The PS2 is still alive, with sony publishing first party games, and outselling the Xbox 360 (as of January 2007). It was launched in Japan in 1999, so it's creeping up on 8 years.
2012 for the next Xbox will mean Microsoft has done a good job. The thing is, at this point system power is not a selling point compared with system functionality. Console manufacturers have a distinct advantage in extending system lifetime because of the online upgradability built-in. Eventually they will need new GPUs but not before they've exhausted adding a lot of functionality on top of the current tech. It's way cheaper and more appealing than a whole new hardware launch nightmare.
There is a balance. "Either/or" is an argument meant to lure people to think only in extremes.
In particular, this isn't an issue that makes a blogger the bitch of a corporation. Such an issue might be if Sony were using child labor to make PS3s, or if PS3 were causing a surge of heart attacks among its owners. This is about stupid Achievements. The only thing Kotaku did was affect announcement timing by 3 days. They didn't save anything, there's no moral victory here.
Unfortunately, Kotaku went with the 'scoop' mentality. Scea blamed Kotaku, reconsidered, and decided that it can only blame itself for the leak. So undoubtedly, there is a major witch-hunt brewing, and Kotaku has really only succeeded in outting its best source. Over nothing.
Really, nothing. Good job. Way to draw the line, Kotaku.
Certain problems like transforming vertices and shading pixels, where programming efficiently is easily achieved with HLSL.
These aren't meant for supercomputers. Those aren't DP flops they are talking about, and it doesn't seem like it is Intel's intent to change that. Of course, there are already GPUs doing SP teraflop. Sony bragged that RSX in PS3 was 1.8 TFlop, and newer GPUs are even faster. But hotter, probably.
Does SSE128 mean some significant departure from the doomed SSE instruction set?
I'm not kidding. In SSE I'm familiar with, one of the input registers is always an output register, which means its contents are destroyed. Another flaw is that there aren't enough registers... SSE uses 8, where 32 are commonly not enough when latency is longish (especially with SoA-style progamming, where pragmatically a single vec3 occupies 3 128-bit registers).
... or Madd. You know, multiply-add. Does it have that?
With the other chips, you have to load the first part(if it's a full 128bit instruction, or if it's multiple instructions added together), save, load, save, add, execute.
Please explain this. Do I understand correctly that you think some SSE instuctions are 16 bytes? Issuing is one thing, and latency another. In most cases I've found AMD/Intel can issue 1 mulps/shufps/adds per cycle, the *ss instructions at 2 per (AMD sometimes 3 per cycle). If you mean that only the first 64-bits, 2 components, are computed in a cycle and the next 2 components in the next cycle, okay. Except that vmx does 4 component multiply-add in a single cycle, which would mean that SSE sucks at its GHz.
There's nothing else in the world where you can destroy $4000 of someone else's property without their permission and then brag about it without having to worry about cops.
Glad I never played this corrupt game. The developer-sponsored cheating that has happened might inspire real anger, and it sounds like the company is liable.
It's a good way to sink, destroying property when you have an anchor named 'Fraud' attached to your ankle. And then it's not even the cops you worry about... it's those damned civil suits. Less burden of proof, more money. You start praying you're clean when those wolves come sniffing around.
If people actually sank that kind of money into this economy, it's not because they couldn't afford lawyers. And, more importantly, it wouldn't be surprising if that valuation didn't evaporate in a flash after this incident.
I suspect that right now nVidia's engineers are working very hard on device drivers with support all the new features of their video cards and will probably have them available in a few days or a week or two.
I think this is the crux of the matter. According to TFA, the suit would be dropped with just a timeline of the fixes and an apology. NVidia touted the 8800 as being the first DX10/Vista ready GPU. Now Vista is here and the drivers aren't (and among people who care about GPUs, the user-mode driver layer that Vista allows is a major selling point), instead of doin' some 'spalinin, nVidia is doing some banning, deleting, and general disrespect to people who read the label.
Now, in Nvidia's defense, the problems they are facing can be really hard to find and fix. So probably any timeline has high variability, and the worst case may be really bad. And faced with a mob shrieking 'Lawsuit!' the less you say, the less you have to deny later in front of a judge.
Around half of a successful game's budget is actually marketing. So while next-gen ups the remaining budget allocation towards engines pushing high-end graphics and artists to create the content, there is an opportunity now for innovative games to undercut a marketing $$$ disadvantage.
