Twilight of the GPU

[FarCry]

Tim Sweeney

 

Goodbye, graphics APIs

 

 

JS: I'd like to chat a little bit about Larrabee and software rendering. I'm sure you're NDA'd on it, but Intel just did a pretty substantial reveal so we can talk in more detail about it. So first off, I'm wondering if you're looking at any of the Larrabee native stuff. What do you think about the prospects of this whole idea of not doing Direct3D or OpenGL, but writing directly to Larrabee's micro-OS?

 

TS: I expect that in the next generation we'll write 100 percent of our rendering code in a real programming language—not DirectX, not OpenGL, but a language like C++ or CUDA. A real programming language unconstrained by weird API restrictions. Whether that runs on NVIDIA hardware, Intel hardware or ATI hardware is really an independent question. You could potentially run it on any hardware that's capable of running general-purpose code efficiently.

 

JS: So you guys are just going to skip these graphics APIs entirely?

 

TS: That's my expectation. Graphics APIs only make sense in the case where you have some very limited, fixed-function hardware underneath the covers. It made perfect sense back with the 3Dfx Voodoo and the first NVIDIA cards, and the very first GeForces, but now that you have completely programmable shaders, the idea that you divide your scene up into triangles rendered in a certain order to a large framebuffer using fixed-function rasterizer features is really an anachronism. With all that general hardware underneath, why do you want to render scenes that way when you have more interesting possibilities available?

 

http://arstechnica.com/articles/paedia/gpu-sweeney-interview.ars/1

[tmjuju]

Ένα σχεδόν συναφές…

 

http://uk.pc.ign.com/articles/911/911791p1.html

 

GC Asia 2008: Crytek Hearts PlayStation 4

The next generation of consoles will be the "catalyst for a renaissance" in gaming graphics.

by Cam Shea, IGN AU

 

Australia, September 19, 2008 - During yesterday's GC Asia keynote speech on the future of gaming graphics, the president and CEO of Crytek, Cevat Yerli emphasised several times how important the next generation of consoles (i.e. PS4 and Xbox 360 II: Electric Boogaloo) will be to a renaissance in gaming graphics that he predicts will arrive in 2011-2012. Not only will the company link its technology development with the console cycles, but it sees the PlayStation 4 as the most important machine of the coming generation, with Yerli stating that "frankly, we are linking ourself to what the PS4 will take up as [its hardware] position; that is the console makeup that will drive all our research, so we cannot wait to know what the [machine] will be about. That's probably the most important missing information we have right now in our company."

 

So, what's this graphics renaissance he's talking about? Yerli believes that – given that the Xbox 360 and PS3 will be around for a few more years – it will be today's technological framework that drives graphics for the next three years, but come 2011-2012 – and new consoles - there will be a huge shift. Sure, processing power and hardware sophistication will increase as they always do, but it's the ability to move beyond today's rasterisation rendering approach that Yerli believes will spark a renaissance. The games of the future will mix and match between different graphics technologies, whether that be ray tracing, ray casting, point-based rendering, voxel-based approaches or something else. In five years Yerli predicts that real-time gaming graphics will match CGI offline rendering.

 

Interestingly, Yerli also emphasised the importance of style and a unique visual approach, as opposed to brute force rendering, in the future of videogames. He also said that developers would be better off making their games "believable, but not photorealistic"; that they should bet on physics, simulation and AI.

 

He also took a moment to comment on the negative reaction of gamers who found that they simply couldn't run Crysis with its settings maxed out, telling the audience that running at those settings at launch was never the team's intent – those settings were meant to future proof the game so that it would continue to look cutting edge as new hardware emerged. Contrary to popular belief, the team actually built the game to run on PCs from 2004 onwards.

[De@th L0rd]

Ενδιαφέρον....Λέτε να δούμε εκρηκτικές επιδόσεις με φθηνές κάρτες;Η θα γίνουν σαν τα windows οι drivers 500 ΜΒ bloat ο καθένας ;

[Ailuros]

Tim Sweeney

 

Goodbye, graphics APIs

 

 

JS: I'd like to chat a little bit about Larrabee and software rendering. I'm sure you're NDA'd on it, but Intel just did a pretty substantial reveal so we can talk in more detail about it. So first off, I'm wondering if you're looking at any of the Larrabee native stuff. What do you think about the prospects of this whole idea of not doing Direct3D or OpenGL, but writing directly to Larrabee's micro-OS?

