Out of all the world's laptops, how many are *really* being used primarily for 3D video games?
Out of all the world's laptops, how many are *really* being used for wordprocessing, spreadsheets, e-mail and other simple data
processing?
Right. So why, oh why, do the dumb knobs keep focusing on stupid things like clockspeed, 3d video, dvd players and shit like that?
Well you obviously have a point that a laptop oriented to the business world will generate more revenue, but that's simply because that business world has more cash to spend on such things. The reason that no laptops are primarily used for 3D is that there are no laptops suitable to gaming. Just ask anyone who ever tried to get a reasonably hightech 3D gaming going on a laptop. Seeing as the games industry just keeps growing (it eclipsed the movie industry quite a while ago already), I can see how a laptop with decent 3D support finds its market both with developers and gamers..
Half the performance of a real G2 still gets you the register combiner extensions (for hardware spotlights etc), on-card geometry (even if at half the T&L speed, they still cut on DMA transfer), accelerated render-to-texture, full 32bpp, 8 bit stencil buffer (shadow/mirror effects), and the best drivers (in terms of raw speed, features and crossplatform support) on the market.
And I think T&L is more than twice as fast as a common CPU doing the same tasks, especially considering the typical CPU in a laptop, and also considering that T&L covers more with every new release (more lightsources, vertex blending, vertex programming, etc). So you still win.
No, it's not great because there can now be a laptop with a Linux supported accelerator.
Yes, it is indeed totally pointless for business applications
And yes, relatively simple 3D games already run fine on the current chips.
But, have you ever tried to take your cutting edge 3D game, development tool, or engine to a tradeshow like ECTS or E3, to give a demonstration ? Currently, you can choose between either taking your full tower (right), a laptop with crappy 3D support, or a couple of demo CDs -- hoping for the best concerning publisher's hardware & driver uptodate-ness. A laptop with a cutting edge 3D chip with proper driver support would rock, which is exactly what NVidia has been delivering, save the "laptop" part.
Granted, it won't generate the revenue of a business model (well, maybe when VRML kicks off or something), but there are many (would-be) game developers waiting for this thing..
And about the screen part; when giving a demonstration a decent screen or even projector is usually available. It's the hardware+drivers that are the risk.
So.. you're getting a job which requires that you travel a lot, and in return you get a 50 buck voucher for making your calls ? I'd rather look for a new company that just refunds all business calls, no questions asked, instead of a new plan:)
Yes, but on the other hand, in certain countries (eg : Belgium), it's not legal to advertize using benchmarks to begin with. I think it'd be okay for Larry in a keynote speech, but he definitely couldn't do an ad on TV or radio or whatever saying "x% faster and y% cheaper than [competitor]"
Yeah right. In theory, sure. In practice.. come on. If you submit a perfectly valid patch to a project the size of, say, Apache, and it is refused for obvious political reasons, will you fork ? Ofcourse not;
Such projects have a whole team behind them, so you would need at least an equally sized team to take over. Assembled based on what, "they rejected my patch" ? Don't think so
Until you take over, all patches and other work will continue to flow to the "real" project. Will you monitor all diffs and apply them to your fork ? Will you update and redistribute diffs that interfere with the changes your fork has made (which might get exponentially worse in time) ?
For any project worth your time, a code fork is a huge effort, so in practice, a rejected patch means Game Over. Forget all that mumbo jumbo about "taking over" and "Darwinistic environments" (other posts)...
No, he's not saying they can't code, he's saying that there's no point in releasing those hobby projects as open source; coz, if you decided to write it all again from scratch, then why would anybody not do the same and ignore your efforts ?
That's a line of thought that's blatantly missing from the cheering descriptions of the bazaar model, it was about time somebody wrote it down.
Yes, exactly.
His point is not that Open Source has this flaw which Closed Software hasn't, it is indeed precisely that Open Source, in this regard, does not improve upon the Closed Source model. At the end of the day, it's still about the politics ! And this despite the nice theories put forward by the gurus, the zealots, the GPL loving crowds.
