AMD's Radeon HD 5870: Bringing About the Next Generation Of GPUs
by Ryan Smith on September 23, 2009 9:00 AM EST- Posted in
- GPUs
More GDDR5 Technologies: Memory Error Detection & Temperature Compensation
As we previously mentioned, for Cypress AMD’s memory controllers have implemented a greater part of the GDDR5 specification. Beyond gaining the ability to use GDDR5’s power saving abilities, AMD has also been working on implementing features to allow their cards to reach higher memory clock speeds. Chief among these is support for GDDR5’s error detection capabilities.
One of the biggest problems in using a high-speed memory device like GDDR5 is that it requires a bus that’s both fast and fairly wide - properties that generally run counter to each other in designing a device bus. A single GDDR5 memory chip on the 5870 needs to connect to a bus that’s 32 bits wide and runs at base speed of 1.2GHz, which requires a bus that can meeting exceedingly precise tolerances. Adding to the challenge is that for a card like the 5870 with a 256-bit total memory bus, eight of these buses will be required, leading to more noise from adjoining buses and less room to work in.
Because of the difficulty in building such a bus, the memory bus has become the weak point for video cards using GDDR5. The GPU’s memory controller can do more and the memory chips themselves can do more, but the bus can’t keep up.
To combat this, GDDR5 memory controllers can perform basic error detection on both reads and writes by implementing a CRC-8 hash function. With this feature enabled, for each 64-bit data burst an 8-bit cyclic redundancy check hash (CRC-8) is transmitted via a set of four dedicated EDC pins. This CRC is then used to check the contents of the data burst, to determine whether any errors were introduced into the data burst during transmission.
The specific CRC function used in GDDR5 can detect 1-bit and 2-bit errors with 100% accuracy, with that accuracy falling with additional erroneous bits. This is due to the fact that the CRC function used can generate collisions, which means that the CRC of an erroneous data burst could match the proper CRC in an unlikely situation. But as the odds decrease for additional errors, the vast majority of errors should be limited to 1-bit and 2-bit errors.
Should an error be found, the GDDR5 controller will request a retransmission of the faulty data burst, and it will keep doing this until the data burst finally goes through correctly. A retransmission request is also used to re-train the GDDR5 link (once again taking advantage of fast link re-training) to correct any potential link problems brought about by changing environmental conditions. Note that this does not involve changing the clock speed of the GDDR5 (i.e. it does not step down in speed); rather it’s merely reinitializing the link. If the errors are due the bus being outright unable to perfectly handle the requested clock speed, errors will continue to happen and be caught. Keep this in mind as it will be important when we get to overclocking.
Finally, we should also note that this error detection scheme is only for detecting bus errors. Errors in the GDDR5 memory modules or errors in the memory controller will not be detected, so it’s still possible to end up with bad data should either of those two devices malfunction. By the same token this is solely a detection scheme, so there are no error correction abilities. The only way to correct a transmission error is to keep trying until the bus gets it right.
Now in spite of the difficulties in building and operating such a high speed bus, error detection is not necessary for its operation. As AMD was quick to point out to us, cards still need to ship defect-free and not produce any errors. Or in other words, the error detection mechanism is a failsafe mechanism rather than a tool specifically to attain higher memory speeds. Memory supplier Qimonda’s own whitepaper on GDDR5 pitches error correction as a necessary precaution due to the increasing amount of code stored in graphics memory, where a failure can lead to a crash rather than just a bad pixel.
In any case, for normal use the ramifications of using GDDR5’s error detection capabilities should be non-existent. In practice, this is going to lead to more stable cards since memory bus errors have been eliminated, but we don’t know to what degree. The full use of the system to retransmit a data burst would itself be a catch-22 after all – it means an error has occurred when it shouldn’t have.
Like the changes to VRM monitoring, the significant ramifications of this will be felt with overclocking. Overclocking attempts that previously would push the bus too hard and lead to errors now will no longer do so, making higher overclocks possible. However this is a bit of an illusion as retransmissions reduce performance. The scenario laid out to us by AMD is that overclockers who have reached the limits of their card’s memory bus will now see the impact of this as a drop in performance due to retransmissions, rather than crashing or graphical corruption. This means assessing an overclock will require monitoring the performance of a card, along with continuing to look for traditional signs as those will still indicate problems in memory chips and the memory controller itself.
