120+ GeForce FX Reviews Collected

Peter writes "We just finished at 8Dimensional our list of GeForce FX reviews. It tries to show all reviews of these video cards currently online, 120+ are listed at the moment." Hmmm, time to upgrade from an Xpert@Play98 ...

Comments

To bad they will all have to be redone

[Megor1]

3dmark released a patch today that avoids nvidias cheating in their benchmark, so all the reviews that used 3dmark need to rerun their tests.

Re:To bad they will all have to be redone

[Goalie_Ca]

It's true a lot of reviewers have relied on futuremarks benchmark. For me and others it means nothing because its "too" synthetic and uncharacteristic of current games. But for others this means everything but then again, they've probably already know that. It's not everyday your "favorite" gpu drops 24% :P

It's worth to note that ATI also "cheated" but they still correctly rendered the scene. All they had done was re-order the shader instructions so they were optimized for their architecture. It only boost their performance by 3% IIRC.

Re:To bad they will all have to be redone

[domninus.DDR]

how is that cheating? re-writing code to make it better VS lowering precision to make it faster and hope you dont notice! HMMMMMM

Re:To bad they will all have to be redone

[Performer Guy]

ATI did not cheat (although they have cheated in the past), they reordered the instructions in a shader but it remained mathematically and functionally identical. This is what optimizing compilers do all the time. It's called optimization, not cheating, and it is legitimate, they looked at the shader INSTRUCTIONS to see if they were suitable for this optimization. It made about a 2% difference, and IMHO ATI should leave this optimization in and broaden the scope if possible.

If it was a really narrow path optimization then it's borderline, but far different from the wholesale cheating of NVIDIA who: rewrote shaders with completely different results, didn't clear the screen at certain points and added hidden clip planes to eliminate pixel fill. All very underhanded and why you would make excuses for this I don't know. Sure it's a synthetic benchmark, but if it don't matter then don't cheat.

Re:To bad they will all have to be redone

[mESSDan]

If it wasn't cheating, and they were only looking at shader instructions, then why did the results change when they renamed the file from Quake3.exe to quack3.exe? It is also important to note that the Quake3.exe results were of less quality than the quack3.exe results.

Sounds like cheating to me.

Re:To bad they will all have to be redone

[Distinguished Hero]

If it wasn't cheating, and they were only looking at shader instructions, then why did the results change when they renamed the file from Quake3.exe to quack3.exe? It is also important to note that the Quake3.exe results were of less quality than the quack3.exe results.

Um... he was talking about 3dMark, not the Quack issue (btw, there are no shaders in Quake3), which although people insist on constantly bringing, up was most likely not a cheat; ATi not only fixed the 5 problematic textures, but also increased the performance past the level of performance attained using the "cheat," in the next driver release.

Re:To bad they will all have to be redone

[mczak]

no, ATI also cheated, though on a MUCH smaller scale than nvidia. They also exchanged the complete shader (recognizing it by name or something like that), even though the exchanged shader delivered the same result (unlike nvidia, their shaders deliver different results, plus they have those really evil clip plane / buffer clear cheats).
You could argue that replacing complete shaders is optimizing (as long as the shader delivers the same output)and you could be right if this would be a game. However, the 3dmark03 licence clearly says that this is a cheat.
So, no, IMHO ATI shouldn't leave this optimization in there (they have already stated it will be gone in the next driver). If you mean with "broaden the scope of the optimization" that they put in an optimizer which could reorder instructions in a generic way then yes, that would be nice (AFAIK they already have such an on-the-fly optimizer, but it's probably not going to achieve as good results as manual tuning).

Re:To bad they will all have to be redone

[Anonymous Coward]

... because ATI doesn't cheat. Quack III anyone?

Re:To bad they will all have to be redone

[OzRoy]

The thing that doesn't make sense to me is from the couple of reviews I have read the new NVidia card outperforms the Radeon card on almost all the game tests. Not by much, but still a little bit.

And yet when it comes to the 3D Mark test ATI Creams NVidia when that new patch is applied.

To me that makes the 3DMark benchmark Very sus. The two possibilities I can see are NVidia's original complaints against the benchmark are justified, and yes they cheated to boost the score. OR 3DMark are bitter at NVidia and decided to release a patch that deliberatly made NVidia cards perform badly.

Now I doubt it's the second, and maybe there is another possibility that I missed, but in the end I think that the new NV card does perform Extremely well in real world applications no matter what 3DMark says.

Re:To bad they will all have to be redone

[Kaldaien]

Actually... This is easily explainable.

John Carmack released a document YEARS ago explaining to hardware vendors how to optimize their OpenGL implementation for the Quake 3 engine. Such things as Vertex Array Client States (i.e. GL_VERTEX_ARRAY) never changing (always enabled), specifics for multi-texturing, vertex structure size, etc...

