Wich card is the king of the hill (nv40 or R420)

[IgnorancePersonified]

Heh i wouldn't care to much for those valve survey results, you do realise most valve customers are cs freaks, the whole community plays this same game over and over and it's comprised of very young kids who use antique hardware and usually can't afford to buy and play the latest pc games. Comparing the hardware base for HL1 based mods to say people who play battlefield or UT2k3/2k4 and other online games built in the last 2 years and you'd get a vastly different picture of the norm in hardware. If i was a game dev i certainly wouldn't be targeting the cs audience for my game, it would be futile and pointless.

Yeh well my 9800Se plays the game well. When I get bored of Painkiller single and farcry single cod single blak hawk down multi bf1942 multi quake3 lol and jeeez most other Single Player games I play against some humans in a fast paced game that takes less than 5 mins for a rnd with small maps so I don't have to hunt/watch ppl hunt friggin campers for ever. I know alot of ppl the same + the prizes on the ladder I play on are heaps bigger for CS??? Dunno why its - such a shit game












noob

 

[Bouncing Zabaglione Bros.]

38% of the dominating cards to play online games with are powered by DX6-DX7 class video cards...with NO pixel shader support at all. So when I see comments about ATI holding back technology, I just simply laugh.

There ya go...in a nutshell THIS is what is holding back the industry, not ATI nor nVidia.

People buy hardware when they need to upgrade. If their games run slowly, they buy better hardware. If their games run fine, they stay how they are. If all you play is Counterstrike, then you don't need to upgrade. A survey of people running an old engine that doesn't need good hardware will show you that the majority of those people don't need good hardware.

Run the same survey against UT2K4 or Far Cry players, and they will have much higher specs because the game they choose to play requires it.

 

[Killua]

Unlike those well known developers with sponsers from monster companies like nvidia and ATI, most game developers still uses quake 3 engine for their games and that's because most cards (*coughs*gf2mx*coughs*) can run on that engine. Doesn't make sense for them to make a game which most people can't play... no sales, no survive.

 

[gunblade]

nVidia, however, should be releasing the rest of the NV4x line by the end of the year, including a value part that will make SM 3.0, including floating point buffer support, available to the masses.

And in 2 to 4 months ATI will have their value part packed and shipped and being sold while Nvidia do's what? Sell what? While they get their value part together for a hypothetical huge support for PS 3.0? at the end of the year? I do hope you understand the concept of first to market! It’s simple business process that says the first one on store shelves makes money, and ties up retailers ability to shelf something else. What the competitor doesn’t make now they cannot make up! Do you see the 2 fold benefit to the principle “it makes you money and hurts your competitorâ€￾.

That is a huge gamble Nvidia is taking for PS 3.0, but that’s just my opinion.

I think Nvidia should be praised instead of being blamed for getting SM3.0 parts out. In fact, if not for their gamble on SM3.0 we may have to wait another half year or so before the new cutting edge game come out with SM3.0. It is always good that company takes initiative to move the technology forward. We certainly don't want to be dictated on whether a certain featureset is needed or not(That is also the beauty of having competitors). For a consumer, Sm3.0 may not matter now but they provide something for the programmers to work on and so enable the round of new games.

I would say Ati wins performance wise with the XT card. Nvidia wins with the better featureset. So, would that be a draw? It depends on what each person need is. I think it is all about the price point now. I think I will take whatever best in the $299 price point. Anything above is way over my limit.

I wonder what can be expected on their video encoding capability tho. It seems nv's sub-processor is more flexible to me, but that flexibility is in the control of nv's driver team. I hate to wait a long time for new feature to come, since the video part probably don't get any priority over the driver performance.

 

[jvd]

think Nvidia should be praised instead of being blamed for getting SM3.0 parts out. In fact, if not for their gamble on SM3.0 we may have to wait another half year or so before the new cutting edge game come out with SM3.0. It is always good that company takes initiative to move the technology forward. We certainly don't want to be dictated on whether a certain featureset is needed or not(That is also the beauty of having competitors). For a consumer, Sm3.0 may not matter now but they provide something for the programmers to work on and so enable the round of new games.

