Messages in this thread | | | Date | Sun, 14 Mar 1999 21:35:44 +1100 | From | Richard Gooch <> | Subject | Re: MTRR/NOMTRR comparison |
| |
Tigran Aivazian writes: > Hi, > > You said: > > Some operations were slighly slower with a MTTR, most about the same, > > and some significantly faster. > > Of course, it is interesting to run x11perf*(1) and analyze the > output. However, in very practical concrete sense, I am running an > OpenGL-based solar system simulator (ssystem) and seeing a huge fps > improvement added by MTRRs on a system almost identical to yours - > about 20-30% speed improvement. (the 2.5x improvement claimed in the > docs is if all you do is bitblt-ing your stuff without doing any > calculations, whilst "real-world" apps spend most of the time > multiplying matrices and calculating sin()/cos() (unless they are so > clever as to use pre-calculated tables))
Ahem. http://www.atnf.csiro.au/karma/ Plenty of real-world applications which get the full 2.5x speedup. Yep, they do a lot of bitblitting. Pure bitblitting *is* used in the real world.
Regards,
Richard....
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |