A warm sound is where the lower midrange and midbass are a bit more pronounced.
A cold sound is when the lower mids are recessed, a dark sound is when the extreme treble is muted, and treble is less prominent. A forward sound implies the speakers hit harder, and have a mroe prominent midrange, a boomy sound means midbass is heavily exxagerated.
Don't take the terms as a face value, they were just adapted to audio.
There was another recent Zelda hack that, amongst other things, allowed you to make your own map. I am not sure if it is played on an emulator or not, however. It goes by the name Zelda Classic, and I think it runs on DOS. Unlike this game, it is an exact duplicate of the original.
Consoles have gotten cheaper recently. The 3DO was $700 when it came out. The PSX was $300, I think the Saturn was too, and $300 is considered a max price for a console, and a Dreamcast is now $100. In a price competitive world, if something is a ripoff it dosen't last very long, so don't expect to see console prices get out of hand. The Gamecube is probably going to be $200, the Xbox $300, so they aren't exactly getting more expensive.
Nvidia never used unfair marketing tactics. They have used some questionable ones when it came to a few websites, but that has been fixed.
Nvidia achieved market dominance by providing good products at good prices, and coming out with new ones so fast to overwhelm their competitors. This is not an unfair business tactic, they are just completely ruthless competitors.
When Nvidia came out with the Riva 128, it had one advantage over the Voodoo. It could do 2D. And everyone had a 2D card, so it really didn't matter. It was slower with the early drivers, it had crap image quality, and it ran Quake 2 at 10fps or so on my P133.
The Voodoo2 came out and Nvidia had nothing but a slightly improved Riva 128, but at this point, people still didn't really care, but the 3d only thing was starting to have an effect. People would run systems with Riva 128s and Voodoo2s for good 2d and great 3d, and the option of real openGL.
With the TNT, Nvidia had a performance competitive product that had much better image quality too, and 3dfx's Banshee was a rehased, weakened Voodoo2. When the Voodoo3 came out, Nvidia promptly took the performance lead with the TNT2/TNT2 Ultra, albeit by a small margin. From then on 3dfx was way too slow to stand a chance. ATI started to come back into the picture here, and they have been tagging along since.
Nvidia releases new products too quickly for others to remain competitive, and they work quickly towards incorporating features OEMs want, lowering their prices, and look at the extremely dominant set of video chips Nvidia has today. Everything they have out is best of class, or damn near it.
MS won the market using completely ruthless tactics, and now they are their own greatest threat. Nvidia is still pumping out products like there is no tomorrow, and they are aware that if they get lax like 3dfx did, they will fall very quickly. ATI, while lagging behind, could jump right back into the front with one botched product release on Nvidia's behalf.
Also, if ATI gets their act together quickly and writes some decent drivers, and gets a comprable card out at a significantly lower price, then ATI will be able to get some of that lost marketshare back. It will take a better deal to beat Nvidia, however, because the established brand can always charge more for the same thing. And ATI has a huge disadvantage there too because they are competing with a ton of smaller boardmakers that can charge whatever they want.
That would probably work, and given the extremely low cost of, say, a PIII 700 or something of that order, it wouldn't add too much to the cost of the system. But, I think that then Intel would need to design a CPU that works like a router and sends the 64 bit instructions to the Itanium and the 32 bit instructions to the PIII.
Then again, that routing type CPU could also act as an instruction re-orderer to help make up for any compiler defencies, as the Itanium is essentially an in-order CPU. There are a lot of possibilites with where Intel could take this, and they aren't stupid, but this(Itanium) is just a very strange design.
Look at http://www.c0re.net. This web site is run by students at a high school in New Hampshire, and they put a disclaimer on the web site to emphasize that the site is not meant to be taken seriously. Maybe the person from Salem needs to do the same. But, not having seen the site, I can't say - he might be advocating to kill police like our recent lawmaker, Tom Alciere.
To top it off, HP calculators tend to be so much more durable than the offerings of TI, Casio, and Sharp and the keyboards can't be beat for feel and durability
I accidentally ran over my TI-83 (I kid you not, it fell out of my truck), and while it is now rather scoffed up, it still works alright. It may not be as powerful as an HP, but I barely even utilize the TI, so I don't have a need for one of the nice HP ones.
Slashdot was telling me my post was junk when I was naming the components of my future stereo, so I had to take it out.
