FPU VAMPIRE2 Beat All Available AMIGA Accelerators | page 1 2 3
|
---|
|
---|
| | Gunnar von Boehn (Apollo Team Member) Posts 6254 30 Sep 2017 19:29
| Continous improvement
| |
| | Gregthe Canuck
Posts 274 30 Sep 2017 21:46
| Very nice jump in performance once again. According to an earlier post by Samuel Devulder that is very close to 68040@40 speed.
| |
| | Stefano Briccolani
Posts 586 30 Sep 2017 21:50
| Amazing!!!
| |
| | Thierry Atheist
Posts 644 30 Sep 2017 22:47
| This is TERRIFIC!!!! But, I'm confused. Why is the MIPS rating 104 when we saw a photograph showing 160 MIPS a short while ago???
| |
| | Philippe Flype (Apollo Team Member) Posts 299 30 Sep 2017 23:42
| 104 MIPS, at moment, is for the next x11 public core. 160 MIPS was surely a higher-clocked internal test core (x13+). Little improvement : EXTERNAL LINK
| |
| | Gregthe Canuck
Posts 274 01 Oct 2017 00:12
| 7.46 -> 7.85Another nice boost. Now faster than a 68040@40.
| |
| | Philippe Flype (Apollo Team Member) Posts 299 01 Oct 2017 09:10
| Well, faster in SysInfo. AIBB BeachBall for example shows that 040 is still much faster, but the work on femu shows it have still many room for improvement, even outside of the math libs calls.
| |
| | Peter Heginbotham
Posts 214 01 Oct 2017 18:01
| Whats the goal for its release with Gold 2.7, beat all performance 040 tests?
| |
| | Rod March
Posts 119 01 Oct 2017 23:44
| Awesome! Gimme! Gimme! Gimme!
| |
| | Manuel Jesus
Posts 155 27 Oct 2017 13:59
| EXTERNAL LINK Here is a test using Mircobe and the Wazp3d 68k FPU based Warp3d library replacement. Since I got a Vampire I have wanted to test this out, sadly an FPU was required but now it actually load objects, of course the really low poly stuff gets a better frame rate.
| |
| | Vojin Vidanovic
Posts 770 27 Oct 2017 14:26
| [quote="Manuel Jesus" Here is a test using Mircobe and the Wazp3d 68k FPU based Warp3d library replacement | Nice development, glad to see FPU is used to solve Warp3D problem :-) Will this library be somewhat availiable?
|