How Far We Have Come

FrgMstr

FrgMstr

Just Plain Mean
Staff member
2FA
Joined
May 18, 1997
Messages
58,590
Please feel free to discuss our latest editorial here.


How Far We Have Come: The world surrounding the PC video card industry has changed greatly in the last decade. Most of it for the better...

Now that Red and Green have learned not to cheat as badly as they used to, does the consumer see it the same way? The fact of the matter is that now the gamers and hardware enthusiasts expect to see game specific optimizations for the "big" games. You want better performance for your hard-earned money and you deserve it. Along with these optimizations, we have to realize that no two images will ever be the same now.
Click to expand...
 
Sweet. Another article for me to read on my day off!
Awesome.

Very good article.
I agree witht the statement that games are not as FPS centered as they use to be.
Now-adays it is about lighting and image quality which I think is a great step in the correct direction.
 
Graphics are getting better.
Games are getting less and less fun.

Of all the problems I have with (PC) games, graphics is the least of my worries.

I suppose you cannot fault ATi or NV for the poor state of fun in games but in general people need to shift from the ooohs and aaaaahhs of graphics to adding style, substance and simplicity to games. If I were in the red or green camps, I would be squarely focused on lowing the power requirements, heat generation, and overall production costs. Most mid range GFX cards are powerful enough to push many fps on the 'average' LCD with some features turned on. Does having 8xaa and 16xAF make for a more immersive experience than 4xaa and 8xaf? I just bought a 7800GTX and I can say that I even though I can crank up the detail, CS:S is no more fun that it was with my 9800Pro. I hate to say it...but unless the quality of titles shoots through the roof...I will be a console gamer from now on. No amount of physics ot graphics enhancements will cause me to think otherwise.

XBX 360 Here I come
PS3 I wait with open arms
Revolution....if you cost $49 i will buy you.

SLI, AGEIA, ....not a chance!

Out!
 
Mysogonist said:
Graphics are getting better.
Games are getting less and less fun.

Of all the problems I have with (PC) games, graphics is the least of my worries.

I suppose you cannot fault ATi or NV for the poor state of fun in games but in general people need to shift from the ooohs and aaaaahhs of graphics to adding style, substance and simplicity to games. If I were in the red or green camps, I would be squarely focused on lowing the power requirements, heat generation, and overall production costs. Most mid range GFX cards are powerful enough to push many fps on the 'average' LCD with some features turned on. Does having 8xaa and 16xAF make for a more immersive experience than 4xaa and 8xaf? I just bought a 7800GTX and I can say that I even though I can crank up the detail, CS:S is no more fun that it was with my 9800Pro. I hate to say it...but unless the quality of titles shoots through the roof...I will be a console gamer from now on. No amount of physics ot graphics enhancements will cause me to think otherwise.

XBX 360 Here I come
PS3 I wait with open arms
Revolution....if you cost $49 i will buy you.

SLI, AGEIA, ....not a chance!

Out!
Click to expand...
You'll change you tune once something new and shiny comes out for PC that those next gen consoles can't do. I promise you that.
 
You know, this article actually makes me feel guilty for critisizing the [H] reviews for not using 3DMarks or solid fps versus fps @ this resolution comparions. :( Oh well, thanks for the write-up...your review style makes alot more sense now and is a little more appriciated....at least by this guy. :p
 
I think that as both Red and Green make better/more powerful cards, the lifetime of these cards seems to last longer and longer. I think back I feel like I went from a Diamond Monster, Diamond Monster II, Riva TNT, Riva TNT2 Ultra, GeForce 1, GeForce 2, Radeon 9700 Pro seemingly very vast. When I got to the 9700 pro, it lasted nearly two years and I still wasn't completely sure I needed to upgrade.

I guess the point here is that it is taking longer and longer for games to take full advantage of what power is out there now. Both Red and Green have the time to make their cards more complete rather than just "faster". I don't mind spending $500 when I know that the card is going to last me 24-36 months.
 
Great article man, really showing off what a good writer you are. It is REALLY nice to hear something positive and reassuring about the graphics world in general. All we get in the forums is "green is lame! no red is lame!" Not many people have stepped back and done a reflective article like this, and I commend you on it.

I think im gonna go play some video games instead of going to school now :D
 
I like the idea of custom gameplay evaluations, but the potential is there for abuse, at least by some sites. If you are using custom, non-released run-throughs, there is no way for others to check your work.

For example, Hardware Analysis published benchmarks of a card they claimed was the R520 days before it released.

Later, when the card was released, those benchmarks looked to be misleading; in some cases, they were 40% too low. It was impossible to prove the benchmarks were in error, though, because they were custom time demos or benchmarks that no one could reproduce.

Custom evaluations have more value, but they also have the potential to be abused by a reviewer with an agenda. That's why it's good to have a mix of both kinds of review methods, and to read lots of sites. Keeps everyone honest.
 
I hope that we can work towards reducing the amount of raw processing power and RAM bandwidth required to deliver high IQ and solid performance. I don't how much farther they can go by simply adding more pipes/clocks/RAM speed. AMD has had great success with reducing or at least holding the line on thermals/RAM bandwidth requirements durign the transition from K7 to K8, and Intel is looking to achieve similar goals with their 2006-7 lineup. ATI's memory controller flexibility on R5xx boards, and the gains that it has allowed under IQ levels in OGL titles is a promising step towards this, IMO. The 6600GT being competitive with (often besting) Radeon 9800Pro despite the handicap of a 128-bit bus is also a similar exploration in this direction.
 
Jonsey said:
I like the idea of custom gameplay evaluations, but the potential is there for abuse, at least by some sites. If you are using custom, non-released run-throughs, there is no way for others to check your work.

For example, Hardware Analysis published benchmarks of a card they claimed was the R520 days before it released.

Later, when the card was released, those benchmarks looked to be misleading; in some cases, they were 40% too low. It was impossible to prove the benchmarks were in error, though, because they were custom time demos or benchmarks that no one could reproduce.

Custom evaluations have more value, but they also have the potential to be abused by a reviewer with an agenda. That's why it's good to have a mix of both kinds of review methods, and to read lots of sites. Keeps everyone honest.
Click to expand...

Just to note, HardOCP does NOT use "Custom evaluations" and quite frankly using something that is not released and "custom" goes against our entire direction with gameplay evaluation.
 
I was glad for the change in the way of producing vid card reviews when you made the change. Sure, it took me a few reviews before I completely understood the presentation of the graphs and being able to read them properly, but it was well worth it. I have found value in the reviews even though I am still running a Radeon 8500. I'll have money to upgrade one of these days.

The main thing I would like to know is if game developers have been of any help or are going to help in regards to "a new way of benchmarking". I remember in a previous article that you were hoping they would develop tools included with the games that would help with the benchmarking. Considering the way you do the reviews, I'm not sure if there is anything they could do that would help. I see the main problem with the current reviewing methodology is the number of man hours required to produce the results. Have you seen any indication that anything is being developed that would help with this?

 
Kevl4R said:
I think that as both Red and Green make better/more powerful cards, the lifetime of these cards seems to last longer and longer. I think back I feel like I went from a Diamond Monster, Diamond Monster II, Riva TNT, Riva TNT2 Ultra, GeForce 1, GeForce 2, Radeon 9700 Pro seemingly very vast. When I got to the 9700 pro, it lasted nearly two years and I still wasn't completely sure I needed to upgrade.

I guess the point here is that it is taking longer and longer for games to take full advantage of what power is out there now. Both Red and Green have the time to make their cards more complete rather than just "faster". I don't mind spending $500 when I know that the card is going to last me 24-36 months.
Click to expand...

Although i agree that the cards seems to have a little more life to them, spending $500-600 on a video card is still crazy to me. I remember when i thought $300 was a good amount to spend on a card. I guess i don't mind buying a card for that price but it damn well had better give me 2-3 years of real solid performace in games.

my real problem is how the tech is changing so quickly. Last year i got the 6800 ultra...now given that is still a real good card, but AGP is dead, and in my plans to move to socket 939 i can't really even use this card anymore...and now i am hearing that 939 will be over in a year and m2 will be the new "thing"

Ya know what would be real nice? If there was one common socket...so intel chips and amd chips could work in it. Then SLI/Crossfire that lets you either go ATI or nVidia.
this would be one of the best things these corps could do...I know it will more than likely never happen, and we will be stuck with hardware that won't work with each other, and paper launches...but i guy can wish right ?
 
SmokeRngs said:
I was glad for the change in the way of producing vid card reviews when you made the change. Sure, it took me a few reviews before I completely understood the presentation of the graphs and being able to read them properly, but it was well worth it. I have found value in the reviews even though I am still running a Radeon 8500. I'll have money to upgrade one of these days.

The main thing I would like to know is if game developers have been of any help or are going to help in regards to "a new way of benchmarking". I remember in a previous article that you were hoping they would develop tools included with the games that would help with the benchmarking. Considering the way you do the reviews, I'm not sure if there is anything they could do that would help. I see the main problem with the current reviewing methodology is the number of man hours required to produce the results. Have you seen any indication that anything is being developed that would help with this?

Click to expand...

I finally decided that looking to game devs to solve our evaluation problems was not the way to go about it. We need to leave them be and focus on making solid content IMO. Looking back, all the tools that we originally asked them for were just going to perpetuate the focus on frame rates. If we are not focused on the game as it is played then we are looking at the wrong thing.
 
It's a good article Kyle. Very true. There definatley have been alot of changes in the video card industry over the last 4 or 5 years. Certainly it isn't what it was 10 years ago either. Indeed, it's come along way.

I also totally agree with ditching the synthetic benches and optimized time demos.
 
I always find myself looking real close at the comparative screenshots of the games [H] reviews. The Difference in FPS is important to an extent but the image quality is also very importatant. Since IQ is subjective to the reviewer (and that is why I try to find a reviewer I agree with more often than not) screenshots are the only way that we, the consumer can make a better IQ based decision.
 
dtess17 said:
Although i agree that the cards seems to have a little more life to them, spending $500-600 on a video card is still crazy to me. I remember when i thought $300 was a good amount to spend on a card. I guess i don't mind buying a card for that price but it damn well had better give me 2-3 years of real solid performace in games.

my real problem is how the tech is changing so quickly. Last year i got the 6800 ultra...now given that is still a real good card, but AGP is dead, and in my plans to move to socket 939 i can't really even use this card anymore...and now i am hearing that 939 will be over in a year and m2 will be the new "thing"

Ya know what would be real nice? If there was one common socket...so intel chips and amd chips could work in it. Then SLI/Crossfire that lets you either go ATI or nVidia.
this would be one of the best things these corps could do...I know it will more than likely never happen, and we will be stuck with hardware that won't work with each other, and paper launches...but i guy can wish right ?
Click to expand...

I remember paying $600 for Voodoo II SLI, so in reality, what you can get for $600 today is far and away much better than what you got for your dollar then. At the time, I think that was $175 more than my rent....

Well, changes in interface have been far and few between, so you just got a bit screwed at that point. I would not expect you to face the same problem for a while.

As for things coming together to be that generic, I don't think it is likely to happen in our lifetime if ever. You are basically asking for Ford and Chevy to use the water pumps, pistons, rods, etc. And just due to basic architectural differences, this is not going to happen. That said, you can expect and upcoming Intel chipset SLI equipped board to accept both SLI and CrossFire, or that is what I have been told directly by the company building it. We will see. So a little of what you want is being delivered.
 
TheBluePill said:
I always find myself looking real close at the comparative screenshots of the games [H] reviews. The Difference in FPS is important to an extent but the image quality is also very importatant. Since IQ is subjective to the reviewer (and that is why I try to find a reviewer I agree with more often than not) screenshots are the only way that we, the consumer can make a better IQ based decision.
Click to expand...


FYI - We will be moving to PNG screenshot format here before the end of the year so our screen shots will not be subject to jpeg compression artifacts. (Although we do not use heavy jpeg compression now to try and avoid as much artifacting as possible.)
 
Good article, but not much more to say than what you stated.

One thing to note, though: Hasn't nvidia ALWAYS been on top less the FX series? ATi's gem against nvidia was the 9700 series and prior to it, ATi was struggling- and now in the post-9700 era, ATi is back to struggling...
 
Jonsey said:
That's true. You say what level or map of a game you run through.
Click to expand...

And we try our best to pick a run through in the game that will likely represent a worst-case-scenario in terms of graphics intensity. So that way it is highly likely that you will get "better" results than ours for much of the overall gameplay in any specific game, and therefore not be disappointed if you used our evaluation as a basis for your purchase. Obviously there are other factors that come into play here, but I am just sharing with you some more of the logic behind our decisions.
 