Ideally there would be a more absolute and expedient way to check for errors than looking at overall performance, but at this time AMD doesn’t have a way to deliver error notices. Maybe in the future they will?
Wrapping things up, we have previously discussed fast link re-training as a tool to allow AMD to clock down GDDR5 during idle periods, and as part of a failsafe method to be used with error detection. However it also serves as a tool to enable higher memory speeds through its use in temperature compensation.
Once again due to the high speeds of GDDR5, it’s more sensitive to memory chip temperatures than previous memory technologies were. Under normal circumstances this sensitivity would limit memory speeds, as temperature swings would change the performance of the memory chips enough to make it difficult to maintain a stable link with the memory controller. By monitoring the temperature of the chips and re-training the link when there are significant shifts in temperature, higher memory speeds are made possible by preventing link failures.
And while temperature compensation may not sound complex, that doesn’t mean it’s not important. As we have mentioned a few times now, the biggest bottleneck in memory performance is the bus. The memory chips can go faster; it’s the bus that can’t. So anything that can help maintain a link along these fragile buses becomes an important tool in achieving higher memory speeds.
Facebook
Twitter
RSS
327 Comments
View All Comments
JarredWalton - Thursday, September 24, 2009 - link
And what was the 8800 GTX Ultimate other than a pathetic clock-speed bump? After that we waited for the GT200 series which launched at $600. It took ATI to bring the price down, just like it took NVIDIA to bring the ATI prices down.NVIDIA stagnated while they were on top, just like ATI with the 9700/9800. NVIDIA made a huge misstep with the FX 5800 series, and ATI did the same thing with the X1800 series (and to a lesser extent the X800 parts). All companies have good and bad times. (Pentium 4 ring a bell? What about the Phenom?)
Your posts on this article have contributed nothing whatsoever other than ranting. Paper or hard launch? Paper is when *nothing* is out for a few weeks (or longer). If NVIDIA "launched" GT300 today, that would be paper. ATI has 5870 parts, albeit in limited quantities. GTX 275 certainly wasn't any better than this, but long term it all evens out.
And who cares about how long a company produced the better product? What matters is what they have now. Pentium 4 stunk in comparison to Athlon 64; does that mean no one should even consider Core 2 or Core i7? According to your "logic" that's exactly what we should do. Give it a rest; when NVIDIA launches GT300, we'll see what it can do. We'll also see if it can compete on pricing. Being fastest is only part of the battle, and anything over $300 is going to be a lower selling part.
SiliconDoc - Thursday, September 24, 2009 - link
Well you are ABSOLUTELY LYING about the GTX275 availability, PERIOD.Next, you didn't refute a single thing I said, but more or less came closer to agreement in many ways, but were WRONG, too.
Now you've decided you can half heartedly claim both sides do the same thing, and even throw in AMD and Intel, let's get to your continuing bias.
You couldn't resist "pathetic clock increase" for the GT8800 Ultimate (would love to see where you said that about the 4890, or the HD2900XTX) , failed to note the OVERPRICED ati card I pointed out, and in your absolute ignorance and CYA, think "stagnation" is something that occurs when "on top" instead of just the natural time it takes to move forward with new technology, after having just completed a round of it.
Once a company makes it "on top" they HAVE SPENT their latest and greatest new tech, and IT TAKES TIME TO GET TO THE NEXT LEVEL.
However, in YOUR MINDS, that is "stagnation". You offer ABSOLUTELY NO TIMETABLE TO EVEN REMOTELY "PROVE" your insane assertion.
You simply want it "ACCEPTED", which is about the DUMBEST theory one can imagine anyway, and I already pointed out EXACTLY why it is SO STUPID.
Let me tell you again, so IT CAN SINK IN FELLA!
When a company "makes it on top" they have just spent their latest greatest newest wad of technology" - AND IT TAKES TIME TO IMPROVE ON THEIR OWN ACHIEVEMENT !
