hi,

the benchmark is nice, but the page is a little uninformative for the
interrested programmer :)
could you tell us a little about the dirty details of the tracer used?
things like:
- does it use sse/sse2/3dnow?
- if so, how do you exploit it? coherent seems unlikely as you seem to be
using full solids instead of tris.
- what kind of spatial subdivision scheme does it use? (seems to have
portals in there, but what for the tracing core?)
- how far did you go into raw assembly or is it all plain c?
- what other surprising insights did you get from making the tracer fast?
any neat & non-obvious tricks you want to share? :)


/shiva^kolor

----- Original Message -----
From: "wurstbrot20002000" <contact@...>
To: <realtime_raytracing@yahoogroups.com>
Sent: Wednesday, August 07, 2002 2:09 AM
Subject: [realtime_raytracing] Realtime Raytrace Benchmark


> Hello,
>
> we launched our new website, about realtime raytracing. at
> www.realtimeraytrace.de you fill our new rtrt-benchmark at the
> downloads section.
>
> If you find the time to run the benchmark, i would be pleased if you
> send the file data\database.rdb back to my email:
> contact@...
>
> thanks for you time,
>
> Michael Piepgras
> (TTS/F.A.N./Oxyron)
>
>
>
>
> Community email addresses:
> Post message: realtime_raytracing@onelist.com
> Subscribe: realtime_raytracing-subscribe@onelist.com
> Unsubscribe: realtime_raytracing-unsubscribe@onelist.com
> List owner: realtime_raytracing-owner@onelist.com
>
> Shortcut URL to this page:
> http://www.onelist.com/community/realtime_raytracing
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>