 |
|
| View previous topic :: View next topic |
| Author |
Message |
leo0207 (RT.X2)
Joined: 04 Jul 2005 Posts: 45
|
Posted: Thu Jun 22, 2006 9:24 pm Post subject: Improving performance |
|
|
Hello,
I have a validated system which is working properly. Some effects or effect combinations seemed to require rendering. I was wondering what bench mark tests is the matrox team using and if they are available to us to test our system to ensure proper operation. I also would like to know what would improve my preformance, more ram, adding more disk to the raid video drive etc, overclocking cpu or GPU ? How is one to know which area is causing the redline?
Can you describe the validating process ?
Also the ranking of systems isnt specific enough...what the difference between a system rank 2 versus 4 ?
Your answer to the question above could help us out improve performance and make sound purchasing decisions.
Regards,
Leo  |
|
| Back to top |
|
 |
Chris Y (Moderator)
Joined: 04 Jul 2005 Posts: 7
|
Posted: Wed Jun 28, 2006 1:52 pm Post subject: |
|
|
The system ranking is a relative one, 1 works better than 2 and so on.
It is difficult, if not impossible, to say exactly what each system does as everyone’s application is different. For example, if multilayer compositing is what you do best, then what you may be interested is how many layers can be played back in real time. If this was the only measure, than we could definitely tell you which system is best and why. (System 1 is the best because it did 7 layers; system 2 did 6 and so on …)
However, this means nothing to the guy who is only interested in doing 2 layers and 3-d effects between them, or the guy who has simple effects on HD. What we are trying to say is there is no universal measuring stick.
What we have done instead is generated several timelines, in different video formats, including different effects. All test results are compiled against their ability to pass without dropping frames, as well as the recorded CPU usage at the time of testing. This is then what is used to generate the resultant table.
The GPU table is a little different. We purposely create an effects intensive timeline that we know will need rendering. Then we test how long it takes to render. The faster it does it the better.
We recommend reading the entire Designing your Matrox RT.X2 system guide as this should provide you some insight.
Some highlights you may want to look at are …
- Performance RAM vs. Value ram
- Better GPU card
- Better CPU
I hope this helps. I realize this is not the black and white response you were probably hoping for, but as I said there is no universal measuring stick, so we cannot provide the exact measurement of each system.
What we can tell you is that all systems have been tested and that the ranking is not arbitrary, but that in fact we did use a logical system. |
|
| Back to top |
|
 |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|