The biggest disservice to the American people is the American Dream and that one day, they will achieve it. And we all know by statistics above, that conservatively 99% won't. So why are there so many people supporting tax breaks for the rich? Because they can't accept the reality that they most likely won't be rich one day.
Personally, I'd be more than happy if I was taxed at the special $1M tax bracket -- that means I actually made at least $1M that year.
The argument between GC vs. manual memory management a la malloc/free is partly why there exists a debate if Java is as fast as C/C++.
First of all, there's no clear line between "GC" vs. "manual memory management". At some point, writing your own GC for your data structure is both "GC" and "manual memory management" at the same time. Unless you're talking about the default GC, otherwise GC is just a form of memory management.
Secondly, anyone who knows how to do heavy multithreaded programming on C/C++ knows to not rely on the global malloc and free. Either you use the platform-specific memory allocate and commit routines, or you use malloc and free very sparingly. After you get the chunk of memory from the system, you manage it yourself, using the appropriate memory allocator suitable for your data requirements. There is absolutely no one-size-fits all solution that runs fast.
As for handcrafted asm seldomly outperforming compiler code, that's probably because most asm writers are just doing it for the knowledge and experience, and not for real optimization purposes. Real asm code in tight loops done by pros usually outperform their compiler counterparts on the orders of 10~100x. At work, one of our pros even got to 1000x once...and that pro is no slouch when it comes to plain C. Of course, modern Intel CPU architectures throws a huge wrench in that analysis...but you pay for that in terms of power and die size...something not visible to the average programmer.
The moral of the story is, if you take the worst practices of one language and pit it against the best practices of another language, then sure, you may get "surprising" results.
Yeah, I remember games that ran on 640KB or less as well. Heck, some games that people consider great ran on 80 bytes. But that's not the point. Devs will always ship their games in-memory -- otherwise, you don't ship at all, since it won't run. The point being, having more RAM makes life WAY easier, which means faster to-market and better games for relatively little cost.
Beyond the rumored specs, there's really only one thing devs really want: more RAM. Don't fucking skimp out on it -- shove loads of it in there. Even if they have to tier it into faster and slower RAM, do it (we'll learn to stream/ping pong/double buffer). Just don't give us shit like 512MB anymore (not in this day and age anyway). RAM is probably the cheapest, easiest, and fastest way to get thanks from devs, which translates to better games in every single way. Give us a serving of at least 1024 MB for main RAM, and 512 for GPU RAM (or at least 1.5GB if it's unified). If they can make it 2GB+ total, all the merrier.
The next on the list would be something bigger than the rumored 8GB flash, if they want to go anywhere near online distribution or content.
Wait, are you saying that BR players can't play DVDs? How does buying a BR player has anything to do with rebuying your DVD collection. The summary talks about even split in the sales of the players, not the discs themselves.
Of course, if you're just being funny by semi-quoting the article (the article talks about people not knowing they can play DVDs in their BR players), then just ignore this comment.
When it was just that, it wasn't a problem. Most of my old games would happily install on your machine if I sold you the disc and you typed in the key. The keys were stickers on the jewel cases, and there was none of this draconian "You can only install this game 5 times, and only on Tuesdays"
Really? Entering a code is equivalent to install this game 5 times and only on Tuesdays? I understanding it's tongue-in-cheek, but let's not get too carried away. You're comparing apples and computer games here.
Because it's not a subscription service that you could download the client for free anyway unless someone scammed you as in your example? You're comparing apples and steaks here.
Last I recall, WoW's latest expansions weren't free. The original WoW when it was released wasn't free either.
How do I access that profile once I sell the game? I bought it, right?
Yes, you bought the physical CD. No, you didn't buy the usable sticker. It's like buying the candy wrapper from someone else and complaining that it didn't come with the candy. Maybe if you didn't know that going into the transaction, that'd be something else.