I'm sorry . But they shouldn't be praised for what they are supposed to be doing .

I have a dog . I don't praise him when he shits outside.

Why would i praise nvidia for supporting new features.

Mabye its the fact that they have single handly held back p.s 2.0 developement .

That they held back p.s 1.4 development ?

Honestly what should we praise them for ?

They made a good solid card . Isn't that what they are supposed to do ?

Now mabye if they cut the rest of the crap they have been doing i might consider buying it .

But praise is the farthest thing from my mind

 

[Maverick]

This debate about SM3.0 vs 2.0 reminds me of arguments about whether games should be sold on DVDs instead of CDs. It's nice to only have to deal with 1 DVD instead of a few CDs, and maybe have some higher res textures or something, but it's not compelling enough to force everybody to upgrade, and it's nowhere near as important as the previous floppy to CD transition. Not everybody has a DVD drive; everybody (at least everybody who is going to be playing a remotely modern game) has a CD drive, so publishers will target CD drives. A few might make DVD versions, but they are few and far between, and don't cause people to buy DVD drives.

Similarly, SM 3.0 has a few advantages over 2.0, but pretty much everything can still be done in 2.0. Most devs will target 2.0, some may write additional 3.0 versions of a few key shaders. End user won't really give a crap about the whole thing, except perhaps in a "My card has SM 3.0! It's better than yours!" playground kind of way.

Nvidia have probably "advanced the state of 3d technology" or whatever a bit more this generation, but I think ATI have judged it better from a business perspective, unless Nvidia can persuade consumers that SM 3.0 is a "must have" feature, which, quite frankly, I don't think they will be able to do.

Cue people tearing giant holes in my lame analogy...

 

[L233]

People buy hardware when they need to upgrade. If their games run slowly, they buy better hardware. If their games run fine, they stay how they are. If all you play is Counterstrike, then you don't need to upgrade.

That's why I think that CS is one of the worst things that happened to the PC gaming industry. CS is played by millions of people and for many it's the one game they play most. When a new game is released that won't run on their ancient hardware they just won't buy it. As long as CS runs fine they see no need to upgrade.

The second worst thing would be the Nvidia GeForce 4 MX.

 

[Maverick]

Damn them for not buying a new 3d card when they don't need to!

 

[Big Bertha EA]

38% of the dominating cards to play online games with are powered by DX6-DX7 class video cards...with NO pixel shader support at all. So when I see comments about ATI holding back technology, I just simply laugh.

There ya go...in a nutshell THIS is what is holding back the industry, not ATI nor nVidia.

1. People buy hardware when they need to upgrade. If their games run slowly, they buy better hardware. If their games run fine, they stay how they are. If all you play is Counterstrike, then you don't need to upgrade. A survey of people running an old engine that doesn't need good hardware will show you that the majority of those people don't need good hardware.

2. Run the same survey against UT2K4 or Far Cry players, and they will have much higher specs because the game they choose to play requires it.

1. True. However, you have a very LARGE majority of the mass market that is still using older technology. When they decide they want to play something more advanced technically and/or graphically, they will upgrade.

2. UT2K4 and Far Cry players, while in significant numbers at a glance, only represent a very small cross-section of eligible PC users. What the industry needs are games that are compelling enough to convince the mass market users to invest in graphics card upgrades to play the latest and greatest. THEN and ONLY then, will you see a step-change in the demographics that developers use as a baseline for graphics quality in games. Keep in mind, these developers have FAR better and more accurate access to their target demographics that you and I do...

 

[L233]

Damn them for not buying a new 3d card when they don't need to!

