One thing I didn't notice asked. Is there a reason the 9600 Pro shows nearly no slowdowns from 1024x768 to 1280x1024 while even the 9800 shows a siginificant fall. Is it the suposed improvements in the new 9600 shades or are the numbers wrong?
LOL BFG. But...but...Doom3 is a "significant" DX9 game, and John Carmack (a somewhat significant programmer) says the FXs will run it faster than R350s?!?!!?
Oh cruel fate! Who to believe? John Carmack, unknown lunatic, or wise BFG, who tells us of "concrete proof" 9600Pros are better?!?!
LMFAO.
Offer to anyone on board foolish enough to believe BFG10K:
I will trade any and all of you NIB 9600Pros of your choice for GF FX5900Us that are NIB. Want to do 1? Let's go. Want to do 100? Let's go.
C'mon boys, step right up and let the trading begin!
Your pal, honest Rollo, has lots of room on his credit cards, money in the bank, and stocks he'll sell to help any of you poor saps with inferior 5900Ultras upgrade to superior 9600Pros!
1. Are there even any DX9 games out yet? So we have 1000s of games that the FX5900s are as fast or faster at than R300/350s, and some pending games that may or may not be faster by the time they're out on ATI.
2. HL2? Pfft. HL grahics suxored, I thought the game was waaaayyy overrated. Didn't even come close to finishing. Doom3? Well, I played Doom 1 and 2 all the way through more than once. I played Quake 1 and Quake 2 all the way through. I played Quake 3 online a LOT. So, Jabba the Gabe can keep his paid for pimping, I don't care if they EVER release HL2. That guy Gabe is looking like he's ready to leave the planet anyway, I think ATI paid him in Twinkies.
3. Aquamark is a DX9 game that FXs seem to run fine.
4. As Rogo would say, it's a logical fallacy to assume the Det50 drivers will contain cheats to make the hardware look better just because past drivers have. Carmack says there's no need for 24bit FP now, he seems knowledgeable.
I guess most of all, you guys are NUTS.
Let's say HL2 comes out tomorrow, and it is a great game, not just a graphics showcase as we've seen often enough on highly anticipated games. You would have to be RETARDED to want a 9600 Pro over a 5900. Who the he!! buys their card to run ONE game faster?! I have a lot a bunch of games, and any of the shooters I can play online I'm as likely to play as the next, so why would I so seriously CRIPPLE myself with a card that's an unredeemable POS like the 9600?! Even it's HL2 performance suxorz, I don't even think my 9800Ps performance is goign to be good enough for that game from the looks of things.
BTW- for those of you about to say, but Rollo, there will be more DX9 games coming out that the 9600s may be faster on, save your breath. I bought a FX5800 and a 9800Pro this year, and that is the fewest video cards I've bought in one year for the last 6 years! Last year I bought GF3, a Ti4400, a V5, a 8500, and a 9700P, and the year before that more cards than that. So what? Well, it's stupid to buy cards for future games. Most of us buy cards at least annually, so by the time there's enough games out for it to make a difference, we've upgraded anyway to cards that make the current crop look like junk. I would be buying any crippled up junk 9600Pro thinking you'll have soem future proof dynamo based on these preliminary benchmarks.
Doom3 is not a significant DX9 game or DX at all for that matter. It is an OpenGL game.
HL1 graphics were pretty decent for its day. I never played Doom 1 or 2. I did play Q1/2/3. As a singleplayer game, none of them were even close to being in the same league as HL. That goes without saying.
Doom/Quake gamplay = Shoot bad guys-find key/button/door-finish level-go to next level-repeat. YAWN. I did have a lot of fun playing Q2 online though!. HL was all about immersive gameplay. It broke new ground in many ways. 5 years later, it is still regarded by the majority of the PC gaming community as one of, if not THE best FPS ever made. I expect HL2 will follow in its footsteps.
And what exactly is the difference between a game and a benchmark? Not polling the input devices (keyboard, mouse joystick, wheel, whatever) for input? Collision detection? Last time I checked video cards didn't play a part in those.
Returning my new unopened GeForce 5900 Ultra. Today the last day to return this thing.
IMO the management at Nvidia doesn't have their act together, don't care, or don't have competitive skills. Or all three. They got blind sided on this and have probably executed damage control. Updated drivers aside, there is no reason to keep thier top of the line card for results like this. Even though I purchased this card on sale at $400, I am expecting Nvidia (or any company for that matter at this price) to take "care" of it's product during it's life. And to take 'care' of it's products the company must have an excellent managemt team. IMO this is Nvidia's REAL problem and that is what needs to be addressed for future revenues.
This doesn't mean I'm jumping to ATI. I'm sure ATI has their managemt issues too. That said, I also believe they have worked better as a management team. It shows in their products.