 

TS: I expect that in the next generation we'll write 100 percent of our rendering code in a real programming language—not DirectX, not OpenGL, but a language like C++ or CUDA. A real programming language unconstrained by weird API restrictions. Whether that runs on NVIDIA hardware, Intel hardware or ATI hardware is really an independent question. You could potentially run it on any hardware that's capable of running general-purpose code efficiently.

 

JS: So you guys are just going to skip these graphics APIs entirely?

 

TS: That's my expectation. Graphics APIs only make sense in the case where you have some very limited, fixed-function hardware underneath the covers. It made perfect sense back with the 3Dfx Voodoo and the first NVIDIA cards, and the very first GeForces, but now that you have completely programmable shaders, the idea that you divide your scene up into triangles rendered in a certain order to a large framebuffer using fixed-function rasterizer features is really an anachronism. With all that general hardware underneath, why do you want to render scenes that way when you have more interesting possibilities available?

 

http://arstechnica.com/articles/paedia/gpu-sweeney-interview.ars/1

 

Πριν αποχαιρετήσουμε οποιοδήποτε ΑΡΙ θα πρέπει πρώτα οι δημιουργοί να μπορούν να γράφουν κωδικό σε C++ ή οποιαδήποτε εναλλακτική γλώσσα προγραμματισμού. Προφανώς αυτή τη χρονική στιγμή δεν είναι εφικτό και δεδομένου ότι δεν ξέρουμε ακόμα τι θα προσφέρουν τα μελλοντικά ΑΡΙ (πέραν του D3D11) είναι λίγο δύσκολο να κάνουμε οποιεσδήποτε προβλέψεις. Ο Sweeney έχει πέσει τόσες φορές έξω στις προβλέψεις του που δεν είναι ούτε καν αστείο πια. Αν είχαν βγει σωστοί παλιότεροι ισχυρισμοί του θα είχαμε επιστρέψει προ πολλού στο software rendering.

-----Προστέθηκε 25/9/2008 στις 10 : 40 : 44-----

Ένα σχεδόν συναφές…

 

΄Αλλο όμως η κονσόλα και άλλο το PC. Στην πρώτη περίπτωση έχεις μια "κλειστή" παιχνιδομηχανή. Και πριν μπω σε περισσότερες λεπτομέρειες, το XBox360 έχει κερδίσει τις καρδιές των δημιουργών παρότι του "μιλάνε" με DirectX εν αντιθέση με το PS3, το οποίο χρησιμοποιεί οτιδήποτε σε OpenGL/OGL_ES και παρότι φέρει τον κατά πολύ ελπιδοφόρο Cell του παρελθόντος μας αποδείχτηκε για άλλη μια φορά πιο πολύπλοκος στον προγραμματισμό.

 

During yesterday's GC Asia keynote speech on the future of gaming graphics, the president and CEO of Crytek, Cevat Yerli emphasised several times how important the next generation of consoles (i.e. PS4 and Xbox 360 II: Electric Boogaloo) will be to a renaissance in gaming graphics that he predicts will arrive in 2011-2012. Not only will the company link its technology development with the console cycles, but it sees the PlayStation 4 as the most important machine of the coming generation, with Yerli stating that "frankly, we are linking ourself to what the PS4 will take up as [its hardware] position; that is the console makeup that will drive all our research, so we cannot wait to know what the [machine] will be about. That's probably the most important missing information we have right now in our company."

 

Πολύ τολμηρή άποψη δεδομένου ότι στην ΙΔΙΑ παράγραφο παραδέχονται ότι δεν έχουν ΙΔΕΑ τι θα περιέχει η κάθε κονσόλα. ΄Εχω ακούσει παπαριές και παπαριές αλλά τέτοια γελοία οξύμωρα σχήματα θα πρέπει να τα εκτυπώνω σε πανό για να έχω να γελάω.

 

So, what's this graphics renaissance he's talking about? Yerli believes that – given that the Xbox 360 and PS3 will be around for a few more years – it will be today's technological framework that drives graphics for the next three years, but come 2011-2012 – and new consoles - there will be a huge shift. Sure, processing power and hardware sophistication will increase as they always do, but it's the ability to move beyond today's rasterisation rendering approach that Yerli believes will spark a renaissance. The games of the future will mix and match between different graphics technologies, whether that be ray tracing, ray casting, point-based rendering, voxel-based approaches or something else. In five years Yerli predicts that real-time gaming graphics will match CGI offline rendering.

