 3-4-4-8 with ease. Isn't the OCZ 3700 rated for 2-3-3-7 (possibly single channel though). Disappointing considering the cost they will charge.

 

[rommelrommel]

Originally posted by: Vegetto

Originally posted by: gramboh
3-4-4-8? That is garbage, you could probably get the XMS 3500 or OCZ 3700 to do 466 at 3-4-4-8 with ease. Isn't the OCZ 3700 rated for 2-3-3-7 (possibly single channel though). Disappointing considering the cost they will charge.

 

[Corporate Thug]

twinX in overpriced, i'd rather get XMS and it'll run in dual channel just fine

 

[wicktron]

Low latencies don't matter as much as top speed past DDR400 on Intel systems. It matters plenty on AMD systems, which is why the blurt about the recommendation for the TwinX 3200LL instead of the new 3700's for nForce2 systems. My only question is will thse sticks hit DDR500+. I wonder what -ns rating the chips used are on these Corsair sticks. I have a friend with the OCZ PC3700 Gold that got to DDR500+ with ease. Plenty of other people got thse results with the OCZ Gold, too, so I'm deciding between the two. They have these modules in stock at Googlegear.com, and they're not that expensive, less than the OCZ Gold even.

 

[wicktron]

I'm ||
That close to becoming your guys' guinea pigs for the Corsair PC3700

 

[mechBgon]

*/me "accidentally" bumps wixt0r's elbow as he passes his mouse pointer over the "Submit Order" button

Ooops!

 

[YBS1]

wixt0r, this stuff will do 250mhz in dual channel, passes memtest86 with flying colors at that speed, comes in 512mb modules (vs. OCZ Gold's 256mb), and cost less than the OCZ Gold to boot.

 

[wicktron]

Originally posted by: YBS1
wixt0r, this stuff will do 250mhz in dual channel, passes memtest86 with flying colors at that speed, comes in 512mb modules (vs. OCZ Gold's 256mb), and cost less than the OCZ Gold to boot.

-5ns chips? I'll pass.
Rumor has it the Corsair modules are -5ns as well. Meaning that they're probably using some hand picked Winbond CH-5 modules and decrease the latencies so that they can hit DDR466. I doubt that there is much more room for OC'ing on these chips.

AFAIK, the OCZ Gold modules are at least -4.5ns chips.

 

[Corporate Thug]

extreme DDR sucks

 

[YBS1]

extreme DDR sucks

I guess that's why mine is running 256mhz 1:1 right now.

 

[Lyfer]

Must cost an arm and a leg.

 

[mechBgon]

Aye, Captain Lyfer, it be rather expensive

 

[Tetsuo]

Originally posted by: rommelrommel

Originally posted by: Vegetto

Originally posted by: gramboh 3-4-4-8? That is garbage, you could probably get the XMS 3500 or OCZ 3700 to do 466 at 3-4-4-8 with ease. Isn't the OCZ 3700 rated for 2-3-3-7 (possibly single channel though). Disappointing considering the cost they will charge.

...thats way too much money for that

 

[Khanivore]

A friend and I were waiting for corsair to annouce their new twinx ram and so we put our orders on old. When we found out it was going to be CAS3 ram, we just ignored it and purchased 1Gb of XMS PC3500C2 each. Haven't tried PC3700 speeds yet but we'll be testing that speed out soon, prolly at 3844 settings.

 

[shoman94]

Originally posted by: Khanivore
A friend and I were waiting for corsair to annouce their new twinx ram and so we put our orders on old. When we found out it was going to be CAS3 ram, we just ignored it and purchased 1Gb of XMS PC3500C2 each. Haven't tried PC3700 speeds yet but we'll be testing that speed out soon, prolly at 3844 settings.

Doesn't work...I tried it....I can get ddr460 stable at 2.8v cas 2.5,7,3,3 Unstable at ddr464......can't post further with this ram.

SHO

 

[oldfart]

Originally posted by: gramboh
3-4-4-8? That is garbage, you could probably get the XMS 3500 or OCZ 3700 to do 466 at 3-4-4-8 with ease. Isn't the OCZ 3700 rated for 2-3-3-7 (possibly single channel though). Disappointing considering the cost they will charge.