One of the highlights of the article was your personal revelation when the "light turned on" - How does this card actually handle the game? And you couldn't answer the question. I lost faith in benchmarks as evaluation tools when I was managing midrange systems (IBM S/36's and AS/400's) in the late 80's and early 90's - in that industry the big thing was transaction processing - and IBM, Oracle, Sybase, et al, had thier one up manship on a particular "standard" set of data in a "standard database" to see who could process things through with new releases and new hardware. Well, the real world results of actually using the DB engine and hardware was completely different - they Optimized the hell out of thier code to get higher numbers in benchmarks. And we the users and system managers were let down many times.

You did a great job in explaining "why" you do things this way - this will be of help to those who haven't been around here reading and participating for 5 years (so I reg'd a long time after I started reading.... guess I didn't have anything to add at the time ;) ).

Like many others, I work hard for my money, have a family to support, bills to pay, and want the best bang for my buck when it comes to computer hardware - for home and for work. There is real value in "hands on" and "everyday usage" reviews - you don't see reviews for cars where they slap the car on a big treadmill and run it through a series of tests do you? Hell no, the get that puppy out on the road and drive it! They drive it to work, take a trip down the freeway, around a winding country road, and often to the track for some specilazed handling and braking tests.

Hands on evaluaitions of real world usage have value.

Peace,
Tim
 
jebo_4jc said:
I wonder if these worst case scenarios might benefit one card vs another? What if one card's pixel shader something or another can do a better job rendering one specific part of a game, whereas another card can run the entire rest of the game better? Let's take that window under the turret in HL2 that really killed cards. Let's say nvidia can render that window better than anybody because of some feature in their hardware, but ATi runs 10% faster in general in the rest of the game? This would skew the worst case benchmark away from the "best" card for the game, wouldn't it?
Click to expand...

Exactly what you are stating here is why we also try to take into a account a run through that does not focus on a particular shader. So yes, if we played a level that had nothing but the "bad window" in it, it would likely give you a skewed result. We make sure not to do this.
 
Dillusion said:
Good article, but not much more to say than what you stated.

One thing to note, though: Hasn't nvidia ALWAYS been on top less the FX series? ATi's gem against nvidia was the 9700 series and prior to it, ATi was struggling- and now in the post-9700 era, ATi is back to struggling...
Click to expand...

not really, nVidia was actually a late bloomer. But if you are referring to the last 6 years then yes they have been very aggressive But they still hold less market share than ATI.

IMHO, it all boils down to game play and what it takes to "get there". As some other folks have pointed out, you don't need a 7800 to have fun, just a good game and decent card.
What is a decent card ? one that you can afford and can play you favorite game just fine.
I hope games and graphics don't get to the point that it looks like a movie, I much prefer games that have a different reality. An alternate reality. More interesting. And it doesn't take too much to get there really.
 
IanG said:
...but I still prefer to visit a few different sites to get more opinions before spending my $$$
Click to expand...

As well you should - Anyone that reads the site and the forums will agree with you there. Everyone evaluates hardware a little differently, and thier varied experiences help all of us out in the long run.

Peace,
Tim
 
IanG said:
That's the tricky part though... how does a reader know that you are being fair in what you sample? [H] is one of the more balanced websites out there, but I still prefer to visit a few different sites to get more opinions before spending my $$$
Click to expand...

We describe pretty well what exactly we do, so obviously you can look for yourself if you have the game and are versed well enough on what to look for. If you are not versed, it simply comes down to trusting the source, or not trusting the source to know what they are talking about. As for visiting a few websites, we link a few hundred every week that cover the same products we do and make sure to give you links to other reviews on launch days that cover the exact same products.
 
Kevl4R said:
I think that as both Red and Green make better/more powerful cards, the lifetime of these cards seems to last longer and longer. I think back I feel like I went from a Diamond Monster, Diamond Monster II, Riva TNT, Riva TNT2 Ultra, GeForce 1, GeForce 2, Radeon 9700 Pro seemingly very vast. When I got to the 9700 pro, it lasted nearly two years and I still wasn't completely sure I needed to upgrade.
Click to expand...
I dunno, I see it the exact opposite. Bought an X800XL a couple months back, and even "out of the box", it couldn't hack modern games (and I'm not even talking 'FEAR' here - take "Pacific Fighters", for example) at 1680x1050 with 4xAA, 4xAniso, adaptive/transparency AA, and in-game settings cranked with acceptable framerate.

So I thought, well, heck, I have an otherwise fairly powerful system...why not try getting a 7800 and see how it goes? Same deal - in Morrowind, Battlefield 2, Pacific Fighters, Call of Duty 2, etc - high res, high FSAA, high in-game settings, and I can't see how anything short of SLI offers acceptable framerate in CURRENT games...forget future titles!

