Carmack on Parhelia
- Thread starter MfA
- Start date
well, that nails last hopes about getting Parhelia powerful under higher polycounts on future games...
I hope Matrox has guts to react fast enough to this situation. If they are going to G400 route, (means veryyy slow price drops or no price drops at all.) then we can kiss good bye at least Matrox gaming cards... But I am hoping for that Matrox could show being able to pull another rabbit from the hat if the first one is totally blind.
*sigh*
Carmack's words can be your best selling advertisements or your worst enemies and I don't think you need help to figure out which one category this one goes from Matrox's point of view...
I hope Matrox has guts to react fast enough to this situation. If they are going to G400 route, (means veryyy slow price drops or no price drops at all.) then we can kiss good bye at least Matrox gaming cards... But I am hoping for that Matrox could show being able to pull another rabbit from the hat if the first one is totally blind.
*sigh*
Carmack's words can be your best selling advertisements or your worst enemies and I don't think you need help to figure out which one category this one goes from Matrox's point of view...
Johnny Rotten
Regular
The performance was really disappointing for the first 256 bit DDR card. I
tried to set up a "poster child" case that would stress the memory subsystem
above and beyond any driver or triangle level inefficiencies, but I was
unable to get it to ever approach the performance of a GF4.
Sad.
tried to set up a "poster child" case that would stress the memory subsystem
above and beyond any driver or triangle level inefficiencies, but I was
unable to get it to ever approach the performance of a GF4.
Sad.
Doomtrooper
Veteran
JC has spoken
LittlePenny
Regular
I wonder what he is using more than 2 alpha bits for? I am also curious how this thing handles the stencils. Lets see some FableMark 
Hellbinder
Banned
And this is beared out as truth in exactly what benchmarks... or comments by JC. Even the radeon 8500 beaths the GF4 in one or two areas but *no one* is suggesting GF4 class performance.
It just kiils me when people can read plain clear english and make up whatever they want out of it....
JC= "even when i tried to favor it... it was not as fast as a GF4"
You say "this proves that it has GF4 class perfomance"
Am I on another planet or what....
Doomtrooper
Veteran
With eye candy on it does, and I thought that is what we bought $400 dollar video cards for. Yes its not blowing a Ti4600 out of the water(yet beats it) but impressive FSAA numbers for a debut. Not to Mention Matrox has been out of the high end game for a long time, I will wait for a few driver releases to say its a failure.
I think HYPE has killed this card (Nappe) being the 1st 256 bit card the expectations were too high...like a movie...hype..hype and you see it and say "that wasn't that great".
I look at this way, if you bought a Ti4600 for FSAA performance whats stopping someone for doing the same with a Parhelia.
http://firingsquad.gamers.com/hardware/parhelia/page12.asp
What about image quality, I think it looks amazing..look at the sky
http://firingsquad.gamers.com/hardware/parhelia/page9.asp
I think HYPE has killed this card (Nappe
) being the 1st 256 bit card the expectations were too high...like a movie...hype..hype and you see it and say "that wasn't that great".I look at this way, if you bought a Ti4600 for FSAA performance whats stopping someone for doing the same with a Parhelia.
http://firingsquad.gamers.com/hardware/parhelia/page12.asp
What about image quality, I think it looks amazing..look at the sky
http://firingsquad.gamers.com/hardware/parhelia/page9.asp
Doomtrooper
Veteran
Fixed..grrr hate that
I seem to recall saying months ago (in a B3D thread) that efficiency problems could kill this card, but other fans were saying "even if 30% less efficient, it has 2x the bandwidth, so it will still be way faster!"
Can we finally put the nail into the coffin on these "bandwidth is king" arguments that have been going on over the past 4 years on B3D? You know, you take the specs of 3D card X, and 3D card Y, compare bandwidth, # pipelines, texture units, and clock, and pronounce the winner?
Of course, these comparisons were used to consistently declare that NVidia's new card would lose its performance crown. And we know how successful those predictions were.
Can we finally put the nail into the coffin on these "bandwidth is king" arguments that have been going on over the past 4 years on B3D? You know, you take the specs of 3D card X, and 3D card Y, compare bandwidth, # pipelines, texture units, and clock, and pronounce the winner?
Of course, these comparisons were used to consistently declare that NVidia's new card would lose its performance crown. And we know how successful those predictions were.
Well, it is not 2x the bandwith compared too GF4 Ti4600
16GB/11.5GB= 1.39 or 39% more bandwith.
Many simultaneous reasons at work:
- new drivers (buggy, not optimized)
- LMA and cache eficiency of GF4
- Lower internal latency of the GF4 core (50% higher frequency) and Amdhal´s law.
- Lots of tricks (occlusion culling, etc)
- Higher image quality ???
- Too much confidence on 256bits and not enough hardware optimizations
- Lower latency GF4 memory.
16GB/11.5GB= 1.39 or 39% more bandwith.
Many simultaneous reasons at work:
- new drivers (buggy, not optimized)
- LMA and cache eficiency of GF4
- Lower internal latency of the GF4 core (50% higher frequency) and Amdhal´s law.
- Lots of tricks (occlusion culling, etc)
- Higher image quality ???
- Too much confidence on 256bits and not enough hardware optimizations
- Lower latency GF4 memory.
arjan de lumens
Veteran
Umm, given 650 MHz RAM for GF4 and 550 MHz for Parhelia, that would bepascal said:
(550*256/8 ) / (650*128/8 ) = 17.6/10.4 = ~1.69 => 69% more bandwidth.
Agree about the over-confidence in the 256-bit bus, though.
Another possibility is hardware bugs which are causing stalls/increased latency in their pipeline. Management could of decided they could not afford another cycle/delay and had to go with the broken hardware to get it out before the next gen from ATI and nVIDIA. If true, the next revision or "spring refresh" could perform better then what can be attributed to clock speed increases alone (assuming a process improvement).
Jerry Cornelius
Newcomer
The Parhelia is not always going to be able to leverage all 4 TMUs in many standard benchmarks. How it stacks up with AF is what interests me, aside from the obvious FAA feature which remains damn impressive.
I'm a little surprised it only faired so well in AF test, yet it's up against a frugle implementation from ATI and NVIDIA's magic TMUs, so it faired pretty well considering it's clock speed.
I'd be interested in measuring it's performance when a alpha slash momentarilly covers the entire screen, completely flushing the texture cache and making little reuse of texels in the process. This is the sort of thing the card is supposed to excel at IIUC.
I'm a little surprised it only faired so well in AF test, yet it's up against a frugle implementation from ATI and NVIDIA's magic TMUs, so it faired pretty well considering it's clock speed.
I'd be interested in measuring it's performance when a alpha slash momentarilly covers the entire screen, completely flushing the texture cache and making little reuse of texels in the process. This is the sort of thing the card is supposed to excel at IIUC.