In fact, they, having just OUTDONE the competition, are to be expected to "NOT COME UP WITH SOME MASSIVE NEW WIN" for the second time, in a row, and "quickly" - the SECOND time, as you fools expect, and even SAY SO, without direct words, because of course, you are FOOLISH and have BOUGHT THE SPIN, like 3rd graders who cannot think for themselves.
You basically "expect the impossible" - another leap forward right after the one just accomplished, before anyone else can even catch up.
YES, IT IS IMMENSLY IDIOTIC! Now you know!
---
You finally come to your senses a bit with: " All companies have good and bad times."
YES THEY DO. But not in your paranoid, conspiractorial, world of "stagnation" - once the top is reached. No, you expect a second miracle, in short order, and say so.
----
You also excuse ATI's bad times I pointed - by kicking yourself in the face doing it, negating your OTHER conspiracy rant " And who cares about how long a company produced the better product? "
Well, if that were actually the case for you, you wouldn't have screamed about stagnation once a company is on top, because obviously YOU DEEPLY CARE ABOUT WHO HAS THE BETTER PRODUCT, AND FOR HOW LONG.
Not only that, you claim, once they are there, they turn flaccid and lazy.... and boy it burns you up !
ROFL, you CONTRADICT YOURSELF, and haven't got a clue you're doing it. That of course, means, that I have just made a major contribution TO YOU, straightening out your wacked conspiracy thinking, that no doubt was induced and locked in by the constant red fan hatred for nvidia, here at this site, over several years, and on the net widely, as well. Not like here is unique.
---
Now, if you had sense, you'd be more likely to wonder why when some company is on top, that their competition cannot pass them up, or equal them, not "why they sit there stagnating" - meaning, in another sense, one we all relate to, it just drives you nuts the next thing isn't here already - because you, we, everyone wants the next greatest, and so, you BLAME the top dog for not fufilling your wish immediately, when, they just had, in fact, done so....
Yeah, there is NO END to how insane that rant of yours is, that the reds, widely repeat against Nvidia, and there is absolutely NO BASIS FOR IT AT ALL in reality.
Voo - Thursday, September 24, 2009 - link
Come on Jarred arguing with someone who actually believes"If either company dies, the other can move on, and there's very little chance that the company will remain stagnant, since then they won't sell anything, and will die, too."
won't do any good. I mean even my 13 year old nephew understands the basics of economy better than this guy.
I think every one in their right mind agrees that competition always leads to lower prices and more innovation.
Also I can't see where there should be any bias - things like the temp of the 2 ati cards are clearly stated in the article and everyone who can read graphs and the text should be able to get a clear picture of the new card.
Just because some people just read every other sentence doesn't mean the review is biased..
SiliconDoc - Thursday, September 24, 2009 - link
Here, let me point out another problem with your "basic understanding", which is the point you start at, remain at, and finish at:" won't do any good. I mean even my 13 year old nephew understands the basics of economy better than this guy.
I think every one in their right mind agrees that competition always leads to lower prices and more innovation. "
LET'S TAKE A CURRENT EXAMPLE: PhysX vs Havok vs Bullet Physics vs Pix - all various forms of in game "physics".
Well, what competition done with this ?
You might call it "innovation", but in this case, it should be called FRAGMENTATION, and STAGNATION - due to your "basic understanding" in economy, you can't fathom such a thing, because it doesn't apply to your pat cleche, which you can ATTACK unfairly with.
Now, if there was a MONOPOLY, ( which is what the red fans have been screaming for, a SINGLE STANDARD, thrust down the throats of all the card makers and game makers, they claim, "open standard" is the very best!), a real monopoly, not an EDICT from a "standards board", why we'd already havce advancement far beyond what we currently do with the fragmented players and implementations.
So, NO competition does not always lead to BETTER END USER expereince or FASTER technological implementation.
So much for you and your 13 year old's "understanding".
In this case, competition has led to fragmentation, and lack of implementation in games, and slower advancement, due to the competing players.