Given that anyone who cares to search Google for a minute or two can pull up this document (The original doc is gone :-\ - Still fun reading material though :)), I'd hardly call using it cheating. The performance DOES relate to real-world situations, as the majority of people playing Q3 Engine games don't change the name of their executable :)

What would be nice, however, is if developers had some access to this driver layer. I'm not saying anything as major as Q3's optimizations, but if a developer could tell the drivers that it won't be making any state changes to a particular attrib stack subset (i.e. GL_ALL_ATTRIB_BITS wouldn't actually push the entire stack), it'd increase render-time without any fancy 1st party vendor magic.

Also, using languages such as Cg, shader optimization is possible at run-time. I've never used it myself, I'm biased to hardware specific ASM at the moment.

Re:To bad they will all have to be redone

[Performer Guy]

Well it really gets into the realms of opinion here. I'm not going to debate this, like I say it's borderline and your conclusion depends on your outlook we agree at least that any infraction was minor compared to NVIDIA and that ATI are already tainted by earlier cheats. I'm no fanboy of either, although I love their work.

Re:To bad they will all have to be redone

[Performer Guy]

Oh I agree that earlier example was a cheat, the Quake->Quack IMHO was a cheat although there are apologists, that's what I meant when I said they've been caught cheating in the past. What's being discussed here is a separate issue w.r.t. ATI cheats in Futuremark where a ~2% difference with shaders was uncovered where they detect instructions and 'optimize'. As I said, ATI were caught cheating earlier (IMHO, I'm not interested in an argument, you're entitled to disagree). This is purely about whether they were cheating in optimizing the futuremark shaders by reordering instructions to something that was mathematically and functionally identical. Like I said it's borderline, it's a close call and I understand the position of anyone on either side of the debate. However, anyone would agree that this is not as flagrant as the NVIDIA cheats. I don't really want to take sides here, but we need to call a cheat a cheat to discourage this stuff in future, that's all, and NVIDIA cheated. ATI... well make your mind up, I don't think they really cheated this time, although they did in the Quake->Quake thing IMHO.

Re:To bad they will all have to be redone

[Anonymous Coward]

If they only did it for the benchmark, then it's a cheat. If they did it for the entire class of applications the benchmark is supposed to represent, then it's an optimization.

They only did it for the benchmark. It's a cheat. Simple!

Re:To bad they will all have to be redone

[Performer Guy]

I believe they trigger their optimization on the shader CODE itself, so it is not a cheat. This is a close call because ideally you want those optimizations to be as general purpose as possible. So I'm still on the fence on this one, it's not as clear cut as the NVIDIA cheats or triggering based on the application name as ATI did with Quake->Quack.

And they all prove...

[da3dAlus]

That the GeForce FX sounds like a leaf blower...

Re:And they all prove...

[Anonymous Coward]

CoolingFan_FUN_small.wmv

A wmv file? Are you kidding me?

Could you be more of a pain in the ass!

Re:And they all prove...

[rehabdoll]

dont you just love webmasters that dont know how to configure their webservers?

Re:And they all prove...

[drwtsn32]

Only the GeForce FX 5800, you clod. I have a 5600 FX and it isn't a hair dryer. The 5900 FX is also supposed to be much better than the 5800.

Re:And they all prove...

[Anonymous Coward]

That was the fx5800, which never saw large scale production. The fx5900 was quickly rushed to market to replace it and uses DDR-I instead of the hotter DDR-II.

Re:And they all prove...

[Billly Gates]

Try 55 decibals

You know, slashdot itself would be more useful

[The Terrorists]

to us if it contained such review compendia in itself, rather than making me go to 100 different sites to see them. It'd also be a way to counter the various technology zealotries that arise here. I'm willing to see these ads if the value of this site goes up commensurately.

Re:You know, slashdot itself would be more useful

[cmburns69]

Good idea.. But is it possible to /. slashdot?

Why now?

[Anonymous Coward]

Why suddenly (after 'cheating' and 'dawn' incident)? Is this some kinda PR from NVIDIA?

Here's a good question

[gazbo]

Why did they not include the multitude of Radeon reviews? They are better performing at the top end, and arguably cheaper at the mid/low end. The only eason I can think of is lack of Linux support, but as the Slashdot story the other day pointed out, they are currently putting together a team to work with the Free Software community to create a GPL driver (in a similar architecture to nVidia's unified driver architecture I believe, so as to ensure the GPL driver will support all ATI cards from the Radeon on).

Re:Here's a good question

[Anonymous Coward]

its because nothing controversial has come out from ati recently
Mr.Big

is it really an upgrade?

[User 956]

Hmmm, time to upgrade from an Xpert@Play98

Hmmm.. I don't know if a GeforceFX is actually an *upgrade* from the Xpert@play98. What benchmark are you using?

Re:is it really an upgrade?

[Evil-G]

What benchmark are you using?

The "decibels produced by the video card" benchmark.

Re:is it really an upgrade?

[drwtsn32]

Wolfenstein 3D is a great benchmark.

Re:is it really an upgrade?

[Vivieus]

What about upgrading from my Voodoo3 2000 PCI? Even if not perfect (as was said, why make 120 links?), it's still nice to see such initiatives.

Re:problem

[Anonymous Coward]

America's Army sucks. And the game is pretty bad, too.