No, damn Valve, these slow fucks, who can't seem to get anything released in time. If there had been a timely HL2 and a new CS to go with it, then the problem wouldn't exist. Has Valve actually managed to release anything of value since the original Half Life?

It took them a year to program some bots for CS and and slap it into a retail box, TF2 is contenting with DNF for the Vapourware of the Century Award, CS:CZ is a joke and STEAM is a bug-ridden POS that destroyed smaller CS LAN parties.

The only thing Valve has contributed to the gaming industry (other than procrastinating progress) was to hire Marc Laidlaw who made HL a story driven game.

 

[Zengar]

It seems to me that the 6800 is really a developers card as it pushes the envelope of features that they want now for the current development of future releases. The 420 seems like more of a consumer card since offers the highest performance at the highest image quality settings for current titles, but would not likely be the desired development platform for a game schedule for release in the fall of '05.

I would be willing to guess that most people that favor the nv solution here are the Joe programmer/developer types. While the ati solution is favored by Joe gamer. Perhaps we are seeing the beginning of a new market segment? Maybe the companies should be specifically making a card that is very forward looking that sacrifices power, convenience, stability or whatever, as well as a gamer oriented card that is targeted for current day performance that sacrifices features and future compatability? It looks like the current "workstation" class cards are maybe not filling this niche properly?

Well noticed!

Wokstation cards are a way too expensive for a poor student

But, the majority of people are gamers, that's why X800 will win.

I would also like to notice that X800 is superiour only if it comes to AF. With small AF Nvidias bord running 50 Mgz lower with 32-bit precision have the same speed as 24-bit Radeon. I think there should be no doubt that Nvidias core is faster as ATI's

Ang again, I don't want to be called Nvidia fanboy. I'm not. I'm just thinking loud. Anyway, we don't know hive do this cards really perform.

 

[Bouncing Zabaglione Bros.]

1. True. However, you have a very LARGE majority of the mass market that is still using older technology. When they decide they want to play something more advanced technically and/or graphically, they will upgrade.

2. UT2K4 and Far Cry players, while in significant numbers at a glance, only represent a very small cross-section of eligible PC users. What the industry needs are games that are compelling enough to convince the mass market users to invest in graphics card upgrades to play the latest and greatest. THEN and ONLY then, will you see a step-change in the demographics that developers use as a baseline for graphics quality in games. Keep in mind, these developers have FAR better and more accurate access to their target demographics that you and I do...

That's true of everything - cars, TV's, A/V gear, etc. You have to give people a compelling reason to upgrade. You can't claim "we won't make advanced games because everyone only plays Counterstrike", and then complain that it's these people that are holding back development. If you make the games, people will upgrade, not before. Why should they upgrade if you're not producing compelling games that require better hardware?

What you're suggesting is akin to the film studios expecting consumers to go out and buy DVD players that sit idle before the studios start publishing movies on DVD. What happened is that the studios had to start putting movies on DVD *before* people went out and bought players for their living rooms. Why should the game industry be any different?

Quite simply it's the games industry that is not (for the most part) willing to convince people to upgrade by producing better games that require more advanced hardware. The scalable games engines that allow backwards compatability with lower hardware are a double edged sword that prevent customers from upgrading and stop customers from enjoying your games to the fullest.

That's fine, you make compromises and you sell more games, but ultimately it's not in the consumer's interest to upgrade before you've given them a reason to do so. Any business model that expects them to is doomed to fail.

 

[RostokMcSpoons]

People buy hardware when they need to upgrade. If their games run slowly, they buy better hardware. If their games run fine, they stay how they are. If all you play is Counterstrike, then you don't need to upgrade.

That's why I think that CS is one of the worst things that happened to the PC gaming industry. CS is played by millions of people and for many it's the one game they play most. When a new game is released that won't run on their ancient hardware they just won't buy it. As long as CS runs fine they see no need to upgrade.

The second worst thing would be the Nvidia GeForce 4 MX.