Just ignore that:)
"The numbers I have noted are costs for the pairs of speakers, and while the center channel isn't as good as it could be, I care about music more then movies so I would prefer to spend more on my front L/R speakers."
Paradigm speakers are extremely good for the money. Their "low end" speakers (The Atom in particular) sound extremely good. They have a pretty good punch for small bookshelf speakers, image very well, and have an excellent tweeter. The smaller ones (Atom, Micro) would need a sub to augment them, but Paradigm has some pretty good floorstanding speakers too, like the Phantom.
The numbers I have noted are costs for the pairs of speakers, and while the center channel isn't as good as it could be, I care about music more then movies so I would prefer to spend more on my front L/R speakers.
Other companies that you may consider purchasing speakers from are NHT, Energy, JBL, or B&W. Don't bother with Bose. Bose might have 22% of the audio market, but this is because they advertise very well. They are not very accurate speakers at all, and their "Acoutimass module"s are bandpass designs, so they basically play a few notes very loud and the rest of what they play is very innacurate. My family has a pair of Bose speakers (They are direct reflecting bookshelves with an 8" woofer, 3" midrange, and 2" "tweeter" - they don't sound awful, but the treble is very harsh, and they are priced a lot higher then comprable speakers from other companies (They cost $280 for the pair and sound a good deal worse then the Paradigm Atoms). The system that I have made will sound a lot better then any Bose setup, at a much lower cost. When you buy Bose you are paying for the name, and for the small size. If you like the Bose sound, check out Cambridge Soundworks smaller line of speakers. They sound very similar to the Bose speakers at a much lower cost. Their Movieworks system is a fairly good choice if you don't listen to much music, and don't want to spend more then $500.
I would reccomend picking up a few issues of Sound&Vision, it is a fairly easy to read magazine that covers everything from about the $500 for a whole system to $3000 per speaker.
Also check out www.audioreview.com and read some of the reviews there. Ignore the ones for Bose speakers because many immature people just give them 1 star ratings without regard to their actual performance, which would usually warrant 2 stars or so:).
If all you are doing is watching movies, then spend more on the center speaker (The majority of the sound in movies is from the center speaker), but don't skimp on the front channels. Having cheap rear speakers as long as they sound pretty good is usually a good cost cutting measure (The Paradigm Atoms are $190 a pair at retail price), but that is really the only area you can skimp much.
Hsu research makes the best subwoofers out there for the price, but they don't have anything on the very low end side of things (Their cheapest sub is $500). Unless you want to spend 1/3 of your budget on a sub, do more research:).
IBM may have had SOI and copper interconnects first, but Intel has them now (And AMD has had them for a while), and has the added benefit of being in heavy competition with AMD. This recent competition has sparked a clock speed race that IBM has not had to partake in, and it has left both Intel and AMD on top of the manufacturing world. Both companies are selling 1Ghz+.18 micron CPUs which are very competitive speed wise to most high end solutions, and catching up pretty fast. Sure, an Athlon is no Alpha, but for under $500 you will soon be able to buy two 1.1Ghz Athlons and a dual motherboard that are faster then any Alpha, and a lot cheaper too.
The reason the framerates are so low is because the screenshots are taken with 4x FSAA on. If FSAA were off the numbers would be more like 40-50fps. I would imagine that the motion blur would have a framerate hit, but it is not going to drop it to 1/4 what it should be in the first place.
Also, not only were they running Quake 3 with 4x FSAA on, they were doing it in 1280x960. The poor Voodoo5 is heavily strained when running in that resolution anyway, and then the load is increased some more by the motion blur, and quadrupled through the FSAA, you are going to see crappy framerates.
Take it from an Anime newbie (All I have seen was Akira, some Fist of the North Star, and a couple others I can't seem to recall the name of). Anyway, I would advise against getting the Fist of the North Star for a "newbie" because quite frankly it was a bore. At least Akira could keep me partically interested, and the animation was good.
Ok, fine, this is coming from a biased point of view as I don't like Anime in general, but FOTNS was horrid by my standards. The plot was dull, the animation bad and done at like 4fps, everything was so bleak, and boring.
I guess Anime is an acquired taste, like Linux (Just had to put that in somewhere =) as I most certainly have not acquired the taste for it. Akira had some entertainment value, but FOTNS was worse than an episode of Full House.