So what he says is that despite all those hallelujahs, OSS is in a large way as closed as true Closed Software. You didn't counter him, you supported him:)
I also hear what he's saying from experience.. It's only in the largest projects that code is accepted on the basis of code merit, quality, usefulness. Which is basically why those projects have become large; 90% of the projects out there are led by people who are just in it for the learning experience. Their code is not too amazing, they don't accept patches, and they manage their projects as badly as they code, so they stay small. That's fine for them, but it's not exactly the Bazaar model. That's all the article says, really.
I agree. Exactly the same post can be deserving of a +1, Interesting if it was at 1, or a -1, Overrated if it was at 4. In metamod, displaying the change in score would be a tremendous help in deducing what the moderator was trying to do.
Also, your point of banality and negative, but well-thought out commentary, may be a consequence of the moderation categories. Essentially, by using the term "Informative" you're ruling out a +1 for such a negative comment, it's Flamebait by definition -- from the mod's point of view. Perhaps it is possible to encourage mods to think in terms of how well is a post formulated, does it make sense, is it rational or emotional, by using different terms : "Rational", "Overemotional", or something like that. Not sure if overemotional wouldn't trigger the same kneejerk reaction as Flamebait, though. Another random idea is a -1 option for obvious karmawhoring, there's only Overrated for this now.
Huh ? From the fact that all interviews with The Carmack are about 3D APIs and videocards, you conclude that gaming has been reduced to videocards and benchmarks ?
Ofcourse, I agree that most games put way too much emphasis on the flash and bang, but otoh don't underestimate the importance of a good engine, either. You can hate the industry's focus on fillrates, but as long as those fillrates aren't high enough, you can simply forget about Doom-like rooms with literally 70-or-so imps in it -- unless you go back to 2D sprites. Yeah, that's an option, but no publisher'll buy it. So, for every 10 games that come out that are all about the tech & gfx but offer no gameplay, there is one game out there that has a designer who really wants the tech for his gameplay ideas. That's probably a fair ratio. You can argue in what category id's products fall -- but at the very least it's clear into which some of their licensees go (HalfLife anyone ?). So give 'm a break;)
Does anybody who's read the article have any idea what the vague references to Doom 3's tech are supposed to mean ? Between the lines it almost sounds like he's going the Unreal way : no more precalculated PVSes, but editor-placed portals, for more dynamic scenery and lower vis times.
Which would be pretty strange, since the PVS constant-time culling was arguably one of the breakthroughs enabling Quake's framerate. Is he saying Sweeney was right after all ? It's obvious that the Q3A engine is too static for a single-player focused Doom3, still, Valve did some nice hack-n-patch with the equally static Q1 code... Curious.
If you get spam with "xxx" anywhere in the url you know what it is -- so the whole point is exactly that the URLs don't want to tell you where you're going to. From which it follows that.xxx would never work without government enforcing it on those companies -- the latter wouldn't want it. And that is only possible with ICANN or some hired watchdog being legally obliged to check every site it registers, periodically. Don't think so.
Yeah, but at some point the net must be connecting back with meatspace, or else it's just an abstract pool of thoughts with no physical reflection (gee, do I sound like Katz yet ?)
What I mean is, you can consider tld's to be a simple "cookie" that partitions the group of sites to your locale. Why would I have to choose between Western & Kanji if the choice is obvious with a.uk appended ? Why ask a country code to avoid showing the list of all sony dealers worldwide on sony.com, when you might as well have typed sony.fr ? Besides, a country tld is still a valid method to avoid name collisions between unrelated, but similarly named companies in distinct countries (not all companies are ThereCanBeOnlyOne multinationals, you know).
Well, that's all I can think of, so it's probably not enough to outweigh your zillion counterpoints, but at least "no good reason" has been rejected;) So although the metaphorical net transcends location, it's actual content often does not.
Obviously that was back in the days when technology would still last 50 years. Why make something indestructable when it'll cost half as much and be twice as good in 6 months. Ok, maybe slightly exagerated, but cell phones etc are not that far off in terms of speech quality, battery life etc. People are glad it breaks, they got a reason for an upgrade.