It seems to me that the lifespan of graphics cards is getting shorter and shorter! True - mostly because they keep raising our expectations. Where 2xAA would have been fine in the past, we want 4xAA now. Nobody who has seen adaptive/transparency AA can do without it, now. Etc.
 
sxr71 said:
I don't know how feasible this is but is it possible to construct an external box that is programmable to store a set of keyboard commands as you play and then replicate that exact set of commands for benchmarking?
Click to expand...

Well, given today's gaming AI, I am not sure how feasible that would be if possible at all. Also, this takes the human experience angle out of evaluating the card and quite frankly I see that as moving backward, not forward. I do understand that you looking for a way to make "every run the same" though. In reality, this has not been much of a problem with most games. Brent has gotten damn good at matching up his gaming over the years, as his graphed per second frame rates are a testament to that.
 
I still have problems trying to get an idea of what the video card would perform like on my system. You test on a more powerful system overall and while you think that an average of 40 fps on certain games is acceptable, I'd much rather turn AA down or off than play under 60fps. Which is why I'm glad you still have the apples to apples section, keep up the good work.
 
dderidex said:
I dunno, I see it the exact opposite. Bought an X800XL a couple months back, and even "out of the box", it couldn't hack modern games (and I'm not even talking 'FEAR' here - take "Pacific Fighters", for example) at 1680x1050 with 4xAA, 4xAniso, adaptive/transparency AA, and in-game settings cranked with acceptable framerate.

So I thought, well, heck, I have an otherwise fairly powerful system...why not try getting a 7800 and see how it goes? Same deal - in Morrowind, Battlefield 2, Pacific Fighters, Call of Duty 2, etc - high res, high FSAA, high in-game settings, and I can't see how anything short of SLI offers acceptable framerate in CURRENT games...forget future titles!

It seems to me that the lifespan of graphics cards is getting shorter and shorter! True - mostly because they keep raising our expectations. Where 2xAA would have been fine in the past, we want 4xAA now. Nobody who has seen adaptive/transparency AA can do without it, now. Etc.
Click to expand...


Yeah this is how I see it too. Everyone keeps talking about how current games can't take advantage of modern hardware and about how our hardware is so good, but I fail to see what they are talking about. In your experience you are having trouble with 1680x1050; what about people running 1920x1200 or higher? The problem is even more pronounced.

Adding to that, your defintion of acceptable framerates may be different from mine, as I'll accept even 30fps as long as it doesn't keep dropping below that. There are games now that with 4xAA and 16xAF at 1920x1200 that will scarcely allow a 7800GTX to run at a consistent 30fps.
 
sxr71 said:
Yeah this is how I see it too. Everyone keeps talking about how current games can't take advantage of modern hardware and about how our hardware is so good, but I fail to see what they are talking about. In your experience you are having trouble with 1680x1050; what about people running 1920x1200 or higher? The problem is even more pronounced.

Adding to that, your defintion of acceptable framerates may be different from mine, as I'll accept even 30fps as long as it doesn't keep dropping below that. There are games now that with 4xAA and 16xAF at 1920x1200 that will scarcely allow a 7800GTX to run at a consistent 30fps.
Click to expand...
Yeah, it's those min framerates that are REALLY the killer. I'd say the *average* is, realistically, largely irrelevant. 30fps-45fps *minimum fps* is the key. If it EVER drops below that...things are not going well.

It's amazing how hard games are on a system - but, graphics cards ARE getting better. I mean, where you can use 2xAA (where it looks acceptable), almost everything plays fine.
 
I'm with you on everything so far Kyle...

And I'm with those that think the current games are unplayable. I have a 3800X2, a 6800GT, 1gb of TCCD ram...

And I can't do WoW at 1600x1200 4xaa 16xaf smoothly. Or even 2x/8x, at times. That's what I'm looking for. That's why I'm going to a 7800GT, and maybe even an SLi setup. That's what I want.
 
Mysogonist said:
Graphics are getting better.
Games are getting less and less fun.
Click to expand...

Sounds like the Movie industry as well. Apparently a lot of industry sectors are focussing on 'features' rather than content a la "My Movie has twice the number of vertices..."

Not sure if this was part of the editorial though.
 
