All it takes is initiative. Which is something kids today have never needed to learn. I think kids are in a situation where they expect success to be handed to them even if they don't bother to try. And this story is just more ammo to blame the teachers, blame the government, blame blame blame.
If your kids are dumb, if they don't feel the need to learn, it's your fault. Teach your kids that things in life aren't just handed away.
At this point, if the PS3 is late and/or the Cell chip is hard to write fast code for then the MS early mover advantage could be really bad news for Sony.
The problem with the 'early mover advantage' is that it really only applies to games, not consoles. The real motivation for being an early adopter to a next-gen console is to play the sickest game 3v4R.
But the comments from EA, that "next generation Xbox titles would ship alongside current generation offerings later this year" implies that whatever games are available for XB2 are going to be available on current consoles as well. People don't pay early adopter prices just to play what everyone else has. (And don't think that EA is going to be releasing any XB2-only games. Their strategy has always been to leverage the expense of producing a game across as many platforms as possible).
The other possibility is that games are going to be rushed to market... Never a good idea. Or they'll have to rely on their first-party titles. It would take a heroic effort to get Halo3 to market before PS3.
The point is that Sony is not under as much pressure as you think. They want to make sure that when they release something, it's perceived to be the highest quality offering (whether it's true or not). Notice that Sony never pressured DP into making Gran Turismo a launch title for any of their consoles. Sony knows that when DP launches a title, everyone in Japan will buy it and the console it was written for, and nothing is worth tarnishing a reputation (again, whether or not the reputation is accurate). Microsoft has never been constrained by such scruples.
these are max figures assuming software is perfectly written to take care of parallelization... this means that most computer programs will hit no where near these rates, but super optimized versions could take advantage of it...just remember how many developers complained about the Emotion Engine from the SP2 and how it was such a bitch to program for, this will be worse
This is essentially what happened with the PS2. 1st gen game teams thought the compiler would handle more of the task of keeping the vector units and GS busy. Didn't happen. However, PS2 teams have learned a lot of valuable lessons in the past 5 years to prepare for this jump. PC developers are going to have a horrible time trying to get performance out of the Cell.
Most notably, devs learned that the PS2 is bus-bound. With only 16kb caches, memory-layout is paramount to avoid requisitioning the bus every 100 cycles to refill both caches just for a vtable look-up and jump.
So the Cell forces programmers to think in this paradigm. No caches, just 256k local storage for threads. So your performance will only suffer if you fail to learn the new principles... no more cache-agnostic coding. No more memory accesses to any random place in memory.
Then programmers figured out that the VU0 was basically sitting dormant the entire time. That's a third of the total proccessing power wasted. Little by little, they started moving tasks to the VU0-- skeletal animation, particle dynamics. The problem was that the VU0 only had 4kb of local memory, so between loading a microprogram, double-buffering memory for DMA in/out, and running the damn thing, the EE couldn't do anything useful besides babysit.
The VU1, OTOH, was a totally different beast. It took over responsibility for the T&L stage in rendering. It had 16kb of memory and could consume a chained DMA stream of microprograms to run and the associated memory (basically a series of display lists). Once you wrote your display-list chain to a buffer and began DMA, it required no babysitting at all. Without a doubt, You can see where the design decisions behind the Cell are coming from. The PS2 was basically just a prototype Cell to see what worked.
And here are the results. They ditched the VU0, multiplied the number of VU1s by 8, and gave each one 16x the memory, jacked up memory bandwidth. The EE (PPC) is now officially the arbiter of threads... Except now the SPEs are capable of generating execution chains-- i.e. one produces, another consumes, so the PPC doesn't need to have all the brains.
Another interesting thing is that while certain large portions of the render loop need to be executed serially (Game Logic, Animation, Collision/Dynamics, then Render), many operations within those category can be parallelized. For instance, devs have resorted to huge hacks to make AIs look like they're running in threads when they're really not. It's simply been the case that games were definitely only running on one processor, and context switches are expensive. It's actually much more convenient to have multiple cores and turn these hacks into actual LWPs. The real question is how is Sonyibm going to handle concurrency? Are they going to write a special pthreads for the PPC threadmaster? Are they going to use chainable microprograms, where the PPC is just a glorified VIF? That is the big question eating at me right now.
