- Joined
- Jun 1, 2002
Overclocker550 said:ok guys you are wrong.
I gain 2 3dmarks per MHz increase in the cpu. never tried the increase in quake3. I gain 1fps in u2004 per 100MHz cpu increase. how bout I try quake3 at 640x480 fastest setting with my cpu at max, at 1GHz and 1.5GHz all using the same fsb? 200 for all cpu speeds tested. I will gain much more going from 1GHz to 1.5 than 1.5 to 2GHz
while I can not quote you about Q3A....
(the thread was deleted.) still, only 1fps gain in UT2004 for every 100mhz seems strange....???
first, is there a demo test in UT2004?????
next, in UT2003 I get about 3fps increase for each 100mhz, CPU increase....and this is with a intel CPU.
this game does far better with a AMD cpu...you should be getting a little more then just an extra frame.
""I will gain much more going from 1GHz to 1.5 than 1.5 to 2GHz ""
this statement shows that your vid card limited....
if your video card had more bandwith in it, then you should be getting the same increase in performance, all the way up to, and far beyond 2ghz.
if your increase in performance takes a nose dive after 1.5ghz then what is the point at running a 2ghz or faster?????
inother words, why buy an A64 and stick that ti4200 in it???
back to Q3A for a moment.....
running the timedemo at 640x480 will basicly only test the CPU.
(and you know this)
run the test at 1024x768, 32bit, 32bit, at least.
btw, if you still play this game, what res do you play it at????
if it's 1600x1200, then that is the res you should be testing the game at....
and it will show more of the vid card in this test then at 640x480.
I realy wish I had the time (and my injored hand didn't hurt so much) to explain bottlenecks to ya OC550....
but you seem to understand it.
I just don't understand why with everything your saying, you still think that a faster CPU will solve your bottleneck???
mica