Posted on 04-Jul-2001 10:27 GMT by Christian Kemp | 24 comments View flat View list |
sauro wrote:
From Apex-Design web site: One new and one updated city are now available for Payback in the Extra Maps section. Also, the PPC upgrade is taking longer than expected because I'm trying to make sure it uses all the available CPU power. I don't expect to release it for another month but I hope you'll agree that the wait was worth it when it finally is ready.
|
|
List of all comments to this article |
Payback news : Comment 22 of 24 | ANN.lu |
Posted by Kent Seaton on 05-Jul-2001 22:00 GMT | In reply to Comment 21 (Menthos): > I will never use benchmarking again (unless impress on lame PC-owners). But
> the Quake test shows that calculations is pretty fast on JIT UAE...
Uhm, I hate to break it to you... the "timedemo" routine is a benchmarking program inside of quake.
Running timedemo demo2.dem from the console of quake in UAE is absolutely no different from running the speed test in Sysinfo. When the fading happens, there is no calculation difference for the frame being updated. To quote from the exact same manual that comes with WinUAE:
"Some benchmarks also use tight loops of DIV or MUL instructions to measure CPU speed, and in "emulation time" these instructions take as much time as any other instructions, and you'll get much too high values. If you enabled the "HAVE_RDTSC" option, running benchmarks makes some more sense. Still not a lot, since there is a minimum speed that the benchmark will report on every machine. The faster the machine, the more accurate will your results be."
On my Amiga running Quake060 (from clickboom) the results I saw were 985 frames 73.0 seconds 13.5fps. Now when I ran the 1GHz system with JIT and no sound, Quake (not the 060 version mind you) came in just about 1 second sooner. As soon as I turned on the sound emulation it was more along the lines 5 seconds later. If you do the math, without sound the most fps I got was about 14fps and with sound it was crawling pretty close to 10.2fps (maybe less). Just imagine how much different that would be if the two systems had the same video card.
> Starting with rev2 (the one I have installed) the sound works like a charm on
> my setup (and it has not done that in previous versions).
The JIT engine supplied with 8.16 r4 is the most stable one I've seen yet on the Athlon T-Bird. The portable I have can't even use the JIT (it's a PIII as well). A friend of mine accross the street has a system where the JIT routines are flawless (same settings file) and my other system 400MHz Celeron can't even load UAE without crashing something fierce. This type of diversity in the outcome of programming is extremely bad and fails the A.C.I.D. tests. In my opinion, the developers should never have released the JIT engine until it did what they intended on a more diverse offering of systems. |
|
List of all comments to this article (continued) |
|
- User Menu
-
- About ANN archives
- The ANN archives is powered by #AmigaZeux. It was updated daily (news last: 22-Oct-2004; comments last: 18-May-2005).
ANN.lu was created, previously owned and maintained by Christian Kemp, www.ckemp.com.
- Contribute
- Not possible at this time!
- Search ANN archives
- Advanced search
- Hosting
- ANN.lu was hosted by Dreamhost. Sign up through this link, mention "ckemp" as referrer and he will get a 10% commission on any account you purchase.
Please show your appreciation for any past, present and future work on ANN.lu by making a contribution via PayPal.
|