JarredWalton - Friday, September 25, 2009 - link
That was the worst "counter" to an argument I've ever read. Standards are not the same thing as a monopoly, and I don't even need to use all caps to get that point across. Standards are what we have with memory types, interfaces, and yes even graphics. A "monopoly" on graphics that has everyone move to one standard can be beneficial; certainly having four competing "standards" doesn't really help.Eventually, the market will select what works best. There used to be a question of OpenGL vs. Direct3D, and that discussion has all but ended. MS put the money and time into DirectX and actually improved it to the point where most programmers stopped caring about using the alternative.
That's why PhysX isn't gaining traction: it has to compete with Havok, which the vast majority of content creators appear to prefer. So NVIDIA can pay companies to use PhysX in games like Batman, but until they actually get people to willingly use their stuff instead of Havok (by improving PhysX), it's not going to "win". And what the companies really want is a standard that works on all hardware, so we're more likely to see OpenCL or Direct Compute take over instead of a proprietary PhysX API. Hence, our discussion in this article about how OpenCL and Direct Compute are promising APIs.
It's not fragmentation, any more than a choice between Chevron, Philips, BP, Texon, etc. is "fragmentation" of the oil industry. Just because one implementation isn't dominant doesn't mean the problem is because of competition. Eventually, some implementation will actually get it right and companies will go that route. Clearly that hasn't happened yet, and your beloved PhysX (two titles where it actually matters so far: Mirror's Edge and Batman) is losing based on merit and nothing else. If it was better, people would use it. End of discussion. I guess all the hyper intelligent programmers making amazing games are too stupid to realize how awesome PhysX is without getting help from NVIDIA. It's so great that they'll pay money to Havok to license that API rather than use PhysX for free.
A monopoly on hardware is a different matter, and again no one is screaming for a monopoly except perhaps for you. Nice job trying to add weight to your position by being a rabid fanboy and accusing the opposition of doing exactly what you're doing. If there is only one hardware vendor, what drives them to improve? Nothing but themselves, which leads to stagnation. It really is basic economics that's apparently too much for a fanboy to grasp. I'd like to see more CPU and GPU vendors (well, *good* vendors), but it's difficult to do properly and thus we remain with the current status quo.
Tell me this: how would it hurt anyone for Company X to enter the graphics market and make something that is clearly superior to ATI and NVIDIA offerings and is 100% compatible with current standards like DirectX and OpenGL? The only people that would potentially hurt would be ATI and NVIDIA employees and shareholders. Similarly, how would it hurt for Company Y to come out with a new API for physics that is clearly superior in every way to PhysX, Havok, etc? If it's better, it would become the new de facto standard. Having competition isn't the problem; the problem is competition between lousy options (i.e. GMC, Chrysler, Ford, and Chevy) when what we want is something better.
Now go ahead and use half-coherent ranting and capitals while you ignore everything meaningful in this post and put up another tirade about how stupid and horrible I am with no clear comprehension of reality. I'm done.
SiliconDoc - Friday, September 25, 2009 - link
What brought this on, dummy ?" A "monopoly" on graphics that has everyone move to one standard can be beneficial; certainly having four competing "standards" doesn't really help. "
Your sainted competition brought it on, you fool. You go on to claim "eventually one standard will be adopted", but by then the MONOPOLY POWER will have won in it's forcing it's choice DOWN EVERYONE ELSE'S THROAT.
The problem YOU HAVE, is you want YOUR MONOPOLY choice, and you want to claim, as you did, the "competition" isn't innovation, it's just plain bad, and for you, that is of course, being the reg rager you are, PhysX, which is clearly superior to any of the others.
BUT, you want YOUR CHOICE FORCED on EVERYONE ( like MSFT and it's xbox pushing that has driven the console implementations with a giant was of cash! LOL ), then you can blandly call it "a standard", and claim it's "the best choice", because "the market decided" in your brainwashed moron manner (because you don't like NVidia pushing but pretend when another player does so "it's innocent and natural" and "happened without such preszure".
ROFLMAO - boy you are sure a tool.
The funniest part of your current STUPIDITY, is that MICROSOFT THE MONOPOLY, has decided to push HAVOK for it's 360 and as competitive lockout against Larrabee competition, hence you LIE as a MONOPOLY uses power to force more crap into the developer channels. roflmao
But of course, the mind controlled by the standard lies is all you've shown in all your commentary.