Epic's main reason to focus on direct 3D is, as they put it, that opengl and direct 3D are diverting too much. Basically direct 3D is still gaining features while opengl is a low level 3d standard that does not change much anymore.
Well, this is true, but as you point out, the features that D3D is gaining over OpenGL are in the high-level domain; D3D is getting built-in support for reasonably exotic stuff like shaders, vertex morphing, continuous LOD etc. So you're right that Epic would in theory be reinventing the wheel if they wanted to support OpenGL, because they'd have to reimplement all that D3D stuff on top of OpenGL.
However, it's important to keep in mind that the higher the complexity goes, the tougher it becomes to write an API, and implement it, in such a way that everybody is happy feature & performance wise. As an example, there are numerous scene-graph libs for OpenGL that actually rock the house -- if you use them for what they were meant for. Like you say, they suck if you don't. Now I admit D3D features like shaders or LOD are still somehow on the borderline, there is probably only a very limited number of ways to implement this, so chances are that if you'd write it yourself you would indeed arrive at what D3D does, so don't bother. But as this trend continues it'll still be worthwile to implement highlevel stuff yourself. I mean, yes, content creation is a substantial burden these days, but that doesn't change the fact that a race game, a FPS and a flightsim still have massively different requirements from the highlevel parts of the 3D engine.
If you think about it, the argument of Epic that D3D is nice because they have a large input in it's development is pretty interesting. It almost seems to suggest that if hard decisions have to be made to trade off features vs performance as D3D gets more highlevel, the balance might swing towards "whatever benefits FPShooters", thanks to Epic..;);)
There are things that just aren't supported by OGL though, like hardware T&L
Um, wrong. OpenGL is an API, the way it works behind the scenes is hidden, and it could be a software rasteriser rendering your frames, it could be a hybrid setup that does the rastering in hardware, but everything else in software, or it could be a 100% hardware setup (such as an expensive SGI rig). I'm afraid you need to read up a bit:) The moment you plug in a T&L board and update the OpenGL driver, most well-written apps will get a speed increment (and that's not just theory, my crappy PII-200 got a new life with a GeF recently).
It would probably be pretty stupid for Epic to start ignoring Linux, now that the market is booming, and them having an edge with their cross platform engine of today. What it'll probably mean (if you read the part about "support for modular code" etc) is that there may be a Linux port, but it'll be seriously inferior to the D3D version. Given that the past problems where mostly with texture swapping, and GM's emphasis on D3D's "better API", I would guess that'll mean less textures, lower resolutions, less procedural effects in geometry and textures, less advanced shaders, and a few more, in the Linux version..
Well, this is only partially true. The thing is that progress in D3D often means nothing more than defining a number of new features in the docs. That's only an illusion of progress : sure, in theory you could say that D3D is more advanced than OpenGL because for instance D3D exposes 8 stage texture combiners since D3D6, while OpenGL still struggled to get multitexturing past the ARB, but on the other hand, as long as the hardware out there doesn't set the capability bits that actually make all the theoretical D3D feature *real*, all that M$ progress is just APIware. And it's just as easy for, say, NVidia, to expose a GL_EXT_thisandthat as to set a capability bit. Basically, I disagree with you saying that D3D > OpenGL, because OpenGL is open to start : any vendor can define extensions for his hardware; with D3D, everything has to come from M$.
I do agree though that D3D is currently the better environment for development when you want to build 3D engines for vast continuous levels; Windows' tendency to trash the videomemory upon a task switch forces OpenGL to waste huge amounts of memory on keeping copies of every texture. Sweeney complained about that quite a while ago, so it was evident back then that he'd either have to find a solution for it (hoping for a GL_EXT_nosystemcopy perhaps ?), or drop OpenGL for being too memory intensive.
