Re:"This is Unix! I KNOW this!"
on
3Dwm Updates
·
· Score: 1
soooo, what does this have to do with the topic at hand?
You did see Jurassic park, didn't you? You may remember the little twerp who sat down at a system and proceeded to zoom her way about the system in true Max Headroom 3D style.
I look forward to the day when we replace/etc/passwd with a VRML file, MD5 sums with collision hulls, and the 31337 h4x0r 0wnz j00 only because of his superior dexterity.
Mark my words - all this Quake playing you do today - this practice of button reflex and mouse coordination - this is going to pay off in SPADES once we enter the world of - 3D SECURITY!
What is the equivalent of what he's done? What's the conventional media equivalent?
If you were to plaster most of the above across a public bulletin board, you'd be sued. You'd be lucky(?) if you could even get the advert placed in the first place.
Creating an entry in the telephone book with any of the above as a company name would also get you in hot water. This is the closest parallel I can draw to registering a domain name.
Is it really so surprising that this has happened? Is it even anything to get upset about? If so... why?
I'm curious to see if this IPO has anything like the wild ride that Redhat enjoyed a year ago. Is Wall Street over the whole "Linux Thing," or is there room for a few more billion dollar lessons?
Websites aren't the only concern here. I have a friend working at Microsoft. Recently, I e-mailed, mistyping the Microsoft domain name. Minutes later, I was inundated with advertisements for "discount" Windows products.
Somebody out there is harvesting e-mail sent to mis-typed domain names and using the addresses for target spamming.
All other things being considered, the Matrox cards continue to have the best output quality, bar none in this price range. From the Matrox Millenium II on forward, this has been true.
If you want a card that really holds up to high refresh rates at high resolutions for your 2D work, the Matrox deserves extra consideration.
Is Bluetooth a free and open standard or is it closed or licensed?
I'm wondering if we're looking at a free-like-USB or a for-pay-like-Firewire situation here. Considering Firewire's technical merits but tiny market share, the above consideration seems to be a pretty good indicator of its relevance.
Can we please not give notice/publicity to announcement-of-intent from companies? Show code, then collect your kudos and marketroid credits.
Unchecked, this could turn into a constant shower of press releases from hundreds of companies wanting a piece of the Linux pie. And when the majority amount to nothing, as most of these seem to, it just makes Linux look bad.
(1)(A) No person shall circumvent a technological measure that effectively controls access to a work protected under this title.
This is still the part that chafes me most; it completely wipes previously legal fair use copying. This would make even dubbing to a tape illegal.
So you go out and buy your new secure digital music CD player and you can't even legally make a copy to play in your car's tape deck.
And of course, this doesn't begin to deal with all the Mac, Linux, etc users who're stranded because some Windows developer doesn't deem their markets worthy of software player ports.
That's fine if you're fill-rate limited, but if instead you're bus-limited, it's a waste. I can't speak for the PS2, but
on the PC platform, saturating PCI or AGP with geometry data can be an issue on a fast machine at high frame
rates. Not sending back-facing polygons can help dramatically.
The rules are different when you've more than one bus and CPU that you're thinking about. If you can offload the checking to an underutilized secondary/tertiary processor and never have to touch any of the vertex data on your main CPU, it's worth it.
* I think they dismissed the vector processors rather lightly. After all, the whole point of the game machines is to make the best games possible - while part of that lies in the graphics, a big, big part lies in the AI and dynamics a game offers. Having a whole bunch of extra processor power around to devote to such things should, I think, translate to the possibility of some amazing games with equally amazing graphics.
Bingo. And by operating entirely off the main bus, you can do some deep, deep analysis without burdening the main bus, even spitting data straight into the graphics hardware without revisiting the main bus.
* As far as anti-aliasing goes, I had thought Sony did have some kind of library now that developers could use - but I'm not sure about that at all. It would be interesting to hear if the description of the TTT anti-aliasing is correct.
Sony have shown how to structure programs for anti-aliasing with zero cost. The kind of trickery and special-case handling other systems use isn't required. It really is spectacular hardware.
I'll say. Other graphics chips must texture the backs of polygons as well as the front facing polygons. Who the hell sends those polys to the renderer?
Back face culling happens after projection into screen space. With modern architectures, you go ahead and dump everything that's on screen into the pipeline and let the hardware take care of the back face culling. The PS2 can chew through triangles so quickly that there's no effective cost in doing it this way.
I'm a big Dreamcast fan, but what would prevent Sony from doing a VRAM memory expansion, similar to what N64 did for its core memory?
The VRAM on the PS2 is akin to the tiny active texture cache on your average 3D card. Texture can be brought in from main memory more quickly than most cards can load the texture cache from video memory.
If you want to make a valid comparison, you need to compare Dreamcast release titles to Playstation 2 release titles.
Early titles never look anything like later titles - that much is beyond debate. If you're comparing currently available titles then this makes sense. If you're comparing the hardware, as this article purports to do, then comparing a first-off PS2 title to a third or fourth generation Dreamcast title is rather foolish.
I'm a game developer, currently working on the Playstation 2. I'm under NDA and can't say much.