First, E3 is dead. Thank the lord. But what has risen from those ashes is the downloadable demo as a way to reach gamers. It's like we've taken all the work that goes into dropping demos on E3 machines and pushed it into the living room for a fraction the cost! Among XBox 360 gamers I know, they all love demos (well, at least they love having the ability to try demos).
The day may come very soon when innovation can compete head-to-head against hype-only games because the battle arena isn't banners on the web and TV commercials, but live on the console with controller in hand.
but its design counts on the programmers to code in 90%+ SIMD instructions to get the absolute fastest performance.
This is an often-repeated misconception. Cell abandons the practice of having different fp, integer, and vector registers... all registers are 128-bit and any instruction can be issued on any of them, and those instructions are generated by a C++ compiler. So saying that programmers code in these SIMD instructions is like saying that "x86's design counts on programmers to shuffle values between the fp stack, integer and vector registers, and code in separate fp, integer, and vector instructions to get the absolute fastest performance".
The reality is that Cell was targeted more at solving the memory problem than just doing SIMD stream processing. Engineers looked around and decided a 32kb L1 cache was silly... not having a cache-snooping DMA engine (or prefetch engine) would be silly. Putting nine cores on a bus with 7 GB/s bandwidth would be silly. Not being able to overlap memory latency with execution is silly. To solve all these problems, you give up having a single coherent address space.
But there is even more power in Java,.NET investments now... It is completely within the realm of possibility to write a runtime that executes your Java thread on SPU, or JITs the.NET to SPU code. It's a nice benefit that these are already handle-based rather than pointer based languages, so the memory-mapping is a task of the runtime and transparent to the code. And IBM is working hard on native C++ code generation that is agnostic of the address space problem.
"Cross Platform" in many cases just means "Lowest Common Denominator" in terms of technology. If I own Console A, I don't want the game to be on Console B because it means that the developer wasted some of their talent on building the B version, and possibly cut features on A they couldn't do on B.
Plus, from a developer perspective there is really no joy in doing ports, neither games nor engines. As a developer you want to target your hardware as specifically as possible and never look back.
What is evil is partial exclusivity, where certain characters or levels are only on one platform, and that is just assinine greed. It is foreseeable a different kind of exclusivity... developers being paid to eliminate features from one platform, to skew side-by-side graphics comparisons. Do you doubt this will happen?
The industry needs to stop emphasizing graphics over gameplay.
It's amazing how often this myth is repeated. There is no such thing as a tradeoff between graphics and gameplay. They are two different things, written by two different groups of people, and the group of gameplay programmers is typically much larger than graphics programmers. If a game has good graphics and terrible gameplay, it's because of the attention to detail of the people who made it.
Clearly you didn't read the blog this "story" links. Its rhetoric is simply the possum argument... defend by being overly attacked, by exaggerating the overness of the attack.
The goal is to make arguing against Climate Change sound like censorship. Scientists are opinionated (even in the face of astounding evidence, unfortunately), so copy-paste one phrase to the web and watch!
The blog linked is just a blog. Just another blogger, 15 minutes and done. To be honest, if there's one number you should listen to, it's your insurance premium. If it goes up, worry more. If it goes down, rest at ease. How do you feel this year?
So in your estimation, both formats sustaining life via dual-format players is 'what is best for the consumer'? Maybe that is true for early adopters who are willing to spend the dime to hedge against picking the loser... But in the long run, for the ordinary consumer it is best if there is a winner.
But calling it a "revolution" is just plain wrong. It's just a better, more integrated version of what we have been doing for years.
See, it's such an integrated version of what we've been doing for years, that I would call it a revolution... not to mention the education possibilities it opens. Technologically, it takes these different facets of computing-- Supercomputers using high-bandwidth interconnects (turns into 25/30 GB/s IOIF and 300 GB/s EIB), GPUs using separate superscalar vector processors + mainstream CPUs growing bigger and bigger banks of low-latency memory (becomes programmable SPUs with 256kb of single-cycle cache), and combines it with high enough bandwidth memory to pump and drain 8 local processors.
My hope is that, with the availability of PS3 Linux, colleges offering "Game Development" curricula will finally begin to offer real professional level training on real console hardware. You wouldn't believe how many people can graduate with the concept of a computer consisting of just 2 components-- CPU and Main Memory in one, and GPU and VRAM in the other.
