Corsair TwinX 4000 vs OCZ Gold 4000

Memory by KeithSuppe @ 2003-09-27

A simple and to the point review. Sandra memory benchmarks will be used MAX-3 version and WCPUID for verification purposes. Which one comes out on top?

  • prev
  • Go to mainpage

Conclusion

Conclusion:

My verdict then is simple. If your platform is the Epox 4PCA3+, and your goal is overclocking, at the highest FSB attainable, then the OCZ does pull ahead. Yet in this comparison, I would not choose one memory over the other. If one were to calculate the bandwidth, one may find that IF the Corsair were able to run at the same FSB it may have resulted in higher bandwidth. So in this case it's really a matter of choice.

To be honest, I like OCZ for overclocking, and I like Corsair for overall stability. Now this may be a result of my pushing OCZ whenever I get it, and the fact it's a true overclocker memory, and always has been is revealed here today.


And in that respect, Corsair performed exactly as it was designed to. I've always seen Corsair as more of a "system" memory, and it's always been stable. Don't get me wrong, Corsair has released some the best overclocking memory on the market. Corsair offers many different memories, to meet the needs of many systems. I find their memory to be more versatile, in its application. My verdict therefore is a dead tie. Both are excellent products, both did what they were designed to do, and far surpassed that design. Overclockers might think, 266FSB isn't that interesting, because we’ve been spoiled.


When Twinmoss, ADATA and other companies release their PC3700, PC4000, and people are posting these high bandwidths in hardware forums, their not taking into consideration the irresponsibility of their acts.

As experienced Overclockers they unfairly influence the average end-user, who may foolishly believe hi/her motherboard is
A) Capable of such high VDIMM
B) That such voltage is safe.

They are not! Many of today's IC's (Integrated (memory) Chip) do have internal voltage regulation, and NO memory, is going to perform error free for life, if your pushing excessive amounts of voltage into it. Many IC's are running a core voltage of 1.8V. Therefore, what it boils down to, and what I'll be showing in the weeks to come, is exactly what's under the heat spreaders of both these modules. Because IC's differ from memory to memory. There are only a handful of Fab's, it’s true and so our choices aren't so large. Hynix rev-B, and Winbond rev-B do not use the internal voltage regulation, ironically this is what's being found on the some of the high bandwidth/Low Latency memory which has been performing so well in Canterwood, and nForce-2/3 motherboards.


I do not believe, Corsair is using these "older" IC's, and in trying to implement a memory that is RATED for 500MHz many Overclockers in their search for the ultimate FSB, eschew these modules. This can have an influence on the Gamer, and end-user who may foolishly think their being taken advantage of, when nothing could be further from the truth.

Corsair hasn't been, perhaps the most respected, and established memory manufacturer on the market, by building memory, which will last 6-months to a year. And here at Madshrimps, while we are Enthusiasts at heart, we aren't led by our emotions. Common sense dictates overclocking, is not the ideal operating environment for any PC.

This will void warranties, and eventually damage memory, as well as other components. Perhaps most frustrating, is the amount of time, and hardware which must be invested, to ameliorate all the unwanted side-effects overclocking produces.

Because I'm an extremist, and scientist at heart, I will be tearing down the heat-spreaders in the weeks to come for my next article. Yet in the interim I want to enjoy both these excellent memory modules until PC4200 arrives, and/or I have the Corsair Pro4000 series, and the OCZ Gold4200 in my TIM stained fingers.

/Liquid3D

I you have any questions or comments please post them here at our forums.
  • prev
  • Go to mainpage