What I can say is that the portion quoted at the head of the Slashdot article alone shows that the author is full of it. VRAM on the Playstation 2 is used for storing textures and the final image, and nothing more. This has nothing to do with triangle sizes.
Further, the PS2 supports a number of compressed vertex formats and types of rapid transforms in hardware which no other system currently supports, making any amateur comparison along these lines laughable at best.
Contrary to the many other posts here, there are indeed reasons for hacking microcode. 2nd and 3rd generation N64 product developers likely know this already.
Inserting one or two new, specialized instructions in your inner loop can make all the difference in the world. Being able to leave registers in an undefined state when you truly don't care about a result, terminating an instruction already partially executed, executing instructions in two pieces - all of these are Good Things when you know nothing else is going to be relying on the processor state.
This could be very useful for performance-critical code in device drivers and specialized applications alike.
I don't know how many fat people I hear saying that they don't know why they're fat. "I don't eat too much." Many alcoholics insist that alcohol isn't their problem, ditto drug addicts and drugs, etc.
Why is this relevant? The whole thing was about surveying people's opinions. I know at least half a dozen co-workers who go home and surf their home internet kiosks until it's time to sleep or turn on Star Trek. Repeat until dead. And it's like digging a tick out of a long-hair to pry some of these people off their systems and get them to join the lunch crowd for some face time. Even more fun to get them to speak more than a few sentences.
Parallel processing is new to console developers, save those who have worked on the Saturn and the Jaguar.
And given the success of Linux, obsucre technical documentation seems to be no hurdle to the average programmer...
The wealth of rich, full Linux games proves your point.
Seriously - the documentation of the Linux kernel is hardly obscure. And you certainly can't claim trouble with the information on man pages. Game developers put up with documentation along these lines, however:
Plus of CMF_HI is failing if CMF_LO most bit setted. (CMF_IQ) Better use of picture for interlacing, unless triangle!
Needless to say, CMF_IQ wouldn't be mentioned anywhere in the text, but you'd find CMF_IQX and CMF_IQR in some header file and take it from there. Six months later you'd get an 80-page errata document that's in no way cross-referenced to the original documentation, plus maybe some sample code.
Right now, MS can 'fork' their code anytime they want to. Take, for example, when word '97 (I think) came out. It was incompatible with earlier versions, and they (willfully, I think) didn't have a module that allowed users to save in the old format. As a result any company which bought the newest version for any of their machines was forced to buy it for all of their machines. If it were open source, people would have just fixed the problem and released it.
I'll indulge in a bit of MS bashing as readily as the next guy.
The problem was not an inability to save in the old format, however. The problem was that old format docs would be saved in the new format unless you specifically told Office to use the old. Microsoft's solution was to have the update select the version in which the file was originally saved, not to add exporters.
MS acted reasonably well here. Your example pretty much amounts to a media event and nothing more.
The right answer, or at least the more amusing one, would be for somebody to take names of folks willing to chip in $10 and $20 apiece for the first person to crack SDMI after deployment. That could certainly add up to more than $10,000 in no time at all.
Intel's response is that, in actuality, the Crusoe uses the same amount of power as the corresponding Intel chip. The Crusoe merely takes twice as long to do so. Another area where Intel is faster...
Slashdot Mirror
You did see Jurassic park, didn't you? You may remember the little twerp who sat down at a system and proceeded to zoom her way about the system in true Max Headroom 3D style.
I look forward to the day when we replace /etc/passwd with a VRML file, MD5 sums with collision hulls, and the 31337 h4x0r 0wnz j00 only because of his superior dexterity.
Mark my words - all this Quake playing you do today - this practice of button reflex and mouse coordination - this is going to pay off in SPADES once we enter the world of - 3D SECURITY!
What is the equivalent of what he's done? What's the conventional media equivalent?
If you were to plaster most of the above across a public bulletin board, you'd be sued. You'd be lucky(?) if you could even get the advert placed in the first place.
Creating an entry in the telephone book with any of the above as a company name would also get you in hot water. This is the closest parallel I can draw to registering a domain name.
Is it really so surprising that this has happened? Is it even anything to get upset about? If so... why?
I'm curious to see if this IPO has anything like the wild ride that Redhat enjoyed a year ago. Is Wall Street over the whole "Linux Thing," or is there room for a few more billion dollar lessons?
It seems like compiled web applications are all but forgotten. For my own use, I still believe in C/C++ cgi apps with various utility libraries.
Why are the various scripting languages in such overwhelming favor?
This is one of the best Slashdot writeups we've had in a long time. A well investigated story like this one is good journalism.
Thanks, jamie. And please - MORE STORIES LIKE THIS!
Websites aren't the only concern here. I have a friend working at Microsoft. Recently, I e-mailed, mistyping the Microsoft domain name. Minutes later, I was inundated with advertisements for "discount" Windows products.
Somebody out there is harvesting e-mail sent to mis-typed domain names and using the addresses for target spamming.
All other things being considered, the Matrox cards continue to have the best output quality, bar none in this price range. From the Matrox Millenium II on forward, this has been true.
