"I believe that 16, for example, is not an important number."
- David Kirk
More pages: 1 ... 5 6 7 8 9 10 11 12 13 14 15 ... 21 ... 31 ... 41 ... 51 ... 61 ... 71 ... 81 ... 91 ... 101 ... 111 ... 121 ... 131 ... 141 ... 151 ... 161 ... 171 ... 181 ... 191 ... 201 ... 211 ... 221 ... 231 ... 241 ... 251 ... 261 ... 271 ... 281 ... 291 ... 301 ... 311 ... 321 ... 331 ... 341 ... 351 ... 361 ... 365
Query FailedJenny Cao
Wednesday, May 6, 2015

Hello Mr Persson (Humus)

My name is JennyCao, we are working on the Crosswalk project (a blink based webview runtime for advanced web technologies) in Intel Open source center.
We have a test suite kit on github to demostrate the usage and effects of new web technologies.
We find you have a good samples for WebCL, and want to integrate it into the test suite usecase.
From the �legal stuff� of https://github.com/uebquite/webcl-demos/blob/master/Deform/textures/readme.txt , we must get your �written approval� if we distribute the sample in Crosswalk test suite.

Would you garentee us the �written approval� for the distribution the sample in Crosswalk test suite?

You could check the integration PR and provide comments, we will merge the PR only after get your �written approval�.
https://github.com/crosswalk-project/crosswalk-test-suite/pull/2073/files#r28686945

Ps: you have a beatiful wife and lovely sons

K�y
Tuesday, April 28, 2015

Awesome! Congrats .

mr
Sunday, April 5, 2015

RSS still contains only blog entry titles.

Humus
Sunday, March 29, 2015

Michael, it's not necessarily a better way to do it. It's a different way to do it, and comes with a different set of pros and cons. On the plus side, it's easier to do light assignment, and you can also do exact culling with very little cost. You also get clusters that are evenly sized and consequently even distribution of lights into clusters.
On the other hand, it doesn't map to pixels as easily, which may add some lookup cost. It also doesn't scale as easily, you may have to use a cascading solution if you have a large view distance.

As for the DX10 icon, yeah I know, I just didn't have a DX11 icon on my site. I'll fix that later.

Whether to do light assignment on the CPU or GPU depends on your needs and workload. Since the CPU doesn't have much to do at all in this demo I expect better performance doing light assignment on the CPU instead of GPU, allowing the GPU to only concern itself with shading.

Nuninho1980
Saturday, March 28, 2015

"Fullscreen" is lower GPU usage than "windowed". lol

I got ~3650 fps at 1024x768@8xAA (windowed at desktop 1280x1024) - ~93% GPU usage

Notice: read my previous posts.

al
Saturday, March 28, 2015

boo

Nuninho1980
Saturday, March 28, 2015

"Fullscreen" is lower GPU usage than "windowed". lol

I got ~3650 fps at 1024x768@8xAA (windowed at desktop 1280x1024) - ~93% GPU usage

Notice: read my previous posts.

Nuninho1980
Saturday, March 28, 2015

Your newer demo hasn't well optimized as almost rest your older demos because its don't use (almost) full GPU usage. Why??

Running Win 8.1 with my same specs (see my previous post)
-- ~2910 fps at 1024x768@8xAA - only ~70% GPU usage
-- ~1445 fps at 1920x1200@8xAA - only ~86% GPU usage

More pages: 1 ... 5 6 7 8 9 10 11 12 13 14 15 ... 21 ... 31 ... 41 ... 51 ... 61 ... 71 ... 81 ... 91 ... 101 ... 111 ... 121 ... 131 ... 141 ... 151 ... 161 ... 171 ... 181 ... 191 ... 201 ... 211 ... 221 ... 231 ... 241 ... 251 ... 261 ... 271 ... 281 ... 291 ... 301 ... 311 ... 321 ... 331 ... 341 ... 351 ... 361 ... 365