Here are the 2 games, BTW.
Game Title Developer Platform
2 Days to Vegas Steel Monkeys PC
10 Balls 7 Cups Graveck iPod
50 Cent: Blood on the Sand Swordfish Studios PC
Adrenalin 2: Rush Hour Gaijin Entertainment PS3, X360
Age of Empires III Distineer Studios PC, Mac
Age of Empires III: The Asian Dynasties Distineer Studios Mac
Age of Empires III: The WarChiefs Distineer Studios Mac
Age of Pirates: Captain Blood 1C: Sea Dog PC, X360
Aliens: Colonial Marines Gearbox Software PC, PS3, X360
Alliance of Valiant Arms Redduck PC
Alpha Prime Black Element Software PC
American McGee's Grimm Spicy Horse PC
APB Realtime Worlds PC, PS3, X360
Army of Two Electronic Arts PS3, X360
Auto Assault Net Devil PC
AutoFans AP-Games PC
B.A.S.E. Jumping Digital Dimentions PC
Backbreaker Natural Motion PC, PS3, X360
Beowolf Ubisoft X360
Bet on Soldier: Blackout Saigon Kylotonn Entertainment PC
Bet on Soldier: Blood of Sahara Kylotonn Entertainment PC
Bet on Soldier: Blood Sport Kylotonn Entertainment PC
Big Fun Racing Decane iPod
Bionic Commando GRIN PC, PS3, X360
Bionic Commando: Rearmed GRIN PC, X360
Bladestorm: The Hundred Years' War Koei PS3, X360
Borderlands Gearbox Software PC, PS3, X360
Bourne Conspiracy High Moon Studios PS3, X360
Brothers in Arms: Hell's Highway Gearbox Software PC, PS3, X360
Buble Bang Decane iPod
Caribbean Legends Seaward.Ru Team PC
Cellfactor: Combat Training Artifical Studios, Immersion Games PC
Cellfactor: Revolution Artifical Studios, Immersion Games PC
Champions Online Cryptic Studios PC
City of Villains Cryptic Studios PC
Clive Barker's Jericho MercurySteam Entertainment PC, X360
Cluth Targem Games PC
Cosmosis Midnight Status iPod
Crazy Machines II FAKT Software PC
Crusaders: Thy Kingdom Come Neocore PC
Cryostasis Action Forms PC
Dark Sector Digital Extremes PC, PS3, X360
Dark Void Airtight Games PC, PS3, X360
Darkest of Days Phantom EFX PC, X360
Debris Midnight Status iPod
Destroy All Humans! Path of the Furon Sandblast Games X360
Divinity 2: Ego Draconis Larian Studios PC, X360
Dracula Origin Frogwares PC
Dragon Age: Origins EA PC
Dragonshard Atari PC
Driver :: Test Squad Interactive Media PC
Drop Point: Alaska Bongfish Interactive Mac
Dungeon Hero Firefly Studios PC, X360
Dusk 12 Orion PC
Empire Above All IceHill PC
Empire Earth III Mad Dog Software PC
Empire Total War The Creative Assembly PC
Entropia Universe MindArk PC
Evil Resistance: Morning of the Dead Openoko Entertainment PC
Fahr Simulator 2009 Astragon Software PC
Fairy Tales: Three Heroes Cats Who Play
Fallen Earth Icarus Studios PC
Fatal Inertia KOEI PS3, X360
Frontlines: Fuel of War Kaos Studios PC, PS3, X360
Fury Auran Games PC
G.B.R. The Fast Response Group OPenoko Entertainment PC
Gears Of War Epic Games PC, X360
Gears of War 2 Epic Games X360
Gluk'Oza: Action GFI Russia PC
GooBall Ambrosia software Mac
Gothic 3 Piranha Bytes PC
Grind WebGames3D.com iPod
GTown Interactive Community 2.0 9you.com PC
Gunship Apocalypse FAKT Software PC
HAZE Free Radical Design X360
Heavy Rain Quantic Dream PC
Helldorado: Conspiracy Spellbound Entertainment PC, PS3
Hero's Journey Simutronics PC
Hour of Victory nFusion Interactive X360
Hunt, The Orion PC
Huxley Webzen, Inc PC, X360
I-Fluid Exkee PC
Infernal Metropolis Software PC
Inhabited Island: Prisoner of Power Orion PC
