274 posts on 10 pages  12345678910
 General Discussion
2015-10-27, 00:10
634 posts

Registered:
Nov 2008
mission wrote:
This article shows that "900 FPS to 450 FPS" performance loss is exactly the same as "60 FPS to 56 FPS" (pretty much negligible difference). For non-programmers that's not obvious at all.

I use 1155 fps. If it drops to 1000 even once, I feel it very well. It feels like a small lag spike, even though there is no spike.
2015-10-27, 00:33
Member
22 posts

Registered:
Oct 2015
Renzo: thanks for excellent reply. Will probably save some bucks on not going for active sync, but rather lots of fps and regular 144 hz. Maybe ill even wait for more hz tech to come to TN panels.
2015-10-27, 09:24
Member
43 posts

Registered:
Apr 2012
That article was bullshit. The only thing it explains is that from a computing cost comparison (not relevant). Also it's first grade math.
2015-10-27, 10:54
1024 posts

Registered:
Apr 2006
mission wrote:
This article shows that "900 FPS to 450 FPS" performance loss is exactly the same as "60 FPS to 56 FPS" (pretty much negligible difference). For non-programmers that's not obvious at all.

That is just plain stupid way of lying with numbers.. Dunno where to start explaining:

First example goes from 1.11ms rendering time to 2.22ms, a factor 2 difference! If you can do _everything_ in 1.11ms as I explained in my previous post then adding something to make it take 2.22ms then your addition takes up as much time as _everything_ else combined.

Second example goes from 16.66ms to 17.77ms! That's not a factor 2, it's way less! Here it makes more sense that your addition has a 6-7% performance hit!

So even if the extra rendering time is the same, in one case it makes makes much more sense than in the other... It can be the same thing and at the same time definitly not the same thing depending on what you compare to. The reasonable thing to do in most cases is to compare to total time and then it's not the same thing.
274 posts on 10 pages  12345678910