tomshardware review online!

[Rags]

If you are referring to the 3Dmark2k1 benchmark with multitexture, I can't really put any weight on that since most of us would agree that 3dmark scores do not translate into real world numbers.

Well, since it's the only thing he used that actually stresses quad texturing, it's the only thing you can go by in that review. The multitexture fill rate gives an idea of the type of performance you will get with those games, and upcoming games like UT2003 will reveal this to be true as well.

As far as "VERY playable framerates" are concerned, I would be most interested to see what the low framerates are for all 3 cards with aniso & AA(FAA vs FSAA).

I cannot comment on what specific framerates I get in games, I will tell you that playing UT and Q3 in triple mode 2400x600 does net higher framerates than playing with a GF4 in 1600x1200 with everything on. In addition, with everything turned to max and on, the Parhelia has a smoother feel to it in games, meaning the framerate doesn't spike down as far as the GF4 does.

If the P drops to below 30 frames, I would not call that "VERY playable framerates". And what do you mean exactly by saying "whereas you don't with the 8500 or the Ti4600."?

It's exactly as it's written. 8500 and GF4 are NOT playable with 16XAA and aniso on, even at 1024 IMHO. Parhelia is VERY playable at that resolution, and I am going by YOUR definition of playable.

I see slightly less numbers with the 4600 in those scenarios with Quake3.

Anyone who has bothered to actually use a GF4 compared back to back with an 8500 and a Parhelia will see that the benchmarks are very decieving. The GF4 gets large pauses and skips when going through portals and large scenes in Q3 with lots of bots. Neither the 8500 nor the Parhelia drop down to the same level of choppiness at those times. IMHO, the Parhelia is the best of the bunch.

Rags, I'm not trying to single you out, but your opinion, so far on this thread, is the most valuable to me since you have all 3 cards and I see you as trying to be unbiased.

I'm a straight shooter, you know this. I am critical of Matrox just as much as the next person, but I criticize what needs attention, and benchmarks are not the tell all that many think they are. Ask as many questions as you wish, I will try my best to answer them.

Also, do you agree the aniso is not as good as the 4600?

In some cases, yes I agree with you. Driver versions make a big difference in quality. Keep this in mind, that's all I am going to say, as you will see soon enough for yourself.

And the most important questions to me are, how does this card play on UT2k3 performance test? It will be interesting to see how P performs with next gen games. Do you think it will surpass the 4600 on next games?

Yes, I have played UT2K3, and yes on all of my cards. There is no comparing. Parhelia is the best all around card IMHO. If this is any indication, Parhelia is the best on the market right now for those games.

What about iamge quality, I know it's a stupid question, but how does everything look in windows?

Awesome. Of course.

Text?

Awesome, of course.

How much better(or worse) does it look than a G400? G450? G550?

It's better than all three. It's the best I have use to date, and you know how picky I am about text and 2D quality. 2D performance is stupendous.

you tried 1600x1200x32 with aniso and FAA on 3 monitors? Is the game still playable? It seems that with the current benchmarks, games will nto be playable wiht 3 monitors? Please educate us BBs if you can.

Parhelia won't do 1600x1200 on three monitors.

Rags

[flaimo]

don't forget that this was only the first of two reviews from tom's mainly focusing on 3D features. in the second one he will compare all the other features like 10bit color, DH/TH, ...

[Indiana]

The Chip review does indeed look Much better. The Parhelia is beating the GF4Ti4600 in all hires AA benchmarks by a large margin except the crappy Codecreatures one.
Shows that I was correct with Toms carefully chosing the test setting to let "his" Nvidia cards shine..

[Betelgeuse]

HardOCP posted a blurb about the benchmarks posted by Tom (I guess) and all they say it's slow and Matrox are morons (to paraphrase.) Anyway, they mention NOTHING about anti-aliasing nor anisotropic filtering scores. Selective journalism at its best!

Maybe someone's bitter about not getting a board for testing?

[Kosh Naranek]

Originally posted by Indiana
The Chip review does indeed look Much better. The Parhelia is beating the GF4Ti4600 in all hires AA benchmarks by a large margin except the crappy Codecreatures one.
Shows that I was correct with Toms carefully chosing the test setting to let "his" Nvidia cards shine..

This is such BS !!

The benchmarks at THG doesn't lie. They can be duplicated.