About OpenGL being geared towards id. That's true, but it's probably a good thing. The 3D APIs of today are so complex that it's probably daydreaming to think a driver writer can optimise *everything*. The combined internal state is so huge that you cannot build an optimal path for each and every combination. One influential game that uses a limited set of the API gives driver writers an opportunity to max out performance on that path, which in turn allows other developers to write their code with that path in mind -- instead of just reading the docs and picking one of the 35 possibilities to specify their geometry, hoping it'll come out right and speedy. You're right that this situation is not without danger of getting stuck, though...
Your reply suffers from a lack of vision. You see "private schools" and you think of a private monopoly, as our "public schools" are a monopoly now. But what we are talking about is de-monopolising.
I think I see what you mean.. You're assuming that a strong set of guidelines which can be enforced if necessary, is a bad thing. Well, I'd agree if those guidelines were a bunch of crap. But I disagree, so I'll try to show that (a) the current guidelines are Good, (b) those guidelines are better than what would evolve by natural selection from a totally deregulated schoolsystem.
As for (a), I'm afraid I find the Taylor readings a bit over the top. It's suggesting that our schoolsystem is designed as a Borg/Sparta crossover to pump out soldiers instead of humans. Even if it were true, it might not have been such a bad thing to educate people into soldiers around 1806, or into assembly workers around 1900 (Mann). After all, you have to make a living, so if school brings abilities you can use to stay out of the sewers, why not. Ofcourse, if those abilities are thought at the expense of everything else (intelligence, creativity etc) there's a problem, but still, schools can teach you specific needs for the world of that day. As I'm good at math and code I think of myself as well-educated, but perhaps in 100 years when creativity is the most important property to develop people will laugh at our schools for thinking we were "educated" with all our math and crap. In short, a document which shows how 1806 schools wanted to build soldiers is not really much foundation for claiming that present-day schools have the same agenda.
As for (b)... Some statistic somewhere (dunno which or what) has claimed that a substantial portion of American people is technically analphabetic : either they can't read at all, or they read miserably, or they can read pretty ok but are unable to digest a solid text (like from a decent newspaper), extracting facts and logic as they go. True or not, it could be a simple result of the fact that most messages are delivered visually, or by TV. Does this mean the ability to read is passe ? I hope you agree the answer is Hell No. But I think that's the kind of conclusions non-regulated schools could jump to too easily (although ditching reading is kindof extreme) : take a short-sighted look at the needs of the present day and invent some guidelines for just that.. who cares about Asian geography, 14th centuries poets, numerical analysis, or electrons. Can you think of a world where everybody teaches their kids what they think is appropriate ? Note I said teaching, not raising : feel free to raise a kid with the morale you deem appropriate, but we're talking about forming intelligence, background, the ability to memorise and categorise large amounts of data, etc. Imho nobody will care for these abilities (you don't need them to watch TV or get groceries), and Civilisation Will Fall !;)
There's no contradiction between (a) and (b), as it's a matter of timescale.. Sorry for the long post:)
This entire -- entire -- argument boils down to exactly that: those who see WAVE as an aberation, and those who see WAVE as characteristic.
True. I believe we basically agree on the ultimate goal of school : create people which are well educated on a number of technical skills (math, history), but with unaffected non-technical abilities (creativity, interests,..).
The situation can be compared to general free-market policies : do you want a market which is completely free (the ideological Libertarian market), or a market which is free up to a certain limit, where the govt sits on it as a watchdog to avoid social or economic excesses ? I hope you go for the 2nd, as we all know what reckless exploitation of scarce resources, child labour, slavery etc have brought us. Well, I guess it's the same thing here; if the goods to trade are the children's minds, do you want a totally free market where corporate (ie non-state) schools can deliver any goods they want (no warranty, no liability, no guarantees) at any cost they think is ok ? Maybe you can point out a flaw in this analogy, but imho non-state schools are exactly the equivalent of a total-freedom free market.
Again I don't have much arguing to substantiate, so I'll pick up your challenge:) My mom is in fact a teacher in a state-governed school, and I'll ask her a detailed account on how policies are set, to what extent history could be warped if it better fits the current puppetmasters way up the govt chain, etc.