Re:GeForce FX is a Good Thing(TM)

[Anonymous Coward]

Sounds like a nice quantitative analysis of the two cards. Just saying that Geforce is "king" at precise colors but giving absolutely no evidence to back that up truly means nothing.

uhhh

[Tweakmeister]

"We just finished at 8Dimensional our list of GeForce FX reviews." ...and? now what?

While the idea of a site that shows all the reviews in one place is noble...unless you have huge amounts on content it's easier to search google for the reviews. The good ones usually end up being on top as well.

Re:uhhh

[Anonymous Coward]

Are the top google links there because they are good? Or because everyone just clicks/links to the top links?

Hmmmm...

Re:uhhh

[Anonymous Coward]

Ever tried with Google..?

For card X + company Y + review ...Google gives you 1000+ results...finding those 5 that are reviews takes about 1-2 hours..

Some have the time, some don't...

8Dimensional was created for the later ones...:)

Peter
http://www.8dimensional.com

Re:uhhh

[Tweakmeister]

Quite the contrary, the Google sites typically offer reviews from the big-name sites like Anandtech, etc. Your page only lists links...many of which are to sites that are sub-par quality reviews. Adding a little description or more info would save time.

Re:Once upon a time in a park...

[Anonymous Coward]

You're telling it like it is, brother!

mark on a bench

[Anonymous Coward]

My benchmark is a long bench I put beside my computer. I make a mark on it for how far away I have to go before I can't hear my Xpert@play98. 0 meters; it has passive cooling. I make another mark on it that shows how far away I have to go before I can't hear the GeforceFX I am considering. 19 meters; it's a long bench.

The GeforceFX has a higher score on my benchmark by 19 meters. Thus it truly is an upgrade for me.

Re:I'm calling for your head on a platter, Timothy

[Anonymous Coward]

Why, this is hardly offtopic... given that it's a Timothy article and all...

Re:hey

[Anonymous Coward]

silly troll... the politically correct term is "geek"

Re:i was partying all weekend....

[Anonymous Coward]

sorry to correct you again, sir, but you're confusing "huge horse cock" with "pathetic little weener" and "ripping your body into two" with "being laughed at by the girls".

[snicker]

[A_Non_Moose]

/obligatory sniping

quote: "...120+ are listed at the moment".

So, production quotas went better than expected, huh? /end sniping

I have to admit, I knew the horse-power game was one that Nvidia was going to trip over eventually when the Radeons were *really* pushing the Z-axis occlusion.

Face it, Nvidia became top dog by pure horse-power, and to some extent, deeper color depths and kick ass drivers.

ATI finally got their hardware on track not by being able to "outshow" Nvidia's muscle, but by outclassing them by being smarter with the bandwidth and 'getting a clue' with drivers.

(witness the Dawn demo on a 9800 running *faster* than on the FX series.)

Speaking of the Dawn demo, does anyone else remember that this was to showcase the power of the FX, yet the 5200 is a *SLIDESHOW*?

Final thought: Did they use 3Dmark in all the benchmarks? (/low blow, sorry)

.

Re:[snicker]

[Distinguished Hero]

I have to admit, I knew the horse-power game was one that Nvidia was going to trip over eventually when the Radeons were *really* pushing the Z-axis occlusion.

That is faulty logic; just because ATi was using Z-axis occlusion and ATi beat nVidia (which is also using z-occlusion nowadays) does not mean that that is why they beat nVidia. The Kyro II used tile based deferred rendering which was far superior to the z-occlusion ATi and nVidia were and are still doing, and look where PowerVR is today.

ATI finally got their hardware on track not by being able to "outshow" Nvidia's muscle, but by outclassing them by being smarter with the bandwidth and 'getting a clue' with drivers.

If being "smarter with the bandwidth" was all that was required, we'd all be using Kyro IIs. Making a successful GPU is not so black and white...

Re:[snicker]

[A_Non_Moose]

True, very true...I was being overly broad, but the simplistic/broad view of the Ati/Nvidia battle is:

Nvidia used brute force and overtook 3dfx (barring 3dfx's stumbles and fall).

Ati tried to "brute force" the Radeon series and could not keep up with Nvidia, until, they (Ati) got smart with the hardware and software optimizations (i.e. finesse).

The point is that Ati overtook Nvidia because of points we both brought up:
Smarter with bandwidth, not overdrawing a scene, texture compression and a whole host of other things that, I suppose amounts to "working smarter, not harder".

You are correct, even if it was a slight nitpick.

Thanks, tho.

Re:[snicker]

[Osty]

Nvidia used brute force and overtook 3dfx (barring 3dfx's stumbles and fall).

Not exactly. 3dfx (3Dfx, before they changed their name, bought STB, and went crazy, wasn't that bad at all) fucked themselves. If anything, 3dfx was guilty of brute-forcing. The Voodoo2 was not much more than a brute-force step from the Voodoo1. The Voodoo3 wasn't much better, with the addition of fairly poor 2D. "22bpp", blah. Gamers wanted 32bpp, nVidia gave them 32bpp. 3dfx had their eyes firmly closed. And let's not even get into the Voodoo4/5 fiasco (nVidia almost pulled a 3dfx with the FX, which isn't very surprising given that nVidia bought out 3dfx). No, 3dfx screwed themselves. nVidia was just in the right place at the right time to take advantage (and they almost weren't there, but Sega bailed them out during their rough times, offering them the contract to make the Saturn video chipset).

Ati tried to "brute force" the Radeon series and could not keep up with Nvidia, until, they (Ati) got smart with the hardware and software optimizations (i.e. finesse).

Except that ATI is still struggling with building rock-solid drivers like nVidia. Yes, they've gotten better (to the "good enough" point), but I'll still choose nVidia over ATI even if ATI gets a few more 3DMarks simply because I know I can trust nVidia to build rock-solid drivers for the life of my card and beyond (hell, nVidia was able to squeeze a good amount of speed out of the old TNT2 with a driver update during the late-GeForce2/early-GeForce3 time frame).

Re:out of touch

[Anonymous Coward]

amen to you! i realized that i had wasted many hours of my precious life rotting away at a computer while i could have been having sex, making friends, and enjoying some good ol' fresh air. it makes me shiver to think of those gaming days (fallout 2, AoE, etc)

time?

[buddha42]

Hmmm, time to upgrade from an Xpert@Play98 ...

Nope. The most stressful video application most users do is DVD playback, and even that is loooooong past the point where hardware-assistance is needed.

The video card market has gotten absolutly rediculous in the last 2 years. Its strange, when Intel and AMD fight it out, prices plummet. While nVidia and ATI have been fighting it out prices have skyrocketed. Sure, so have features, but they're so random and game or api dependant that most people don't even know how to turn them on in different games or in the drivers advanced settings.

Up until UT2K3 it was completely absurd, because anything with DDR could play any game just fine. Now with the new crop its even worse because modern cards still can't play the very-new and upcoming games well. So buying a high-end card now is overkill for older games, and underpowered for upcoming games.

Re:time?

[Anonymous Coward]

While nVidia and ATI have been fighting it out prices have skyrocketed.

Huh? What the heck are you talking about? I just picked up a GeForce4 MX card for $40 and before that I picked up an ATI Radeon 7000 for under $40. Prices seem plenty cheap to me. The only people being bilked are the ones who are willing to pay $500 for a stupid video card to get another 20 fps. Every single "modern" game I've tried works just fine on my "old" GeForce 2 in my gaming machine. I don't know what the heck games people are playing to get such crappy performance where you could justify such an expensive card.

Re:time?

[Pulzar]

Try Ut2003, new Wolf demo, Splinter Cell.. Well, any relatively new FPS game, in any resolution over 1024x768, with any kind of image-improving feature (AA, AF).

Do you honestly believe that a million people bought R9700 in the first few months after it came out just to get "extra 20fps", or to win a "pissing contest" (another slashdot favourite)?

Re:time?

[Anonymous Coward]

So are you saying that we should stop developing software that pushes the limits of hardware, which creates a greater demand for developing new technologies?

I'm in!!

Re:time?

[BrookHarty]

Actually, I had a GF3ti500, and U2K3 and BF1942 really started to drag on the FPS at 1024x768+ Res. And then if I turned on AA/AF it dropped to sub 20's. Time to upgrade.
Picked up a ATI 9700Pro, and it has been about 9 months of great use for about another year before I need to upgrade.

BTW, to see if your system is worth an upgrade, check out future marks online browser, you can search for a CPU near yours and then check to see if a faster video card will improve your FPS. Good reference.

Re:time?

[Anonymous Coward]

Could you please give me the link to that site? It's probably obvious...but, I am a noob.

Thanks

Re:time?

[OzRoy]

You CANNOT compare the CPU market (AMD and Intel) with the GPU market (ATI and NVidia).

How often do CPU manufacturers release completely new chip designs? Once every couple years maybe. I know it has certainly been a few years since the Athlon was released, and it has been a while since the P4 was released, and there is no sign of the next CPU from Intel yet.

Now think about how often new GPU's are released. Once a year at the most. The Radeon 9700 card hasn't even been out for a year, and ATI are already bringing out the Radeon 9800 which has a new core. It's only been a couple months for NVidia before they decided to scrap the NV30 and bring out the NV35.

Re:time?

[Anonymous Coward]

www.futuremark.com

I was using 3dMark 2001, but 2003 is out. Download the demo (if they still offer it) and sign up with the profile manager.

Then you can do 2 neat things, Compare by CPU and Compare by GFX Card. So if can see what a faster CPU does with the same GFX card, or the same cpu with Faster GFX cards. Really ace.

Re:time?

[Jeremy Erwin]

IIRC, the Radeon 7000 doesn't support shaders and the GeForce4MX only supports vertex shaders to a limited extent. Neither card supports pixel shaders. Games that take advantage of these (admittedly esoteric) features won't look as nice, or will run more slowly on these cards.

The GeForceFX and Radeon 9700 support the latest incarnations of DirectX 9, and therefore appeal to rich Windoze gamers.