Joint Task Force Most Wanted Entertainment PC
Kingdom Under Fire: Circle of Doom Blueside Inc. X360
Kran Simulator 2009 Astragon Software PC
Kuma\WAR Kuma Reality Games PC
Landwirtschafts Simulator 2008 Astragon Software PC
Landwirtschafts Simulator 2009 Astragon Software PC
The Last Remnant Squre Enix PC, X360
Legend: Hand of God Anaconda Games PC
Legendary Spark Unlimited PC, PS3, X360
Lost Odyssey Mistwalker X360
Lost: Via Domus Ubisoft PC, PS3, X360
Mafia 2 Illusion Softworks PC, PS3, X360
Magic ball 3 Alawar Entertaiment PC
Magic ball 4 Alawar Entertaiment PC
Mass Effect BioWare PC, X360
Medal of Honor: Airborne EA Los Angeles PC, X360
Metal Knight Zero Online ObjectSoftware Limited PC
Metro 2033 4A Games PC
Minotaur China Shop Flashbang Studios PC
Mirror's Edge DICE PC, PS3, X360
Mobile Suit Gundam: Crossfire BEC PS3
Monster Madness: Battle for Suburbia Artificial Studios PC, X360
Monster Madness: Gravedigger Artificial Studios PS3
Monster Truck Maniax Legendo Entertainment PC
Mortal Kombat vs. DC Universe Midway PS3, X360
Mascow Racer IRS Games PC
Myst Online: URU Live Cyan Worlds PC
Need for Speed: Shift Electronic Arts PC
Nights: Journey of Dreams SEGA Wii
Night of a Million Billion Zombies PowerUP Studios PC
Nurien Nurien Software PC
Open Fire BlueTorch Studios PC
Parabellum ACONY PC, PS3, X360
Paragraph 78 Gaijin Entertainment PC
Physix Michael Wuhrer iPod
Pirate Hunter DIOsoft PC, X360
Pirates of the Burning Sea Flying Lab Software PC
Point Blank Barunson Interactive PC
Prey 2 Human Head PC, X360
PT Boats: Knights of the Sea Akella PC
Pyroblazer Eipix PC, Wii
QQ Speed Tencent Inc. PC
Rail Simulator Kuju Entertainment Ltd PC
Red Steel Ubisoft Paris Wii
Rise Of Nations: Rise Of Legends Big Huge Games PC
Rise of the Argonauts Liquid Entertainment PC, PS3, X360
Roboblitz Naked Sky Entertainment PC, X360
Rocket Bowl 21-6 Productions X360
Rock'n'Roll Dice 3DA Interactive iPod
Rush Hour: Streets of Moscow Gaijin Entertainment PC
Sacred 2 ASCARON Entertainment PC
Shadow Harvest Black Lion Studios PC, X360
Shadowgrounds Survivor Frozenbyte PC
Shattered Horizon Futuremark Games Studio PC
Sherlock Holmes vs. Arsene Lupin Frogware Games PC
Sherlock Holmes: The Awakened Frogwares Game Development Studio PC
Showdown: Scorpion B-COOL Interactive PC
Silverfall Monte Cristo PC
Silverfall: Earth Awakening Monte Cristo PC
SkylineBlade Midnight Status iPod
Sledgehammer Targem Games PC
Sovereign Symphony Ceidot Game Studios PC
Sonic and the Black Knight SEGA Wii
Sonic and the Secret Rings SEGA Wii
Space Race SARGE Games iPod
Space Siege Gas Powered Games PC
Spectraball Flashcube Studios PC
Speedball 2 Kylotonn Entertainment PC
Squashem Jelly Biscuits iPod
Stalin Subway, The Orion PC
Star Tales QWD1 PC
Stoked Bongfish Interactive Entertainment X360
Stoked Rider: Alaska Alien Bongfish Interactive Entertainment PC
Streets of Moscow Gaijin Entertainment PC
Strike Ball 3 Alaware Entertainment PC
Stuntmanbob potatocows.com iPod
Supersonic Acrobatic Rocket-Powered Battle Psyonix Studios PS3
The Swarm Targem Games PC
Switchball Atomic Elbow PC
Tank Universal Dialogue Design PC
Tension Ice-pick Lodge PC
Terminator Salvation GRIN PC
Tom Clancy's Ghost Recon Advanced Warfighter GRIN PC, X360
