Why are frame rates of Jaguar 3D games so low?

[agradeneu]

Just some measures from myself (average):

 

Checkered Flag: 9 fps

Edit: Skyhammer: 9-12fps

Hover Strike CD: 9-11fps

Alien vs Predator: 8-9 fps & lower

Cybermorph: 12fps

Edit: Supercross 3D 6fps

I-War: 12fps

World Tour Racing: 9fps

 

Edited February 15, 2007 by agradeneu

[The_Laird]

Mainly because the engines were not good enough, except HS:UL of course

 

AVP is not really 3D either.

[Goochman]

Are you sure Supercross 3D gets that many ;

[The_Laird]

Are you sure Supercross 3D gets that many ;

 

I was thinking that too LOL!

[EmOneGarand]

AVP is not really 3D either.

I think it has some to do with the depth of the ray tracing, the distance goes on pretty damn far, especially the lighting sources.. even the sprites gradually fade in unlike some games with popup, they also arn't a blurred mess like in DOOM at far distances. With some optimizing it probly could have been just as fast as Doom and Wolfenstein 3D. But the game still looks great, the frame rate never bugged me.

[Punisher5.0]

I'm going to say at best crappy support from Atari and not enough time to learn how to use the hardware efficiently. Just look at BattleSphere which has a very smooth framerate that gets as high as 60fps. Also judging by the videos of Gorf3D the framerate looks very good.

[agradeneu]

I'm going to say at best crappy support from Atari and not enough time to learn how to use the hardware efficiently. Just look at BattleSphere which has a very smooth framerate that gets as high as 60fps. Also judging by the videos of Gorf3D the framerate looks very good.

 

 

It's 60fps when there is no object. But it's one of the best.

But I rate rate both ISoldiers easily the best 3D ever created on the Jaguar.

Then comes Battlemorph which has many objects, good looking textures and gourad shading and runs at respectable 15fps average.

[+LS650]

Because the Jaguar was designed to be first and foremost a sprite-pushing engine, and 3D was not the primary goal. Take a look at games like Super Burnout, Val d'Isere, Atari Karts, etc.

 

It's unfortunate that more of the developers didn't try to build their games with scaling sprites instead of polygons.

 

Supercross 3D: they should have named this game Supercross 3FPS...

Edited February 15, 2007 by ls650

[agradeneu]

Because the Jaguar was designed to be first and foremost a sprite-pushing engine, and 3D was not the primary goal. Take a look at games like Super Burnout, Val d'Isere, Atari Karts, etc.

 

It's unfortunate that more of the developers didn't try to build their games with scaling sprites instead of polygons.

 

 

 

That was the Panther. The Jaguar was actually a design to do both 2d and 3d gfx, though the hardware supports gourad shaded polygons and z-buffer and texture mapping is limited.

[Zerosquare]

The Jaguar was actually a design to do both 2d and 3d gfx, though the hardware supports gourad shaded polygons and z-buffer and texture mapping is limited.

Correct, but the hardware support for 3D in the Jag is pretty minimal compared to, for example, the PlayStation ; a lot of things have to be handled in software.

Of course, the Jag was released during the early days of 3D in consoles.

[Atari_Owl]

Forget it

Edited February 16, 2007 by Atari_Owl

[+LS650]

How the heck do you measure the number of frames per second, anyhow?

[Video]

 

Supercross 3D: they should have named this game Supercross 3FPS...

LOL

 

Yeah, of course, the Jaguar was built when 2D was still WAY bigger than 3D, it has some limited 3D built in, but it was built largely as a 2D system.

 

Also, for the day it was made, 93-96 or so, 10 FPS for a Poly game was actually quiet respectable.

 

Just compare the Jaguar to say, Genny or SNES poly games.

[JagChris]

How the heck do you measure the number of frames per second, anyhow?

 

That's a good question. What techniques are you both using to come to these numbers?

[Atari_Owl]

Forget it

Edited February 16, 2007 by Atari_Owl

[agradeneu]

Just some measures from myself (average):

 

Checkered Flag: 9 fps

Edit: Skyhammer: 9-12fps

Hover Strike CD: 9-11fps

Alien vs Predator: 8-9 fps & lower

Cybermorph: 12fps

Edit: Supercross 3D 6fps

I-War: 12fps

World Tour Racing: 9fps

 

 

 

Your frame rates do not agree with my measurements:

 

HS CD You say 9-11 - i measure 12-15 (Edit: your values are reasonable for the Cart version however)

WTR You say 9 - i measure 12-15 (Edit: perhaps you meant 2 player, if so you should state this.)

AvP you say 8-9 - i measure 9-11.

 

We agree on Battlemorph being 15fps though.