Great article Kyle. I know that in the not so recent past I was guilty of basing my vid card purchase on the numbers that were put up by some of the overclockers on the xs forum. I got so wrapped up in it that I began to not even play the new games that were out. I still have not completed Doom3 or Half Life2 yet. Just recently have I gotten out of that mentality and just stuck to playing games. Thank god for BF2.
It does baffle me sometimes how some forumers will put H down for being biased. It is about 50/50 on how people think this site is either an Nvidiot or a FanATIc. Maybe I just doent see it but I think you go right down the middle. You call each company out when they need to be called out. Congrats guys on the evolution of your vid card reviews and hopefully more sites will follow. As for me I have to get back to Karkand. ;)
 
Bravo Kyle. I have kept my mouth shut and bit my tongue quite a bit on the forum about recent Nvidia vs. ATI wars that have ensued. THis article hits the nail on the head of chose what is right for you and do not bash others decisions. I want to immerse myself in a game as much as the next guy and individual tweaking will get you there. IF you are into a gmae that much you will find the ways to make your games look the way you want them to. I think your reviews are perfect for any logic-minded person who really thinks ahead and plans to spend their money not just having a wad of cash that will be spent on the newest toy just for bragging rights. Materialism has become an evil in our society and I feel it consumes too much of our day rather than just enjoying ourselves and shooting up the baddies in our fun time. Bleed be the name of sane people in this world.
 
Is it all about who can get the most frames in a DOOM 3 or Half-Life 2 benchmark? Is it really about benchmarks noting frames per second in some game’s demo level that you have never seen and will likely not ever play? Hardly..
Click to expand...

This is how Green and Red want us to think. The marketing of computer hardware has become a priority. Look at the "Fatal1ty" hardware line Abit produced. This kind of marketing is designed to put a product on a pedestal by relating it to a respected person. I really don't understand it. Why not just spend the extra money on a business model based on product quality. I could probably rant forever...

[H]ard OCP has been loyal to it's fan base for years. I feel that I get a no bull shit rundown of new hardware. Thankyou, and keep doing what your doing.
 
I know its against the policy and I think its a great that your site uses real world benchmarks...however when used correctly I still think some synthetic benchmarks have their place (albeit small) in a review. What they are usefull is to show the architectural differences in video cards. Some of these may show up in the benchmarks that Hard runs. Some may not. What you CAN NOT DO IS EXCPET TO FIND OUT HOW YOUR CARD WILL RUN GAMES based on them.

A case in point was that whole FX thing. Here we saw that using the games (mainly DX7/8 based) at that time showed them to be not so bad. However once the next gen games that came out, we really saw how poor the FX series was at running DX9 games. In fact Brent linked articles in that said Developers were treating the FX as DX8 hardware to save their time/money. What makes this interesting is that 3Dmark/Shadermak told us more or less the same thing about 1 year earlier.

So I don't expect Hard to change their stance on Synthetic...just was trying to point out in some cases they do add some value to those of us that like to know all that we can about the cards. :)
 
sxr71 said:
Yeah this is how I see it too. Everyone keeps talking about how current games can't take advantage of modern hardware and about how our hardware is so good, but I fail to see what they are talking about. In your experience you are having trouble with 1680x1050; what about people running 1920x1200 or higher? The problem is even more pronounced.

Adding to that, your defintion of acceptable framerates may be different from mine, as I'll accept even 30fps as long as it doesn't keep dropping below that. There are games now that with 4xAA and 16xAF at 1920x1200 that will scarcely allow a 7800GTX to run at a consistent 30fps.
Click to expand...

This is a joke right. For the most part the thought of running any game at 1600 was joke, and AA would absolutely kill your machine. Now we have cards that are so fast and so close to doing 1600x1200 8xAA 16xAF that people think that it should be atainable and that anycard should be able to handle it. I think this start with the low cost Super large LCDs, Hoenstly on a CRT the difference between 1600 and no AA and 1024 with 8xAA is very simular. But with LCDs you have to play at 1600 or 1920 because otherwise it doesn't look right.
 
Jbirney said:
I know its against the policy and I think its a great that your site uses real world benchmarks...however when used correctly I still think some synthetic benchmarks have their place (albeit small) in a review. What they are usefull is to show the architectural differences in video cards. Some of these may show up in the benchmarks that Hard runs. Some may not. What you CAN NOT DO IS EXCPET TO FIND OUT HOW YOUR CARD WILL RUN GAMES based on them.