BTW, If you want an example of why the Radeon 7000 and GeForce4MX are considered obsolete by some, check out this table of results from Tom's Hardware.

Personally, I still use a Rage128 chipset (I have an iBook). T&L would be nice for Data Explorer, though.

Re:time?

[evilviper]

While nVidia and ATI have been fighting it out prices have skyrocketed.

Just a while ago, I was looking for an Nvidia card with TV-out (nothing else is even likely to work under Linux/FreeBSD). Searched pricewatch and found one for $20... Do you really need a videocard to be creaper?

There is PLENTY of blame to go around for videocard prices:

Stores try to only stock the most expensive items, because that means higher margins. You don't walk into Best Buy/Circuit City and see SIS videocards, because they make them very inexpensive, which means less profit for the stores.

Game developers almost seem like they are getting funding from ATI/Nvidia... It seems that they needlessly require incredibly high-end videocards. Come on, do you really need 5 billion Frames Per Second to play a first-person shooter? They also seem to intentionally deny gamers the ability to operate at lower quiality modes, as if they want to force people to upgrade their systems... Which seems counter productive to me. No wonder consoles kill PCs when it comes to gaming.

ATI and Nvidia can get away with raising prices because everyone along the lines (except consumers) wants the prices as high as possible, and consumers have just gotten used to bending over. If major stores stocked SIS videocards, equivalent to their ATI/Nvidia counterparts, prices would drop dramatically. If videogame makers stopped forcing number-crunching monsters upon the public, most people wouldn't need a faster videocard, and would couldn't tell the difference between their $10 videocard, and a top-of-the-line card.

Re:time?

[Magila]

Oh please, stores tend to stock the more expensive cards because the kind of person that wants a new videocard is going to want something fast. Guess what? fast videocards are expensive! There's no retail market for slow cards, people who don't care about how fast thier card is are happy with the one that came with their OEM system. There's the people building a new system from parts, but they tend to want a fast videocard too and they almost always buy all their parts online anyways.

As for the great game dev/chip maker conspiracy, you may think game devs "needlessly" require high-end cards, but you couldn't be farther from the truth. One of the banes of people who buy high end cards is that games don't take full advantage of their shiny new toy. Beyond that, almost all games can be set to a "fugly mode" that will run on all but the most obsolete system. The thing is the fugly mode is just that, fugly. Gamers want their games to look good, they expect this years games to look better than last's. Better graphics require, you guessed it, a faster videocard! Nobody's forcing people to buy new cards, you want to stick with 320x200 sprite based games then so right ahaead. Just don't expect the rest of the world to feel the same as you do.

Finally, that consoles will/are killing PC gaming line has been thrown around since the SNES. It hasn't/isn't, PC gaming isn't going anywhere. You don't like it, stick with consoles.

Re:time?

[fr0dicus]

However you could have saved yourself a stack of cash and got something like a GF4 4200, which plays UT2k3 perfectly well, and will be just as inadequate as the 9700pro when Doom 3 and Half Life 2 comes along.

Re:time?

[Anonymous Coward]

Actually, i wanted to enable 4x AA and 8x AF, which the 4200 couldnt push at 1280x1024, and the ATI 9700 could. In fact, I played around with a gf4-4600 for the same price as the 9700, took the 9700 back.

While I miss nvidia's driver options, running most games at 1280x1024 4xAA/8XAF is worth the tradeup. In fact most lan parties I goto people are using ATI cards. (Well most)

The only bitch, seems like every 2 years, I spend 400 bux on a video card.

Re:time?

[fr0dicus]

AA is pointless at higher resolutions IMO, but AF is nice admittedly, however neither is really worth the 300% premium that the top cards currently carry. The graphics market needs someone like AMD to shake it up with some decent low cost solutions. I suppose card manufacturers can get away with it because the differences are much more obvious to the eye than things like SSE or 3DNow!

Re:i was partying all weekend....

[Anonymous Coward]

well, anyway. i think i am going to go play outside now. have fun typing on your keyboard to other geeks across the globe.

Re:i was partying all weekend....

[Anonymous Coward]

i think you may be confusing "playing outside" with "doing your job as a garbageman". have fun anyway.

Prevision

[Anonymous Coward]

ATI can't compete with nVidia on prevision.

Prevision - A technique that allows one to see the scene before it is rendered.

Wow, I love technology!

Re:hey

[Anonymous Coward]

THEY SHALL FACE THE TRUTH!

120 reviews...

[Capt'n Hector]

Are you guys telling me that every single person with a geforce fx wrote an online review?

Re:120 reviews...

[gearheadsmp]

No, just the people who couldn't afford a leaf blower after shelling out for their 5800FX.

ati

[labratuk]

Hmmm, time to upgrade from an Xpert@Play98 ...

Whoah, really? Probably time for me to upgrade to an Xpert@Play98...

Re:problem

[Anonymous Coward]

It's the most realistic shooter out there if thats what you mean by sucks...

I mean if i wanted to run around with an imaginary plasma rifle jumping 10 feet in the air and getting magic healthpacks or other such nonesense then i would...hmmm, i would never want to do that...oh well...