Agree. Not impressed. Those timings are pathetic. And latency certainly DOES matter! On my IC7 875 rig, I can get the SAME performance running 5:4 ratio 2-2-2-6 as running 1:1 ratio 2.5-3-3-7. The lower latency makes a big difference.

 

[wicktron]

Not according to these results. Not that in depth as far as going through all the possible permutations of latencies, but it sure does show that CAS2 vs CAS2.5 @ DDR400 does NOT matter.

CAS2 vs CAS2.5

(results taken from Zroc)

The timing that matters the most (on i865/i875) is RAS to CAS. A simple change from 2 to 3 or vice versa in this yields the biggest difference in REAL benchmarks (eg. UT2K3, JK2, Q3). A lower CAS will ALWAYS help Sandra/Aida as they test for theoretical maximum bandwidth.

 

[oldfart]

Here is some testing I did on 1:1 Vs 5:4 with my config.

My system can hit one of two walls

1) Mem speed wall. My ram maxes out ~ DDR 426 with slow 2.5-3-3-7 timings. This equals 213 FSB max in this config.

2) CPU speed wall. My 2.6C can go to 257 FSB @ 1.6 Vcore.

This gives me a choice of

1) 213 FSB | 2.77 GHz | 1:1 | DDR426 2.5-3-3-7
2) 257 FSB | 3.34 GHz | 5:4 | DDR 412 2-2-2-5

LAME MP3 Encode
(lower is better)
257 FSB | 5:4 | 2-2-2-5
Encoded 11 files in 0:02:17

213 FSB | 1:1 | 2.5-3-3-7
Encoded 11 files in 0:02:45

213 FSB | 5:4 | 2-2-2-5
Encoded 11 files in 0:02:46

213 FSB | 3:2 | 2-2-2-5
Encoded 11 files in 0:02:46

Obviously, MP3 encode is all about CPU speed. 1:1, 5:4, 3:2 made no difference at all.

MPEG2 Video encode benches
TMPGEnc AVI -> SVCD CBR (lower is better)

257 FSB | 5:4 | 2-2-2-5
2:10

213 FSB | 1:1 | 2.5-3-3-7
2:35

213 FSB | 5:4 | 2-2-2-5
2:36

213 FSB | 3:2 | 2-2-2-5
2:40

Again, CPU speed is king. Also 1:1 slow timings and 5:4 fast timings are about the same. 3:2 is a little farther behind.

Q3 benches:

FSB | Ratio | cas settings | performance mode | Q3 FPS
257 FSB | 5:4 | 2-2-2-5 | disabled | 390.6
257 FSB | 5:4 | 2-2-2-5 | fast | 391.3

Next, 213 FSB since this is my max 1:1 FSB setting due to the limit of my ram. To get there I have to use slow timings:

FSB | Ratio | cas settings | performance mode | Q3 FPS
213 FSB | 1:1 | 2.5-3-3-7 | disabled | 335.8
213 FSB | 5:4 | 2-2-2-5 | disabled | 336.2
213 FSB | 5:4 | 2-2-2-5 | fast | 338.3

Didn't test 3:2 on that test.

All this talk about "MUST RUN 1:1" is WAY overstated. Every bench I've seen and run myself certainly says otherwise.

 

[wicktron]

All this talk about "MUST RUN 1:1" is WAY overstated. Every bench I've seen and run myself certainly says otherwise.

That's because you run i875, for i865, 1:1 is the ONLY way to go, or else the pseudo-PAT enhancements are lost.

Regardless, we need a test for the various permutations of latencies at different speeds to give a definitive answer to the latency mystery, but AFAIK, they do NOT matter as much as they used to.

 

[oldfart]

PAT (MAM, GAM, etc) is normally off on all 875 and 865 boards when using a ratio. Once Asus figured out how to enable it on the 865 chip, things changed. If the mobo mfgr decides to make a BIOS that will do it, PAT can be enabled on the 875 or 865 when using a 5:4 ratio. CPUz version 1.18c will test for PAT disabled/enabled.