Re: MultiGen DB speed

New Message Reply Date view Thread view Subject view Author view

Russell Suter (russell++at++ctasim.com)
Tue, 1 Apr 1997 07:45:14 -0800

On Apr 1, 6:45am, Steve Baker wrote:
> Subject: Re: MultiGen DB speed
> > From: russell++at++ctasim.com ("Russell Suter" )
> > Subject: MultiGen DB speed
> >
> > Hey all,
> >
> > I've been working on fixing a large MultiGen database using MultiGen
> > version 14.0. The database had, among other things, problems with
> > detail textures. I went through the entire database and cataloged
> > all textures used and selected an appropriate detail texture (where
> > necessary). Now, when I perfly the database on a RE system using
> > IRIX 6.2 and performer 2.0.2, and OpenGL my frame rate is about 1/5
> > of the performance than when no detail textures were used. I
> > expected a performance hit but really! This performance hit is even
> > in places where I wouldn't expect it -- when the eye point is far
> > enough away from the terrain so that the detail texture is not seen.
> > I can't find much useful info in the performer documentation. Is there
> > some adjustments I can make somewhere?
> >
> > Any info is appreciated!
>
> It sounds like you are running out of texture memory.
>
> Usually detail texture comes at no extra cost (or at least negligable
> extra cost) - so if the only change you made was to the textures,
> that must be the cause.
>
> When you run out of texture, GL starts paging it into and out of the hardware
> and the frame rate takes a nose dive.
>
> If you turn on the Performer GFX stats, it will tell you how many texture
> transfers are happening. If the answer is anything other than zero then wave
> goodbye to your frame rate.
>

Thanks Steve! That's the conclusion that I reached as well. I was looking
for paging in perfly stats and didn't see any. By a stroke of luck, I was
able to talk with a former employee here who now works for SGI. He said
that since I was running on an old IR with performer 2.0.2, performer
stats might not be complete. He said that just because I wasn't seeing any
texture paging in the stats, didn't mean that it wasn't happening. The
good news is that the target system is an IR. I look forward to verifying
this.

Thanks once again.

> I hope this helps...
>
>
> Steve Baker 817-619-1361 (Vox-Lab)
> Hughes Training Inc. 817-619-8776 (Vox-Office/Vox-Mail)
> 2200 Arlington Downs Road 817-619-4028 (Fax)
> Arlington, Texas. TX 76005-6171 Steve++at++MrEd.bgm.link.com (eMail)
> http://www.hti.com (external) http://MrEd.bgm.link.com/staff/steve
(intranet)
> http://web2.airmail.net/sjbaker1
    (external)
>
> "You can't destroy the Earth - that's where I keep all my stuff!" - The Tick.
>
>-- End of excerpt from Steve Baker

-- 
Russ
_______________________________________________ ______________________________
Though my eyes could see                       | Russell Suter
        I still was a blind man.               |    Voice : (303) 889-1262
               Though my mind could think      |      Fax : (303) 889-1210
                       I still was a mad man.  | Internet : russell++at++ctasim.com
_______________________________________________|______________________________
=======================================================================
List Archives, FAQ, FTP:  http://www.sgi.com/Technology/Performer/
            Submissions:  info-performer++at++sgi.com
        Admin. requests:  info-performer-request++at++sgi.com

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:55:00 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.