But if you look at the latest games-- Jak3, GT4... they're hitting pretty much near the theoretical limit of the PS2, no matter how unlikely that seemed just 5 years ago. The first half of the Cell learning curve has already been traversed by those brave PS2 freaks, it's up to the rest of us to learn from where they've been.
The fact is that this will be a much more difficult processor to program efficiently for. This is the same situation that faced developers when the PS2 came out. It's taken game developers 4 years to finally tame the beast, and this chip is everything that made PS2 programming difficult, times 8.
But look at the graphics in PS2 games now compared to 1st gen titles. The improvement is incredible! The hardware hasn't changed: it's still just a 300Mhz cpu with 4MB graphics and no pixel shading. I think we'll see the same maturation process with Cell/PS3, where the 1st gen games don't live up to the hype but more and more of the Cell's enormous potential is realized with successive generations.
The question is whether Sony decides that part of the slow evolution in efficient PS2 programming was because of the small, exclusive development community. I would love to see Sony push a Linux PS3 similar to the version of Linux PS2 they released.
Well you can't say it isn't news for nerds. And this article has enough added information in it that I thought it to be worth posting. Most Cell news stories are dumbed down for the nonnerds, whose most pressing question is "Does it run Windows?" This article is the best source I've seen of all the info we know about Cell, without a painful amount of editorializing.
It seemed there was a lot of misinformation/confusion going around because some people heard it supported DP floats and some people heard it used Altivec (which doesn't support DP). So half the people extrapolated that IBM had ditched Altivec (i.e. VMX), and the other half assumed there was no DP support... both of which angered people. The truth (according to this article) is that it uses BOTH: A version of VMX that supports DP. whew!
The article also points out that the SP floats aren't truly 754-compliant, as they round-toward-zero on cast to int. This makes it compatible with that horrible C/C++ truncation cast (If anyone knows why C opts to round-toward-zero, please let me know!). However, rest assured, DPs are 854-compliant.
Also, the article suggests that there is a memory limit (at least initially) of 256MB:
The maximum of 4 DRAM devices means that the CELL processor is limited to 256 MB of memory, given that the highest capacity XDR DRAM device is currently 512 Mbits. Fortunately, XDR DRAM devices could in theory be reconfigured in such a way so that more than 36 XDR devices can be connected to the same 36 bit wide channel and provide 1 bit wide data bus each to the 36 bit wide point-to-point interconnect. In such a configuration, a two channel XDR memory can support upwards of 16 GB of ECC protected memory with 256 Mbit DRAM devices or 32 GB of ECC protected memory with 512 Mbit DRAM devices.
This is similar to the 'scratchpad' RAM that Sony used in the PS2 and PS1. It's 16kb of on-chip (super-fast) memory that can be loaded and manipulated by the programmer, completely separate from the jurisdiction of the cache (which can cause big headaches-- think cache writeback with stale data).
We'd do our skeletal animation skinning with this. DMA a bunch of verts to scratchpad, transform and weight them on the VU, DMA back to a display list. The thing is, there's really no high-level language support for this... the onus is on the programmer to schedule and memory map everything, mostly in assembly.
The design of the cell-- it's incredible. It's every game programmer's wet dream. I just don't see how it's going to be as useful in other areas though. It's going to be a compiler-writer's nightmare, and to get real performance frome the SPEs is going to take a lot of assembly or a high-level language construct that I haven't seen yet.
"When a player cheats in a game, they are choosing a battlefield that is broader in context than the game itself."
This is totally false. The context of the game is the restrictions that make the game challenging. How hard you have to work to acquire a certain weapon, how careful you have to be to conserve ammo... how many enemies you have to kill to get to level 20.
Those challenges are really the only things separating 'playing a video game' from 'pressing buttons on a controller'. That's probably why whenever I've cheated in a game in the past, it's gotten really boring really fast. The value of the goal becomes diminished along with the challenge.
I don't think is necessarily limited to gaming, either. I think it's built into human nature.
Is it morally acceptable for a group of people to require their fellow citizens for fork over tax dollars at the point of a gun to pay for a service they don't all want to use?
This is the same argument that people try to make against "Student Fees" at colleges that go to support the Athletic Center and the Club Sports teams that students organize. They're open to everyone, and their only purpose is to enrich the lives of students and faculty. Yes, of course some people don't take advantage of it, there are people that don't take advantage of anything college has to offer them (even an education!)... but $150 of fees gives everyone an opportunity they wouldn't otherwise have to participate in virtually anything you could think of. It just takes an inkling of initiative.