 

Χαχαχαχαχα για το τελευταίο. Δεν θα αμφισβητήσω ότι τα παιχνίδια μετά το 2011-12 πιθανότατα να έχουν μια μικρή δόση ray tracing για αντανακλάσεις για παράδειγμα όπως χρησιμοποιεί σήμερα η Pixar αλλά αυτό δεν σημαίνει σε καμία περίπτωση ότι για τότε το real time gaming graphics θα έχει καμία σχέση με CGI offline rendering.

 

Xαμηλώστε λίγο τους τόνους να χαρείτε Κύριοι. Εδώ το μπουρδο-Crysis που βγάλατε μόλις που θα τρέχει στα μέγιστά του εκείνη την περίοδο (χάρη στην ανικανότητά σας να εξοικονομείτε πόρους στον κωδικό), και όχι δεν θέλω υποσχέσεις του τύπου ότι το Wall-E θα τρέχει σε πραγματικό χρόνο σε GPU ή κονσόλα του μέλλοντος.

 

Interestingly, Yerli also emphasised the importance of style and a unique visual approach, as opposed to brute force rendering, in the future of videogames. He also said that developers would be better off making their games "believable, but not photorealistic"; that they should bet on physics, simulation and AI.

 

Δεν θα διαφωνήσω εδώ καθόλου. Διαφωνώ κάθετα ωστόσο με τυχαίες προβλέψεις που αφορούν το λιγότερο προβλεπόμενο μέλλον ΧΩΡΙΣ ίχνος στοιχείων από τι θα αποτελείται μελλοντικό hardware. Αν είχαμε σήμερα ένα δείγμα αυτών το οποίο θα είχαν αναλύσει και θα είχαν τρέξει δοκιμαστικά θα έλεγα ναι οι άνθρωποι έχουν δίκιο. Με τόσα ερωτηματικά και αστάθμητους παράγοντες οποιαδήποτε πρόβλεψη είναι παρακινδυνευμένη και γελοία το λιγότερο. Και για του λόγου το αληθές η Microsoft δεν έχει αποφασίσει ακόμα τι θα βάλει σε μελλοντικές κονσόλες πόσο μάλλον η SONY για την οποία υποψιάζομαι ότι και πάλι θα είναι πολύπλοκη στον προγραμματισμό όπως παραδοσιακά όλες οι κονσόλες τους.

 

He also took a moment to comment on the negative reaction of gamers who found that they simply couldn't run Crysis with its settings maxed out, telling the audience that running at those settings at launch was never the team's intent – those settings were meant to future proof the game so that it would continue to look cutting edge as new hardware emerged. Contrary to popular belief, the team actually built the game to run on PCs from 2004 onwards.

 

Να τρέξουμε τότε το Crysis σε μια Χ800 ή 6800 να δούμε με τι ρυθμίσεις θα καταφέρει να βγάλει έστω 30fps μέσο όρο και σε τι γελοία ανάλυση. Θα ήταν πολύ πιο έντιμο να παραδεχτούν ότι το παραχέσανε λιγάκι με τις προσδοκίες τους και υπάρχουν πολλά μα ΠΑΡΑ ΠΟΛΛΑ που θα μπορούσαν να είχαν αποφεχθεί.

-----Προστέθηκε 25/9/2008 στις 10 : 45 : 46-----

Ενδιαφέρον....Λέτε να δούμε εκρηκτικές επιδόσεις με φθηνές κάρτες;

 

Μια λέξη: ΠΟΤΕ

 

Η θα γίνουν σαν τα windows οι drivers 500 ΜΒ bloat ο καθένας ;

 

΄Οτι θα υπάρχει μια ελαφριά μετατόπιση του "κέντρου βάρους" στο λογισμικό αρχής γενομένης με το D3D11 θέλω να το θεωρήσω δεδομένο. Η διαφορά στο τέλος είναι αν κάποιος πατάει με τα πόδια του γερά στη γη ή φαντάζεται ότι είναι στο σύννεφο νούμερο 9.....