Stating that they did something to make the GF4 look better is such crap.

If P was 20% better than the GF4 you wouldn't even consider saying something like that.

I'm so tired of all this *blip* bashing.

And of course ... 16xFAA works in all games.

[Greebe]

No shit... The guys that wrote Code Creature is now defunct, you can't even address their bugs, so why attempt to fix it so it does score better. IMHO it would be a great waste of energy and money that should go to optimizing something else that really has merit.

Tom = nVidiot

[Rags]

Originally posted by Kosh Naranek


This is such BS !!

The benchmarks at THG doesn't lie. They can be duplicated.

Stating that they did something to make the GF4 look better is such crap.

If P was 20% better than the GF4 you wouldn't even consider saying something like that.

And of course ... 16xFAA works in all games.

Keep telling yourself that, and maybe you will someday believe it.

Mr. Rivastation dropped the ball on his review of the card, and in true THG fashion, ends up looking the fool.

More reviews are coming around the corner, and remember the drivers are still pretty raw, and there is a LOT of room for improvement.

Rags

[Greebe]

No Kosh, nobody has ever claimed 16xFAA works in all games and they won't.

[Kosh Naranek]

Originally posted by Greebe
No Kosh, nobody has ever claimed 16xFAA works in all games and they won't.

My point exactly !

You all say that P rulez in high res with all eye candy on !

So when 16xFAA isn't used then what ?

Originally posted by Rags
Keep telling yourself that, and maybe you will someday believe it.

Mr. Rivastation dropped the ball on his review of the card, and in true THG fashion, ends up looking the fool.

More reviews are coming around the corner, and remember the drivers are still pretty raw, and there is a LOT of room for improvement.

Rags.

Ok .. what did they do to make the GF shine ?

What bothers me most is OpenGL performance which isn't up to par.

[Kruzin]

Look at the other review that has been posted, and you'll see what they did to make their favorite shine.
They ommited every test that their card lost in.
If that's not biased, I don't know what is.

[Indiana]

Kosh, can't you read or what?
I did never say that Tom is making up the benchmark numbers and I'm sure they can be duplicated when you use the same settings as he does.
However I was suspecting (and I'm quite sure that I'm right there) that he carefully adapts the used tests and the used test-settings to make the Nvidia cards look good.
He does this by just omitting the "bad" results and only picking out the "good" ones for the review.
This is why you never saw a anisotropic filtering comparision between the R8500 and the GF4, this is why he uses the Codecreatures and Aquanox and Comanche benchmarks, this is why he does put less emphasize on 3DMark (with the Radeon being quite good at that nowadays) than he did earlier.

And you can't say anything against the argument that he did try to hide that the Parhelia simply is faster at AA than the GF4 by clumsing the AA graphs with GF4-scores with inferior quality.
The way the chip graphs are made everyone sees the obvious pattern, THG trys to hide this with the above mentioned stealth-tactics.

[Snake007uk]

all the beta testers have said they have all the candy on all the time in every game and they get a smooth game no probs now why THE HELL would u want to turn it OFF ???

can u do that with GF4 have everything on max and even play the game!


Snake

[Kosh Naranek]

Originally posted by Kruzin
Look at the other review that has been posted, and you'll see what they did to make their favorite shine.
They ommited every test that their card lost in.
If that's not biased, I don't know what is.

Yes ...it's biased.

My point in all this is ...


When a review shows selected tests to make P shine .. it's a good review.

When a review shows selected tests where the GF4 shines it's a bad review.

My opinion is ... neither is a bad review, they supplement each other by showing P from it's worst and it's best side thus giving a complete picture of P.

[Rags]

Originally posted by Kosh Naranek


Yes ...it's biased.

My point in all this is ...


When a review shows selected tests to make P shine .. it's a good review.

When a review shows selected tests where the GF4 shines it's a bad review.

My opinion is ... neither is a bad review, they supplement each other by showing P from it's worst and it's best side thus giving a complete picture of P.

No, it's a countering review. Most sites are very lazy in their reviews and simply run a suite of benchmarks on their tests. They fail to realize the unique features and advantages a particular piece of hardware has. Running two or three benchmarks is not enough.

Rags

[xortam]

Benchmarks are only useful if they reflect how you'll be using the system. You're best off running a comparison on your system with your apps driven the way you want to use them.