And granted, it's not at the point of a gun... but just go ahead and try NOT paying it:)
By my calculations, this method is approximately 2,522,880 times more efficient than Knuth's
Why are you getting so worked up about an improvement by only a constant factor?
Theoretically, the methods are equivalent... In fact, as the number of Knuth's books goes to infinity, the overhead of having to call the typesetters each time will overcome the one-time expense of writing the typesetting language.
Hah, I was scanning the "Total Viewers" column for season 1, and saw a huge jump for the last row, and I'm like "Wow, that's a pretty good bump for the season finale."
Then I look closer at the row heading-- "Season Average"
Riiight. Except that Sony's about to plug x0 million NVidia GPUs into the PS3. I don't think flurried coding standards are going to be the end of NVidia, and especially not while it's Sony's bitch.
Re:Biggest Market for $100 PC? Developed World
on
The Hundred-Buck PC
·
· Score: 1
...or building up a distributed compile farm.
Increasing processor speeds have not done nearly as much to improve compile times as distcc and some cheap PCs. Why companies haven't recognized this and made the small investment (which pays for itself in no time), I'm not sure... but with such a low potential price tag, and no real need for extra peripherals (mouse, keyboard, monitor), surely this will start happening soon.
Oh man, that's what I thought til I moved to Seattle, the realm of 90.3 kexp. It's what everyone wants their radio stations to be... you can listen for hours and hear no songs you recognize, but still be thoroughly entertained the entire time.
No commercials, DJs who know their job is to spin tunes and not blah, blah... Anyway, you can listen over the internet from their site
People always use speed as the metric to compare the layouts, and that's just not very useful. I use Dvorak, even though I don't feel appreciably faster than when I type qwerty... for me it's all about comfort.
Dvorak incorporates a number of features that make it more comfortable. Besides a wise choice of keys to include in the home row, the position is very important. Common key combinations generally either go outside-in (like strumming your fingers), or are typed by non-adjacent fingers.
The big problem is also words that are typed mostly with one hand, which QWERTY is really bad about. T-R-E-A-D-S-C... it's a Wheel-of-Fortune finalist's wet dream. And who thought to put j,k, and ; on prime real-estate?
It's elementary... QWERTY is a very fatiguing keyboard layout for your wrists; even if you keep those particular muscles in pretty good shape already.
How do you even pronounce 'pwn3d' ?
Google is not a tool to study speech patterns, and there's nothing to say that speech even resembles written text.
The article addresses this in a weird way, where it first draws attention to the distinction, but once it reaches its crux, where google is used as a tool, the distinction is ignored entirely; instead it opts to focus on stranger things.
Slashdot Mirror
All it takes is initiative. Which is something kids today have never needed to learn. I think kids are in a situation where they expect success to be handed to them even if they don't bother to try. And this story is just more ammo to blame the teachers, blame the government, blame blame blame.
If your kids are dumb, if they don't feel the need to learn, it's your fault. Teach your kids that things in life aren't just handed away.
It says no updates available... Do I need to actually update from the site?
The problem with the 'early mover advantage' is that it really only applies to games, not consoles. The real motivation for being an early adopter to a next-gen console is to play the sickest game 3v4R.
But the comments from EA, that "next generation Xbox titles would ship alongside current generation offerings later this year" implies that whatever games are available for XB2 are going to be available on current consoles as well. People don't pay early adopter prices just to play what everyone else has. (And don't think that EA is going to be releasing any XB2-only games. Their strategy has always been to leverage the expense of producing a game across as many platforms as possible).
The other possibility is that games are going to be rushed to market... Never a good idea. Or they'll have to rely on their first-party titles. It would take a heroic effort to get Halo3 to market before PS3.
The point is that Sony is not under as much pressure as you think. They want to make sure that when they release something, it's perceived to be the highest quality offering (whether it's true or not). Notice that Sony never pressured DP into making Gran Turismo a launch title for any of their consoles. Sony knows that when DP launches a title, everyone in Japan will buy it and the console it was written for, and nothing is worth tarnishing a reputation (again, whether or not the reputation is accurate). Microsoft has never been constrained by such scruples.