All I know is that I came out of school pretty much the way I like it, and I am a nerd/geek. I also don't know any other nerds or geeks which came out of a state-run highschool with traumas, supressed creativities or any other personality modifications that'd suit a Big Brother. That's in Belgium, though, and from what I gather here on/., USA schools must have a totally different attitude towards creativity and being different. Which is exactly why I still think of any problems as aberrations : it seems to work over here.
Well, maybe I was just conditioned into thinking that Our School Is Good;)
Hum, well, I seem to be slightly misunderstood here:) Ofcourse this patch is much more expensive to produce than "just food", is in a way impractical, makes the 3rd world dependent on 1st world hightech, and is most definitely not a long term solution.
But that's why it's called a Patch : I was specifically referring to it as a quick fix for particular disasters like the Ethiopia dryness of 1983. As you point out the army wouldn't feed their soldiers permanently on this, it's a one-stop solution for a particular situation. Similarly I thought of these patches as a non-structural quickfix for particularly nasty fuckups of mother nature. One C130 flying over dropping one crate of these sure helps surviving a whole lot more people than a crate of whatever else.
Hence also the link to milkpowder. I agree that Nestle's vision of powder as a permanent solution is a disaster -- so is this patch. Ok, so nevermind the political implications etc, it's literally a patch. Hope that clears it up ?:)
Slashdot Mirror
What you say !!
Out of all the world's laptops, how many are *really* being used for wordprocessing, spreadsheets, e-mail and other simple data processing?
Right. So why, oh why, do the dumb knobs keep focusing on stupid things like clockspeed, 3d video, dvd players and shit like that?
Well you obviously have a point that a laptop oriented to the business world will generate more revenue, but that's simply because that business world has more cash to spend on such things. The reason that no laptops are primarily used for 3D is that there are no laptops suitable to gaming. Just ask anyone who ever tried to get a reasonably hightech 3D gaming going on a laptop. Seeing as the games industry just keeps growing (it eclipsed the movie industry quite a while ago already), I can see how a laptop with decent 3D support finds its market both with developers and gamers..
And I think T&L is more than twice as fast as a common CPU doing the same tasks, especially considering the typical CPU in a laptop, and also considering that T&L covers more with every new release (more lightsources, vertex blending, vertex programming, etc). So you still win.
Yes, it is indeed totally pointless for business applications
And yes, relatively simple 3D games already run fine on the current chips.
But, have you ever tried to take your cutting edge 3D game, development tool, or engine to a tradeshow like ECTS or E3, to give a demonstration ? Currently, you can choose between either taking your full tower (right), a laptop with crappy 3D support, or a couple of demo CDs -- hoping for the best concerning publisher's hardware & driver uptodate-ness. A laptop with a cutting edge 3D chip with proper driver support would rock, which is exactly what NVidia has been delivering, save the "laptop" part.
Granted, it won't generate the revenue of a business model (well, maybe when VRML kicks off or something), but there are many (would-be) game developers waiting for this thing..
And about the screen part; when giving a demonstration a decent screen or even projector is usually available. It's the hardware+drivers that are the risk.
So.. you're getting a job which requires that you travel a lot, and in return you get a 50 buck voucher for making your calls ? I'd rather look for a new company that just refunds all business calls, no questions asked, instead of a new plan :)
Yes, but on the other hand, in certain countries (eg : Belgium), it's not legal to advertize using benchmarks to begin with. I think it'd be okay for Larry in a keynote speech, but he definitely couldn't do an ad on TV or radio or whatever saying "x% faster and y% cheaper than [competitor]"
- Such projects have a whole team behind them, so you would need at least an equally sized team to take over. Assembled based on what, "they rejected my patch" ? Don't think so
- Until you take over, all patches and other work will continue to flow to the "real" project. Will you monitor all diffs and apply them to your fork ? Will you update and redistribute diffs that interfere with the changes your fork has made (which might get exponentially worse in time) ?
For any project worth your time, a code fork is a huge effort, so in practice, a rejected patch means Game Over. Forget all that mumbo jumbo about "taking over" and "Darwinistic environments" (other posts)...That's a line of thought that's blatantly missing from the cheering descriptions of the bazaar model, it was about time somebody wrote it down.
