games are supposed to be FUN with a capital F. Think a minute about the games you TRULY love, the ones you go back to, do you go back to them for the graphics?
At the moment I have both Oblivion and Skyrim games going on the PS3, and to be honest, Oblivion is more fun even though Skyrim is way shinier. Why? It's hard to say, maybe Bethesda just dumbed down the attributes system too much and the abilities tree is way too close to FFXIII, evoking bad memories for me. The new "finishing moves" - though impressive - are anti-immersive. It's like the computer is having fun instead of you. The magic system - yes, dual casting, great. Sucky way of selecting spells, even worse than Oblivion's sucky way. Boxes... just don't look like boxes inside, not at all.The character designer is way dumbed down, why did they do that? And the world, though bigger and prettier, just isn't as densely populated with interesting, emergent behavior. In Oblivion it is always fun to go wandering off the beaten path, in Skyrim not so much.There are a lot more walls in your way, and insurmountable obstacles. And cliffs where you just die just by going out of bounds, why is that supposed to be fun? Well I don't know, maybe it is just what I got used to, but I don't really care whether I finish Skyrim or not, whereas Oblivion was completely addictive. Progress? Hmm. I couldn't wait for Elder Scrolls five, it was the only reason I kept the PS3. But I can wait for Elder Scrolls six. Forever, thanks.
If you're saying that a $325/year investment in multi-purpose hardware is too much, but dropping $60 at a shot to play modern games is cheap, you have very, very weird budgeting.
Only for older games with simplistic lighting models. As Sweeny says, just load up your shader with subsurface scattering logic and see if your GPU isn't the bottleneck.
Libraries, and the commons. These algorithms will slowly but surely enter the public domain. How long before we see the equivalent of Speed Tree as a GPL library? Hard to say. Not more than 5 years from now, that is safe to say. Maybe a lot less.
Next gen consoles will be sufficiently more powerful than a low end PC, and hopefully easier to use...
Not easier to use than Android or iOS, with which the entire target market will already be familiar.
...such that they will retain the market of "I just want it to work".
So you think the last generation of consoles "just worked"? Reality says otherwise, ranging from mysterious hardware failures to awkward configuration to frequent network outages that are still common. There is no way Xbox or PS/3 ever "just worked".
...And mobile is never going to match the experience of the big AA titles.
Famous last words. Big Media goes where the money is, and there is way more money in the pockets of people with expensive smartphones than Gen Y's playing Halo every day on the recroom sofa.
Mobile is the place to be if you're small, indie dev, but the money is on consoles...
Today it is, or it was. Truthfully, which franchise do you think is more valuable today, Elder Scrolls or Farmville? (Can you spell "network"?)
no[t] on PC (easier to provide a consistent experience for).
The PC game market has shrunk (not overall, but as a fraction of the whole) for one reason only: Microsoft. Microsoft is on the verge of losing control of the PC market because those hangers on who can't imagine switching away from Microsoft's dull and awkward user interfaces will in future not be able to imagine switching away from their smart phone interface. The rise of ARM desktops running Android will be the big story over the next 5 years, by coincidence, the same period in which Microsoft and Sony will be arming for their next console war. Which by the way, they will both lose. They came awfully close to crashing last generation and only infinite bank accounts augmented by boneheaded ego-driven strategic misdirection from the executive suite allowed either of Xbox or PS3 to survive. Close to the edge last time... straight over the edge this time.
If you can't get console sales for a Action ish title (skyrim/mass effect/any FPS/sports games etc.) then you can't get enough money to warrant having made the game.
What your are missing is the fact that the market segment that is willing to put 100 hours into a game is a small fraction of the market. Yes, Elder Scrolls 6 will be a big title, but whatever Bethesda brings out on smart phones will be much, much bigger. Notice John Carmack, who started all that, has already read the writing on the wall, can you?
if you're not moving a million copies on Xbox2/PS3 combined you're not making a game at this point.
A million copies? Feh. Future game distribution numbers will be measured commonly in the tens of millions, and sometimes, hundreds of millions. That is only possible in the the mobile market, big clunky consoles tied to furniture-style displays will lag far behind. Now, if you think that next year's smart phone won't be able to run Skyrim, you are just out of touch with hardware capabilities. Plus, Skyrim isn't where the money is, Farmville is.
...Indie titles will never be able to compete on average with big titles.
More famous last words. One word in answer: Steam.
Big money means voice acting, professional scoring, huge amounts of art...
...all of which the indie community has proved beyond a shadow of a doubt it is capable of producing itself without help from Big Media. If you doubt me then you better surf on over to Youtube and watch Sintel.
