"The race is not [always] to the swift, nor the battle to the strong."
- Ecclesiastes 9:11
More pages: 1 ... 11 ... 21 ... 31 ... 41 ... 51 ... 61 ... 71 ... 81 ... 91 ... 101 ... 111 ... 121 122 123 124 125 126 127 128 129 130 131 ... 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 Failedicastano
Sunday, March 22, 2009

Time to buy an NVIDIA board? :P
Michael
Sunday, March 22, 2009

R.I.P. (Rest in Pixels)
Groovounet
Saturday, March 21, 2009

GPUs get better performane only if the amount of data is large enought and it depents on whether you are using it on the gpu. Good to have a good sse implementation

It's quite easy to demonstrate improvement on a specific experiment, I'm looking forward for real use case test. Mat4 product and inverse are such great topics
Michael
Saturday, March 21, 2009

nice gain!

I'd love to see a fleshed out article about this- perhaps adding a GPU implementation for comparison
Humus
Saturday, March 21, 2009

Sorry, I guess the post was a bit unclear. I actually implemented all the SSE stuff in MSVC first. But I also wanted to make sure it works in Linux before I go to great lengths with this idea since I want to keep cross platform compatibility. I only needed a few minor changes to make it compile and run fine under GCC in Linux so it looks like I'll keep this code.
Michael
Friday, March 20, 2009

ah - scratch last comment. you never said you needed to install linux in order to experiment with intrinsics... I misread.
Michael
Friday, March 20, 2009

why do you think you needed linux for using sse intrinsics... ?
the Windows sdk comes with the xmmintrin.h header file which contains declarations for MMX/SSE/SSE2 intrinsics.

btw - i played around with gentoo years ago as well. It seems that most linux users I know spend an awful lot of their time configuring their system/desktops....
Arseny Kapoulkine
Friday, March 20, 2009

There is MinGW, which should have support for all SSE intrinsics you might need, so Linux as a SSE testground is a bit strange
More pages: 1 ... 11 ... 21 ... 31 ... 41 ... 51 ... 61 ... 71 ... 81 ... 91 ... 101 ... 111 ... 121 122 123 124 125 126 127 128 129 130 131 ... 141 ... 151 ... 161 ... 171 ... 181 ... 191 ... 201 ... 211 ... 221 ... 231 ... 241 ... 251 ... 261 ... 271 ... 281 ... 291 ... 301 ... 311 ... 321 ... 331 ... 341 ... 351 ... 361 ... 365