For now and the next 3 months, I'm waiting until the dust settles.
IMO the management at Nvidia doesn't have their act together, don't care, or don't have competitive skills. Or all three. They got blind sided on this and have probably executed damage control. Updated drivers aside, there is no reason to keep thier top of the line card for results like this. Even though I purchased this card on sale at $400, I am expecting Nvidia (or any company for that matter at this price) to take "care" of it's product during it's life. And to take 'care' of it's products the company must have an excellent managemt team. IMO this is Nvidia's REAL problem and that is what needs to be addressed for future revenues.
This doesn't mean I'm jumping to ATI. I'm sure ATI has their managemt issues too. That said, I also believe they have worked better as a management team. It shows in their products.
For now and the next 3 months, I'm waiting until the dust settles.
You guys can be bitter all you want, and wait for Doom III, just know that you'll be missing a better game in Half-Life 2. I guess some people prefer mindless action and stuff, that's on you.
Actually there are *many* significant differences between a benchmark and a game.
1. Many/most benchmarks don't run with sound enabled. Last I checked, most games are played with sound.
2. Many synthetic benchmarks, though they are created with good intentions, are a poor indication of a real-game engine. For example: in 3dmark 2k3, pretty much all of the benchmarks are flawed. Most notably, the shader heavy game demos are very poorly coded. The method they use to light everything in the scene is grossly inefficient, at least according to some game developers. John Carmack I believe talked about this, and how Doom3 for example, uses many clever tricks to cut down on memory bandwidth and wasted GPU power. The game demos in 3dmark 2k3 are just essentially whipped-up game demos with lots of shader usage. They aren't coded to be playable, just to run without crashing.
3. Many games are based upon the same family of engines: the Quake 3 Engine, the Unreal Engine (different builds), the Source Engine (upcoming), etc. Therefore if a game runs well with one game based on a particular engine, you can expect good performance out of another game using that engine. Most benchmarks are not based on these engines, despite the fact that many/most of the popular games are.
4. Benchmarks can be tampered with, as is evident with the slew of optimizations ATI and Nvidia have enabled for them. One of the most obvious ones was where Nvidia saved a TON of memory bandwidth by cutting out rendering except on the visual plane that some game benchmark would run at (I forget which one). Running the benchmark, everything looks fine, however if you were to actually PLAY the demo you would have most of your visual field cut out at certain angles. This can be spotted MUCH easier with a game, where you can actually play/run the game to check that everything is working. For expamle, Valve quickly detecting that Nvidia didn't render fog/smoke in the new 50.xx drivers for HL2.
5. Despite the fact that video cards have nothing to do with input devices, etc, those devices must be factored into the equation of game performance, because they will be taking CPU cycles away from the GPU, running the AI, etc.
6. Benchmarks can be created with biases. For example, they can be coded more optimally to run on one architecture or another, or (in a worst case scenario) they can actually have code in them to make them run worse on another platform (on purpose).
7. Games are created (usually) to run optimally on all video cards (assuming the company wants to maximize sales). Benchmarks are often not tweaked as well as they can be to get optimal performance out of every card.
8. We should not encourage video card manufacturers (or more specifically, their driver teams, marketing teams, etc) to optimize their cards for benchmarks. Nvidia has an incredible driver team that has essentially worked miracles in the past when focusing on one particular game. For example, they have optimized so well for Quake 3 that it runs at ridiculously high framerates, even on older cards. ATI also has an excellent (and ever improving) driver team that can get very good performance out of their cards. Remember how poorly the Radeon 8500 ran at launch? It was slower than the GeForce 3 despite many significant technical advantages. The Radeon 8500 is now in a class above the GF3, just because the driver team put so much effort into their Catalyst drivers.
Would you prefer that Nvidia and ATI focus their efforts on making themselves look good on all the synthetic benchmarks out there (so people buy their cards), or would you prefer that they work as hard as they possibly can on making then newest/greatest games run as fast as possible? Again, they only have so much developing resources, and any resources they use to optimize for benchmarks could have been used somewhere better.
LOL @ myself Old Fart. Good point, I forgot. Actually I'd say my OGl games outnumber my DX games at least 5X1, if not more. They're all based on Quake engines, except for the Mechs, Unreals, and Max Payne. (and even the Unreals go OGl)
This is a VERY interesting point the more I think about it. How relevant is DX9 for the fps player? Everyone uses Carmack's game engines. Damn. I better cancel my "superior 9600 Pro" order.
We'll have to agree to disagree there. Everything was pretty angular, and the lighting effects were cheesy. All it really had for me was the fun of interacting with the characters. Didn't think the monsters were scary. The Fiend in Q1 was a damn fine monster, when those things first jumped you and started slicing away, pretty tense. The chainsaw ogres were good to, as was the shambler.