No, the NV20 does not use Tiling, but it uses something called hidden surface removal, which uses the standard rendering techniques that most cards use today, but it does to a small extent what tiling does. It is not nearly as efficent, but it will probably increase performance 20-30% realistically. The 7x performance gain number is in environments specifically made to take advantage of the NV20 over the GeForce2 Ultra (A lot of overdraw, high res textures, lots of polygons all at the same time), similar to the inclusion of T&L allowed the GeForce to walk all over the TNT2 Ultra in treemark.
One other big feature of the NV20 is the programmable T&L unit. That way you can add in small features you want to what the video card processes instead of relying on the CPU.
Another performance advantage that people will see is from the increased theoretical max fillrate. The GeForce 2 runs at 200Mhz, and has 4 pipelines each capable of processing 2 textures per clock, which gives you a fillrate of 800 Megapixels/second or 1.6 Gigatexels/second. The NV20 will likely run around 250Mhz with 4 pipelines that can handle 3 textures per clock, which will give fillrates of 1 Gigapixel/second and 3Gigatexels/second. This would allow for a theoretical performance increase of about 30% with single and dual textured games and a performance increase on the order of 100-130% in games that use 3 textures per pixel and more. This is of course assuming that there is enough memory bandwidth to push all of those pixels left.
Price wise I would expect a 32MB version with ~200Mhz DDR memory for $300-$350 when it comes out, and a 64MB version for $600 with perhaps 233Mhz DDR memory.
Uh, no. Platinum is worth more along the lines of $500 an ounce. Still, if it costs $9000 an ounce to bring this back to Earth, there is no way prices would be competitive, unless we came up with a much more efficent way of transporting the goods. If anyone here has read 3001, the elevators were quite the idea, and the thing is that they would be feasible in the not too distant future. Probably not in any of our lifetimes, but I can see it happening in 200-300 years, minus the fast accelerating without any G-Forces part (But that wouldn't matter if you transporting raw materials much if they were packaged correctly anyway)
The bottom line is that mining asteroids is a feasible thing to do, but I don't see it happening untill the 23rd century or so in an efficent manner. We need a lot more technology to be able to get a manufacturing plant to the asteroid, then we have to build it and make robots that will do everything one way or another, and then we will need a semi-efficent way of transporting it back, to the point that costs are decent enough that people/companies will be able to afford them.
Re:The Best Goddamn Keyboard in the World, Ever
on
Keyless Keyboard
·
· Score: 1
I must disagree. Try one of SGI's PS/2 keyboards and you will be an SGI keyboard person for life:). The Extended Keyboard II is definantly nice, I have used them myself, but I find the feel to impede my typing ability. As far as typing speed goes with me, I type very quickly on the new Apple keyboards, but the keys are a bit springy for me so I am slightly more prone to error.
I happen to use an RF tag to get into work when the alarm is on. It goes on my keyring, and I wave it in front of the RF reciever. It needs to be very close to it to work, like a couple inches or so. And while yes you could put it inside clothes, when you put them in the wash you will ruin the poor thing because it is made of paper:).
In the past I have been a cashier and I would have to say that the barcodes work just fine for all intents and purposes, in my opinion this technology would be a waste until it got to the point where it could be read from a few feet away. Imagine walking into a supermarket, and on your way out the door you don't need to go through a cash register, but instead it just automatically bills you?
Many biologists believe that animals don't have the ability to think on their own. They could possibly use this research to help determine whether they are right or wrong, as if every single penguin leaned over to the point of falling it would side with the "They act completely on instinct" side, and if some lean to the point of falling over and others don't, then it would support the "Animals have a mind of their own" side.
I hate to say it, but no matter how powerful Indrema is, it will fail as a mass market game console. Why? Marketing. Ask your favorite consle gaming friends what they think of the Indrema, and they will usually respond with a "huh?". Hell, most people have no idea what the X-Box is yet.
Another reason it will fail is because it is missing support of major developers, because they are working on more profitable platforms. Yes, there will be the open source community behind Indrema, but that just means that there will be some free games. Not that that is a bad thing, but with the X-Box's 160+ developers, and numerous ones for PS2 and Dreamcast, it stands no chance on the games front. I have not heard once of one major multi-platform game that will be ported to Indrema, and I don't think it will happen (Other then some ports of PC games like Quake 3).