Tom Clancy's Ghost Recon Advanced Warfighter 2 GRIN, Ubisoft Paris PC, X360
Tom Clancy's Rainbow Six Vegas Ubisoft Montreal PC, PS3, X360
Tom Clancy's Rainbow Six Vegas 2 Ubisoft Montreal PC, PS3, X360
Tom Clancy's Splinter Cell: Double Agent (multiplayer) Ubisoft Shanghai PC, X360
Tortuga: Two Treasures Ascaron Entertainment PC
Trine Frozenbyte PC, PS3
Tunnel Rats Replay Studios PC
Turning Point: Fall of Liberty Spark Unlimited PC, PS3, X360
Turok Propaganda Games PC, PS3, X360
Two Worlds Reality Pump PC, X360
Two Worlds: The Temptation Reality Pump PC, X360, PS3
Underwater Wars Biart Studio PC, X360
Ultra Tubes Eipix PC
Unreal Tournament 3 Epic Games PC, PS3, X360
Unreal Tournament 3: Extreme Physics Mod Epic Games PC
Urban Empires Radioactive Software PC
U-WARS Biart Studio PC, X360
Valkyria Chronicles SEGA PS3
Virtual Tennis 3 SEGA PS3, X360
Viva Pinata: Party Animals Krome Studios X360
W.E.L.L. Online Sibilant Interactive PC
Wanted: Weapons of Fate GRIN PC, PS3, X360
Warfare GFI Russia PC
Warmonger: Operation Downtown Destruction Net Devil PC
Watchmen: The End is Nigh Deadline Games PC, PS3, X360
Way of the Samurai 3 Aquire X360
Welkin 4591 Outpop Digital PC
Winterheart's Guild Zelian Games PC, X360
WorldShift Black Sea Studios PC
X-Razer Rayd GmbH iPod
X-men Origins: Wolverine Raven Software PC
--
ROFLMAO 2 games...
--
Larrabee will use the x86 instruction set with Larrabee-specific extensions
Larrabee will include very little specialized graphics hardware, .... using a tile-based rendering approach
---
Larrabee's early presentation has drawn some criticism from GPU competitors. At NVISION 08, several NVIDIA employees called Intel's SIGGRAPH paper about Larrabee "marketing puff" and told the press that the Larrabee architecture was "like a GPU from 2006".[8] As of June 2009, prototypes of Larrabee have been claimed to be on par with the nVidia GeForce GTX 285.[9]
http://en.wikipedia.org/wiki/Larrabee_(GPU)">http://en.wikipedia.org/wiki/Larrabee_(GPU)
---
So in this case we have 3 warring parties (your beloved "beneficial" competition), and endless delays, lack of game developement and content because of that, and the score won't be settled till the MONOPOLY POWER sets "the standard" (opencl you hope it seems/ aka JAVA for physx, or anything so long as it isn't PhysX, right?) as you call it, and even then, with the nature of game coding, it is highly likely that more than one type and implementation will widely survive. The "market competition" picked VHS over BETA, and nearly everyone calls that a mistake to this day (psst, their were powerful players behind the scene just like in the physics game wars).
What really happpens in what we're talking about is POWER picks what is brought forth for all, and you should well know instead of pretending the lie that you have, that OFTEN in this computing world something worse is shoved down everyone's throats because of that.
Your infantile "pure minded rhetoric" is just that, a big pile of BS, as usual.
SiliconDoc - Friday, September 25, 2009 - link
PS - I quite understand in your "only framerates matter" deranged high end video red rager gaming card mindset, THE ONLY GAMES THAT MATTER FOR YOU IN YOUR BS ARGUMENT are PC games that wind up on ANANDTECH PC videocard reviews. ROFLMAOYES IT'S TRUE!
