Personal computing discussed

Moderators: Flying Fox, Thresher

Gerbil Elite
Topic Author
Posts: 917
Joined: Fri Jan 20, 2006 12:37 pm
Location: The void which binds

CAS timings

Thu Dec 20, 2007 8:18 am

So I always kind of wondered what they actually mean. I read the wiki article on it, but


So yeah, are lower numbers better or higher ones? :D
Flying Fox
Gerbil God
Posts: 25408
Joined: Mon May 24, 2004 2:19 am

Thu Dec 20, 2007 8:23 am

Ever since this it has been shown it does not make a discernable (other than benchmarks) difference. But lower is better. Plus on Intel platforms this matters even less.
The Model M is not for the faint of heart. You either like them or hate them.

Gerbils unite! Fold for UnitedGerbilNation, team 2630.
Grand Admiral Gerbil
Posts: 10000
Joined: Thu May 08, 2003 12:42 pm
Location: Cincinnati, OH

Thu Dec 20, 2007 9:43 am

In super duper dumb terms, CAS latency numbers indicate the time it takes to do something, so:

5 units of time is more than 4 units of time so 4 is faster (less time) than 5.

But that's hugely simplified because there are a bunch of factors involved.

As Flying Fox said, CAS timings are not worth messing with unless your some Liq-Nitro freak hell bent on beating someone else's benchmark score even if by only 1/100th of a percent. Sometimes those timings can have 1-3% impact and in rare instances the impact can be higher but those are rare and very specific instances.

Who is online

Users browsing this forum: No registered users and 1 guest