If you want an Indrema to develop for, if you could come up with something decent you could put in in a portfolio for potential employers to see, but this system will in all likelyhood be a commercial failure with a small, tight following after the system dies.
Hopefully these guys prove me wrong, but everything I know about the console market and the business world screams "failure" pretty loudly.
Well, I stole them from Ars Technica, because I couldn't think of anything better. Try to do it yourself without spending an inordinate amount of time on making a post:)
The PS2 and PC are two completely animals (The DC is more like a PC then a PS2 architecturaly). Metaphorically, we will call memory a "water body" (Bucket, pool, ocean) and bandwidth a "pipe". The PS2 has buckets that are connected to sewer lines, whereas the DC has pools connected to normal 4" pipes, and the PC has the pipe used to get water to your kitchen sink sucking water out of the pacific ocean.
In games, the same instructions are made over and over on different data (i.e. rendering). This data is constantly changing, and it takes a lot of memory bandwidth to support this. By designing the PS2 with small amounts of memory and a lot of bandwidth, it is more difficult to program for, because using the PC mentality (Load all data into memory and pull out what you need when you need it) does not work. The PS2 does not have the memory to do so. Instead, you need to load the data into memory more dynamically, so over the course of generating a frame of graphics on screen the whole 4MB of video ram may be used several times over, instead of trying to do everything in it.
The bottom line is that programming for the PS2 requires much more dynamic memory managment, whereas programming on the PC utilizes a more static style.
Ars Technica recently had a very interesting article on the two completely different architectures somewhere (I can't seem to find it).
Slashdot Mirror
Of course, that has absolutely nothing to do with security.
A warm sound is where the lower midrange and midbass are a bit more pronounced. A cold sound is when the lower mids are recessed, a dark sound is when the extreme treble is muted, and treble is less prominent. A forward sound implies the speakers hit harder, and have a mroe prominent midrange, a boomy sound means midbass is heavily exxagerated. Don't take the terms as a face value, they were just adapted to audio.
There was another recent Zelda hack that, amongst other things, allowed you to make your own map. I am not sure if it is played on an emulator or not, however. It goes by the name Zelda Classic, and I think it runs on DOS. Unlike this game, it is an exact duplicate of the original.
Consoles have gotten cheaper recently. The 3DO was $700 when it came out. The PSX was $300, I think the Saturn was too, and $300 is considered a max price for a console, and a Dreamcast is now $100. In a price competitive world, if something is a ripoff it dosen't last very long, so don't expect to see console prices get out of hand. The Gamecube is probably going to be $200, the Xbox $300, so they aren't exactly getting more expensive.
Nvidia never used unfair marketing tactics. They have used some questionable ones when it came to a few websites, but that has been fixed.
Nvidia achieved market dominance by providing good products at good prices, and coming out with new ones so fast to overwhelm their competitors. This is not an unfair business tactic, they are just completely ruthless competitors.
When Nvidia came out with the Riva 128, it had one advantage over the Voodoo. It could do 2D. And everyone had a 2D card, so it really didn't matter. It was slower with the early drivers, it had crap image quality, and it ran Quake 2 at 10fps or so on my P133.
The Voodoo2 came out and Nvidia had nothing but a slightly improved Riva 128, but at this point, people still didn't really care, but the 3d only thing was starting to have an effect. People would run systems with Riva 128s and Voodoo2s for good 2d and great 3d, and the option of real openGL.
With the TNT, Nvidia had a performance competitive product that had much better image quality too, and 3dfx's Banshee was a rehased, weakened Voodoo2. When the Voodoo3 came out, Nvidia promptly took the performance lead with the TNT2/TNT2 Ultra, albeit by a small margin. From then on 3dfx was way too slow to stand a chance. ATI started to come back into the picture here, and they have been tagging along since.
Nvidia releases new products too quickly for others to remain competitive, and they work quickly towards incorporating features OEMs want, lowering their prices, and look at the extremely dominant set of video chips Nvidia has today. Everything they have out is best of class, or damn near it.
MS won the market using completely ruthless tactics, and now they are their own greatest threat. Nvidia is still pumping out products like there is no tomorrow, and they are aware that if they get lax like 3dfx did, they will fall very quickly. ATI, while lagging behind, could jump right back into the front with one botched product release on Nvidia's behalf.