His point is not that Open Source has this flaw which Closed Software hasn't, it is indeed precisely that Open Source, in this regard, does not improve upon the Closed Source model. At the end of the day, it's still about the politics ! And this despite the nice theories put forward by the gurus, the zealots, the GPL loving crowds.
So what he says is that despite all those hallelujahs, OSS is in a large way as closed as true Closed Software. You didn't counter him, you supported him :)
I also hear what he's saying from experience.. It's only in the largest projects that code is accepted on the basis of code merit, quality, usefulness. Which is basically why those projects have become large; 90% of the projects out there are led by people who are just in it for the learning experience. Their code is not too amazing, they don't accept patches, and they manage their projects as badly as they code, so they stay small. That's fine for them, but it's not exactly the Bazaar model. That's all the article says, really.
Also, your point of banality and negative, but well-thought out commentary, may be a consequence of the moderation categories. Essentially, by using the term "Informative" you're ruling out a +1 for such a negative comment, it's Flamebait by definition -- from the mod's point of view. Perhaps it is possible to encourage mods to think in terms of how well is a post formulated, does it make sense, is it rational or emotional, by using different terms : "Rational", "Overemotional", or something like that. Not sure if overemotional wouldn't trigger the same kneejerk reaction as Flamebait, though. Another random idea is a -1 option for obvious karmawhoring, there's only Overrated for this now.
PS, hide the karma :)
Ofcourse, I agree that most games put way too much emphasis on the flash and bang, but otoh don't underestimate the importance of a good engine, either. You can hate the industry's focus on fillrates, but as long as those fillrates aren't high enough, you can simply forget about Doom-like rooms with literally 70-or-so imps in it -- unless you go back to 2D sprites. Yeah, that's an option, but no publisher'll buy it. So, for every 10 games that come out that are all about the tech & gfx but offer no gameplay, there is one game out there that has a designer who really wants the tech for his gameplay ideas. That's probably a fair ratio. You can argue in what category id's products fall -- but at the very least it's clear into which some of their licensees go (HalfLife anyone ?). So give 'm a break ;)
Does anybody who's read the article have any idea what the vague references to Doom 3's tech are supposed to mean ? Between the lines it almost sounds like he's going the Unreal way : no more precalculated PVSes, but editor-placed portals, for more dynamic scenery and lower vis times. Which would be pretty strange, since the PVS constant-time culling was arguably one of the breakthroughs enabling Quake's framerate. Is he saying Sweeney was right after all ? It's obvious that the Q3A engine is too static for a single-player focused Doom3, still, Valve did some nice hack-n-patch with the equally static Q1 code... Curious.
If you get spam with "xxx" anywhere in the url you know what it is -- so the whole point is exactly that the URLs don't want to tell you where you're going to. From which it follows that .xxx would never work without government enforcing it on those companies -- the latter wouldn't want it. And that is only possible with ICANN or some hired watchdog being legally obliged to check every site it registers, periodically. Don't think so.
What I mean is, you can consider tld's to be a simple "cookie" that partitions the group of sites to your locale. Why would I have to choose between Western & Kanji if the choice is obvious with a .uk appended ? Why ask a country code to avoid showing the list of all sony dealers worldwide on sony.com, when you might as well have typed sony.fr ? Besides, a country tld is still a valid method to avoid name collisions between unrelated, but similarly named companies in distinct countries (not all companies are ThereCanBeOnlyOne multinationals, you know).
Well, that's all I can think of, so it's probably not enough to outweigh your zillion counterpoints, but at least "no good reason" has been rejected ;) So although the metaphorical net transcends location, it's actual content often does not.
It could've been .rms
We already are machines. Perhaps you need to check Bladerunner and Ghost in the Shell again ? ;)
Obviously that was back in the days when technology would still last 50 years. Why make something indestructable when it'll cost half as much and be twice as good in 6 months. Ok, maybe slightly exagerated, but cell phones etc are not that far off in terms of speech quality, battery life etc. People are glad it breaks, they got a reason for an upgrade.
male.Recipients.Add(malead)
male.Subject = "ILOVEYOU"
"male" ? Omg ! It's JeffK that's behind it !!!!