Re:out of touch

[Anonymous Coward]

you're still not outside? now leave already, mr. small-weener.

Re:i was partying all weekend....

[Anonymous Coward]

i will and thanks. flame with ya later

did you notice...

[Anonymous Coward]

that in several of the tests, they severely overclocked the fx, which resulted in no performance increase? that shows that it is simply the crappy drivers holding it back....

Mr.Big

Sure ATI cards are great and wonderful

[Anonymous Coward]

but their drivers have a history of being crap. I have never encountered a single game that wouldn't run on my Geforce 4 4600. My girlfriend owns a 128 mb ATI card and I can think of over 10 games that wouldn't play, strictly due to the video card.

Benchmarks are meaningless if you can't play games with the thing

Re:problem

[Anonymous Coward]

If you want to "run around" with a realistic rifle shooting realistic bullets and realistically killing people, why don't you join the Army?

If you want reality, it's right there, outside your door.

120+ reviews?

[earthforce_1]

Nice piece of research, but I don't think I have time to read 120 reviews on anything, even the next vehicle I plan to buy. Can they just put up an executive summary?

Re:120+ reviews?

[Ko5mo]

You think 120 reviews is bad. Just imagine how many screen shots/benchmark-bar-graphs they have combined...

Re:120+ reviews?

[Anonymous Coward]

Sure you would read just about the one you consider to buy...no one reads all 120...;)

Peter
http://www.8dimensional.com

120 reviews in one place...

[nickgrieve]

But why... give it a month and they will be out of date... and for fucks sake... hardware is _cheap_ just but the thing and be done with it.

Re:120 reviews in one place...

[Anonymous Coward]

$300 is cheap? Can I have your job? Please?

Re:120 reviews in one place...

[Ko5mo]

When you go by time = money, by the time you read all 120 reviews, you would have made more than $300, thus the poster is saying just buy the damn thing and be done with it.

Of course it helps if you have a job that pays good.

Sorry.

[MooKore (675835)]

Im a linux user. My cheap Rage 128 is enough for tuxracer.

Re:Sorry.

[LittleLebowskiUrbanA]

I'm a Linux user. My cheap geForce ti4200 is enough for Unreal Tournament 2003.
Just because you run Linux doesn't mean you have to give up gaming. It usually means that you have alot of heartache :)

No...

[dark-br]

... I bet some wrote 2 or more :P

Xpert@Play98

[tunabomber]

I'm using an Xpert@Play98 and I love it. Yeah, it doesn't support OpenGL very well, but look how fast it can display B's in a /. post:

BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BB BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB BBBBBBBBBBBBBBBBBBBBBB

Wow, that's some card!

Re:Xpert@Play98

[Anonymous Coward]

yuo are two skillz for slahsdot!

Re:Xpert@Play98

[Pingular]

Hmmm, time to upgrade from an Xpert@Play98 ... (for those of you that've seen the hl2 demo) Hmm.. maybe I'll have to upgrade my 486 afterall :/

and bootie!

[Anonymous Coward]

AND BOOTIE!

Re:hey

[gearheadsmp]

You forgot to include 'goose' in that statement.

Re:hey

[Anonymous Coward]

sorry, but i do not understand why goose is missing?

uncharacteristic of current games...

[TrekkieGod]

Heck yes, it is.

You don't need to buy the top of the line card to play the current games well. So you want a benchmark that'll tell you how your card will deal with the features that are new to the card and will make their way down to the games later.

If you're buying a top of the line card, either you just want the status of having a top of the line card, in which case you want it to pass all benchmarks with flying colors, or you want a card that you won't be replacing for a while...in which case you want the benchmark to show that it can handle a whole bunch of new features future games might make use of.

Re:out of touch

[Anonymous Coward]

You mean you aren't capable of programming, reading, playing outside, AND playing a game now and then? Yeesh, wankers like you are frightening.

Fastest card in existance?

[rmdyer]

Everyone knit picks about a few percentage points here and there when comparing Nvidia and ATI cards. But, for people with money to spend, there are more expensive CAD Pro level cards out there. I am wondering, what the fastest card you can purchase is for the PC AGP bus? Anybody know? And, how much faster are they than the FX or Radeon?

Another annoying thing... Looks like Nvidia and ATI are now price tiering cards. Up until recently, the most you would pay retail for the best consumer level card was around $400. Now it looks like Nvidia and ATI want to push us into the $500 card level. What is next year card $600?

Re:Fastest card in existance?

[gm-7]

The Radion 9800 Pro... For game performance that is..

Re:Fastest card in existance?

[gm-7]

Radeon with an E!

Re:Fastest card in existance?

[Anonymous Coward]

If you want something that'll raise your 3dmarks a little higher, go for ATI. But for actual games, it's all Nvidia.

Re:Fastest card in existance?

[Anonymous Coward]

The CAD targetted cards aren't very good game performers. They're far more concerned with accuracy.

Re:Fastest card in existance?