Yes, you can have the odd indie title that does crazy well (think minecraft) but the vast majority of them will be aiming to break even by flooding the market.
Only a few years back IBM said Intel is ahead of everyone else by 1-2 years and will remain so for the foreseeable future.
My point is, there is a real chance of a 1-2 year hiatus this time round where TSMC is shipping 20 nm while Intel is still at 22 nm and does not have 14 nm fully ready. Sure, Intel might still be "ahead" technically, but during such a hiatus that is just pie in the sky.
The real question is who hits mass production first. Intel is a powerhouse when it comes to fabs.
No question about it. But so is the Asian semiconducter business, and so far Intel never had to compete head to head because AMD has the only credible superscalar x86 design outside Intel, and Intel had effective means of preventing AMD from building the fabs it needed to compete effectively. Two words: air supply. That game just changed.
I guess NVidia must be getting awfully close to [open-sourcing their drivers]
Why did you put words in my mouth, then do the straw-man thing? What I meant - for anyone who did not see it clearly from context - is that NVidia must be close to opening their register specs.
They are doomed. Their multibillion development costs will never be recovered when they get slaughtered in the market, crushed between mobile platforms and vastly more capable discrete GPUs. The rise of AAA idie titles won't help much either.
Hats off to your incisive analysis. I certainly will not make the mistake of buying an Atom-based computer again. Atom runs too hot for the amount of computing it does. Against ARM it is just no contest. The only thing it has going for it is x86 compatability, and I guess I would prefer to get that more of that for less money from AMD, bundled with a decent GPU.
One thing to add: OpenCL is a game changer. It shifts the multi-core equation onto the GPU. Four cores? Feh, how about 80, or 800, all cranking single or double precision floating point in SIMD. When that truly takes hold, as in common applications restructured to utilize parallel resources automatically, the role of the monolithic central CPU will be further compartmentalized. Basically, you will only need monster firebreathing massively superscalar cores for legacy code that hasn't been updated to support parallelism, or those rare applications that just can't be serialized or care a lot about submicrosecond latency. Yes, they exist (financial transaction processing anyone?) but they are rare. I look forward to gcc -j1000 on my GPU.
Not true at all. But when they do fall behind, which has happened several times (e.g. the slow move to serial interconnect) they will spend any amount of money and employ any means, fair or foul to catch up.
This time round though, AMD has a very real chance to be on a 20nm process while Intel spends a year or two at 22nm. That is because TSMC has already taped out ARM15s at 20 nm, while Intel has a long way to go on its 14 nm process. That will go a long way towards levelling the playing field. Hmm, interesting that AMD has also mumbled about going "ambidextrous". I would not mind having an ARM core in an AMD part, perhaps to keep the box running in ultra low power mode without actually suspending.
Does that mean Nvidia gonna open source the driver for the graphic cards using Nvidia chips?
I don't think they will ever open-source their drivers. It would be embarrassing for them when others discuss their code, they are protective of their work, etc. All you can hope for, and what you should be demanding, is that they give more specs to the nouveau team.
Of course that is all that the community wants. Why should NVidia have all the fun of writing kick-butt driver code?:-p
I guess NVidia must be getting awfully close to taking that step. I would say, just waiting for a suitably high profile occasion to announce it now. Stranger than fiction: they have some strong OSS advocates on the inside.
Oracle is (partly) a Linux company: they sell and support their own distro.
They try to. It would seems the vast majority of their DB customers do not really want to buy LInux contracts from them, I wonder why that would be. In truth, Oracle is not a Linux company, they are a DB company that relies on Linux plaform installations for a large and increasing share of their revenue.
Furthermore, most people don't even know that Oracle has dedicated team of paid staff that does nothing but work on FOSS [oracle.com]. One of these projects is OCFS2, which I have personally been involved with (as a user & community member, not a developer) for 2-3 years now and has recently become part of the mainline Linux kernel.
As I understand it, OCFS2 got sidelined for no good reason by some idiot at Oracle, of which there seems to be no shortage.
They were the only ones who made a GPU driver that actually both worked and performed well. Whether or not it's open source is of secondary consideration - give me a fucking GPU driver that actually pumps pixels!