It's not *now*. And therein lies the problem. It's an end run around the first sale doctrine by basically saying, "We didn't sell you that, we "licensed" it to you". Imagine if you couldn't buy a used car without paying Ford a "transfer fee" for the keys.
Personally... I've never sold a game in my life, or bought one used for that matter... the few bucks just isn't worth the hassle. But many, many people do - because they can't afford to buy everything they want new. There is a fairly huge secondary market with console games, and the game companies want to eliminate it because they somehow think people will magically have more money to spend.
And therein lies the misinterpretation of the first sale doctrine. The game studio sold you a game with a $15 voucher. And what's wrong with that?
I'm not against used games, especially the single player component. But think about it this way: why should Gamestop profit (hugely) from these sales while the original content makers don't? How is that helping the people making the games for you in the first place? How does that logic even make sense? If anything, you should be criticizing the incorporated used games sellers for profiting with such a high margin. You should demand that their used games come down in price by at least the same cost as the "replacement voucher" (if it's used up...and it's up to them to find out if it's used or not). That way, the public (you) wins, the game makers win, and the people that shouldn't be winning don't win.
No, you got it completely wrong. China is criticizing the US for being hypocrites, not censorship.
As a matter of fact, the author even has it wrong; it's not the pot calling the kettle black at all. That specifically only applies when one criticizes another for doing something that he himself does. In this case, China is calling the US a hypocrite when it comes to freedom, while China has never purported to be freedom-embracing to begin with. Like someone else said, US == hypocrite, China != hypocrite, and therefore, in that regard (especially pertaining to internet freedom), China is not the same as the US, and therefore it's not pot calling kettle black.
You're equating games to engine technology, and that's fundamentally the distinction that many people miss. Carmack doesn't make games, he makes game engine technology. Whether or not id can use that engine technology to its fullest to make their games is a completely different question.
In terms of engine technology, Carmack is still good at finding low level insights and even high level algorithms. So that's why he's still relevant, even if people don't necessarily want to buy into his megatexture tech (there are huge trade-offs going down that path).
And to answer your question about his relevancy to mobile games, remember that every ounce of performance you can extract means you can either pick between better battery life or more . And in that respect, that's why he's relevant. Besides, there's nothing that he said in the article that isn't true, him being relevant or not.
Unfortunately, that's an extremely short-sighted point of view. For the maximum dissemination of information, the long-term survivability of Wikileaks comes into play. So holding some cards close as a form of insurance while revealing others will maximize the chances of survival, and thus the amount of information released.
Really? Time Warner vs. AT&T? Oh, great choices there.... Or do you mean how Verizon customer service responded to me, "Sorry, your address is within the AT&T area, and we aren't allowed to put FiOS there."
Then again, I'm not sure if that was meant to be tongue-in-cheek....
Other than the whole Ship of Theseus thing, I would assume if I lost half of my memories, I'd be half of my former self in terms of memory. Similarly, if I lost half of my "personality", I'd be half of my former self in terms of personality.
This is all assuming that I can't "fill in the gaps" reliably, of course. Then again, I'm no philosopher.
If DRM doesn't improve the sales of a game, I'm pretty certain that developers won't bother taking the time to painstakingly implement and test DRM. In other words, shooting yourself in the foot is stupid, and developers that survive to implement DRM aren't (usually) stupid.
(Obviously, this requires the developers to have reliable information. But that's a topic for another day.)
The kinds of programming gods I know of (ones whose only concern is speed within the memory constraints) all know quite a few languages.
But when it gets down to it, they will hand roll their own assembly code. Now you may think that compilers could do all this for you...but then that just shows how naive that assumption is. Compilers actually do a pretty shitty job. Just think of this simple C code:
uint8_t foo = function_param_1;// function_param_1 is something passed into the current function scope
foo++;
Now, what do you think the compiler for a 32-bit RISC processor would do? Most likely involving something to do with loading the memory values into a register and do some add immediate. And then it will do another AND against 0x000000ff. Why? Because the compiler doesn't know what foo is, and will need to mimic the wrapping behavior on an 8-bit variable. But clearly if you, the programmer, know that you'll never get 255 as function_param_1, then that extra AND is totally useless.