I think it's not that Cell is particularly difficult to program for or in the realm reserved for genius. It's just that people aren't taught to conceptualize the flow of data or its latency, or how the internal pipelines and dependencies of algorithms map onto hardware. With Cell, you have to address this complexity explicitly and I think students of this will learn that it's complicated... but not too complicated. You just have to learn to think "how does the hardware want to process my algorithm."
Your Xbox 360 will Never, ever have HDMI outputs, while the PS3 downscaling to 480p can be fixed with a simple firmware update. Online. And will likely be included in the firmware before you have a chance to even buy one at a store.
If the biggest downside can be be fixed in firmware... that's really kind of an upside.
rsqrtss xmm1, xmm0
about 5 cycles. And it can pipeline.
Not a fan of x86? Maybe altivec...
vrsqrtefp V2, V1
depends, but 12 cycles probably and pipelined.
On PS3's SPU it's rsqrte (6 cycles), on 3dNow it's pfrsqrt (8 cycles) both pipelined. Even PS2 had rsqrt (13 cycles). There's just no reason for software reciprocal square root. It's a cool trick, but it's not even useful anymore.
It's nice that Sony will let us run Linux, but I highly doubt the graphic chip will ever work on Linux
A lot of people highly doubted that Sony was serious about this Linux thing, and here it is on launch day. It's likely that Sony hasn't had time to write graphics chip drivers yet since they've been busy with... other things. For console devs they probably just provided bare-metal access and moved on.
From the looks of it, Sony built support for non-game OS from the very beginning, and most of the work to bring the OS to Cell was by third parties. However, Sony will definitely not let third parties do stuff like gfx drivers, so we'll have to wait til it bubbles to the top of somebody's priority list. I think it will happen though.
Regardless, this is a seriously cool development. Think of something as simple as a 7-node compile farm. Depending on how this progresses, there may be a lot more value in that $600 than anyone was expecting.
Overall FFXII is visually impressive but coming out when it did it's showing just how weak the PS2 is to the next gen systems, and totally invalidating Sony's claim that it could have lasted 10 years.
FF XII is not the best the PS2 will ever look. It's worse graphically than games like Jak2 (came out 3 years ago) which used techniques like super-sampled antialiasing, tri-linear mipmapping, full day-night cycles, no loading times, vehicle physics (albeit annoying at times), and all at mostly 60 frames per second.
How can this guy can call his game the best... running at 30 fps, with horrible texture sparkling, tons of loading screens, no day/night effect. Don't get me wrong, I love the game, but technically it wouldn't even have been the best 3 years ago.
And there are other polish issues, like terrible camera control, poor player controls, poor animations... the same thing that made GTA feel so sloppy despite being really fun. The truth is that they have great artists, a compelling story, and an addictive game format. Enjoy it for that.
One key piece here is the duplication of game data. See, while the disc capacity and the amount of RAM to be filled have increased 15x, the disc bandwidth and seek times have improved only ~2x. So there is this huge bottleneck getting data into the game.
Now, you commonly have models that reference the same textures or normal maps, and these models might be very far away from each other in the game world. You could seek around scooping up all the shared resources, but that would be really slow and loading times would be attrocious. What you really want to do is load up a giant chunk of data pre-packaged, and the only way to do that is to duplicate the shared resource. With giant disc capacity, there really is no downside except that some data gets squished further toward the "slow-read" inner ring.
Higher capacity helps gameplay by improving load times, allowing denser data to be loaded and flushed more frequently, and making the game world richer. As far as 25 gigs of pre-rendered movies goes, I don't think you'll see that. It's just not cost effective. Those cinematics cost an ass-full of money, and maybe a few games will go nuts with it. But it certainly won't be the state of the industry in 2 years or anything.
Sure, that's a good point. However, the study found that obeciles not only started off performing more poorly, but they had further declined after 5 years (while the non-obese performance was unchanged).
So if there is a third factor causing both obesity and low IQ, it is also causing the deterioration. You can explain low-income resulting in a poor education and the worse starting ability... but it doesn't explain the deterioration. Someone else suggested lower energy as being the third factor, which makes sense given that the body and brain are both "use it or lose it" in terms of ability.
Or it could be that the foods are the cause. Diets high in hormone-saturated beef, grease, and alcohol could certainly cause people to be both dumb and fat and to become even dumber and fatter. But anyway, establishing a causal link in well beyond the scope of the study... and nobody was making that claim anyway. Establishing the correlation is an important first step.