Yes, damn people for sticking with a game they enjoy. Damn it, just line 'em up against a wall and shoot them. And those bastards that play Solitaire or Pinball on their 2D cards - shoot them too. No, shooting's too good for them - gut them with a spoon and hang their corpses upside down!

Honestly, you've got to remember - these graphics cards are just a means to an end - they're *just to play games*

Ah, my first post, and I'll probably get banned for saying some of that

Heretic! Stone the unbeliever!


Anyway - I'm one of those 1 million people in that survey. I was suprised to find my machine is relatively 'high spec' (Athlon 2500, Radeon 9700non-pro). It runs my fave game (Day of Defeat) pretty well - I only need to upgrade for HL2 / Stalker / Doom3... and for DoD2 when it eventually comes out. I got all excited reading the recent reviews, but if I think about it sensibly - do I really want to spend 3 or 4 hundred pounds just to buy more fps in games that aren't out yet? Ok, if I was addicted to Far Cry, that'd be some justification, but I haven't got it (yet). Painkiller runs beautifully on what I have. I play UT2004 (where it occasionally chugs) but I don't play that as much as DoD, so I'm not overly worried.

I'm thinking I'm going to hold out for PCI-E, and the *next* gen cards... probably an R500. Which means I'll be getting an Athlon64 cpu too... no doubt SATA Raid would be 'nice' I'm going to need time to save up for a *total* upgrade. I just know I'll kick myself if I buy another AGP card now, with PCI-E so close...

(I reserve the right to change my mind about all of that, if HL2 actually does come out soon, and my pc can't get over 40fps!)



Btw... I wonder what impact the wide-spread adoption of LCD screens will have on the future of graphics cards... where the displayable frame rate is often limited to 60 or at best 75fps... it gets harder to justify buying a mammothly powerful card when half the frames it renders can't ever be displayed... ?

 

[L233]

Yes, damn people for sticking with a game they enjoy. Damn it, just line 'em up against a wall and shoot them. And those bastards that play Solitaire or Pinball on their 2D cards - shoot them too.

While I am definately in favour of lining people up against walls and shooting them for all kinds of reasons or even just for shits and giggles, I don't really think that would solve the problem in this particular case.

Anyway - I'm one of those 1 million people in that survey. I was suprised to find my machine is relatively 'high spec' (Athlon 2500, Radeon 9700non-pro). It runs my fave game (Day of Defeat) pretty well - I only need to upgrade for HL2 / Stalker / Doom3... and for DoD2 when it eventually comes out.

I play DoD frequently but I haven't heard anything about DoD2 for like 18 months... will DoD2 now use the HL2 engine or something? Are there any previews?

 

[Big Bertha EA]

1. That's true of everything - cars, TV's, A/V gear, etc. You have to give people a compelling reason to upgrade. You can't claim "we won't make advanced games because everyone only plays Counterstrike", and then complain that it's these people that are holding back development. If you make the games, people will upgrade, not before. Why should they upgrade if you're not producing compelling games that require better hardware?

2. What you're suggesting is akin to the film studios expecting consumers to go out and buy DVD players that sit idle before the studios start publishing movies on DVD. What happened is that the studios had to start putting movies on DVD *before* people went out and bought players for their living rooms. Why should the game industry be any different?

3. Quite simply it's the games industry that is not (for the most part) willing to convince people to upgrade by producing better games that require more advanced hardware. The scalable games engines that allow backwards compatability with lower hardware are a double edged sword that prevent customers from upgrading and stop customers from enjoying your games to the fullest.

4. That's fine, you make compromises and you sell more games, but ultimately it's not in the consumer's interest to upgrade before you've given them a reason to do so. Any business model that expects them to is doomed to fail.