Also, if ATI gets their act together quickly and writes some decent drivers, and gets a comprable card out at a significantly lower price, then ATI will be able to get some of that lost marketshare back. It will take a better deal to beat Nvidia, however, because the established brand can always charge more for the same thing. And ATI has a huge disadvantage there too because they are competing with a ton of smaller boardmakers that can charge whatever they want.
Then again, that routing type CPU could also act as an instruction re-orderer to help make up for any compiler defencies, as the Itanium is essentially an in-order CPU. There are a lot of possibilites with where Intel could take this, and they aren't stupid, but this(Itanium) is just a very strange design.
Look at http://www.c0re.net. This web site is run by students at a high school in New Hampshire, and they put a disclaimer on the web site to emphasize that the site is not meant to be taken seriously. Maybe the person from Salem needs to do the same. But, not having seen the site, I can't say - he might be advocating to kill police like our recent lawmaker, Tom Alciere.
Slashdot was telling me my post was junk when I was naming the components of my future stereo, so I had to take it out. Just ignore that :)
"The numbers I have noted are costs for the pairs of speakers, and while the center channel isn't as good as it could be, I care about music more then movies so I would prefer to spend more on my front L/R speakers."
The numbers I have noted are costs for the pairs of speakers, and while the center channel isn't as good as it could be, I care about music more then movies so I would prefer to spend more on my front L/R speakers.
Other companies that you may consider purchasing speakers from are NHT, Energy, JBL, or B&W. Don't bother with Bose. Bose might have 22% of the audio market, but this is because they advertise very well. They are not very accurate speakers at all, and their "Acoutimass module"s are bandpass designs, so they basically play a few notes very loud and the rest of what they play is very innacurate. My family has a pair of Bose speakers (They are direct reflecting bookshelves with an 8" woofer, 3" midrange, and 2" "tweeter" - they don't sound awful, but the treble is very harsh, and they are priced a lot higher then comprable speakers from other companies (They cost $280 for the pair and sound a good deal worse then the Paradigm Atoms). The system that I have made will sound a lot better then any Bose setup, at a much lower cost. When you buy Bose you are paying for the name, and for the small size. If you like the Bose sound, check out Cambridge Soundworks smaller line of speakers. They sound very similar to the Bose speakers at a much lower cost. Their Movieworks system is a fairly good choice if you don't listen to much music, and don't want to spend more then $500.
I would reccomend picking up a few issues of Sound&Vision, it is a fairly easy to read magazine that covers everything from about the $500 for a whole system to $3000 per speaker.
Also check out www.audioreview.com and read some of the reviews there. Ignore the ones for Bose speakers because many immature people just give them 1 star ratings without regard to their actual performance, which would usually warrant 2 stars or so :).
If all you are doing is watching movies, then spend more on the center speaker (The majority of the sound in movies is from the center speaker), but don't skimp on the front channels. Having cheap rear speakers as long as they sound pretty good is usually a good cost cutting measure (The Paradigm Atoms are $190 a pair at retail price), but that is really the only area you can skimp much.
Hsu research makes the best subwoofers out there for the price, but they don't have anything on the very low end side of things (Their cheapest sub is $500). Unless you want to spend 1/3 of your budget on a sub, do more research :).
IBM may have had SOI and copper interconnects first, but Intel has them now (And AMD has had them for a while), and has the added benefit of being in heavy competition with AMD. This recent competition has sparked a clock speed race that IBM has not had to partake in, and it has left both Intel and AMD on top of the manufacturing world. Both companies are selling 1Ghz+ .18 micron CPUs which are very competitive speed wise to most high end solutions, and catching up pretty fast. Sure, an Athlon is no Alpha, but for under $500 you will soon be able to buy two 1.1Ghz Athlons and a dual motherboard that are faster then any Alpha, and a lot cheaper too.
Also, not only were they running Quake 3 with 4x FSAA on, they were doing it in 1280x960. The poor Voodoo5 is heavily strained when running in that resolution anyway, and then the load is increased some more by the motion blur, and quadrupled through the FSAA, you are going to see crappy framerates.