No, gamers don't need HDTV... however gamers with HDTVs do need HD games for their TVs.
See, HDTVs displaying a low-def signal look really, really terrible. Lo-def CRTs actually benefit games because they blur out the jaggies a little bit, and give a very bright responsive image with good contrast. HDTVs are downsampling and displaying non-native, so the pixelation and smearing just looks awful. Also many flaws, like one pixel cracks between geometry pieces which are normally hidden on CRT, stand out like a sore-thumb on hi-def.
So, get a HDTV for watching HD movies and playing HD games, but keep that old CRT for playing low-def games!
For his part Mr. Lim cultivates a relatively low-key image. He knows that at 27 he is nearing the end of his window as an elite player... But he said experience could make up for a few milliseconds of lost reflexes.
"The faster you think, the faster you can move," he said. "And the faster you move, the more time you have to think. It does matter in that your finger movements can slow down as you age. But that's why I try harder and I work on the flexibility of my fingers more than other players."
So it appears to be a delicate balance between strategy and twitch, like any televised sport. It has to be slow enough for strategy to exist and a generally unskilled audience to appreciate. But fast enough that the intensity can draw viewers. The other TV qualification that I didn't see addressed is the "comeback factor" wherein a devious opponent can wrestle victory from the jaws of defeat, without relying wholly on luck.
Slashdot Mirror
The PS2 is still alive, with sony publishing first party games, and outselling the Xbox 360 (as of January 2007). It was launched in Japan in 1999, so it's creeping up on 8 years.
2012 for the next Xbox will mean Microsoft has done a good job. The thing is, at this point system power is not a selling point compared with system functionality. Console manufacturers have a distinct advantage in extending system lifetime because of the online upgradability built-in. Eventually they will need new GPUs but not before they've exhausted adding a lot of functionality on top of the current tech. It's way cheaper and more appealing than a whole new hardware launch nightmare.
There is a balance. "Either/or" is an argument meant to lure people to think only in extremes.
In particular, this isn't an issue that makes a blogger the bitch of a corporation. Such an issue might be if Sony were using child labor to make PS3s, or if PS3 were causing a surge of heart attacks among its owners. This is about stupid Achievements. The only thing Kotaku did was affect announcement timing by 3 days. They didn't save anything, there's no moral victory here.
Unfortunately, Kotaku went with the 'scoop' mentality. Scea blamed Kotaku, reconsidered, and decided that it can only blame itself for the leak. So undoubtedly, there is a major witch-hunt brewing, and Kotaku has really only succeeded in outting its best source. Over nothing.
Really, nothing. Good job. Way to draw the line, Kotaku.
These aren't meant for supercomputers. Those aren't DP flops they are talking about, and it doesn't seem like it is Intel's intent to change that. Of course, there are already GPUs doing SP teraflop. Sony bragged that RSX in PS3 was 1.8 TFlop, and newer GPUs are even faster. But hotter, probably.
I'm not kidding. In SSE I'm familiar with, one of the input registers is always an output register, which means its contents are destroyed. Another flaw is that there aren't enough registers... SSE uses 8, where 32 are commonly not enough when latency is longish (especially with SoA-style progamming, where pragmatically a single vec3 occupies 3 128-bit registers).
... or Madd. You know, multiply-add. Does it have that?
Please explain this. Do I understand correctly that you think some SSE instuctions are 16 bytes? Issuing is one thing, and latency another. In most cases I've found AMD/Intel can issue 1 mulps/shufps/adds per cycle, the *ss instructions at 2 per (AMD sometimes 3 per cycle). If you mean that only the first 64-bits, 2 components, are computed in a cycle and the next 2 components in the next cycle, okay. Except that vmx does 4 component multiply-add in a single cycle, which would mean that SSE sucks at its GHz.
Glad I never played this corrupt game. The developer-sponsored cheating that has happened might inspire real anger, and it sounds like the company is liable.
It's a good way to sink, destroying property when you have an anchor named 'Fraud' attached to your ankle. And then it's not even the cops you worry about... it's those damned civil suits. Less burden of proof, more money. You start praying you're clean when those wolves come sniffing around. If people actually sank that kind of money into this economy, it's not because they couldn't afford lawyers. And, more importantly, it wouldn't be surprising if that valuation didn't evaporate in a flash after this incident.
I think this is the crux of the matter. According to TFA, the suit would be dropped with just a timeline of the fixes and an apology. NVidia touted the 8800 as being the first DX10/Vista ready GPU. Now Vista is here and the drivers aren't (and among people who care about GPUs, the user-mode driver layer that Vista allows is a major selling point), instead of doin' some 'spalinin, nVidia is doing some banning, deleting, and general disrespect to people who read the label.
Now, in Nvidia's defense, the problems they are facing can be really hard to find and fix. So probably any timeline has high variability, and the worst case may be really bad. And faced with a mob shrieking 'Lawsuit!' the less you say, the less you have to deny later in front of a judge.
First, E3 is dead. Thank the lord. But what has risen from those ashes is the downloadable demo as a way to reach gamers. It's like we've taken all the work that goes into dropping demos on E3 machines and pushed it into the living room for a fraction the cost! Among XBox 360 gamers I know, they all love demos (well, at least they love having the ability to try demos).
The day may come very soon when innovation can compete head-to-head against hype-only games because the battle arena isn't banners on the web and TV commercials, but live on the console with controller in hand.
This is an often-repeated misconception. Cell abandons the practice of having different fp, integer, and vector registers... all registers are 128-bit and any instruction can be issued on any of them, and those instructions are generated by a C++ compiler. So saying that programmers code in these SIMD instructions is like saying that "x86's design counts on programmers to shuffle values between the fp stack, integer and vector registers, and code in separate fp, integer, and vector instructions to get the absolute fastest performance".
The reality is that Cell was targeted more at solving the memory problem than just doing SIMD stream processing. Engineers looked around and decided a 32kb L1 cache was silly... not having a cache-snooping DMA engine (or prefetch engine) would be silly. Putting nine cores on a bus with 7 GB/s bandwidth would be silly. Not being able to overlap memory latency with execution is silly. To solve all these problems, you give up having a single coherent address space.
But there is even more power in Java, .NET investments now... It is completely within the realm of possibility to write a runtime that executes your Java thread on SPU, or JITs the .NET to SPU code. It's a nice benefit that these are already handle-based rather than pointer based languages, so the memory-mapping is a task of the runtime and transparent to the code. And IBM is working hard on native C++ code generation that is agnostic of the address space problem.
Plus, from a developer perspective there is really no joy in doing ports, neither games nor engines. As a developer you want to target your hardware as specifically as possible and never look back.
What is evil is partial exclusivity, where certain characters or levels are only on one platform, and that is just assinine greed. It is foreseeable a different kind of exclusivity... developers being paid to eliminate features from one platform, to skew side-by-side graphics comparisons. Do you doubt this will happen?
It's amazing how often this myth is repeated. There is no such thing as a tradeoff between graphics and gameplay. They are two different things, written by two different groups of people, and the group of gameplay programmers is typically much larger than graphics programmers. If a game has good graphics and terrible gameplay, it's because of the attention to detail of the people who made it.
The goal is to make arguing against Climate Change sound like censorship. Scientists are opinionated (even in the face of astounding evidence, unfortunately), so copy-paste one phrase to the web and watch!
The blog linked is just a blog. Just another blogger, 15 minutes and done. To be honest, if there's one number you should listen to, it's your insurance premium. If it goes up, worry more. If it goes down, rest at ease. How do you feel this year?
So in your estimation, both formats sustaining life via dual-format players is 'what is best for the consumer'? Maybe that is true for early adopters who are willing to spend the dime to hedge against picking the loser... But in the long run, for the ordinary consumer it is best if there is a winner.
See, it's such an integrated version of what we've been doing for years, that I would call it a revolution... not to mention the education possibilities it opens. Technologically, it takes these different facets of computing-- Supercomputers using high-bandwidth interconnects (turns into 25/30 GB/s IOIF and 300 GB/s EIB), GPUs using separate superscalar vector processors + mainstream CPUs growing bigger and bigger banks of low-latency memory (becomes programmable SPUs with 256kb of single-cycle cache), and combines it with high enough bandwidth memory to pump and drain 8 local processors.
My hope is that, with the availability of PS3 Linux, colleges offering "Game Development" curricula will finally begin to offer real professional level training on real console hardware. You wouldn't believe how many people can graduate with the concept of a computer consisting of just 2 components-- CPU and Main Memory in one, and GPU and VRAM in the other.