Well, this is true, but as you point out, the features that D3D is gaining over OpenGL are in the high-level domain; D3D is getting built-in support for reasonably exotic stuff like shaders, vertex morphing, continuous LOD etc. So you're right that Epic would in theory be reinventing the wheel if they wanted to support OpenGL, because they'd have to reimplement all that D3D stuff on top of OpenGL.
However, it's important to keep in mind that the higher the complexity goes, the tougher it becomes to write an API, and implement it, in such a way that everybody is happy feature & performance wise. As an example, there are numerous scene-graph libs for OpenGL that actually rock the house -- if you use them for what they were meant for. Like you say, they suck if you don't. Now I admit D3D features like shaders or LOD are still somehow on the borderline, there is probably only a very limited number of ways to implement this, so chances are that if you'd write it yourself you would indeed arrive at what D3D does, so don't bother. But as this trend continues it'll still be worthwile to implement highlevel stuff yourself. I mean, yes, content creation is a substantial burden these days, but that doesn't change the fact that a race game, a FPS and a flightsim still have massively different requirements from the highlevel parts of the 3D engine.
If you think about it, the argument of Epic that D3D is nice because they have a large input in it's development is pretty interesting. It almost seems to suggest that if hard decisions have to be made to trade off features vs performance as D3D gets more highlevel, the balance might swing towards "whatever benefits FPShooters", thanks to Epic.. ;) ;)
Um, wrong. OpenGL is an API, the way it works behind the scenes is hidden, and it could be a software rasteriser rendering your frames, it could be a hybrid setup that does the rastering in hardware, but everything else in software, or it could be a 100% hardware setup (such as an expensive SGI rig). I'm afraid you need to read up a bit :) The moment you plug in a T&L board and update the OpenGL driver, most well-written apps will get a speed increment (and that's not just theory, my crappy PII-200 got a new life with a GeF recently).
It would probably be pretty stupid for Epic to start ignoring Linux, now that the market is booming, and them having an edge with their cross platform engine of today. What it'll probably mean (if you read the part about "support for modular code" etc) is that there may be a Linux port, but it'll be seriously inferior to the D3D version. Given that the past problems where mostly with texture swapping, and GM's emphasis on D3D's "better API", I would guess that'll mean less textures, lower resolutions, less procedural effects in geometry and textures, less advanced shaders, and a few more, in the Linux version..
I do agree though that D3D is currently the better environment for development when you want to build 3D engines for vast continuous levels; Windows' tendency to trash the videomemory upon a task switch forces OpenGL to waste huge amounts of memory on keeping copies of every texture. Sweeney complained about that quite a while ago, so it was evident back then that he'd either have to find a solution for it (hoping for a GL_EXT_nosystemcopy perhaps ?), or drop OpenGL for being too memory intensive.
About OpenGL being geared towards id. That's true, but it's probably a good thing. The 3D APIs of today are so complex that it's probably daydreaming to think a driver writer can optimise *everything*. The combined internal state is so huge that you cannot build an optimal path for each and every combination. One influential game that uses a limited set of the API gives driver writers an opportunity to max out performance on that path, which in turn allows other developers to write their code with that path in mind -- instead of just reading the docs and picking one of the 35 possibilities to specify their geometry, hoping it'll come out right and speedy. You're right that this situation is not without danger of getting stuck, though...
I think I see what you mean.. You're assuming that a strong set of guidelines which can be enforced if necessary, is a bad thing. Well, I'd agree if those guidelines were a bunch of crap. But I disagree, so I'll try to show that (a) the current guidelines are Good, (b) those guidelines are better than what would evolve by natural selection from a totally deregulated schoolsystem.