"The graphics, what it can do, the way things look, huge innovation... What is 'Halo 2?' It's the greatest video game ever written."
"...And furthermore, we fully expect Halo 3 to be the most innovative game ever written!"
This is essentially what happened with the PS2. 1st gen game teams thought the compiler would handle more of the task of keeping the vector units and GS busy. Didn't happen. However, PS2 teams have learned a lot of valuable lessons in the past 5 years to prepare for this jump. PC developers are going to have a horrible time trying to get performance out of the Cell.
Most notably, devs learned that the PS2 is bus-bound. With only 16kb caches, memory-layout is paramount to avoid requisitioning the bus every 100 cycles to refill both caches just for a vtable look-up and jump.
So the Cell forces programmers to think in this paradigm. No caches, just 256k local storage for threads. So your performance will only suffer if you fail to learn the new principles... no more cache-agnostic coding. No more memory accesses to any random place in memory.
Then programmers figured out that the VU0 was basically sitting dormant the entire time. That's a third of the total proccessing power wasted. Little by little, they started moving tasks to the VU0-- skeletal animation, particle dynamics. The problem was that the VU0 only had 4kb of local memory, so between loading a microprogram, double-buffering memory for DMA in/out, and running the damn thing, the EE couldn't do anything useful besides babysit.
The VU1, OTOH, was a totally different beast. It took over responsibility for the T&L stage in rendering. It had 16kb of memory and could consume a chained DMA stream of microprograms to run and the associated memory (basically a series of display lists). Once you wrote your display-list chain to a buffer and began DMA, it required no babysitting at all. Without a doubt, You can see where the design decisions behind the Cell are coming from. The PS2 was basically just a prototype Cell to see what worked.
And here are the results. They ditched the VU0, multiplied the number of VU1s by 8, and gave each one 16x the memory, jacked up memory bandwidth. The EE (PPC) is now officially the arbiter of threads... Except now the SPEs are capable of generating execution chains-- i.e. one produces, another consumes, so the PPC doesn't need to have all the brains.
Another interesting thing is that while certain large portions of the render loop need to be executed serially (Game Logic, Animation, Collision/Dynamics, then Render), many operations within those category can be parallelized. For instance, devs have resorted to huge hacks to make AIs look like they're running in threads when they're really not. It's simply been the case that games were definitely only running on one processor, and context switches are expensive. It's actually much more convenient to have multiple cores and turn these hacks into actual LWPs. The real question is how is Sonyibm going to handle concurrency? Are they going to write a special pthreads for the PPC threadmaster? Are they going to use chainable microprograms, where the PPC is just a glorified VIF? That is the big question eating at me right now.
But if you look at the latest games-- Jak3, GT4... they're hitting pretty much near the theoretical limit of the PS2, no matter how unlikely that seemed just 5 years ago. The first half of the Cell learning curve has already been traversed by those brave PS2 freaks, it's up to the rest of us to learn from where they've been.
But look at the graphics in PS2 games now compared to 1st gen titles. The improvement is incredible! The hardware hasn't changed: it's still just a 300Mhz cpu with 4MB graphics and no pixel shading. I think we'll see the same maturation process with Cell/PS3, where the 1st gen games don't live up to the hype but more and more of the Cell's enormous potential is realized with successive generations.
The question is whether Sony decides that part of the slow evolution in efficient PS2 programming was because of the small, exclusive development community. I would love to see Sony push a Linux PS3 similar to the version of Linux PS2 they released.
It seemed there was a lot of misinformation/confusion going around because some people heard it supported DP floats and some people heard it used Altivec (which doesn't support DP). So half the people extrapolated that IBM had ditched Altivec (i.e. VMX), and the other half assumed there was no DP support... both of which angered people. The truth (according to this article) is that it uses BOTH: A version of VMX that supports DP. whew!
The article also points out that the SP floats aren't truly 754-compliant, as they round-toward-zero on cast to int. This makes it compatible with that horrible C/C++ truncation cast (If anyone knows why C opts to round-toward-zero, please let me know!). However, rest assured, DPs are 854-compliant.