True, but Aquanox2 and Spellforce will be games with the same engine, so is there any difference in this case?
Don't do it Parrothead! I'll trade you a superior 9600Pro for it, unopened also! You'll be all set for DX9 and better off, just like BFG says! He is wise and respected, listen to him. PM we'll exchange addresses and Heat, and I'll set you up right with some (POS) ATI 9600 goodness!
LOL, look who's talking about "mindless action" , the guy with the baseball stuff in his sig, NYHOUSTONMAN. My dad just retired, said he's even started watching baseball again. I laughed and said he must like watching guys stand around a lot. He found that amusing, and said I'm probably right as all the "action" within a 3 hour game probably happens in 15 minutes. "Mindless" is watching millionaires scratch themself and spit for three hours to see 15 minutes of athletic competition NHM.
BTW- HL2 isn't exactly "cerebral". I don't think the guys at Mensa are anxiously awaiting it to expand upon their knowledge base. These are just kids video games, comic book diversions.
And what game had better graphics at that time? Q2 and Unreal were also games of that same era. HL certainly looked better than Q2. Unreal was pretty but like Q2, a VERY boring SP game. The monsters in HL were OK, but fighting the Marines was very good. No other game at that time had the enemy AI that HL did.
All shooter games are just that. Games. I dont see your point about them not being "cerebral". They are for entertainment.
I was pointing out they are all just "mindless action". The poster seemed to be saying HL2 is something more.
Unreal looked a lot better, and was more fun.
I am PISSED at nVidia, period. I am not an nVidia person. I sold my FX5800 at a huge loss due to their mishandling of the nV30, and not even bothering to return my emails about it, and wasted money on a 9800Pro.
They can go out of business today for all I care, their customer service sucks, as does their driver cheating in UT2003, my favorite game.
I am a gamer however; and this ATI circle jerk over the benchmarks of one unreleased game is ridiculous.
You are lumping all of them into one category. Its interesting the games that are the most "mindless action" types are the ones you prefer. Q1/2/3/Doom/unreal were far more mindless than Half-Life. Half-Life was the only one that actually had a plot. The others were just shoot/find secret key/button/go to next level. No thinking at all.
Unreal looked a lot better, and was more fun.[/quote]
We will have to (strongly) disagree on that one as well. Unreal looked good, but was one of the most boring SP shooters I've played. Unreal 2 is even worse. HL was a FAR more enjoyable game. The gaming industry/community agree pretty much universally on that one. This is the reason for all the hoopla over HL2.
Exactly...From the starting scene, HL sucked in my attention, and HL2 looks poised to do the same. The interactivity and feel of the game is just more fun to me than games like Quake II (granted, not the same game, but Doom III looks to be the same, albeit with more suspense) .
And don't knock baseball, first it has nothing to do with the argument at hand and second everybody has their own preferences as far as sports (some people just don't watch some sports for one reason or another).
And don't knock baseball, first it has nothing to do with the argument at hand and second everybody has their own preferences as far as sports (some people just don't watch some sports for one reason or another).
Err, sure. The kind of "thinking" involved in HL is still just doing what the characters have told you to do, and a variation on the find key/open door theme. What's the difference in finding a key to open a door and "turn on power so train can take you to next level"? Yep, Einstein himself would get lost in HL.
The whole point is that DX9 is the platform, not ATI or Nvidia. If both had folowed the DX9 spec we would probably see more comparable performance between the different classes of cards from either company.
It's actually kinda scary when people refer to the video card as the platform, makes me think of being stuck with propriatary API's again, different patches needed to make the game work correctly with different cards. Like, let's just throw out all the advances made in the last 6 years.
Or maybe you've touched on the root of the problem. Is Nvidia starting to think of it self as THE PLATFORM? They've been the unchallenged performance leader for so long, and have such a large user base/market share, I think it's time to start asking questions.
I mean, why should Valve have had to spend so much time optimising for NV? I realise any video card manufacturer will impliment the DX9 spec differently, but the whole point of the API being the platform is so the game only has to be written once to run well on any card that follows the spec. (of course taking into consideration variables like CPU speed and class of vid card etc..)
Remember NV's first big hit? The Riva 128 I think it was, and if I'm wrong , I'm sure someone will be glad to correct me. NV touted it as being "API Agnostic", meaning NO PROPRIATARY API. It was indeed the fastest DX accelerator of it's day and the critics loved it. Of course everybody was playing GLQuake at the time so who cared.