I think it's not that Cell is particularly difficult to program for or in the realm reserved for genius. It's just that people aren't taught to conceptualize the flow of data or its latency, or how the internal pipelines and dependencies of algorithms map onto hardware. With Cell, you have to address this complexity explicitly and I think students of this will learn that it's complicated... but not too complicated. You just have to learn to think "how does the hardware want to process my algorithm."
If the biggest downside can be be fixed in firmware... that's really kind of an upside.
Soooo... They didn't "get it" with the iPod?
rsqrtss xmm1, xmm0
about 5 cycles. And it can pipeline.
Not a fan of x86? Maybe altivec...
vrsqrtefp V2, V1
depends, but 12 cycles probably and pipelined.
On PS3's SPU it's rsqrte (6 cycles), on 3dNow it's pfrsqrt (8 cycles) both pipelined. Even PS2 had rsqrt (13 cycles). There's just no reason for software reciprocal square root. It's a cool trick, but it's not even useful anymore.
A lot of people highly doubted that Sony was serious about this Linux thing, and here it is on launch day. It's likely that Sony hasn't had time to write graphics chip drivers yet since they've been busy with... other things. For console devs they probably just provided bare-metal access and moved on.
From the looks of it, Sony built support for non-game OS from the very beginning, and most of the work to bring the OS to Cell was by third parties. However, Sony will definitely not let third parties do stuff like gfx drivers, so we'll have to wait til it bubbles to the top of somebody's priority list. I think it will happen though.
Regardless, this is a seriously cool development. Think of something as simple as a 7-node compile farm. Depending on how this progresses, there may be a lot more value in that $600 than anyone was expecting.
FF XII is not the best the PS2 will ever look. It's worse graphically than games like Jak2 (came out 3 years ago) which used techniques like super-sampled antialiasing, tri-linear mipmapping, full day-night cycles, no loading times, vehicle physics (albeit annoying at times), and all at mostly 60 frames per second.
How can this guy can call his game the best... running at 30 fps, with horrible texture sparkling, tons of loading screens, no day/night effect. Don't get me wrong, I love the game, but technically it wouldn't even have been the best 3 years ago.
And there are other polish issues, like terrible camera control, poor player controls, poor animations... the same thing that made GTA feel so sloppy despite being really fun. The truth is that they have great artists, a compelling story, and an addictive game format. Enjoy it for that.
There are some villains in the world that would pay handsomely for this... meteorite.
Now, you commonly have models that reference the same textures or normal maps, and these models might be very far away from each other in the game world. You could seek around scooping up all the shared resources, but that would be really slow and loading times would be attrocious. What you really want to do is load up a giant chunk of data pre-packaged, and the only way to do that is to duplicate the shared resource. With giant disc capacity, there really is no downside except that some data gets squished further toward the "slow-read" inner ring.
Higher capacity helps gameplay by improving load times, allowing denser data to be loaded and flushed more frequently, and making the game world richer. As far as 25 gigs of pre-rendered movies goes, I don't think you'll see that. It's just not cost effective. Those cinematics cost an ass-full of money, and maybe a few games will go nuts with it. But it certainly won't be the state of the industry in 2 years or anything.
So if there is a third factor causing both obesity and low IQ, it is also causing the deterioration. You can explain low-income resulting in a poor education and the worse starting ability... but it doesn't explain the deterioration. Someone else suggested lower energy as being the third factor, which makes sense given that the body and brain are both "use it or lose it" in terms of ability.
Or it could be that the foods are the cause. Diets high in hormone-saturated beef, grease, and alcohol could certainly cause people to be both dumb and fat and to become even dumber and fatter. But anyway, establishing a causal link in well beyond the scope of the study... and nobody was making that claim anyway. Establishing the correlation is an important first step.
See, HDTVs displaying a low-def signal look really, really terrible. Lo-def CRTs actually benefit games because they blur out the jaggies a little bit, and give a very bright responsive image with good contrast. HDTVs are downsampling and displaying non-native, so the pixelation and smearing just looks awful. Also many flaws, like one pixel cracks between geometry pieces which are normally hidden on CRT, stand out like a sore-thumb on hi-def.
So, get a HDTV for watching HD movies and playing HD games, but keep that old CRT for playing low-def games!
So it appears to be a delicate balance between strategy and twitch, like any televised sport. It has to be slow enough for strategy to exist and a generally unskilled audience to appreciate. But fast enough that the intensity can draw viewers. The other TV qualification that I didn't see addressed is the "comeback factor" wherein a devious opponent can wrestle victory from the jaws of defeat, without relying wholly on luck.