Claiming open source or not makes no difference just shows that you have no first hand experience. I am currently running the open source Radeon driver, and for the first time[1] ever in my 3D accelerator history I have a platform that never segfaults (any more) handles text mode properly (looking at you NVidia) and doesn't break on every kernel upgrade. This is a huge deal-maker for me because at this point I value stability over throughput, and over being able to run OGL 3+, which is the only reason I will boot the Catalyst driver on occasion. I do not disagree that the Catalyst driver pumps more pixels - and it also has other goodies like proper antialiased lines and FSAA - but that does not matter as much to me as being hassle-free. By the way, I can do 50K triangles/frame at 60 FPS with the Radeom driver on a fanless 4830 using 50% of one Phenom II core. Did I mention, I also value quiet? It's true.
[1] Except for Intel GMA, which is also open for too underpowered for serious development work.
Laptops are heavy. They are after you lug them through an airport for hours. An iPad is light. It is a great device to use to check email and surf the web, as long as you don't need to use the keyboard much. I think it makes a great device for short trips.
This is all true, however if this is what you need then you probably should get an Android tablet, better suited to this task because of more and better laptop-replacement software available and better connectivity (e.g., built in USB for your camera). On my last roadtrip I brought along a Xoom+USB keyboard just as an experiment and also brought a standard netbook. To my surprise, I managed the entire trip without ever turning on the netbook. Mind you, that is because I did not need to do any heavy duty document editing that time, nor any coding. Just maps, reading documents, light editing, and... SSH to my home server, which is where Android really pulls ahead of Apple. There are a bunch of excellent free (as in Freedom) apps for that.
Next roadtrip I guess I will still bring the netbook but I fully expect to leave it in the bag. After LibreOffice arrives and some miscellaneous fixes to things like cut and paste, I guess I will stop packing the netbook around.
The "60 days away from bankruptcy" is what you would except a new CEO to say; Gil Amelio probably wouldn't agree with it. Apple had a good billion dollars in cash when Microsoft invested $150 million.
Slashdot Mirror
Actually, his whole slide deck has that "one cuppa coffee too many" feeling.
I want millions. I want the entire population of New York City, all eight million people, turned into shambling, flesh-eating monsters.
Sounds like you haven't been to New York lately.
games are supposed to be FUN with a capital F. Think a minute about the games you TRULY love, the ones you go back to, do you go back to them for the graphics?
At the moment I have both Oblivion and Skyrim games going on the PS3, and to be honest, Oblivion is more fun even though Skyrim is way shinier. Why? It's hard to say, maybe Bethesda just dumbed down the attributes system too much and the abilities tree is way too close to FFXIII, evoking bad memories for me. The new "finishing moves" - though impressive - are anti-immersive. It's like the computer is having fun instead of you. The magic system - yes, dual casting, great. Sucky way of selecting spells, even worse than Oblivion's sucky way. Boxes... just don't look like boxes inside, not at all.The character designer is way dumbed down, why did they do that? And the world, though bigger and prettier, just isn't as densely populated with interesting, emergent behavior. In Oblivion it is always fun to go wandering off the beaten path, in Skyrim not so much.There are a lot more walls in your way, and insurmountable obstacles. And cliffs where you just die just by going out of bounds, why is that supposed to be fun? Well I don't know, maybe it is just what I got used to, but I don't really care whether I finish Skyrim or not, whereas Oblivion was completely addictive. Progress? Hmm. I couldn't wait for Elder Scrolls five, it was the only reason I kept the PS3. But I can wait for Elder Scrolls six. Forever, thanks.
If you're saying that a $325/year investment in multi-purpose hardware is too much, but dropping $60 at a shot to play modern games is cheap, you have very, very weird budgeting.
You are obviously not married.
GPUs aren't the bottleneck anymore
Only for older games with simplistic lighting models. As Sweeny says, just load up your shader with subsurface scattering logic and see if your GPU isn't the bottleneck.
Sweeny and company need to get a clue...
Sweeny and company need to keep doing exactly what they are doing, we all benefit from it. The rest of us can get a clue.
Libraries, and the commons. These algorithms will slowly but surely enter the public domain. How long before we see the equivalent of Speed Tree as a GPL library? Hard to say. Not more than 5 years from now, that is safe to say. Maybe a lot less.
Next gen consoles will be sufficiently more powerful than a low end PC, and hopefully easier to use...
Not easier to use than Android or iOS, with which the entire target market will already be familiar.
...such that they will retain the market of "I just want it to work".
So you think the last generation of consoles "just worked"? Reality says otherwise, ranging from mysterious hardware failures to awkward configuration to frequent network outages that are still common. There is no way Xbox or PS/3 ever "just worked".
...And mobile is never going to match the experience of the big AA titles.
Famous last words. Big Media goes where the money is, and there is way more money in the pockets of people with expensive smartphones than Gen Y's playing Halo every day on the recroom sofa.