[paradesign]

try at 3d labs theyre wild cat series is amazing for high end apps that take advantage of it. i have many friends that use fire gls and quadros, and i hear that they are not the best gaming cards, but they destroy gaming cards where they need to, in DCC apps and other specialized apps. features like hardware overlay planes and line anti aliasing arent needed by gamers, but developers would surely cry without them.

i believe there are some sites that use gaming benchmarks to review these cards, try highend3d.com, they usually are a good place to start looking into it. if not, google is your friend.

Re:Fastest card in existance?

[rmdyer]

Seems a lot like the ATA vs. SCSI debates. You've got this class of people who must buy whatever the "best" cards are, and are recognized to be the "best" cards for doing certain things by their peers. If their peers are gamers, then you'd probably buy and FX, or Radeon. If their peers are engineers, then you wouldn't be caught dead not spending thousands of dollars for a Quadro or something. Interesting that the Quadros would be slower at rendering.

The only thing you seem to be getting out of 3D labs and Quadro level cards are industry certifications for high level CAD/CAM software. This seems to be a bit of an click group where'in they are self supporting. You buy in by purchasing a high cost card, then you are in the click, so to speak.

But don't the gamer level cards also perform for CAD/CAM? Is this an ATA vs. SCSI type of debate?

+1 penny.

Re:Fastest card in existance?

[Anonymous Coward]

Definately the Bitboys Desktop 3D Processor. It gets 20,000,000,000 3DMarks. As a matter of fact, The Phantom is supposed to be shipping with that chipset... XBox... Gamecube... Prepare to die.

Re:Fastest card in existance?

[murgee]

If memory serves, the difference is the Quadros and FireGL (etc.) cards are better for CAD/CAM because they're more exact, whereas your Radeon and GeForce series chips are engineered to be fast. You don't care too much if a wall in QuakeIII is misrendered for a second, but it'd be a big problem if you were rendering something like an office building or such. There was an article on Slashdot semi-recently that covered the differences, in fact. Gist is, you might be able to render things fine on a GeForceFX , but if you're doing it for your livelihood, then it's a Nice Thing to know that the rendered result is accurate, and that's what the workstation cards try to provide.

Re:Fastest card in existance?

[gm-7]

I hear that 3dmarks are pretty high on the Nvidia cards ;P at least if you dont patch!

Re:Fastest card in existance?

[Luminair]

No. The add-in video card market is driven by Nvidia and ATI, Gamer card, CAD card, or otherwise.

3dlabs is barely holding on to their tiny niche. While CAD cards need to do different things than gamer cards, this is basically how it works: Nvidia/ATI build the best cards they can. Then they tweak them and triple the price for the CAD market.

He means...

[Anonymous Coward]

it has skyrocketted for people who don't want a piece of shit.

Re:hey

[gearheadsmp]

You silly goose! ;)

Re:problem

[Anonymous Coward]

Nah, I don't have anything against the people America tends to invade and conquer.

Besides actually dying is pretty annoying.

mod parent down

[Magila]

please, this dead horse gets wheeled out every time a story that has to do with graphics cards gets posted. It been rebutted a hundred times before, I'm not going to bother posting another one.

Re:mod parent down

[Kircle]

this dead horse... has been rebutted a hundred times before, I'm not going to bother posting another one.

Then a link perhaps?

I wish nvidia

[Anonymous Coward]

would change the way they name their cards. You nearly have to be one of their employees to tell one card from another. It was so confusing, I bought an ATI instead.

RTFA -- the 5200 doesn't even have a fan

[Morgaine]

And both the 5200 Ultra and 5600 Ultra have much smaller fans than the attrocious 5800 Ultra that gave the FX series such a bad start, and they don't take up the ridiculous 2 slots that the 5800 Ultra did either, nor cost such a fortune as they've dropped the DDRII memory.

But going back to your inappropriate comment ... the non-Ultra 5200 doesn't sounds like a leafblower, since it doesn't sound at all. It's a great card for the high-volume, medium performance but still leading edge, DX9 market.

Re:RTFA -- the 5200 doesn't even have a fan

[Dr. Spork]

What? the non-Ultra 5200 is much slower than the ancient Radeon 8500 (and more expensive). There isn't a single DX9 game now which would be playably-fast on it, and they'll only get more intense in the future, so the DirectX9 compatibility is a red herring.

SmellyHardware

[czion3]

That would be great if they had a site that compiled all the reviews for a product on the net and gave it a present score. Like http://www.rottentomatoes.com I would call it Smelly Harware.

120+ reviews... in other news...

[Dr. Spork]

Nvidia recently announced they have shipped their 130th FX 5900 card.

Liquid cooling

[DigiShaman]

I've never been a die-hard PC/case modder myself. But, I think now would be a good time to research liquid cooling. I know Swiftech makes all sorts of watercooling acessories. You simply take all the cooling block and connect them in serial and eventually back over to the radiator. I'm sure it would be alot less noisey too. Of course, I would only have a reason to do this if I had an FX card and someone made a water block for that card too. Check out some of the stuff here. http://www.pccasegear.com/category45_1.htm

This is news?