Hence "two games!", only for PC, nothing else, is "your standard".
And IT'S DERANGED, given the facts.
JarredWalton - Friday, September 25, 2009 - link
Unlike you, obviously, I've tried NVIDIA and ATI, and both are fine. I've never suggested I want ATI to win, and I don't know why you continue to think that. PhysX is "used" in tons of games... where does it actually matter? How many of the games you list sold more than 100K copies? How many are actually good games? How often does it make a discernible and positive difference? (More trash flying around isn't really better.)Let's also not count chickens before they hatch and remove games that haven't even shipped. You know, sort of like removing GT300 from benchmark comparisons until it's actually available.
That list includes games that had super lame PhysX (all the Tom Clancy titles for sure), games that are completely trivial (skeeball anyone?), games where it degrades performance relative to not enabling it (umm, that's most of the titles). Unreal Tournament 3 has PhysX support... but only on the released-after-the-patch levels, and even then only the Tornado level is actually impressive visually. Almost no one played/plays these levels.
Since you've got so much time to promote NVIDIA, tell us all which games on this list are "Must Haves" and make good use of PhysX. I said there were "two games where it has mattered: Batman and Mirror's Edge". Now put your fanboy hat on and tell us which games in that list. I'm sure that 50 Cent, 10 Balls 7 Cups, Jericho, Cellfactor, Rock'n'Roll Dice, and Untra Tubes are at the top of the sales/preorder charts!
SiliconDoc - Friday, September 25, 2009 - link
The other SEVERE PROBLEM your massive bias holds is this:-
Just a few months ago, HERE, PhysX was given a run in Mirror's Edge, and NEVER BEFORE SEEN or IMPLEMENTED effects were present.
Anand loved it, couldn't get away...from the computer, as he said.
The Master declared it.
--
But, when EMERGING TECHNOLOGY from the card company you must absolutely HATE comes forth, for you as a gamer, an advantage even, you have nothing but a big pile of dung for it.
---
Get over there to the other fellow in the discussion and point out how competition brings innovation, right, and that PhysX IS INNOVATION !
---
( Oh, that's right, after your tagteam preached it, you already breached it, and blew cookies all over your economic lessons.)
LOL -= hahahha -
--
I guess this is another case of "NVidia stagnating" in BOTH your minds. (Yes, of course it is)
I do hope your conditions clear up.
JarredWalton - Friday, September 25, 2009 - link
Actually, Anand said Mirror's Edge with PhysX was the first (and at the time only) title where it made a palpable difference to the point where turning it off made him miss it somewhat. Actually, http://www.anandtech.com/video/showdoc.aspx?i=3539..." target="_blank">here are his exact words. He also goes on to discuss such things as Havok porting to OpenCL and how that won't happen for PhysX. Thanks master of hyperbole; taking things out of context is the sure sign of a weak argument. But yes, NVIDIA was highly "innovative" when they bought out a competitor because they couldn't do any better -- a competitor that to date had released hardware no one wanted and a few titles that didn't matter.You're so set on making me an ATI fanatic and throwing about words like hate and sadness and whatever. It's pathetic and funny that you're so delusional that you could even pretend to think that way. I mean, obviously you don't really think that and you're just some troll trying to stir up crap, but it boggles the mind that you have this much energy to put into spewing vitriol.
Love ATI? Hardly. I've ripped on their mobile components quite thoroughly for the past two years. After all, I review laptops so that's my area of expertise, and up until HD 46xx they had nothing compelling on laptops for years. Even the 4000 series on laptops is marred by their lack of mobile reference drivers, something I've praised NVIDIA for releasing (after saying it was absolutely necessary for the year or two before it happened).
So yes, put your blinders on and act as though you have any idea whatsoever about what people think. Someone disagrees with you and they become spawn of satan, worshiping all that is ATI. It's a reflection of your own insecurities that you can't accept the good of the competitor while at the same time pointing out flaws. Go check into a mental institute, or head back over to nZone and be secure with others that can't be objective when it comes to graphics cards.