Latency for normal use

JonB

Distinguished
Jan 10, 2003
38
0
18,530
How big of a deal is latency for normal applications and some periodic gaming? Which would offer better performance: 512MB (DDR333) CAS 2, or 1024MB (DDR333) CAS 2.5? Thanks!

JonB

"Consiousness of the past alone
can make us understand the present" - Herbert Luethy
 

speeduk

Distinguished
Feb 20, 2003
1,476
0
19,280
From what I know, 2.0 latency offers just a little bit more performance than 2.5, probably not even noticeable, only in benchmarks!
 

JonB

Distinguished
Jan 10, 2003
38
0
18,530
I think i've heard that its a bigger deal for hardcore gaming, but I'm not sure. Anybody else? Please? :smile:

"Consiousness of the past alone
can make us understand the present" - Herbert Luethy
 

phsstpok

Splendid
Dec 31, 2007
5,600
1
25,780
There are so many latencies in accessing memory, RAS, CAS, RAS to CAS and couple of others that I can't think of. In the worst case, truly random memory access all the latencies come into play. I think there are a max of something like 9 to 12 clock cycles (memory clock) of latencies. In the best case, fewer latencies are in play and CAS has a bigger impact. In any case, reducing CAS from 2.5 to 2 is only a half cycle and a small part in the total latency.

I've heard claims the net gains of CAS 2 vs CAS 2.5 is as low as 1%-2% or as high as 8%-10% for typical use. Personally I believe the lower estimate.

However, that said, in general CAS 2 memory is of higher quality than CAS 2.5 memory. You can often reduce the other latencies (RAS, tRCD, etc) of CAS 2 memory where you might not with CAS 2.5 memory (especially at high memory speeds). The more latencies you can reduce the better the performance.

<b>99% is great, unless you are talking about system stability</b>
 

juin

Distinguished
May 19, 2001
3,323
0
20,780
some spec benchmark run close to 15% faster but are highly syntetique and old wich i dont trust.

Just next to the lab and the bunker you will find the marketing departement.
 

juin

Distinguished
May 19, 2001
3,323
0
20,780
on a page hit i guess not much difference but on random acces page miss gain can reach 10$ but normaly it will have a page hit excepting if you use a lot of ram as chipset got a fix number of page open.Exception EV7 with it 2000+ open page wich it almost like having a 1 GB of cache it will allwayse hit exception to 1 tetrabytes of ram and the like but that what EV7 is made for.

[-peep-] french
 

juin

Distinguished
May 19, 2001
3,323
0
20,780
it timing is the internal lantency inside the chip. abit like P4 got about 20 cycle lantency on FPU divider.by the way P4 FPU divider dont have pipeligne

[-peep-] french
 

ejsmith2

Distinguished
Feb 9, 2001
3,228
0
20,780
Yep.

The only time you would actually ever notice the latency is if you were running a 4 or 5 hour process. Like, encoding a DVD to mpeg4 or something.

And then, it would depend on your processor and northbridge, as well.

"I personally think filesystems should be rewritten from scratch every 5 years..." --- Hans Reiser
 

phsstpok

Splendid
Dec 31, 2007
5,600
1
25,780
on a page hit i guess not much difference but on random acces page miss gain can reach 10$ but normaly it will have a page hit excepting if you use a lot of ram as chipset got a fix number of page open.Exception EV7 with it 2000+ open page wich it almost like having a 1 GB of cache it will allwayse hit exception to 1 tetrabytes of ram and the like but that what EV7 is made for.
Whoosh! That went right over my head.

Sorry, I'm not a programmer. You lost me there. I was talking about typical use stuff, games, DivX encoding, MP3 encoding, productivity tools like photo editing, maybe some video editing. You know, consumer applications. Do you think these a greatly affected by whether memory is CAS 2.5 or CAS 2?

Did your mean "Terabyte"? "Tetrabyte" would mean four bytes.

<b>99% is great, unless you are talking about system stability</b>
 

Crashman

Polypheme
Former Staff
Game benchmarks have shown me around 7% better performance using Cas2 instead of Cas3 on PC133. I would expect the Cas2.5 to Cas2 switch to be about half that, or just over 3%.

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>
 

umheint0

Distinguished
Feb 18, 2003
947
0
18,980
You'll only really notice any gain if you are a FPS fanatic, or if you are encoding a big honking DivX movie. however, lower CAS often means lower latencies all around, and if you want to overclock, the CAS2s ought to be able to suck more performance than 2.5s.
 

Col_Kiwi

Distinguished
Aug 8, 2002
429
0
18,780
Cas difference seems to be erratic and hard to predict -- back when I had only 256mb ddr333 (samsung) i tried forcing it to cas2 (from default cas2.5) as someone here told me it can usually manage that. my 3dmark score went DOWN.



-Col.Kiwi
 

Crashman

Polypheme
Former Staff
That can be caused by your BIOS secretly changing other memory timings to slower settings, or by memory errors.

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>