View Single Post
  #6 (permalink)  
Old 12-16-2010, 03:41 PM
[ACL]'s Avatar
[ACL]
VIP Member
Offline
Threadstarter
Location: NY
 
Join Date: Feb 2010
Posts: 1,534
Reputation: 6350
[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community[ACL] is a trusted member of the community
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Re: Test with framebuffer

Awesome.. i want to make sure all the Rhod variants are covered. I will commit this since it looks like it works for almost everyone now.

One thing some people noticed is the spike to 50fps. This is what we are working on now. I just got my hands on the novtec documentation 2 days ago so we are still trying to see how we can benefit from this.

The good news is our topaz brethren are also helping out in the quest for 50fps so lets hope one of us stumble on it. For those of you who havent heard of the theory. We cant update the screen until the vsync happens. It happens only 60times per second, so if we miss it by just a few ms, then we have to wait for the next one killing our fps by half pretty much.

I can share the documentation with anyone who can compile and are willing to make a few changes on their own. This limit was discovered by toastcfh when the 30fps cap was changed on the evo.

https://github.com/toastcfh/Superson...705a7e7fb497b5

The link above is the commit that managed to break 30fps on the evo. I already tried and it does not work for us, but maybe im missing something. Sometimes it takes a few eyes to figure this out.

i had a lot of wine so i cant think any more.. thanks fellas (those of you who tested)
__________________
Reply With Quote
This post has been thanked 1 times.