1. Not what I am saying at all. What you don't seem to grasp is that game development these days is a double-edged sword. How you develop a game depends on the genre and commerical viability of the franchise or expectations if a new franchise (like Far Cry). Most game developers walk this double-edged sword because they

a) want to push the technology forward
b) are forced to make their games run with lesser hardware to ensure adequate sales and revenue to continue on because the vast majority of PC gamers do NOT have the latest and greatest

The performance curve IS moving, but it is not moving as fast as the technology is. A compelling game like CS or any other game that is compelling enough to entice those using old technology to upgrade, is what I am talking about. Strike Commander and Origin were VERY well known for doing just that...

2. Again, not what I am saying whatsoever. I think I have been VERY clear that the industry needs the compelling reason (GAME) to be enticed enough to upgrade. Not the other way around...

3. See #1.

4. I never advocated that consumers upgrade before needing a reason to. Not sure where you got that idea but I have never even hinted at such an idea...

 

[Atomahawk]

nVidia, however, should be releasing the rest of the NV4x line by the end of the year, including a value part that will make SM 3.0, including floating point buffer support, available to the masses.

And in 2 to 4 months ATI will have their value part packed and shipped and being sold while Nvidia do's what? Sell what? While they get their value part together for a hypothetical huge support for PS 3.0? at the end of the year? I do hope you understand the concept of first to market! It’s simple business process that says the first one on store shelves makes money, and ties up retailers ability to shelf something else. What the competitor doesn’t make now they cannot make up! Do you see the 2 fold benefit to the principle “it makes you money and hurts your competitorâ€￾.

That is a huge gamble Nvidia is taking for PS 3.0, but that’s just my opinion.

I think Nvidia should be praised instead of being blamed for getting SM3.0 parts out. In fact, if not for their gamble on SM3.0 we may have to wait another half year or so before the new cutting edge game come out with SM3.0. It is always good that company takes initiative to move the technology forward. We certainly don't want to be dictated on whether a certain featureset is needed or not(That is also the beauty of having competitors). For a consumer, Sm3.0 may not matter now but they provide something for the programmers to work on and so enable the round of new games.

I would say Ati wins performance wise with the XT card. Nvidia wins with the better featureset. So, would that be a draw? It depends on what each person need is. I think it is all about the price point now. I think I will take whatever best in the $299 price point. Anything above is way over my limit.

I wonder what can be expected on their video encoding capability tho. It seems nv's sub-processor is more flexible to me, but that flexibility is in the control of nv's driver team. I hate to wait a long time for new feature to come, since the video part probably don't get any priority over the driver performance.

My point of view is based on the business (making money) angle, thus the gamble. Recouping development cost is an important part of any companies bottom line, then you need to make a profit. You can't continue in business by pretending you have a product to sell, but it's not on the shelf, thus the problem. In this business like any other you can have the best, but if you don't have shelf space or no delivered goods it's pointless to even talk about it. Thus the problem for Nvidia this round, what are they going to sell for the next 8 months? What they have now? Maybe if they drop the prices and there's stock at distributors. What do's ATI have to sell ? See my point you can't sell PS 3.0 with no goods.

 

[RostokMcSpoons]

While I am definately in favour of lining people up against walls and shooting them for all kinds of reasons or even just for shits and giggles, I don't really think that would solve the problem in this particular case.

I think I may have missed a smilie

I play DoD frequently but I haven't heard anything about DoD2 for like 18 months... will DoD2 now use the HL2 engine or something? Are there any previews?

As far as I know there'll be a version 1.2 coming out very soon, there maybe one more after that (with a possible mortar class) and after that all effort will be expended on 2.0 (Source engine rewrite). I remember the leaked HL2 included versions of the existing CS and DOD code amended to run under the Source engine. I don't know if that's what DOD2 will be - a simple port which will effectively give us Direct X graphics that work, and re-written netcode, but presumably not making full use of Source - i.e. physics interaction etc, OR if it'll be a ground-up rebuild, which I'd suggest would take a long time to appear, given how long it takes them to fix some of the recent issues :? As usual, the team are being quite tight-lipped about their longer term plans.

