Sign in to follow this  

best way to make game render most efficient

This topic is 1186 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Hi,

I asked this question from some level designer and modeler peoples and forums but non of them gave me a good answer so I thought to ask it from graphic programmers.

 

How can I find maximum possible polycount and texture resolution and particle for a certain hardware or platform? I think my question is obvious but no one gives a good answer.

 

Many people say even great teams start benchmarking and do some test and fail on it. for example when they see the game runs slow on 360 they start to lower polycounts. I think it is a long and hard process. or for pc they make game and after that they start benchmarking it on different configurations and they introduce minimum system. I just want to know what is the best way to do that.

 

Thank you for helping

Share this post


Link to post
Share on other sites

 

how can i find maximum possible polycount and texture resolution and particle for a certain hardware or platform? i think my question is obvios but no one gives a good answer.

Well then you asked realy competent people - trust them.

 

There is a big differeince in a polycount that runs over 50 pixels or over 1000000 of pixels, if this is per poly, not per group, square it up smile.png

There is big difference in pixel function code (how many reads from textures it performs, how many writes to target it drops, how many triangles cover the fill and even how triangles look...).

 

But some general infromation would be :

1 -GPUs can compute unbelievably many instructions with local or registered memory, but access upper memory (read texture, write to target, feed vertex function) very limited.

2 - returnign to 1, GPUs most of the time wait for memory, even if they have supper ballance speculations

3- Newest Intel CPU gpu cores are as much slower as top most gpus as they ram is slower  - compare 24 stream processors with 1333MHz memory to 800 stream processors with 4500Mhz memory, ratio of their perofrmance is much like  4500/1333

Edited by JohnnyCode

Share this post


Link to post
Share on other sites

This topic is 1186 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

Sign in to follow this