Voodoo 3 3000 -vs- Voodoo 4 4500, strange

[Ilwrath]

The methods used on Amigaspeed.de are dated, and un-scientific. The only way to have an accurate comparison is to do the tests on identical systems (ie. same software, patches, memory, etc. The only variable should be the PCI solution)

Yeah... And even then, why so much emphasis on 3d performance?   I mean, I'm interested in a Prometheus, and have never even bothered to install Warp3D as it's basically useless except if you want to play ports of 5-7 year old games at framerates that only look when compared to 5-7 year old x86's.

I'd really like to see some fairly good apples-to-apples comparisons of 2D performance vs. both Mediator and a Z3 graphics card for reference.  The same for network card performance would be cool, too.

I think that would finally put the FUD to rest.

[Ilwrath]

P96Speed is a 2D benchmarking program.

Yes, but none of the P96 tests that I saw on Amigaspeed.de.vu even mentioned the Prometheus?

I dunno, Quake II here makes a pretty playable game. I really don't think a 7 year old PC could touch it.

You forget that the Pentium II 300mhz was released over 7 years ago.  (May 7, 1997)  Toss that Voodoo card in with a P2/300 and an Intel 440LX chipset (available August 1997 -- so ALMOST 7 years), and I bet it'll match or probably even beat the Amiga.  :-(  

And who would use that rig, anymore?  I mean, charitable organizations won't even take them as donations!  In comparison to anything resembling a modern machine, the Amiga has virtually no 3D power, with or without a Voodoo card.  Why try to press something that the Amiga just isn't suited for?

You have one handy? Run P96speed and post the results for comparison.

Which version are you using?  I have a module created for my baseline A4000 reference.  (A4000 Cyberstorm Mk2 CV64/3D)  

If it's the right version, I'll e-mail it to you.  :-)  (If not, let me build a stock workbench, and build one -- I wouldn't want to use my normal workbench, as it's a little corrupted/flaky after a nasty disk corruption.)

I'd also happily donate a baseline A2065 Network baseline.  Just supply me with the test.  :-)  

I'm all for finding actual performace facts.

[Ilwrath]

on pathetic framerates:

Possible, go dig up some benchmarks and let's find out

The RIVA 128 AGP on a 300mhz P2 with 440LX chipset pulls 49 fps in Quake II GL. Source: Tom's Hardware -- December 1997 (And if Tom's had a review board back then, it's a good chance he bought it from the local store.  His website was hardly known.)  So if we go by your 33fps, a nearly 7 year old computer bests the Amiga by 16 FPS, or 48%.

Why would this be a selling point?  It's like trying to sell a new ultra-cool fuel-efficient hybrid car by advertising that it can "accelerate almost as fast as a 15 year old Yugo" instead of playing up the environmentally friendly aspects.  

Contrary to what modern PC benchmarking entails, I think there are performance numbers more useful than 3D benchmarks.  Especially in cases like this, where the 3D power is so woefully behind modern systems.  

Playing Quake II with a solid 33FPS is no slouch and does not handicap someone playing against a PC.

Well, it kind of does.  The PC user can crank some resolution and pick off the Amiga user before the Amiga user can even pick out the PC player from the blotchy wall texture.

And, outside of Amiga users, does anyone play Quake II anymore?  The FPS genre doesn't really have "nostalgia."  I mean, no one wants to go back to the "good ole days" of shooting at poorly formed, low polygon objects with blotchy textures and little to no atmospheric effects.  I don't think anyone really chooses to play Quake II over Quake III, do they?  Especially as the multi-player game goes, Quake III isn't so much as sequel as it is a drop-in replacement.

on P96 benchmarks:
I'm feeling a bit of insomnia.  I'll run the set tonight.  (I can't find my P96 module right now, and I want to post this before my session times out...)

[Ilwrath]

I really think you are missing the point here.

Well, I think we're kind of missing each other's points.  Mine is more like this:  "Why would an Amiga user want to play 5-7 year old FPS games (very badly dated, and look terrible compared to modern offerings) at 5-7 year old speeds (33FPS doesn't fit my definition of smooth, and 640x480 doesn't fit my definition of high resolution.)"  I didn't care THAT much for Quake II when it was new.  I surely have no nostalgia for it when I can play another game that has exactly the same gameplay and looks so much better...

A PC that could perform 3D better than the top Amiga can be found in your local dumpster.  Literally, I wasn't kidding when I said
local charaties won't even take them!

Why torture yourself and your miggy trying to do something that it fails at so miserably?

Why do people restore classic/antique cars? Why do people customize classic/antique cars? They aren't as fast as modern cars, get much worse gas mileage, handles far worse and often it costs alot more to restore an old classic then it is to buy a new production car.

I think you're on the same course as me, here, but came to a totally different conclusion.  It's a good bet that the person who restores a classic car also has a "daily driver."  The car that takes all the abuse of daily wear and tear of city driving, huge potholes, winter salt, etc.  You don't see them trolling for a parking spot at the local mall with their classic.  IF they drive the classic somewhere, it's a safe bet it'll be parked safely away from stray shopping carts, careless kids, etc...  

I see 3D performance as a relatively mundane task.  And if I want to play a 3D game, why torture the Amiga over it, when I have a tremendously more capable PC (that I really don't care much about) sitting here begging for some polygons?

It's like taking your classic restored car up to Home Depot to pick up a load of plywood and lumber.  You just don't do it.

[Ilwrath]

Do you need 150FPS + just to make a game playable? I think not, it's simply bragging rights.

No... But it's hard to argue that 50fps isn't an advantage over 30 (especially when that 30 can drop to 20 or less in high polygon situations).  And is all but impossible to argue that 1600x1200 isn't an advantage over 640x480.  Regardless of games you play.  Personally, (if you couldn't tell from my avatar) I've been into Papyrus NASCAR 2003 for about a year now.  The extra resolution and draw distance are CRITICAL for a game like that.  When you're approaching trouble at 180MPH you've only got a split second to either save your ride, or have an early trip to the pits.  The extra resolution can help you identify that trouble sooner.  I'm sure it's the same for people into first-person shooters, or whatever.  (The sooner you can identify the enemy, the sooner you can neutralize them.)  The extra resolution and draw distance is much more important than any frames over 50.  I can't imagine anyone who would willingly go backwards to having something in the power-class of an old Voodoo 3 or RIVA TNT.  

Let me ask you this, if you feel Amigas should be burried along with PCs of that era, why are you here?

Like the classic cars, the Amiga has plenty of character and nostalgia.  It's the retro computer champ.  It is really a historic masterpiece.  It was multimedia before the word was invented.  Easily 10 years or more ahead of its time.  I'm much more into the preservation of Amiga than trying to make it into a useful every-day platform, anymore.  The Prometheus is an amazing product, built after a long line of previous amazing products for Amiga.  (Like the Golden Gate that could use [then common] ISA cards native on Amiga.)  Here is a simple product that allows the use of common PCI video cards and allows the fastest high-color Workbench ever on Amiga (fastest Windowed environment on any platform, ever?  Maybe -- not sure how to prove that, though), full RTG support, etc, and it all integrates nicely with the original Amiga.  

My own advice would be to let the Prometheus stand on those merits, rather than trying to play up 3D figures that are still very sub-par, any way you look at them.