A case in point was that whole FX thing. Here we saw that using the games (mainly DX7/8 based) at that time showed them to be not so bad. However once the next gen games that came out, we really saw how poor the FX series was at running DX9 games. In fact Brent linked articles in that said Developers were treating the FX as DX8 hardware to save their time/money. What makes this interesting is that 3Dmark/Shadermak told us more or less the same thing about 1 year earlier.

So I don't expect Hard to change their stance on Synthetic...just was trying to point out in some cases they do add some value to those of us that like to know all that we can about the cards. :)
Click to expand...

Here is a quote from OUR conclusion on our first go with the FX5800 based primarily on 3DMark.

At the current US$400.00 price point, the GFFX simply does not seem worth it to us. If NVIDIA can work some driver magic and pull an extra 20% increase in frame rate out of the bag like we have seen in the past, they had best start pulling. Either that or pull out the NV35 chipset, and quick.
Click to expand...

Here is a quote from our FX 5200 & 5600 conclusion, based largely on gaming tests.

Although NVIDIA has taken two steps forward in terms of features and functionality, they have seemingly taken a step backwards for both image quality and performance. As we score things, that would make NVIDIA's current position stagnant at best. Should the "cinematic computing era" be taking place, it is happening with ATI's product line, as we have seen first hand that the current GeForceFX product line simply cannot compete with ATI products released last year.
Click to expand...

And a FX 5900 conclusion based mainly on gameplay.

Both cards offer exceptional performance, but if I have to place my finger on which card I would choose for my primary system that I play my games on, I would have to opt for the 256MB 9800 Pro. It has better AA quality, and can play at 6XAA on a 256MB card very easily. It has also proven to be very strong in shader operations.
Click to expand...

I don't think our evaluation process has caused us to give bad information at any point in time. I will say this about your thoughts though. I think if we sat down and really analyzed how synthetics measure things, we could find just as many specifics in which they have been "wrong" as they have "right" when it comes to real world gameplay performance in hindsight.
 
This comment really struck me:

Along with these optimizations, we have to realize that no two images will ever be the same now. Red and Green will never be apples to apples. They are now as different as… well, red and green. Now add into this mix totally different techniques and forms of implementing antialiasing and anisotropic texture filtering when comparing Red and Green and you end up with images that are nowhere near apples to apples even when you have the IQ quality settings set in the "same" positions.
Click to expand...

I strongly agree, but I wonder where that leaves benchmarks and testing. Right now [H] benchmarks based on "playable framerates" and I think I am like everyone else when those graphs are the first thing I look at. I rarely look at the screenshot comparisons because:
a) I read [H] at work
b) I don't trust the image due to such things as compression artifacts.

But, if moving forward ATI and NVIDIA use AA and AF methodology that is "apples and oranges" how will we test to find the best? Will [H] end needing to say that 2X on one card is equal to 4X on a different card? I actually hope so.... maybe we can finish the move to a hardware environment where we talk about which card the game is prettiest on, because fastest is no longer an issue.

Me.
 
I like HardOCP and I appreciate what you guys try and do with the reviews but I usually skip them and go right to apples to apples or I just don't read them anymore.

Why? The way you guys "cheat the cheaters" is to just find what settings and resolution makes the cards you are reviewing score about the same, and call it good. That is very subjective. I like how you guys don't just give a bar chart and instead use a graph that shows the Min/Max framerate but I don't like how you arbitrarily just put different settings on a graph and call it a day.

Also the message on the apples to apples section sounds very condescending:
Keeping in mind nothing is truly apples-to-apples to begin with, some people still do find value in looking at performance with the resolution, AA, and AF levels being "equal."
Click to expand...

As if those that want to see what two cards score at with the same resolution and AA are ignorant.

I just prefer to make up my own mind, please give me the graphs you guys use, I love the Min/Max numbers, but putting 1024x768 no AA with Card A and 1600x1200 4x AA with Card B then declaring a winner makes little sense.
 
What I am liking about all of this is that game manufacturers finally "figured it out". They now set theoretical boundaries of hardware requirements to not be able to "top out" their game for quite some time after a release.

My last graphics card purchase was my most expensive, and the next one will probably cost me more. I remember when the geforce 3 was making its debut in an apple, and the 600 dollar initial pricetag floored all of us. Today for that same beyond tomarrow performace costs in the neighborhood of a thousand dollars.

Its been a good ride.

Whats next? Who knows?
 
You must log in or register to reply here.
Back
Top