Entries for tag "gpu", ordered from most recent. Entry count: 19.
# Lectures on ETI, Gdańsk University of Technology
Employees of Intel Technology Poland are visiting Gdańsk University of Technology, Faculty of Electronics, Telecommunications and Informatics (known as ETI). On Thursday - 8, 15, 22 January 2015, there will be lectures as part of "Computer Graphics" course. Time: 11:15 - 13:00, place: new ETI building, room NE AUD1L. It's a lecture for students of computer science, but anyone who is interested can come and listen.
Together with Piotr Kozioł, I will be presenting on January 22nd. Our presentation has title "Shaders and their compilation" and will cover:
During 2 hours we will cover lots of topics - basically all what happens to the shader after it's written in high level language and passed to graphics API - how it's processed by the driver and executed by the GPU.
# What do we have from benchmarks?
There was this case some time ago about some graphics vendors cheating in Futuremark benchmark (see this). They basically detected this particular application and raised frequency to increase performance and gain higher score. So some devices have been delisted from the Best Mobile Devices list for cheating and they published this document: Benchmark Rules and Guidelines.
My first thought was: Good, they just want everyone to play fair. But then I read the rules again, especially this one: "The platform may not replace or remove any portion of the requested work even if the change would result in the same output." and I said: Wait, what? Isn't it a generic definition of every optimization? If a developer writes 2+2 in GLSL and the platform just uses 4, is it cheating because it removed requested work (addition in this case) even if result is the same?
And then I started thinking: What do we have from benchmarks after all? Is their importance a good thing for gamers and other customers of graphics technology? In theory, benchmarks should mimick some aspect of real applications to measure and compare how different hardware performs in this type of applications (e.g. games). But it may be that decision makers want to just see good scores in benchmarks (bosses generally like numbers and bars and graphs :) so engineers implement optimizations or even some cheats just for these benchmarks. And then media notice that, devices get delisted, benchmark creators write such rules... and gamers just want to play games.
If performance was measured just in real games, and platform vendors optimized or even cheated for a particular title, then at least we would have a better performing game. Just my personal opinion :)
# Recent Demoscene Parties
There have been several interesting demoscene parties recently. First one is RiverWash in Warsaw, Poland (I have been there by myself :) Next one is Function in Budapest, Hungary. And finally, today was the end of MAiN in Arles, France. After each party it's nice to download prods from compos (productions from competitions) from Pouet.net website (see prods from this year's RiverWash, Function and Main). Demos and intros are worth watching!
Different topic: the new Nvidia Fermi architecture looks very promising. Just have a look at White Paper PDF and watch Next Generation GPU Fluids on YouTube. This architecture is going to be released soon as GT300 graphics card. Intel Larrabee is no longer a revolutionary idea :)
By the way, the whole GPU Technology Conference (San Jose, USA, Sep 30 - Oct 2) looks like an interesting event. I wonder if there will be any papers available for download...