Please test

Community Forums/Showcase/Please test

Bouncer(Posted 2003) [#1]
This is a massive overdraw test (for testing fillrate) using 2000 big particles.

With my radeon 9600pro I got little over 100fps.
Would like to know how badly it crawls with for example GF2 cards.

It would be great to hear some FPS.
Thanks.

www.kotiposti.net/naama/massive.zip
Hold space to get real FPS reading.

Source now available .... scroll down......


fredborg(Posted 2003) [#2]
Very nice!

~35Fps on a P3-733MHz with a GeForce2-GTS


EOF(Posted 2003) [#3]
~26fps. Specs below.
Holding SPACE seemed to make no difference.


koekjesbaby(Posted 2003) [#4]
in case you care, on a radeon 9800 pro (inside an athlon 3000+) is runs between 140 and 157fps.


GfK(Posted 2003) [#5]
~20FPS - Athlon 900MHz, GF2MX.

Couldn't you get more speed by using a tri instead of a quad for each particle?


Caff(Posted 2003) [#6]
45fps, P700, Geforce 3 64mb Ti-200


Sunteam Software(Posted 2003) [#7]
8FPS - Celery 1.8Ghz, NVidia Riva TNT 64Mb


Ross C(Posted 2003) [#8]
99 fps. specs in sig. Win XP Pro, Dx9.0b


Bouncer(Posted 2003) [#9]
Gfk: It's not the particle count or vertex count that makes any signifficant difference in this case. It's purely about fillrate. The particle system itself can handle 10000 particles updated every frame on my machine, if they are smaller.


Ross C(Posted 2003) [#10]
10,000???? At what fps do you get that? That's impressive!


Bouncer(Posted 2003) [#11]
50fps


Ross C(Posted 2003) [#12]
Christ!! Care to share any of the trade secrets? :)


Beaker(Posted 2003) [#13]
I get 15 FPS! :)

GFK - triangles often lead to poorer fill than quads.


Bouncer(Posted 2003) [#14]
I can post source if you want to.... It's not documented though.

Just give me some time to clean it up and seperate it from my game engine....


Henrik(Posted 2003) [#15]
18-20 fps


Bouncer(Posted 2003) [#16]
In the meantime... here's 10,000 particles.
!!warning ... may kill the old GFX cards :)

www.kotiposti.net/naama/10000particles.zip


ChrML(Posted 2003) [#17]
18-22 fps on my GF2 MX 400 :(.


fredborg(Posted 2003) [#18]
Amazingly I get 20-22 fps on the 10000 particle test on the old hog: P3-733MHz GF2GTS


Ross C(Posted 2003) [#19]
Cool, i see your using the same surface for the text output as well, very nice. I'd like to see the source please, if you could be so kind :)

Oh, i get 46 Fps


Bouncer(Posted 2003) [#20]
Here's the source... It can seem a bit weird... but I ripped it straight out of my engine and documented it a bit. Hope you can get something out of it. Nothing advanced is implemented yet. Only basic movement and rotation functionality.

Also the quads don't rotate with camera (because my game is 2D) ... but it's easy to add.

So it's a combined font / particle system.
Here's the source:

www.kotiposti.net/naama/psource.zip

Enjoy


poopla(Posted 2003) [#21]
Looks nice Bouncer, I'm excited to see the work your putting into blitz as a finished product. Keep it up.


Ross C(Posted 2003) [#22]
Thank you very much man!


Magitta(Posted 2003) [#23]
28-30fps
P4m 1.6Ghz 256MB GeForce4MX Go 16MB


Andy_A(Posted 2003) [#24]
31-34fps
AMD 1GHz, GF2GTS, 512MB, 98SE

I get same frame rate holding down space or not. Is it supposed to do that?

BTW Nice.

Andy


Nebula(Posted 2003) [#25]
38-42 - see specs below


Bouncer(Posted 2003) [#26]
Andy A: Yes... if it runs under 50fps then it doesn't matter if you hold space or not.


Tom(Posted 2003) [#27]
First demo (2000): 50fps normal, 83fps holding space

Second demo (10k): 50fps with/without space held

specs in sig


Dustin(Posted 2003) [#28]
Failed to run (on my laptop)! It's a P3-850, 384meg RAM, and a Geforce 2 go but with a weak 16meg of video ram which is what I suspect is the reason I only get a black screen. I like to use my laptop as a "worst case scenario" machine. If it runs on it, it'll run on anything (almost)! After work, I'll try it on a more respectable XP 2200 w/ a Geforce 4.

Dustin


Rob(Posted 2003) [#29]
about 32fps with the first test!


Dragon57(Posted 2003) [#30]
Ran on my work machine with a Matrox G400 card.
15 fps on the first one w/wo space bar
9 fps on the second one w/wo space bar

Ran on my 1.8Ghz machine in my sig:
50 fps on the first one, no space bar. 90 with space bar
45 fps on the second one, no space bar. 55 with space bar


Bouncer(Posted 2003) [#31]
Thanks to all who reported framerates... it gives me some idea of the fillrate on different gfx cards.
I can now safely assume that a few hundred small particles will not affect framerate on even some slow spec machines.


Ethan3850(Posted 2003) [#32]
Better late than never...

On the first test:
50 fps without spacebar.
95 fps with spacebar.

On the second (10000) test:
52 fps without spacebar
62 fps with spacebar.


Apollonius(Posted 2003) [#33]
50 without space bar and 80 with space bar.


Rob Pearmain(Posted 2003) [#34]
On my laptop (See spec below)

50 fps without spacebar
63 fps with spacebar


Perturbatio(Posted 2003) [#35]
2000 particles: ~49 fps / 82fps with space
10000 particles: 48 fps (space makes no difference)


Ross C(Posted 2003) [#36]
Mmmm... i can't really see how you are getting so many particles on screen. I'll need to look closer :D