If you want a card that really holds up to high refresh rates at high resolutions for your 2D work, the Matrox deserves extra consideration.
Is Bluetooth a free and open standard or is it closed or licensed?
I'm wondering if we're looking at a free-like-USB or a for-pay-like-Firewire situation here. Considering Firewire's technical merits but tiny market share, the above consideration seems to be a pretty good indicator of its relevance.
Can we please not give notice/publicity to announcement-of-intent from companies? Show code, then collect your kudos and marketroid credits.
Unchecked, this could turn into a constant shower of press releases from hundreds of companies wanting a piece of the Linux pie. And when the majority amount to nothing, as most of these seem to, it just makes Linux look bad.
This is still the part that chafes me most; it completely wipes previously legal fair use copying. This would make even dubbing to a tape illegal.
So you go out and buy your new secure digital music CD player and you can't even legally make a copy to play in your car's tape deck.
And of course, this doesn't begin to deal with all the Mac, Linux, etc users who're stranded because some Windows developer doesn't deem their markets worthy of software player ports.
Stupid.
The rules are different when you've more than one bus and CPU that you're thinking about. If you can offload the checking to an underutilized secondary/tertiary processor and never have to touch any of the vertex data on your main CPU, it's worth it.
Bingo. And by operating entirely off the main bus, you can do some deep, deep analysis without burdening the main bus, even spitting data straight into the graphics hardware without revisiting the main bus.
Sony have shown how to structure programs for anti-aliasing with zero cost. The kind of trickery and special-case handling other systems use isn't required. It really is spectacular hardware.
Back face culling happens after projection into screen space. With modern architectures, you go ahead and dump everything that's on screen into the pipeline and let the hardware take care of the back face culling. The PS2 can chew through triangles so quickly that there's no effective cost in doing it this way.
The VRAM on the PS2 is akin to the tiny active texture cache on your average 3D card. Texture can be brought in from main memory more quickly than most cards can load the texture cache from video memory.
If you want to make a valid comparison, you need to compare Dreamcast release titles to Playstation 2 release titles.
Early titles never look anything like later titles - that much is beyond debate. If you're comparing currently available titles then this makes sense. If you're comparing the hardware, as this article purports to do, then comparing a first-off PS2 title to a third or fourth generation Dreamcast title is rather foolish.
I'm a game developer, currently working on the Playstation 2. I'm under NDA and can't say much.
What I can say is that the portion quoted at the head of the Slashdot article alone shows that the author is full of it. VRAM on the Playstation 2 is used for storing textures and the final image, and nothing more. This has nothing to do with triangle sizes.
Further, the PS2 supports a number of compressed vertex formats and types of rapid transforms in hardware which no other system currently supports, making any amateur comparison along these lines laughable at best.
Contrary to the many other posts here, there are indeed reasons for hacking microcode. 2nd and 3rd generation N64 product developers likely know this already.
Inserting one or two new, specialized instructions in your inner loop can make all the difference in the world. Being able to leave registers in an undefined state when you truly don't care about a result, terminating an instruction already partially executed, executing instructions in two pieces - all of these are Good Things when you know nothing else is going to be relying on the processor state.
This could be very useful for performance-critical code in device drivers and specialized applications alike.
I don't know how many fat people I hear saying that they don't know why they're fat. "I don't eat too much." Many alcoholics insist that alcohol isn't their problem, ditto drug addicts and drugs, etc.
Why is this relevant? The whole thing was about surveying people's opinions. I know at least half a dozen co-workers who go home and surf their home internet kiosks until it's time to sleep or turn on Star Trek. Repeat until dead. And it's like digging a tick out of a long-hair to pry some of these people off their systems and get them to join the lunch crowd for some face time. Even more fun to get them to speak more than a few sentences.
Parallel processing is new to console developers, save those who have worked on the Saturn and the Jaguar.
The wealth of rich, full Linux games proves your point.
Seriously - the documentation of the Linux kernel is hardly obscure. And you certainly can't claim trouble with the information on man pages. Game developers put up with documentation along these lines, however:
Needless to say, CMF_IQ wouldn't be mentioned anywhere in the text, but you'd find CMF_IQX and CMF_IQR in some header file and take it from there. Six months later you'd get an 80-page errata document that's in no way cross-referenced to the original documentation, plus maybe some sample code.
I'll indulge in a bit of MS bashing as readily as the next guy.
The problem was not an inability to save in the old format, however. The problem was that old format docs would be saved in the new format unless you specifically told Office to use the old. Microsoft's solution was to have the update select the version in which the file was originally saved, not to add exporters.
MS acted reasonably well here. Your example pretty much amounts to a media event and nothing more.
Where?
The right answer, or at least the more amusing one, would be for somebody to take names of folks willing to chip in $10 and $20 apiece for the first person to crack SDMI after deployment. That could certainly add up to more than $10,000 in no time at all.
Intel's response is that, in actuality, the Crusoe uses the same amount of power as the corresponding Intel chip. The Crusoe merely takes twice as long to do so. Another area where Intel is faster...
...aww, fuck it.
Oh my god! They killed Kennedy!! You bastards!