Mobile is the place to be if you're small, indie dev, but the money is on consoles...
Today it is, or it was. Truthfully, which franchise do you think is more valuable today, Elder Scrolls or Farmville? (Can you spell "network"?)
no[t] on PC (easier to provide a consistent experience for).
The PC game market has shrunk (not overall, but as a fraction of the whole) for one reason only: Microsoft. Microsoft is on the verge of losing control of the PC market because those hangers on who can't imagine switching away from Microsoft's dull and awkward user interfaces will in future not be able to imagine switching away from their smart phone interface. The rise of ARM desktops running Android will be the big story over the next 5 years, by coincidence, the same period in which Microsoft and Sony will be arming for their next console war. Which by the way, they will both lose. They came awfully close to crashing last generation and only infinite bank accounts augmented by boneheaded ego-driven strategic misdirection from the executive suite allowed either of Xbox or PS3 to survive. Close to the edge last time... straight over the edge this time.
If you can't get console sales for a Action ish title (skyrim/mass effect/any FPS/sports games etc.) then you can't get enough money to warrant having made the game.
What your are missing is the fact that the market segment that is willing to put 100 hours into a game is a small fraction of the market. Yes, Elder Scrolls 6 will be a big title, but whatever Bethesda brings out on smart phones will be much, much bigger. Notice John Carmack, who started all that, has already read the writing on the wall, can you?
if you're not moving a million copies on Xbox2/PS3 combined you're not making a game at this point.
A million copies? Feh. Future game distribution numbers will be measured commonly in the tens of millions, and sometimes, hundreds of millions. That is only possible in the the mobile market, big clunky consoles tied to furniture-style displays will lag far behind. Now, if you think that next year's smart phone won't be able to run Skyrim, you are just out of touch with hardware capabilities. Plus, Skyrim isn't where the money is, Farmville is.
...Indie titles will never be able to compete on average with big titles.
More famous last words. One word in answer: Steam.
Big money means voice acting, professional scoring, huge amounts of art...
...all of which the indie community has proved beyond a shadow of a doubt it is capable of producing itself without help from Big Media. If you doubt me then you better surf on over to Youtube and watch Sintel.
Yes, you can have the odd indie title that does crazy well (think minecraft) but the vast majority of them will be aiming to break even by flooding the market.
Sure, of course, but do you think that will s
Only a few years back IBM said Intel is ahead of everyone else by 1-2 years and will remain so for the foreseeable future.
My point is, there is a real chance of a 1-2 year hiatus this time round where TSMC is shipping 20 nm while Intel is still at 22 nm and does not have 14 nm fully ready. Sure, Intel might still be "ahead" technically, but during such a hiatus that is just pie in the sky.
The real question is who hits mass production first. Intel is a powerhouse when it comes to fabs.
No question about it. But so is the Asian semiconducter business, and so far Intel never had to compete head to head because AMD has the only credible superscalar x86 design outside Intel, and Intel had effective means of preventing AMD from building the fabs it needed to compete effectively. Two words: air supply. That game just changed.
I guess NVidia must be getting awfully close to [open-sourcing their drivers]
Why did you put words in my mouth, then do the straw-man thing? What I meant - for anyone who did not see it clearly from context - is that NVidia must be close to opening their register specs.
the 'next gen' consoles are in the making now
They are doomed. Their multibillion development costs will never be recovered when they get slaughtered in the market, crushed between mobile platforms and vastly more capable discrete GPUs. The rise of AAA idie titles won't help much either.
"just can't be serialized"
Eh, "parallelized", obviously.
Hats off to your incisive analysis. I certainly will not make the mistake of buying an Atom-based computer again. Atom runs too hot for the amount of computing it does. Against ARM it is just no contest. The only thing it has going for it is x86 compatability, and I guess I would prefer to get that more of that for less money from AMD, bundled with a decent GPU.
One thing to add: OpenCL is a game changer. It shifts the multi-core equation onto the GPU. Four cores? Feh, how about 80, or 800, all cranking single or double precision floating point in SIMD. When that truly takes hold, as in common applications restructured to utilize parallel resources automatically, the role of the monolithic central CPU will be further compartmentalized. Basically, you will only need monster firebreathing massively superscalar cores for legacy code that hasn't been updated to support parallelism, or those rare applications that just can't be serialized or care a lot about submicrosecond latency. Yes, they exist (financial transaction processing anyone?) but they are rare. I look forward to gcc -j1000 on my GPU.
Intel very rarely falls behind anything AMD does.
Not true at all. But when they do fall behind, which has happened several times (e.g. the slow move to serial interconnect) they will spend any amount of money and employ any means, fair or foul to catch up.