Also, the article suggests that there is a memory limit (at least initially) of 256MB:
The maximum of 4 DRAM devices means that the CELL processor is limited to 256 MB of memory, given that the highest capacity XDR DRAM device is currently 512 Mbits. Fortunately, XDR DRAM devices could in theory be reconfigured in such a way so that more than 36 XDR devices can be connected to the same 36 bit wide channel and provide 1 bit wide data bus each to the 36 bit wide point-to-point interconnect. In such a configuration, a two channel XDR memory can support upwards of 16 GB of ECC protected memory with 256 Mbit DRAM devices or 32 GB of ECC protected memory with 512 Mbit DRAM devices.
We'd do our skeletal animation skinning with this. DMA a bunch of verts to scratchpad, transform and weight them on the VU, DMA back to a display list. The thing is, there's really no high-level language support for this... the onus is on the programmer to schedule and memory map everything, mostly in assembly.
The design of the cell-- it's incredible. It's every game programmer's wet dream. I just don't see how it's going to be as useful in other areas though. It's going to be a compiler-writer's nightmare, and to get real performance frome the SPEs is going to take a lot of assembly or a high-level language construct that I haven't seen yet.
Here, you can borrow mine...
"When a player cheats in a game, they are choosing a battlefield that is broader in context than the game itself."
This is totally false. The context of the game is the restrictions that make the game challenging. How hard you have to work to acquire a certain weapon, how careful you have to be to conserve ammo... how many enemies you have to kill to get to level 20.
Those challenges are really the only things separating 'playing a video game' from 'pressing buttons on a controller'. That's probably why whenever I've cheated in a game in the past, it's gotten really boring really fast. The value of the goal becomes diminished along with the challenge.
I don't think is necessarily limited to gaming, either. I think it's built into human nature.
This is the same argument that people try to make against "Student Fees" at colleges that go to support the Athletic Center and the Club Sports teams that students organize. They're open to everyone, and their only purpose is to enrich the lives of students and faculty. Yes, of course some people don't take advantage of it, there are people that don't take advantage of anything college has to offer them (even an education!)... but $150 of fees gives everyone an opportunity they wouldn't otherwise have to participate in virtually anything you could think of. It just takes an inkling of initiative.
And granted, it's not at the point of a gun... but just go ahead and try NOT paying it :)
Why are you getting so worked up about an improvement by only a constant factor?
Theoretically, the methods are equivalent... In fact, as the number of Knuth's books goes to infinity, the overhead of having to call the typesetters each time will overcome the one-time expense of writing the typesetting language.
Then I look closer at the row heading-- "Season Average"
"and using it to talk to the girl on my Mac."
the big deal is that in this version... she talks back!
I remember one card had a Skull...
and the other had a Penguin.
Riiight. Except that Sony's about to plug x0 million NVidia GPUs into the PS3. I don't think flurried coding standards are going to be the end of NVidia, and especially not while it's Sony's bitch.
Increasing processor speeds have not done nearly as much to improve compile times as distcc and some cheap PCs. Why companies haven't recognized this and made the small investment (which pays for itself in no time), I'm not sure... but with such a low potential price tag, and no real need for extra peripherals (mouse, keyboard, monitor), surely this will start happening soon.
The day traders went broke and had to get real jobs
More like had to start doing their real jobs at their real jobs... until those went bust. And then they had to get real jobs.
How about $170? Let's be reasonable here.
Next time your parents step to you with some noise about homework and skipping school, fire back.
Or maybe it's supposed to be the AMD VI0V, and Intel decided to make it one more... you know, so people would be "Hey look, this one goes to II"
No commercials, DJs who know their job is to spin tunes and not blah, blah... Anyway, you can listen over the internet from their site
I'll be impressed when the computer learns to play 'Cat, Tin foil, Microwave'
Dvorak incorporates a number of features that make it more comfortable. Besides a wise choice of keys to include in the home row, the position is very important. Common key combinations generally either go outside-in (like strumming your fingers), or are typed by non-adjacent fingers.
The big problem is also words that are typed mostly with one hand, which QWERTY is really bad about. T-R-E-A-D-S-C ... it's a Wheel-of-Fortune finalist's wet dream. And who thought to put j,k, and ; on prime real-estate?
It's elementary... QWERTY is a very fatiguing keyboard layout for your wrists; even if you keep those particular muscles in pretty good shape already.
The article addresses this in a weird way, where it first draws attention to the distinction, but once it reaches its crux, where google is used as a tool, the distinction is ignored entirely; instead it opts to focus on stranger things.