As for (a), I'm afraid I find the Taylor readings a bit over the top. It's suggesting that our schoolsystem is designed as a Borg/Sparta crossover to pump out soldiers instead of humans. Even if it were true, it might not have been such a bad thing to educate people into soldiers around 1806, or into assembly workers around 1900 (Mann). After all, you have to make a living, so if school brings abilities you can use to stay out of the sewers, why not. Ofcourse, if those abilities are thought at the expense of everything else (intelligence, creativity etc) there's a problem, but still, schools can teach you specific needs for the world of that day. As I'm good at math and code I think of myself as well-educated, but perhaps in 100 years when creativity is the most important property to develop people will laugh at our schools for thinking we were "educated" with all our math and crap.
In short, a document which shows how 1806 schools wanted to build soldiers is not really much foundation for claiming that present-day schools have the same agenda.
As for (b)... Some statistic somewhere (dunno which or what) has claimed that a substantial portion of American people is technically analphabetic : either they can't read at all, or they read miserably, or they can read pretty ok but are unable to digest a solid text (like from a decent newspaper), extracting facts and logic as they go. True or not, it could be a simple result of the fact that most messages are delivered visually, or by TV. Does this mean the ability to read is passe ? I hope you agree the answer is Hell No. But I think that's the kind of conclusions non-regulated schools could jump to too easily (although ditching reading is kindof extreme) : take a short-sighted look at the needs of the present day and invent some guidelines for just that.. who cares about Asian geography, 14th centuries poets, numerical analysis, or electrons. Can you think of a world where everybody teaches their kids what they think is appropriate ? Note I said teaching, not raising : feel free to raise a kid with the morale you deem appropriate, but we're talking about forming intelligence, background, the ability to memorise and categorise large amounts of data, etc. Imho nobody will care for these abilities (you don't need them to watch TV or get groceries), and Civilisation Will Fall ! ;)
There's no contradiction between (a) and (b), as it's a matter of timescale.. :)
Sorry for the long post
True. I believe we basically agree on the ultimate goal of school : create people which are well educated on a number of technical skills (math, history), but with unaffected non-technical abilities (creativity, interests, ..).
The situation can be compared to general free-market policies : do you want a market which is completely free (the ideological Libertarian market), or a market which is free up to a certain limit, where the govt sits on it as a watchdog to avoid social or economic excesses ? I hope you go for the 2nd, as we all know what reckless exploitation of scarce resources, child labour, slavery etc have brought us. Well, I guess it's the same thing here; if the goods to trade are the children's minds, do you want a totally free market where corporate (ie non-state) schools can deliver any goods they want (no warranty, no liability, no guarantees) at any cost they think is ok ? Maybe you can point out a flaw in this analogy, but imho non-state schools are exactly the equivalent of a total-freedom free market.
Again I don't have much arguing to substantiate, so I'll pick up your challenge :) My mom is in fact a teacher in a state-governed school, and I'll ask her a detailed account on how policies are set, to what extent history could be warped if it better fits the current puppetmasters way up the govt chain, etc.
All I know is that I came out of school pretty much the way I like it, and I am a nerd/geek. I also don't know any other nerds or geeks which came out of a state-run highschool with traumas, supressed creativities or any other personality modifications that'd suit a Big Brother. That's in Belgium, though, and from what I gather here on /., USA schools must have a totally different attitude towards creativity and being different. Which is exactly why I still think of any problems as aberrations : it seems to work over here.
Well, maybe I was just conditioned into thinking that Our School Is Good ;)
But that's why it's called a Patch : I was specifically referring to it as a quick fix for particular disasters like the Ethiopia dryness of 1983. As you point out the army wouldn't feed their soldiers permanently on this, it's a one-stop solution for a particular situation. Similarly I thought of these patches as a non-structural quickfix for particularly nasty fuckups of mother nature. One C130 flying over dropping one crate of these sure helps surviving a whole lot more people than a crate of whatever else.
Hence also the link to milkpowder. I agree that Nestle's vision of powder as a permanent solution is a disaster -- so is this patch. Ok, so nevermind the political implications etc, it's literally a patch. Hope that clears it up ? :)