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

What's better DDR3 1333@CAS7 or 1600@CAS9?

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

vsalt2

Member
Joined
Oct 27, 2004
Location
Albany, NY
This mobo/cpu combination has onboard video, so I would assume the tighter timings would benefit the video more than higher frequency.

The ram is currently rated at 1600@CAS9 1.5v, and it seems to run at 1333/CAS7 1.65v

Any assistance would be duly appreciated :)
 
I would run 1600@CAS9, which would be roughly 7% faster. In either scenario, you aren't going to notice the difference.
 
I would run 1600@CAS9, which would be roughly 7% faster. In either scenario, you aren't going to notice the difference.

Thanks for the reply, can you possibly explain how CAS9 at any frequency would be better than CAS7? The frequency must have a lot to do with it. I don't think my APU supports natively at 1600 but the motherboard seems to run at 1600 no problem.

I think the max frequency the APU supports is 1333.
 
Thanks for the reply, can you possibly explain how CAS9 at any frequency would be better than CAS7? The frequency must have a lot to do with it. I don't think my APU supports natively at 1600 but the motherboard seems to run at 1600 no problem.
In very basic terms, the CAS delay is how many cycles (Hz) it will wait before issuing the next command. So if you have a CAS of 7, it will wait 7 cycles before doing the next operation. If you run the memory faster, you have to raise the CAS so that there are no errors. An easy way to compare is to divide the speed by the CAS rating. So 1333@CAS7 is ~190 and 1600@CAS9 is ~177. Assuming everything is fully loaded (and my math isn't faulty), those numbers are millions of operations a second. So, in a perfect world, 1333@CAS7 should be 7% faster, but not noticeable without synthetic benchmarks.
 
In very basic terms, the CAS delay is how many cycles (Hz) it will wait before issuing the next command. So if you have a CAS of 7, it will wait 7 cycles before doing the next operation. If you run the memory faster, you have to raise the CAS so that there are no errors. An easy way to compare is to divide the speed by the CAS rating. So 1333@CAS7 is ~190 and 1600@CAS9 is ~177. Assuming everything is fully loaded (and my math isn't faulty), those numbers are millions of operations a second. So, in a perfect world, 1333@CAS7 should be 7% faster, but not noticeable without synthetic benchmarks.

Doesn't CAS ratings have more an effect in regards to shared onboard video memory? Obviously the faster the memory, the better benchmark I will get with onboard video (which is my #1 concern btw).

These new APU series pack out some pretty decent onboard video capability, and I want to make sure I'm getting the best possible shared RAM to feed it :)

I greatly appreciate your math, however I'm confused. Earlier in a post you thought that the 1600@CAS9 was around 7% faster, do you still believe this to be true?
 
CAS is a delay (in Hz) the memory will wait before doing another operation. The higher this value, the more it waits.

CAS itself has nothing to do with onboard memory. Bandwidth, on the other hand, does.

I greatly appreciate your math, however I'm confused. Earlier in a post you thought that the 1600@CAS9 was around 7% faster, do you still believe this to be true?
I apologize for the confusion, I typed it incorrectly. 1333@CAS7 should be 7% faster.
 
The A8 actually supports up to DDR3-1866.
If that's the case I'm likely going to sell this RAM and get the 1866, i want the best it can get :) Thanks for the info, do you have a source where i can confirm it accepting RAM so high in frequency before I change it out?
 
http://www.overclockers.com/amd-llano-desktop-lynx-platform-review/
They’ve slightly revamped their Stars core (which is found in existing Athlon II and Phenom II CPUs) and given the memory controller a makeover, giving support for up to DDR3-1866. I can attest to the fact that it takes no tweaking at all to run that kind of speed with timings of 8-9-8-24.

http://www.amd.com/us/products/desktop/apu/mainstream/Pages/mainstream.aspx#3

fusionmemory.png
 
Back