Jump to content
Corsair Community

ddr2 800 versus 1066


L4d

Recommended Posts

It depends. With your AMD machine it certainly is since the On CPU memory controller is very fast and transmits data along the HT path very quickly already so additional bandwidth speed is mostly theoretical and timings are more important. With the Intel legacy Memory Controller situated on the motherboard and running at legacy speeds, the additional bandwidth of the memory is more important than the timings. So, since you are running with your machine, I would say that 4-4-4-8-12 @ 800Mhz is faster than 5-5-5-15-22 in reality based tests. Canned bandwidth benchmarks might show that the 1066 is faster, but if you use a digital stopwatch and test the loads and accesses of data you will find the reality.

 

Is 4-4-4-8-12 ddr2 800 faster than 5-5-5-15-22 ddr2 1066?

Thanks.

Link to comment
Share on other sites

Rather than use canned bandwidth benchmarks, I advise you to actually work with the system and use a digital stopwatch. As well, an overclock of the CPU<--> Memory Controller<-->DRAM had better show even more improvement due to the overclock of the CPU and Memory Controller as well. Or there would be none of us overclockers. :p: Even so, using the Sandra bandwidth benchmark as a true statement on the actual performance of the entire subsystem is in error as much of that bandwidth is unused.

 

Given:

 

An Eight lane highway capable of transmitting 10000 automobiles per mile per hour seems very theoretically sound. The numbers bear this out.

 

Now if there are only 1000 automobiles per mile per hour that ever use the highway? I wonder what the reality based and digitally timed output would be versus the canned bandwidth benchmark?

 

Now latency is what gives the system the "Snap". When you click on an icon and how fast it opens up is the proof of the latencies ability. This is where timings will show great benefit. When using Photoshop and a 50MB rendering with twenty backups. Moving back and forth through the backup and forward steps and a final render with a digital stopwatch will show the reality.

Link to comment
Share on other sites

  • Corsair Employees
In real world usage you'll likely never see the difference with the naked eye. And this is especially true if you don't increase the CPU speed. As Derek noted, test it doing what you built the machine to do. If the difference is negligable, go with the settings that you find to be the most stable.
Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...