But over time as DX got better, I think it was Forsaken on DX5 or 6 that shut all our mouths about M$ and their crappy DX, gamers realized they weren't stuck with having to buy 3Dfx. Forsaken looked almost as good as any Glide game. AND you could run it on a G200 or a TnT or a VooDoo, or what ever DX compliant vid card you had. Developers were happy because they only had to write one version of the game, no extra DLL's or EXE's needed to make the game work on a specific vendors card. Extra development time could go into bug fixes and such.
So? What the heck is going on here? Are we seeing NV trying to force the development community to code specifically for their hardware? Let's face it, NV has a lot of users out there. It would be financial suicide for a developer to release a game that didn't run well on NV hardware. Scary when a company gets so big it can exert that kind of pressure.
Or are we seeing a split between DX and OGL? ATI is obviously the card to have for DX9 games, but at this point, it's looking like NV has the hardware you want to run Doom3 on.
I don't like either situation. I don't know how good a game Doom3 will be, but I can guarentee the engine will be used on a hell of a lot of games in the next few years.
Maybe this post is just FUD, but I can't help feeling uneasy about this. I don't need the fastest vid card on the planet, I just want to play the games I like without being forced to buy a specific card. A new card, OK, that's understandable, but not because some monoply has forced me to.
I like my freedom to choose(LINUX FOREVER)
I really hope this is not what we're seeing here. Kinda makes the fanboy ranting seem petty. If it turns out that NV is trying to use it's user base as a means of controlling developers, then we're ALL in trouble.
Again, we will agree to disagree on this. Me and the vast majority of the gaming community dont agree with you, but you are entitled to your opinion.
BFG10K
Lifer
- Aug 14, 2000
- 22,709
- 3,002
- 126
Rollo, it is now clear to me that you're nothing more than a clueless troll. You constantly ignore facts and logic and your posts add absoulutely nothing to the discussions at hand.
And people don't believe me? How can they believe someone like you who went from a 9700 Pro to a 5800, claiming it was an "upgrade"? And then, having realised the utter idiocy of his actions, pretended that nVidia's removal of the card from their site "forced" him to upgrade to a 9800 Pro?
Do you actually know anything about the issues at hand or do you just constantly post mindless drivel?
No actually, it isn't. Doom III utilises a DX7 featureset as a base and but also uses a few PS1.3 shaders via extensions for a few of the lighting effects. Doom III is not an indication of DX9 games, not by a longshot.
Oh dear, not this clown show again. We all know what happened to Rollo's "upgrade" path last time.
And people don't believe me? How can they believe someone like you who went from a 9700 Pro to a 5800, claiming it was an "upgrade"? And then, having realised the utter idiocy of his actions, pretended that nVidia's removal of the card from their site "forced" him to upgrade to a 9800 Pro?
And precisely how is your opinion relevant to any of the issues at hand? How does it disprove any of the technical information that has been coming for the last 6-12 months about nVidia's horrible PS/VS performance?
Do you actually know anything about the issues at hand or do you just constantly post mindless drivel?
Here is the quick and dirty. In the game code, at init time, you ask what level of shaders (pixel and vertex) you (the card) support? The card comes back and says PS2.0 and VS2.0, then you go, "Cool", we can use DX9 shaders. At this point in the code, you have no idea what video card is really out there, unless you specifically test BEFORE starting DX9 up.
It's not Vlave's fault that the NV3x family of parts perform very badly using the shaders thay calim to support.
For Valve to fix this problem, they would have to disable all shaders. Well, the user community has wanted dynamic code in and have been bittching for years, "why can't game devs support the high end cards?"
Now we are doing it and you are bittching. Folks, the NV3x cards suck at DX9 shaders. That is the simple truth. Deal with it. It's not Valve's problem and I, for one, am glad they are taking the stance they are. Why?
Well, it might just make the farkin video card companies take notice that we, the devs, are not going to go quietly into the night anymore and take heat from gamers about features that are not being used. Maybe, just maybe, it might make the video card companies stand up and take notice, that if you put a piece of crap out in the market, we will expose it."
It's not Vlave's fault that the NV3x family of parts perform very badly using the shaders thay calim to support.
For Valve to fix this problem, they would have to disable all shaders. Well, the user community has wanted dynamic code in and have been bittching for years, "why can't game devs support the high end cards?"
Now we are doing it and you are bittching. Folks, the NV3x cards suck at DX9 shaders. That is the simple truth. Deal with it. It's not Valve's problem and I, for one, am glad they are taking the stance they are. Why?
Well, it might just make the farkin video card companies take notice that we, the devs, are not going to go quietly into the night anymore and take heat from gamers about features that are not being used. Maybe, just maybe, it might make the video card companies stand up and take notice, that if you put a piece of crap out in the market, we will expose it."
Why do you need validation from internet users regarding your purchase rollo-Maybe you should take a few more philosophy courses.
rogo
rogo
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter