Sony developed an entirely new CPU architecture for the PS3.
Microsoft went well beyond the current state of the art for desktops: three custom PowerPC cores on one die, running at 3+ GHz.
And honestly, that's where much of the expense is coming from. It's not like SEGA (with the Genesis) or Nintendo (with the SNES or GC or GBA) or even Sony in the days of the PS1 decided to go with custom processors, let alone processors that shoot for the ultra-high end. Consoles have always been about custom hardware for some things, lowish-end commodity parts for everything else.
Modern consoles (PS2 and up) run *hot*. I've often wondered if the awkward styling is to *prevent* stacking. The cooling systems aren't designed for numerous hot boxes stacked in close quarters.
Not sure where the "Is Apple ready to take the plunge?" tag came from. Of *course* Apple is ready to take the plunge. They've already announced the switch to x86 processors, and they even gave a specific time frame (2006). It's not like there's any real question here.
Because, quite obviously, oil didn't cost as much then.
So many geeks clamor for the latest and greatest video card, then upgrade to a new one every year or so. Ditto for CPUs. In reality, if they were concerned about conserving power, they'd make due with what they had. And really, is it worth all the power that goes into creating a high-end video card, plus all the power it takes when in use (yes, they use less power when you're not running 3D apps), just so you can get some high frame rates in Half-Life 2 or F.E.A.R. or whatever? Now oil is 3x more expensive, and all of a sudden there's great worry.
On a side note, the current push it toward multi-core CPUs. Honestly, no one has made a multi-core CPU that uses less power than a single-core equivalent (and if they did, they could just drop the second core and they'd be there). Clearly the drive is toward using more energy, even if there's no clear benefit for many people.
Also, even works such as Encyclopedia Britannica has inaccuracies... I mean, the perfect informational book hasn't been written.
But you don't have Encyclopedia Britannica articles being written by people with no knowledge of a subject, who are unwilling to check facts, and who lean almost entirely on hearsay and personal issues. Many Wikipedia entries are appallingly bad. Even a techy entry like "compiler" appears to have been written by someone without basic knowledge. Sure, I can go in and fix them, only to have them "fixed" again by either the original author or someone else who doesn't see their own myth or incorrect assumption listed as fact.
Five years ago, this was big news. Very few companies were looking to the Solitaire / Minesweeper market. Some were, yes, and some of these were doing extremely well. Since then there has been a proliferation of casual game companies to the point where it's a difficult market to break into. A big reason for this is that the barriers for entry are low: games that involve fairly simple programming, aren't content heavy, aren't 3D, etc. You can toss together a Bewjeweled-like game in a week, plus spend another few weeks polishing it up. What's worse is that there's very little innovation of any kind. Everyone is cloning the same handful of games, usually bubble-popping games (like Bejeweled). And it's unclear that the casual market has any interest in innovation anyway. People want a mindless game like Solitaire or Bejeweled and that's that. They don't want an endless stream of games.
With 20 years and 95% market share they had the time, money and resources to create the most advanced operating system ever. Instead, all they ever produced was "good enough" - never on the leading edge, never innovative.
It dependso on your point of view. Linux, for example, was a clone of 15+ year old software when it was first written. Was it innovative? No. Was it on the leading edge? No. Even today, you still have UNIX/Linux based on tech that stopped advancing in the mid 1980s.
(Not trying to be anti-Linux here. Just showing you that "leading edge" and "innovative" are tough terms to deal with.)
I'm so sick of this. Style and realism are not opposites. Realism is just one of many visual styles that a game could adopt.
And just because you choose realism, that doesn't say much about the visual style or flair of your game. The vast majority of photography is realistic, and no one would argue that all photographers have the same style.
Actually, that paper has been shot down numerous times since its publication. I dont' mean by Slashdotters, but by other published papers. Appel's argument is true in theory, but cache effects make stack allocation faster.
Anyone I interviewed who pulled out the xor-swap would immediately be pinned as a wannabe. The normal code, with a temporary, is both clearer and faster.
One thing which helps immensely is also having a strong type system with "sum" types
You can do the same thing with C++ classes, where State1 is derived from "state" and so on. Having tried things this way, I much prefer having having super-lightweight, cooperative processes. Manually handling state transitions is just bulky.
"Will Autodesk use the inherited expertise and codebase to finally develop their product line for the platforms most of their customer base would prefer"...oh come on. That's your personal view extrapolated to include the whole world. From my point of view, Windows is _the_ platform for 3D game development, and that's one of the largest uses of 3D modelling software. See? Your view is not everyone's.
(I'm already going to be flagged as a troll, so I might as well add that I sincerely wish Slashdot's editors would go for flat or enthusiastic teasers and not loaded, spin-full ones.)
On the one hand, computer technology has progressed by leaps and bounds. Compare what you can get for $500 today with what would have cost $5000 twenty years ago. On the other hand, we're still having loads of trouble making software reliable. There are things that you can do to improve this, yes, but it's certainly not where the money is being poured at the moment. And we're nowhere near making any advances toward "real" AI. We're no closer today than we were in the sixties, except we can solve some brute force problems faster. We're also nowhere near understanding how the brain works. We like to think we have a clue here, but even the best writings on the subject are akin to writings about chemistry from the sixteenth century.
No, it isn't. More than 99% of internet users don't have blogs, yet the tiny minority who do get an inordinate amount of press time - much like PETA in that regard.
More than 99% of internet users don't have websites either, so I guess websites are fad?
Let's face it: IE and Firefox are very similar
on
Firefox Momentum Slows
·
· Score: 4, Interesting
Firefox has always had the following noticible benefits:
1. better security 2. tabbed browsing 3. pop-up blocking 4. more polished UI for certain things (like the options dialog)
#2 and #3 are in IE. #1 is a big deal, but you've always been able to get the same effect by changing security settings in IE. With Microsoft switching to those settings, the difference is negligible. #4 I like, but it's just a tweak; it's not like MS couldn't do this.
So, in all honesty, Firefox has always been, more or less, IE+. The key point is that open source developers managed to get something that worked as well as IE, then add some nice bonus features. In regard to the bonus features, it's not like Microsoft (a) wouldn't borrow good ideas from Firefox, and (b) would have a tremendous amount of trouble adding some additional features to an existing application. From that perspective, anyone who expect Firefox to blow IE out of the water was fooling themselves. There was a nice lull in IE development which allowed some catch-up time, and it's good that the Firefox team could take advantage of it.
But now we're at the stage where FF and IE are essentially equals. Microsoft engineers could very easily have significant improvements to IE in the works; something that could become the standard for web browsing. If that happens, it's going to make the FF team look out of touch, spending all that time just to clone a previous generation product and make some improvements to it. True vision is something that FF development is lacking.
Wow, the submitter seems to think that blogs are worthless, yet it's a huge phenomenon, and seems puzzled as to why. I've seen this attitude before--it's common on Slashdot--but it's misguided. A weblog is simply someone posting their thoughts on a topic that interests them. It could be links to other sites, it could be software development, it could be graphic arts, it could be TV commercials, it could simply be what appear to be mundane details about daily life. The key is that you ignore what you don't care about. The mundane detail blogs are intended for family and friends (but could still be read by anyone who might want to). The graphic arts blogs are likely only of interest to other graphic artists. Slashdot-types might like software development blogs, Linux advocacy blogs, OS X blogs, and so on. There's no need to be cynical just because other people are writing about topics you have no interest in.
Yes, even though only some people need 4WD and high ground clearance, SUVs are clearly the better vehicle and we all should have them.
(Mod away!
Re:What I've learned from Pugs.
on
State of the Onion 9
·
· Score: 2, Insightful
Pugs is a Perl 6 implementation. It is written in Haskell. I recently fooled around with it. What did I learn? Haskell is powerful. Perhaps even more powerful than Perl.
Perl was never about raw power. Perl has always been about providing quick access to stuff you need often: hash tables, regular expressions, plowing through files, and so on. Haskell is a more powerful language on a fundamental level, but not on the day-to-day usability level. They each have their uses.
Most of the responses--and the original question--are along "more, bigger, faster" lines of thought. Here's my take:
1. First, don't bother with a desktop set-up. Get a notebook. As money is no object, get a high-end notebook. The key is that this is something you want for after-hours work, not your day job. You want something that doesn't keep you holed up in an office, something you can take out on the patio, on business trips (when you almost always have a lot of free time in the evenings), or just around town. You'll be lots more productive that way.
2. "Ultimate" appears to hobbled with the underlying premise of "using today's most prosaic and uninspiring technology." Unlesss you have a specific moonlighting gig, do you really want to be writing Java and C++ at home? What you want is something that gives you an edge, something that makes you more productive than entire departments. You should be looking into Lisp, Python, Smalltalk, Haskell...whichever of these floats your boat. Push the limits. Do the stuff that other people think is too hard to approach.
Intel CPU check. Nvidia GPU check. Hard drive check. Runs a version of Windows check. Can run Linux check. Sounds like a PC to me.
First, it doesn't run a version of Windows. It runs a very minimal kernel, something that can barely be called an operating system. 99.99% of Windows calls don't exist. Ditto for Direct3D. It's hardly the same as what people consider Direct3D on the PC. It's a minimal little subset that lets you dump hardware-specific data to the GPU. Running Liunx is irrelevant. You can run Linux on a Palm, but do you really want to?
The big difference is that you don't have virtual memory, so you have to manage memory and prevent fragmentation yourself. That's far and away from PC development.
They've been very lucky, releasing highly polished articles at just the time when consumers, spoiled by choice, are beginning to use quality of a design as a differentiator betweem almost equal rival products.
I'm not sure if you've just realized that design has always been important for high-end products or what. This is hardly a new phenomenon. Look at the entire history of the Mac, for example. Or designer sound systems. Or cars.
Slashdot Mirror
This is just the first time such large multi-core chips have been put in a consumer device
Hence "state of the art."
Sony developed an entirely new CPU architecture for the PS3.
Microsoft went well beyond the current state of the art for desktops: three custom PowerPC cores on one die, running at 3+ GHz.
And honestly, that's where much of the expense is coming from. It's not like SEGA (with the Genesis) or Nintendo (with the SNES or GC or GBA) or even Sony in the days of the PS1 decided to go with custom processors, let alone processors that shoot for the ultra-high end. Consoles have always been about custom hardware for some things, lowish-end commodity parts for everything else.
Modern consoles (PS2 and up) run *hot*. I've often wondered if the awkward styling is to *prevent* stacking. The cooling systems aren't designed for numerous hot boxes stacked in close quarters.
Not sure where the "Is Apple ready to take the plunge?" tag came from. Of *course* Apple is ready to take the plunge. They've already announced the switch to x86 processors, and they even gave a specific time frame (2006). It's not like there's any real question here.
Because, quite obviously, oil didn't cost as much then.
So many geeks clamor for the latest and greatest video card, then upgrade to a new one every year or so. Ditto for CPUs. In reality, if they were concerned about conserving power, they'd make due with what they had. And really, is it worth all the power that goes into creating a high-end video card, plus all the power it takes when in use (yes, they use less power when you're not running 3D apps), just so you can get some high frame rates in Half-Life 2 or F.E.A.R. or whatever? Now oil is 3x more expensive, and all of a sudden there's great worry.
On a side note, the current push it toward multi-core CPUs. Honestly, no one has made a multi-core CPU that uses less power than a single-core equivalent (and if they did, they could just drop the second core and they'd be there). Clearly the drive is toward using more energy, even if there's no clear benefit for many people.
Also, even works such as Encyclopedia Britannica has inaccuracies...
I mean, the perfect informational book hasn't been written.
But you don't have Encyclopedia Britannica articles being written by people with no knowledge of a subject, who are unwilling to check facts, and who lean almost entirely on hearsay and personal issues. Many Wikipedia entries are appallingly bad. Even a techy entry like "compiler" appears to have been written by someone without basic knowledge. Sure, I can go in and fix them, only to have them "fixed" again by either the original author or someone else who doesn't see their own myth or incorrect assumption listed as fact.
Indie games? Direct download? Where has the submitter of this piece been?
Five years ago, this was big news. Very few companies were looking to the Solitaire / Minesweeper market. Some were, yes, and some of these were doing extremely well. Since then there has been a proliferation of casual game companies to the point where it's a difficult market to break into. A big reason for this is that the barriers for entry are low: games that involve fairly simple programming, aren't content heavy, aren't 3D, etc. You can toss together a Bewjeweled-like game in a week, plus spend another few weeks polishing it up. What's worse is that there's very little innovation of any kind. Everyone is cloning the same handful of games, usually bubble-popping games (like Bejeweled). And it's unclear that the casual market has any interest in innovation anyway. People want a mindless game like Solitaire or Bejeweled and that's that. They don't want an endless stream of games.
With 20 years and 95% market share they had the time, money and resources to create the most advanced operating system ever. Instead, all they ever produced was "good enough" - never on the leading edge, never innovative.
It dependso on your point of view. Linux, for example, was a clone of 15+ year old software when it was first written. Was it innovative? No. Was it on the leading edge? No. Even today, you still have UNIX/Linux based on tech that stopped advancing in the mid 1980s.
(Not trying to be anti-Linux here. Just showing you that "leading edge" and "innovative" are tough terms to deal with.)
I'm so sick of this. Style and realism are not opposites. Realism is just one of many visual styles that a game could adopt.
And just because you choose realism, that doesn't say much about the visual style or flair of your game. The vast majority of photography is realistic, and no one would argue that all photographers have the same style.
An article previously mentioned on Slashdot.
You can eliminate one more local variable using an XOR swap.
You're trying to outthink the compiler and failing.
Actually, that paper has been shot down numerous times since its publication. I dont' mean by Slashdotters, but by other published papers. Appel's argument is true in theory, but cache effects make stack allocation faster.
Anyone I interviewed who pulled out the xor-swap would immediately be pinned as a wannabe. The normal code, with a temporary, is both clearer and faster.
One thing which helps immensely is also having a strong type system with "sum" types
You can do the same thing with C++ classes, where State1 is derived from "state" and so on. Having tried things this way, I much prefer having having super-lightweight, cooperative processes. Manually handling state transitions is just bulky.
"Will Autodesk use the inherited expertise and codebase to finally develop their product line for the platforms most of their customer base would prefer"...oh come on. That's your personal view extrapolated to include the whole world. From my point of view, Windows is _the_ platform for 3D game development, and that's one of the largest uses of 3D modelling software. See? Your view is not everyone's.
(I'm already going to be flagged as a troll, so I might as well add that I sincerely wish Slashdot's editors would go for flat or enthusiastic teasers and not loaded, spin-full ones.)
On the one hand, computer technology has progressed by leaps and bounds. Compare what you can get for $500 today with what would have cost $5000 twenty years ago. On the other hand, we're still having loads of trouble making software reliable. There are things that you can do to improve this, yes, but it's certainly not where the money is being poured at the moment. And we're nowhere near making any advances toward "real" AI. We're no closer today than we were in the sixties, except we can solve some brute force problems faster. We're also nowhere near understanding how the brain works. We like to think we have a clue here, but even the best writings on the subject are akin to writings about chemistry from the sixteenth century.
No, it isn't. More than 99% of internet users don't have blogs, yet the tiny minority who do get an inordinate amount of press time - much like PETA in that regard.
More than 99% of internet users don't have websites either, so I guess websites are fad?
Firefox has always had the following noticible benefits:
1. better security
2. tabbed browsing
3. pop-up blocking
4. more polished UI for certain things (like the options dialog)
#2 and #3 are in IE. #1 is a big deal, but you've always been able to get the same effect by changing security settings in IE. With Microsoft switching to those settings, the difference is negligible. #4 I like, but it's just a tweak; it's not like MS couldn't do this.
So, in all honesty, Firefox has always been, more or less, IE+. The key point is that open source developers managed to get something that worked as well as IE, then add some nice bonus features. In regard to the bonus features, it's not like Microsoft (a) wouldn't borrow good ideas from Firefox, and (b) would have a tremendous amount of trouble adding some additional features to an existing application. From that perspective, anyone who expect Firefox to blow IE out of the water was fooling themselves. There was a nice lull in IE development which allowed some catch-up time, and it's good that the Firefox team could take advantage of it.
But now we're at the stage where FF and IE are essentially equals. Microsoft engineers could very easily have significant improvements to IE in the works; something that could become the standard for web browsing. If that happens, it's going to make the FF team look out of touch, spending all that time just to clone a previous generation product and make some improvements to it. True vision is something that FF development is lacking.
Wow, the submitter seems to think that blogs are worthless, yet it's a huge phenomenon, and seems puzzled as to why. I've seen this attitude before--it's common on Slashdot--but it's misguided. A weblog is simply someone posting their thoughts on a topic that interests them. It could be links to other sites, it could be software development, it could be graphic arts, it could be TV commercials, it could simply be what appear to be mundane details about daily life. The key is that you ignore what you don't care about. The mundane detail blogs are intended for family and friends (but could still be read by anyone who might want to). The graphic arts blogs are likely only of interest to other graphic artists. Slashdot-types might like software development blogs, Linux advocacy blogs, OS X blogs, and so on. There's no need to be cynical just because other people are writing about topics you have no interest in.
Yes, even though only some people need 4WD and high ground clearance, SUVs are clearly the better vehicle and we all should have them.
(Mod away!
Pugs is a Perl 6 implementation. It is written in Haskell. I recently fooled around with it. What did I learn? Haskell is powerful. Perhaps even more powerful than Perl.
Perl was never about raw power. Perl has always been about providing quick access to stuff you need often: hash tables, regular expressions, plowing through files, and so on. Haskell is a more powerful language on a fundamental level, but not on the day-to-day usability level. They each have their uses.
Most of the responses--and the original question--are along "more, bigger, faster" lines of thought. Here's my take:
1. First, don't bother with a desktop set-up. Get a notebook. As money is no object, get a high-end notebook. The key is that this is something you want for after-hours work, not your day job. You want something that doesn't keep you holed up in an office, something you can take out on the patio, on business trips (when you almost always have a lot of free time in the evenings), or just around town. You'll be lots more productive that way.
2. "Ultimate" appears to hobbled with the underlying premise of "using today's most prosaic and uninspiring technology." Unlesss you have a specific moonlighting gig, do you really want to be writing Java and C++ at home? What you want is something that gives you an edge, something that makes you more productive than entire departments. You should be looking into Lisp, Python, Smalltalk, Haskell...whichever of these floats your boat. Push the limits. Do the stuff that other people think is too hard to approach.
Intel CPU check.
Nvidia GPU check.
Hard drive check.
Runs a version of Windows check.
Can run Linux check.
Sounds like a PC to me.
First, it doesn't run a version of Windows. It runs a very minimal kernel, something that can barely be called an operating system. 99.99% of Windows calls don't exist. Ditto for Direct3D. It's hardly the same as what people consider Direct3D on the PC. It's a minimal little subset that lets you dump hardware-specific data to the GPU. Running Liunx is irrelevant. You can run Linux on a Palm, but do you really want to?
The big difference is that you don't have virtual memory, so you have to manage memory and prevent fragmentation yourself. That's far and away from PC development.
They've been very lucky, releasing highly polished articles at just the time when consumers, spoiled by choice, are beginning to use quality of a design as a differentiator betweem almost equal rival products.
I'm not sure if you've just realized that design has always been important for high-end products or what. This is hardly a new phenomenon. Look at the entire history of the Mac, for example. Or designer sound systems. Or cars.