This time round though, AMD has a very real chance to be on a 20nm process while Intel spends a year or two at 22nm. That is because TSMC has already taped out ARM15s at 20 nm, while Intel has a long way to go on its 14 nm process. That will go a long way towards levelling the playing field. Hmm, interesting that AMD has also mumbled about going "ambidextrous". I would not mind having an ARM core in an AMD part, perhaps to keep the box running in ultra low power mode without actually suspending.
Ah, excuse me, 50K TPS on a 4350, not a 4830. The latter should do about ten times that but fanless is a bit of a stretch.
Does that mean Nvidia gonna open source the driver for the graphic cards using Nvidia chips?
I don't think they will ever open-source their drivers. It would be embarrassing for them when others discuss their code, they are protective of their work, etc. All you can hope for, and what you should be demanding, is that they give more specs to the nouveau team.
Of course that is all that the community wants. Why should NVidia have all the fun of writing kick-butt driver code? :-p
I guess NVidia must be getting awfully close to taking that step. I would say, just waiting for a suitably high profile occasion to announce it now. Stranger than fiction: they have some strong OSS advocates on the inside.
Oracle is (partly) a Linux company: they sell and support their own distro.
They try to. It would seems the vast majority of their DB customers do not really want to buy LInux contracts from them, I wonder why that would be. In truth, Oracle is not a Linux company, they are a DB company that relies on Linux plaform installations for a large and increasing share of their revenue.
Furthermore, most people don't even know that Oracle has dedicated team of paid staff that does nothing but work on FOSS [oracle.com]. One of these projects is OCFS2, which I have personally been involved with (as a user & community member, not a developer) for 2-3 years now and has recently become part of the mainline Linux kernel.
As I understand it, OCFS2 got sidelined for no good reason by some idiot at Oracle, of which there seems to be no shortage.
Googlers should know that expressing disagreement via mod points is not Googly. Or is it now?
Oh right, I forgot, the "don't be evil" already left the building.
They were the only ones who made a GPU driver that actually both worked and performed well. Whether or not it's open source is of secondary consideration - give me a fucking GPU driver that actually pumps pixels!
Claiming open source or not makes no difference just shows that you have no first hand experience. I am currently running the open source Radeon driver, and for the first time[1] ever in my 3D accelerator history I have a platform that never segfaults (any more) handles text mode properly (looking at you NVidia) and doesn't break on every kernel upgrade. This is a huge deal-maker for me because at this point I value stability over throughput, and over being able to run OGL 3+, which is the only reason I will boot the Catalyst driver on occasion. I do not disagree that the Catalyst driver pumps more pixels - and it also has other goodies like proper antialiased lines and FSAA - but that does not matter as much to me as being hassle-free. By the way, I can do 50K triangles/frame at 60 FPS with the Radeom driver on a fanless 4830 using 50% of one Phenom II core. Did I mention, I also value quiet? It's true.
[1] Except for Intel GMA, which is also open for too underpowered for serious development work.
Tell me that Google couldn't do a better job than that.
5 minutes? What sort of coding knowledge does Google have anyway.
Not as much as the combined wisdom of the community, a fact that permeates slowly through some of the thicker skulls in the land of Oz.
Googlers should know that expressing disagreement via mod points is not Googly. Or is it now?
That's $12 million/hour, more than Larry and Sergey combined :-)
Laptops are heavy. They are after you lug them through an airport for hours. An iPad is light. It is a great device to use to check email and surf the web, as long as you don't need to use the keyboard much. I think it makes a great device for short trips.
This is all true, however if this is what you need then you probably should get an Android tablet, better suited to this task because of more and better laptop-replacement software available and better connectivity (e.g., built in USB for your camera). On my last roadtrip I brought along a Xoom+USB keyboard just as an experiment and also brought a standard netbook. To my surprise, I managed the entire trip without ever turning on the netbook. Mind you, that is because I did not need to do any heavy duty document editing that time, nor any coding. Just maps, reading documents, light editing, and... SSH to my home server, which is where Android really pulls ahead of Apple. There are a bunch of excellent free (as in Freedom) apps for that.
Next roadtrip I guess I will still bring the netbook but I fully expect to leave it in the bag. After LibreOffice arrives and some miscellaneous fixes to things like cut and paste, I guess I will stop packing the netbook around.
The "60 days away from bankruptcy" is what you would except a new CEO to say; Gil Amelio probably wouldn't agree with it. Apple had a good billion dollars in cash when Microsoft invested $150 million.
That reality distortion field again.