Hmmm... DOD on Source... that's going to make me want to upgrade - I need 75fps solid to keep me really happy

Glad to hear your a DoD player - what name do play under - we might have exchanged bullets I'm currently [*UKSF*]Brooker[R] but until recently used to play as [CM]Brooker

Anyhow - we digress... back to the actual topic...

 

[L233]

As far as I know there'll be a version 1.2 coming out very soon, there maybe one more after that (with a possible mortar class) and after that all effort will be expended on 2.0 (Source engine rewrite).

DoD2 was supposed to be a somewhat improved DoD... some new textures, a new crosshair and stuff like that. I remember some previews back in 2002. They said nothing about Source engine. Seems like the original DoD2 won't happen or maybe the new stuff was just put into DoD 1.x.

I don't know if that's what DOD2 will be - a simple port which will effectively give us Direct X graphics that work, and re-written netcode, but presumably not making full use of Source - i.e. physics interaction etc, OR if it'll be a ground-up rebuild, which I'd suggest would take a long time to appear, given how long it takes them to fix some of the recent issues :?

I hope they rebuild DoD for HL2 but don't fuck with the gameplay. I like it the way it's now. New visuals and some minor changes (grenade physics, those totally dark camper spots and some minor weapon balance tweaks) would make me one happy camper (pun intended).

Glad to hear your a DoD player - what name do play under - we might have exchanged bullets I'm currently [*UKSF*]Brooker[R] but until recently used to play as [CM]Brooker

I play as L233, usually on German, Dutch or British servers. I try to avoid French server because the French seem to have a habbit of holding conversation over voice com and I find that rather distracting, especially because I don't understand a word.

 

[pascal]

It seems to me that the 6800 is really a developers card as it pushes the envelope of features that they want now for the current development of future releases. The 420 seems like more of a consumer card since offers the highest performance at the highest image quality settings for current titles, but would not likely be the desired development platform for a game schedule for release in the fall of '05.

I would be willing to guess that most people that favor the nv solution here are the Joe programmer/developer types. While the ati solution is favored by Joe gamer. Perhaps we are seeing the beginning of a new market segment? Maybe the companies should be specifically making a card that is very forward looking that sacrifices power, convenience, stability or whatever, as well as a gamer oriented card that is targeted for current day performance that sacrifices features and future compatability? It looks like the current "workstation" class cards are maybe not filling this niche properly?

Well noticed!

Wokstation cards are a way too expensive for a poor student

But, the majority of people are gamers, that's why X800 will win.

I would also like to notice that X800 is superiour only if it comes to AF. With small AF Nvidias bord running 50 Mgz lower with 32-bit precision have the same speed as 24-bit Radeon. I think there should be no doubt that Nvidias core is faster as ATI's

Ang again, I don't want to be called Nvidia <bleep>. I'm not. I'm just thinking loud. Anyway, we don't know hive do this cards really perform.

I disagree about X800 not for game developers.

First, I am tired of bad looking games poorly using new graphics API.
Now someone that want to sell more games will just say that the game has DX9 support or something like that because:
a - he used some DX9 trick somewhere without any quality improvment or WOW factor
b - he used the DX9 API to do something that could be done with DX7

We are STILL waiting for the first trully pervasive use of the DX7 featureset level game (DOOM3) and Carmack´s last year said that his next base level for development after DOOM3 will be DX9 level. My guess we will have to wait more 4 years to see something really good with DX9

The X800 is about sustained performance with pervasive use of the DX9 API level featureset. I (as a consumer) want more on real world delivery and less of paper featureset or market blablabla

Congratulations ATI for the good work. IMHO now is time for you developers deliver something good.

 

[Evildeus]

Well, Nv should have 1 year of experience before Ati delivers a SM3.0 card, but Ati will deliver. I think that with HLSL, SM 2.0 and 3.0 will be there every time in future games. It will surely be interesting to see the performance consequencies when it's done.