By using this site, you agree to our Privacy Policy and our Terms of Use. Close
selnor said:
HappySqurriel said:
selnor said:

Please dont get me wrong. The GC was a beast of a console. And IMO PS2 was quite away behind. Xbox  just had an advantage with a more capable and feature rich GPU than what GC posted.

Something that will make your mind explode.

Virtua Fighter 5 for PS3 and 360 looks great. It has 40k of polys per fighter and a maximum backfround ploy count of 300k. Per frame VF5 has 380k polys max/sec. Compare that with Rogue 3 which as 20mill polys/sec. It has roughly 60 times more polys than VF5. But looks infinately graphically better.

Poly performance is 100% not a guage for graphics.

380,000 polygons per frame = 22.8 Million polygons per second at 60fps

Edit: In other words, Virtual Fighter 4 pushed slightly more polygons but had greater texture detail, advanced effects, and rendered at a higher resolution resulting in a better appearance.


Sorry Rogue 3 is 20 mill per frame. I worded it wrong.

Rogue 3 is 20 mill per frame. LP is 3mill per frame, Dead Rising is 4 mill per frame.

Lair by contrast is 134 mill per frame.

 

Its interesting to note that Lair is the highest polygon count game this generation for PC or console. Again its made by the same developer Factor 5.

Yet is not the best looking  game this generation. Also is predominantly a flight game just like Rogue 3 was last gen.

 

Mario Sunshine about 110k polys/ per frame

Lost Planet 3mill polys/ per frame

Dead Rising 4 mill polys/ per frame

Rogue 3 20 mill polys/ per frame

Lair  134 mill polys/per frame

 

I'm almost certain you are mixing polys/frame in some games with polys/second in other games.

There is no way that Rogue 3 is 20 million per frame, the game runs at 60fps, that would be 1.2 billion polys/second.  The initial conservative estimates for the GC put performance at about 12 million/sec, conservative certainly, but not out by a factor of 100.

Furthermore the only number I found for Lair was a quote of 134 million polygons for the whole stage/level, again, not per frame.

Agreed on the main point though, using polys per frame or per second as a metric of performance in this day and age is rather redundant.