Basically, the problem is that the compiler never really knows what you're intending to do. It also never knows what you're expecting as the input so it always has to use the safest route. Sometimes you can substitute an even pipe instruction for an odd pipe instruction. And when your inner loop involves only 10 instructions and runs a billion times per second (no need to go into loop unrolling for now), shaving that extra instruction off and/or rebalancing the odd/even pipes so that you can run it in
And for those who don't know assembly and what algorithms to use...well, they've got a lot to learn anyway.
Right, so you counter their "numbers out of their asses" with your numbers, while accurate for your one sample, is most likely also completely "numbers out of your ass" relative to the true value. Since everyone is pulling numbers out of their asses, let me pull another one out of my ass: for every person I know that buys software, there are 5 or 6 that pirate them (I'm only using people I know not from work). All of these people are working now, making enough to buy most apps they pirate.
Do they pirate apps? Yes. Would some (>= 1) of these people buy the software they pirate if they somehow couldn't pirate them? Definitely. So would I say the numbers in the article are correct? Nope, but they sure look like they're underestimating the impact to me!
You are totally correct. In addition, there's also the problem that the JVM (for Java devices) implementations on these phones are complete shit where the libs don't work accord to spec, even for the commonly used libs (think on the level of 1+1=3). Furthermore, the phones' computing power range in the orders of magnitude from each other. This results in code that can't run anywhere other than the target platform that the developer coded on. This is why there are huge porting houses, and why small time developers can't "break into" the market (since they need the funds to port the apps to the plethora of phones).
I know BREW devices (and maybe Windows Mobile) devices are better, but they still have their share of problems to this day. But this is exactly why the iPhone/iPod touch is so much better to work with: essentially a very limited set of platforms that work exactly as advertised, cutting out the costly middleman.
Yes, but this is generally why the government tries to not interfere with the private sector, and that everyone knows the private sector shouldn't be allowed to provide public service. For example, I'm pretty sure that you wouldn't want to drive on roads designed by the 802.11n committee...who knows, you might get a zigzag road with 200mph speed limit signs posted all over!
Also, I like how you say that the private sector policies itself better than the government in many respects, but not most or all respects. C'mon, no one is saying that the private sector sucks, but let's not try to paint the picture as if it's way better than the public sector. I mean, I'm not arguing that the government is terrific, but I'm pretty sure you and I wouldn't want to drink milk from some place where food is also made by the private sector, although a lot less inspected.
The PS3 and 360 PPC elements are identical, yes. But the rest aren't. The SPUs are vastly different to the PPEs ranging from the ISA, to the memory architecture, to the instruction latencies, to the register file size/width, to the local memory latencies, to the...oh boy, they're vastly different on so many levels. I also don't know how those TFLOP numbers came about, because they're totally wrong.
Comparing GTAIV on the 360 vs the PS3, the 360 looks like it's running in 16-bit color depth, shadows are absolutely horrible, and the draw distance isn't even on par with the PS3.
Using GTA to compare the graphics hardware and concluding that PS3 is better? I just hope you don't mention that to the devs, because they'll laugh their ass off about how wrong that comment is.
the PS3 has 256MB of GDDR3 for their GPU, and the 256MB of XDR DESTROYS the 512MB of GDDR3 that the 360 uses for system memory (For one GDDR3 isn't meant to be used as main system memory, XDR is.)
On the XDR front, I don't know how it destroys the GDDR3. Both are pieces of memory and they're just there to support reads and writes. As long as they have the bandwidth, size, and low latency, that's all that really matters to devs (obviously, devs shouldn't have to worry about signal integrity and what not here).
PS3 stomps the 360. The 360 is by far inferior, it's locked down, and it burns itself out more often than not.
And the slim isn't locked down? But true, the original PS3 doesn't burn itself out more than the original 360.
Without breaking NDAs, pretty much every game renders into the eDRAM (I'm not even sure if you can render into main mem directly). That's the piece of on-chip hardware that nets you the free AA and z-writes because of the 256GB/s and ROP units. Also, it has nothing to do with caches or VRAM...it's just a piece of write-only memory.
On the PS3 front, it does have 20/15 R/W performance to the Cell FlexIO, but it also has the 22.4 to the GDDR3. Obviously, it's important to use both at the same time.
As for GTA, they've been very, very concerned about having high quality graphics. There's no doubt about that.
I don't know about you, but where I'm from (games), the devil's in the details. Oftentimes, coming up with a branchless solution for what amounts to 3 lines of code is the determining factor of whether or not a feature makes it into a game.
And yes, we are that strapped for performance. We don't have the leisure that most developers have with their expensive and quadruple nested if-else if-else statements code. For us, it has to be elegant and cheap (counted in hardware cycles spent both working and waiting).
Totally true for the Big Gov part. And yes, totally true for the capitalists part. No arguing about that.
I'd just like to add something to the capitalists part. For substitutable goods and services, this works marvelously (which is what the US government is saying they're trying to stay out of). These tend to be goods and services with either high obsolescence, or low cost to entry into the market. (Of course, this is just a simplification, not meant to be dissected.)
However, for everything else, such as buying that MRI, drug research, health insurance, etc... the "free market" doesn't work so well...since it's, well, not really that free (as in, freely moving and competitive) anymore.
If by high prices you mean I'm paying $5USD or less to go see the doctor and pay for prescriptions, then, well, I don't know what to say. I'm also pretty sure I don't have to wait a month for a basic checkup. Have you even lived in these countries before?
Slashdot Mirror
The biggest disservice to the American people is the American Dream and that one day, they will achieve it. And we all know by statistics above, that conservatively 99% won't. So why are there so many people supporting tax breaks for the rich? Because they can't accept the reality that they most likely won't be rich one day.
Personally, I'd be more than happy if I was taxed at the special $1M tax bracket -- that means I actually made at least $1M that year.
The argument between GC vs. manual memory management a la malloc/free is partly why there exists a debate if Java is as fast as C/C++.
First of all, there's no clear line between "GC" vs. "manual memory management". At some point, writing your own GC for your data structure is both "GC" and "manual memory management" at the same time. Unless you're talking about the default GC, otherwise GC is just a form of memory management.
Secondly, anyone who knows how to do heavy multithreaded programming on C/C++ knows to not rely on the global malloc and free. Either you use the platform-specific memory allocate and commit routines, or you use malloc and free very sparingly. After you get the chunk of memory from the system, you manage it yourself, using the appropriate memory allocator suitable for your data requirements. There is absolutely no one-size-fits all solution that runs fast.
As for handcrafted asm seldomly outperforming compiler code, that's probably because most asm writers are just doing it for the knowledge and experience, and not for real optimization purposes. Real asm code in tight loops done by pros usually outperform their compiler counterparts on the orders of 10~100x. At work, one of our pros even got to 1000x once...and that pro is no slouch when it comes to plain C. Of course, modern Intel CPU architectures throws a huge wrench in that analysis...but you pay for that in terms of power and die size...something not visible to the average programmer.
The moral of the story is, if you take the worst practices of one language and pit it against the best practices of another language, then sure, you may get "surprising" results.
Yes, and it's called "hedging." Southwest doesn't "speculate" -- they "hedge" (for the visible part anyway).
Yeah, I remember games that ran on 640KB or less as well. Heck, some games that people consider great ran on 80 bytes. But that's not the point. Devs will always ship their games in-memory -- otherwise, you don't ship at all, since it won't run. The point being, having more RAM makes life WAY easier, which means faster to-market and better games for relatively little cost.
Beyond the rumored specs, there's really only one thing devs really want: more RAM. Don't fucking skimp out on it -- shove loads of it in there. Even if they have to tier it into faster and slower RAM, do it (we'll learn to stream/ping pong/double buffer). Just don't give us shit like 512MB anymore (not in this day and age anyway). RAM is probably the cheapest, easiest, and fastest way to get thanks from devs, which translates to better games in every single way. Give us a serving of at least 1024 MB for main RAM, and 512 for GPU RAM (or at least 1.5GB if it's unified). If they can make it 2GB+ total, all the merrier.
The next on the list would be something bigger than the rumored 8GB flash, if they want to go anywhere near online distribution or content.
Wait, are you saying that BR players can't play DVDs? How does buying a BR player has anything to do with rebuying your DVD collection. The summary talks about even split in the sales of the players, not the discs themselves.
Of course, if you're just being funny by semi-quoting the article (the article talks about people not knowing they can play DVDs in their BR players), then just ignore this comment.
When it was just that, it wasn't a problem. Most of my old games would happily install on your machine if I sold you the disc and you typed in the key. The keys were stickers on the jewel cases, and there was none of this draconian "You can only install this game 5 times, and only on Tuesdays"
Really? Entering a code is equivalent to install this game 5 times and only on Tuesdays? I understanding it's tongue-in-cheek, but let's not get too carried away. You're comparing apples and computer games here.
Because it's not a subscription service that you could download the client for free anyway unless someone scammed you as in your example? You're comparing apples and steaks here.
Last I recall, WoW's latest expansions weren't free. The original WoW when it was released wasn't free either.
How do I access that profile once I sell the game? I bought it, right?
Yes, you bought the physical CD. No, you didn't buy the usable sticker. It's like buying the candy wrapper from someone else and complaining that it didn't come with the candy. Maybe if you didn't know that going into the transaction, that'd be something else.
It's not *now*. And therein lies the problem. It's an end run around the first sale doctrine by basically saying, "We didn't sell you that, we "licensed" it to you". Imagine if you couldn't buy a used car without paying Ford a "transfer fee" for the keys. ... I've never sold a game in my life, or bought one used for that matter ... the few bucks just isn't worth the hassle. But many, many people do - because they can't afford to buy everything they want new. There is a fairly huge secondary market with console games, and the game companies want to eliminate it because they somehow think people will magically have more money to spend.
Personally
And therein lies the misinterpretation of the first sale doctrine. The game studio sold you a game with a $15 voucher. And what's wrong with that?
I'm not against used games, especially the single player component. But think about it this way: why should Gamestop profit (hugely) from these sales while the original content makers don't? How is that helping the people making the games for you in the first place? How does that logic even make sense? If anything, you should be criticizing the incorporated used games sellers for profiting with such a high margin. You should demand that their used games come down in price by at least the same cost as the "replacement voucher" (if it's used up...and it's up to them to find out if it's used or not). That way, the public (you) wins, the game makers win, and the people that shouldn't be winning don't win.
No, you got it completely wrong. China is criticizing the US for being hypocrites, not censorship.
As a matter of fact, the author even has it wrong; it's not the pot calling the kettle black at all. That specifically only applies when one criticizes another for doing something that he himself does. In this case, China is calling the US a hypocrite when it comes to freedom, while China has never purported to be freedom-embracing to begin with. Like someone else said, US == hypocrite, China != hypocrite, and therefore, in that regard (especially pertaining to internet freedom), China is not the same as the US, and therefore it's not pot calling kettle black.
You're equating games to engine technology, and that's fundamentally the distinction that many people miss. Carmack doesn't make games, he makes game engine technology. Whether or not id can use that engine technology to its fullest to make their games is a completely different question. In terms of engine technology, Carmack is still good at finding low level insights and even high level algorithms. So that's why he's still relevant, even if people don't necessarily want to buy into his megatexture tech (there are huge trade-offs going down that path). And to answer your question about his relevancy to mobile games, remember that every ounce of performance you can extract means you can either pick between better battery life or more . And in that respect, that's why he's relevant. Besides, there's nothing that he said in the article that isn't true, him being relevant or not.
Unfortunately, that's an extremely short-sighted point of view. For the maximum dissemination of information, the long-term survivability of Wikileaks comes into play. So holding some cards close as a form of insurance while revealing others will maximize the chances of survival, and thus the amount of information released.
Really? Time Warner vs. AT&T? Oh, great choices there.... Or do you mean how Verizon customer service responded to me, "Sorry, your address is within the AT&T area, and we aren't allowed to put FiOS there."
Then again, I'm not sure if that was meant to be tongue-in-cheek....
Other than the whole Ship of Theseus thing, I would assume if I lost half of my memories, I'd be half of my former self in terms of memory. Similarly, if I lost half of my "personality", I'd be half of my former self in terms of personality.
This is all assuming that I can't "fill in the gaps" reliably, of course. Then again, I'm no philosopher.
If DRM doesn't improve the sales of a game, I'm pretty certain that developers won't bother taking the time to painstakingly implement and test DRM. In other words, shooting yourself in the foot is stupid, and developers that survive to implement DRM aren't (usually) stupid.
(Obviously, this requires the developers to have reliable information. But that's a topic for another day.)
The kinds of programming gods I know of (ones whose only concern is speed within the memory constraints) all know quite a few languages.
// function_param_1 is something passed into the current function scope
But when it gets down to it, they will hand roll their own assembly code. Now you may think that compilers could do all this for you...but then that just shows how naive that assumption is. Compilers actually do a pretty shitty job. Just think of this simple C code:
uint8_t foo = function_param_1;
foo++;
Now, what do you think the compiler for a 32-bit RISC processor would do? Most likely involving something to do with loading the memory values into a register and do some add immediate. And then it will do another AND against 0x000000ff. Why? Because the compiler doesn't know what foo is, and will need to mimic the wrapping behavior on an 8-bit variable. But clearly if you, the programmer, know that you'll never get 255 as function_param_1, then that extra AND is totally useless.
Basically, the problem is that the compiler never really knows what you're intending to do. It also never knows what you're expecting as the input so it always has to use the safest route. Sometimes you can substitute an even pipe instruction for an odd pipe instruction. And when your inner loop involves only 10 instructions and runs a billion times per second (no need to go into loop unrolling for now), shaving that extra instruction off and/or rebalancing the odd/even pipes so that you can run it in
And for those who don't know assembly and what algorithms to use...well, they've got a lot to learn anyway.
Good job painting all L-1/H-1B employees as potential traitors.
And to the moderators, remind me how Slashdot modding works? Is it "+1 Interesting" and "+1 Underrated" for being a xenophobic?
Right, I forgot. USA! USA! USA! =P
Right, so you counter their "numbers out of their asses" with your numbers, while accurate for your one sample, is most likely also completely "numbers out of your ass" relative to the true value. Since everyone is pulling numbers out of their asses, let me pull another one out of my ass: for every person I know that buys software, there are 5 or 6 that pirate them (I'm only using people I know not from work). All of these people are working now, making enough to buy most apps they pirate.
Do they pirate apps? Yes. Would some (>= 1) of these people buy the software they pirate if they somehow couldn't pirate them? Definitely. So would I say the numbers in the article are correct? Nope, but they sure look like they're underestimating the impact to me!
You are totally correct. In addition, there's also the problem that the JVM (for Java devices) implementations on these phones are complete shit where the libs don't work accord to spec, even for the commonly used libs (think on the level of 1+1=3). Furthermore, the phones' computing power range in the orders of magnitude from each other. This results in code that can't run anywhere other than the target platform that the developer coded on. This is why there are huge porting houses, and why small time developers can't "break into" the market (since they need the funds to port the apps to the plethora of phones).
I know BREW devices (and maybe Windows Mobile) devices are better, but they still have their share of problems to this day. But this is exactly why the iPhone/iPod touch is so much better to work with: essentially a very limited set of platforms that work exactly as advertised, cutting out the costly middleman.
Not sure how much it matters to sales, though.
Yes, but this is generally why the government tries to not interfere with the private sector, and that everyone knows the private sector shouldn't be allowed to provide public service. For example, I'm pretty sure that you wouldn't want to drive on roads designed by the 802.11n committee...who knows, you might get a zigzag road with 200mph speed limit signs posted all over!
Also, I like how you say that the private sector policies itself better than the government in many respects, but not most or all respects. C'mon, no one is saying that the private sector sucks, but let's not try to paint the picture as if it's way better than the public sector. I mean, I'm not arguing that the government is terrific, but I'm pretty sure you and I wouldn't want to drink milk from some place where food is also made by the private sector, although a lot less inspected.
Local RTS games like these never use databases. They all use things like arrays and...arrays. That's it.
As for saving info after the game, such as stats, replay, and whatnot...those are just custom file formats.
The 360 and PS3 are practically identical.
The PS3 and 360 PPC elements are identical, yes. But the rest aren't. The SPUs are vastly different to the PPEs ranging from the ISA, to the memory architecture, to the instruction latencies, to the register file size/width, to the local memory latencies, to the...oh boy, they're vastly different on so many levels. I also don't know how those TFLOP numbers came about, because they're totally wrong.
Comparing GTAIV on the 360 vs the PS3, the 360 looks like it's running in 16-bit color depth, shadows are absolutely horrible, and the draw distance isn't even on par with the PS3.
Using GTA to compare the graphics hardware and concluding that PS3 is better? I just hope you don't mention that to the devs, because they'll laugh their ass off about how wrong that comment is.
the PS3 has 256MB of GDDR3 for their GPU, and the 256MB of XDR DESTROYS the 512MB of GDDR3 that the 360 uses for system memory (For one GDDR3 isn't meant to be used as main system memory, XDR is.)
On the XDR front, I don't know how it destroys the GDDR3. Both are pieces of memory and they're just there to support reads and writes. As long as they have the bandwidth, size, and low latency, that's all that really matters to devs (obviously, devs shouldn't have to worry about signal integrity and what not here).
PS3 stomps the 360. The 360 is by far inferior, it's locked down, and it burns itself out more often than not.
And the slim isn't locked down? But true, the original PS3 doesn't burn itself out more than the original 360.
Without breaking NDAs, pretty much every game renders into the eDRAM (I'm not even sure if you can render into main mem directly). That's the piece of on-chip hardware that nets you the free AA and z-writes because of the 256GB/s and ROP units. Also, it has nothing to do with caches or VRAM...it's just a piece of write-only memory.
On the PS3 front, it does have 20/15 R/W performance to the Cell FlexIO, but it also has the 22.4 to the GDDR3. Obviously, it's important to use both at the same time.
As for GTA, they've been very, very concerned about having high quality graphics. There's no doubt about that.
I don't know about you, but where I'm from (games), the devil's in the details. Oftentimes, coming up with a branchless solution for what amounts to 3 lines of code is the determining factor of whether or not a feature makes it into a game.
And yes, we are that strapped for performance. We don't have the leisure that most developers have with their expensive and quadruple nested if-else if-else statements code. For us, it has to be elegant and cheap (counted in hardware cycles spent both working and waiting).
Totally true for the Big Gov part. And yes, totally true for the capitalists part. No arguing about that.
I'd just like to add something to the capitalists part. For substitutable goods and services, this works marvelously (which is what the US government is saying they're trying to stay out of). These tend to be goods and services with either high obsolescence, or low cost to entry into the market. (Of course, this is just a simplification, not meant to be dissected.)
However, for everything else, such as buying that MRI, drug research, health insurance, etc... the "free market" doesn't work so well...since it's, well, not really that free (as in, freely moving and competitive) anymore.
If by high prices you mean I'm paying $5USD or less to go see the doctor and pay for prescriptions, then, well, I don't know what to say. I'm also pretty sure I don't have to wait a month for a basic checkup. Have you even lived in these countries before?