 

i have not measured CF, Cybermorph, Skyhammer, SC3D or I-War

 

BIWN was a surprisingly high 12-15fps dependent upon angles

Iron Soldier >=25-30fps

 

 

Let us not forget also tht it has been stated that Zelda Majora's Mask on the N64 drops as low as 10fps in some scenes.

These frame rates were not at all unusual for the mid 90s.

 

Well except 6fps is clearly unacceptable.

 

 

Hovertrike CD: I think the framerates depend on which level you play. In more complex levels like the urban levels, the framerate is between 7-12 fps. In the ice level, frame rate mostly runs at 12 fps.

Did you notice that the framerate drops whenever there are more than 2 objects on time on screen?

WTR: Some tracks are running faster some are really choppy. In turns the frame rate gets really choppy. This game is nowhere at 12-15 if you take the average.

AvP: In sublevel 4 I can count the frames by eye.

 

I doubt that Zelda Majora Mask runs at a 10 fps average.

Even Iron Soldier drops to 5-7 frames occasionally.

With the arrivement of Playstation, Saturn and N64 animation speed at least at 24 fps average became usual. The Jaguars animation speed does not compare favourably.

[Atari_Owl]

Forget it

 

Oh except

 

 

I doubt that Zelda Majora Mask runs at a 10 fps average. icon_ponder.gif

 

Did i say average... No i didn't! I said IN SOME SCENES.

Edited February 16, 2007 by Atari_Owl

[Fredifredo]

So we can say that IS2 is the best 3D engine of the Jaguar ?

[Breakpack]

How the heck do you measure the number of frames per second, anyhow?

 

That's a good question. What techniques are you both using to come to these numbers?

 

 

 

If the frame rate is >25or30fps then there is the complication that the acquisition device is probably not capable of capturing that many frames per second so a completely accurate measure is no longer fully certain. You can get a decent idea however if the acquired frames show evidence of different images being interlaced (suggesting that the fps >25-30)

 

 

Which is why I like DVD recorders as opposed to video capture cards.But thats another subject for another time in another forum

[+LS650]

It sure reads to me like you guys are making estimates with a Mark I Eyeball, and not actually taking measurements with any accurate means....

 

If that's the case, then your estimates can be biased, and are pretty much meaningless.

[Tempest]

Just some measures from myself (average):

 

Checkered Flag: 9 fps

 

Does CF actually hit 9 fps? Wow, I would have thought it was around 5 or 6...

 

Tempest

[Atari_Owl]

It sure reads to me like you guys are making estimates with a Mark I Eyeball, and not actually taking measurements with any accurate means....

 

If that's the case, then your estimates can be biased, and are pretty much meaningless.

 

Fine Forget it

The frame rates vary and since agradeneu didn;t explain how his mesures, there may have been a good reason for the differences.

 

If however you'd rather believe hype than objective measures... then do what you want, frankly I don't care to explain it again. I should have known better than to post anything and not expect mocking comments like this.

Edited February 16, 2007 by Atari_Owl

[Breakpack]

It sure reads to me like you guys are making estimates with a Mark I Eyeball, and not actually taking measurements with any accurate means....

 

If that's the case, then your estimates can be biased, and are pretty much meaningless.

 

Fine Forget it

The frame rates vary and since agradeneu didn;t explain how his mesures, there may have been a good reason for the differences.

 

If however you'd rather believe hype than objective measures... then do what you want, frankly I don't care to explain it again. I should have known better than to post anything and not expect mocking comments like this.

 

How many FPS are those comments?

[agradeneu]

It sure reads to me like you guys are making estimates with a Mark I Eyeball, and not actually taking measurements with any accurate means....

 

If that's the case, then your estimates can be biased, and are pretty much meaningless.

 

 

For Hoverstrike there is a built in debugger which shows you exactly the FPS count.

Apart from that, I never claimed that these measures are accurate or scientific proof. However it's obvious that the FPS counts are low and there is no need for scientific research to make rough estimates on that. It may vary within 2-3 frames but its rather objective.

Edited February 16, 2007 by agradeneu

[agradeneu]

It sure reads to me like you guys are making estimates with a Mark I Eyeball, and not actually taking measurements with any accurate means....

 

If that's the case, then your estimates can be biased, and are pretty much meaningless.

 

Fine Forget it

The frame rates vary and since agradeneu didn;t explain how his mesures, there may have been a good reason for the differences.

 

If however you'd rather believe hype than objective measures... then do what you want, frankly I don't care to explain it again. I should have known better than to post anything and not expect mocking comments like this.

 

 

I took Hoverstrikes FPS counter and compared it to the rest of Jaguar games. Mark I eyeball.