[Xtro]

Who cares if someone found 120 reviews to link to, how is this helpful to anyone? Will anyone now go and read all 120? Go use Google, you'll get reviews of ANYTHING, and the best ones will probably rise to the top.

Now if they had really 'collected' them they could have perhaps summarised all the conclusions into one short conclusion to give an overall 'world' view or something.

This page could have been written by a dumb search robot just as easily as a human, where's the human value in it! Where's the humanity !?!?!? This is how the Matrix started you know.

Re:This is news?

[Anonymous Coward]

Lmao...for kids that have no idea what to buy we have a forum, that they don't need to cry alone out there in the void of the Matix...we also have moderator full of humanity to hug our younger readers...:) ...but most know which card they want, just want to check 4-6 reviews, saves them hours of Google work...:)

I play so much here

[AvengerXP]

My video card on this Dell really should be called an ATI Play@Work 98 hahaha just kidding i hope no one is monitoring my web browsing activity from the home office :)

FX 5200

[Vidiot3k]

Here's my review: It's better than my onboard geforce2. It's only 100 bones. It looks SO realistic when I am beating helpless pedestrians to death in GTA:Vice City. thank you...

Re:problem

[oingoboingo]

It's the most realistic shooter out there if thats what you mean by sucks...

More realistic than Operation Flashpoint?

Re:hey

[Anonymous Coward]

oohh, i get it now. thx

The Differing Economics: CPUs vs. Video Cards

[Nova Express]

The video card market has gotten absolutly rediculous in the last 2 years. Its strange, when Intel and AMD fight it out, prices plummet. While nVidia and ATI have been fighting it out prices have skyrocketed.

There are many quite logical reasons high-end video card prices rise, having to do with the differing economics of the businesses involved:

1. First of all, Intel and AMD own their own fabs. A modern, 300mm, .13 micron chip fabrication plant costs in the neighborhood of 2.5 to 4 billion dollars. To make back such a staggering initial investment, CPU manufacturers must achieve remarkable economies of scaleby cranking out as many chips as possible.
   
2. By contrast, both ATI and Nvidia are fabless chip manufacturers; the contract out the actual chip production to foundries like TMSC and UMC in Taiwan. Because they don't own the fabs, they have to pay higher prices so the foundries can make their profit.
   
3. Most of the speed increases in new CPUs tend to come from die shinks (i.e., moving from .18 micron to .13 micron); serious design revisions only happen every few years. By contrast, the upgrade cycle on video chips seems to be every six months.
   
4. In a sense, offloading graphics processing into a separate has given CPU manufacturers one less thing to worry about, allowing space on a chip that would be used for graphics processing to be used for something else, and also freeing them from having to throw resources at designing chips for the latest and greatest games.

So, as usual, it comes down to economics. With higher fab costs and faster product revision cycles, high end graphic chips increase in price compared to CPUs because the economies of scale and structural costs are different. Still, capitalist competition and the wonders of the free market means that last year's $350 high end graphics card can frequently be found for $99 today...

Patches?

[nacs]

Patches? Bah. The true test of video cards is glxgears.

So here now is my benchmark results for the Geforce 2 MX 64MB on a P4 @ 2.4Ghz with Gentoo 1.4rc4:

bash-2.05b$ glxgears
3961 frames in 5.0 seconds = 792.200 FPS
3971 frames in 5.0 seconds = 794.200 FPS
4028 frames in 5.0 seconds = 805.600 FPS
4386 frames in 5.0 seconds = 877.200 FPS

Amazing isn't it?

The one thing they can all agree on:

[DeathPenguin]

Closed source drivers or not, nVidia's Linux performance rocks.

Of course, that's only among those that actually reviewed Linux support...

A short list of errors made in the reviews...

[Xspringe1]

(ripped from Dave H @ beyond3d - http://www.beyond3d.com)

1. Anand posted benches that claimed the 5900U was platform-limited (at 223fps!) running Q3 at 1600x1200 with 4xAA and 8xAF.

2. Lars at Tom's mislabeled the D3 Medium Quality + 4xAA benches as High Quality + 4xAA

3. Kyle and Anand both ran D3 in Medium Quality with 8xAF set in the drivers, despite what seems to be the fact that Nvidia drivers interpret Medium Quality as forcing no AF, while ATI drivers do not

5. ExtremeTech's 3dMark03 build 320 vs. build 330 benchmarks show the Radeon 9800P losing performance in all four game tests with the new build; benchmarks by forum members and Wavey himself demonstrate that ET got it wrong

nvidia quadroFX kick ass...

[Anonymous Coward]

For the CAD pro-level stuff, nvidia quadro's show the highest scores....

http://www.specbench.org/gpc/opc.data/vp71/summa ry .html

(oh yeah, there's an ati firegl and a 3d labs wildcat in the mix, but they fall pretty far back, and the 3d labs system was $5K deblumes)

How is Linux support for these cards?

[Thag]

I bought an ASUS V9520TD FX 5200 card last week and it wouldn't even install on Windows 98...:(

Back it goes today!

Jon Acheson