Take it from an Anime newbie (All I have seen was Akira, some Fist of the North Star, and a couple others I can't seem to recall the name of). Anyway, I would advise against getting the Fist of the North Star for a "newbie" because quite frankly it was a bore. At least Akira could keep me partically interested, and the animation was good. Ok, fine, this is coming from a biased point of view as I don't like Anime in general, but FOTNS was horrid by my standards. The plot was dull, the animation bad and done at like 4fps, everything was so bleak, and boring. I guess Anime is an acquired taste, like Linux (Just had to put that in somewhere =) as I most certainly have not acquired the taste for it. Akira had some entertainment value, but FOTNS was worse than an episode of Full House.
One other big feature of the NV20 is the programmable T&L unit. That way you can add in small features you want to what the video card processes instead of relying on the CPU.
Another performance advantage that people will see is from the increased theoretical max fillrate. The GeForce 2 runs at 200Mhz, and has 4 pipelines each capable of processing 2 textures per clock, which gives you a fillrate of 800 Megapixels/second or 1.6 Gigatexels/second. The NV20 will likely run around 250Mhz with 4 pipelines that can handle 3 textures per clock, which will give fillrates of 1 Gigapixel/second and 3Gigatexels/second. This would allow for a theoretical performance increase of about 30% with single and dual textured games and a performance increase on the order of 100-130% in games that use 3 textures per pixel and more. This is of course assuming that there is enough memory bandwidth to push all of those pixels left.
Price wise I would expect a 32MB version with ~200Mhz DDR memory for $300-$350 when it comes out, and a 64MB version for $600 with perhaps 233Mhz DDR memory.
Uh, no. Platinum is worth more along the lines of $500 an ounce. Still, if it costs $9000 an ounce to bring this back to Earth, there is no way prices would be competitive, unless we came up with a much more efficent way of transporting the goods. If anyone here has read 3001, the elevators were quite the idea, and the thing is that they would be feasible in the not too distant future. Probably not in any of our lifetimes, but I can see it happening in 200-300 years, minus the fast accelerating without any G-Forces part (But that wouldn't matter if you transporting raw materials much if they were packaged correctly anyway)
The bottom line is that mining asteroids is a feasible thing to do, but I don't see it happening untill the 23rd century or so in an efficent manner. We need a lot more technology to be able to get a manufacturing plant to the asteroid, then we have to build it and make robots that will do everything one way or another, and then we will need a semi-efficent way of transporting it back, to the point that costs are decent enough that people/companies will be able to afford them.
I must disagree. Try one of SGI's PS/2 keyboards and you will be an SGI keyboard person for life :). The Extended Keyboard II is definantly nice, I have used them myself, but I find the feel to impede my typing ability. As far as typing speed goes with me, I type very quickly on the new Apple keyboards, but the keys are a bit springy for me so I am slightly more prone to error.
In the past I have been a cashier and I would have to say that the barcodes work just fine for all intents and purposes, in my opinion this technology would be a waste until it got to the point where it could be read from a few feet away. Imagine walking into a supermarket, and on your way out the door you don't need to go through a cash register, but instead it just automatically bills you?
I got 30ms (Not 300). If anyone wants to contend with that just send me an email and I will respond with a picture :)
Wohooo! I have 3 percent of the vote! Vote for me!
Another reason it will fail is because it is missing support of major developers, because they are working on more profitable platforms. Yes, there will be the open source community behind Indrema, but that just means that there will be some free games. Not that that is a bad thing, but with the X-Box's 160+ developers, and numerous ones for PS2 and Dreamcast, it stands no chance on the games front. I have not heard once of one major multi-platform game that will be ported to Indrema, and I don't think it will happen (Other then some ports of PC games like Quake 3).
If you want an Indrema to develop for, if you could come up with something decent you could put in in a portfolio for potential employers to see, but this system will in all likelyhood be a commercial failure with a small, tight following after the system dies.
Hopefully these guys prove me wrong, but everything I know about the console market and the business world screams "failure" pretty loudly.
Well, I stole them from Ars Technica, because I couldn't think of anything better. Try to do it yourself without spending an inordinate amount of time on making a post :)
I would mod myself down if I could. I guess I was a bit too slow hitting the reply button. Oh well, 2nd post!
In games, the same instructions are made over and over on different data (i.e. rendering). This data is constantly changing, and it takes a lot of memory bandwidth to support this. By designing the PS2 with small amounts of memory and a lot of bandwidth, it is more difficult to program for, because using the PC mentality (Load all data into memory and pull out what you need when you need it) does not work. The PS2 does not have the memory to do so. Instead, you need to load the data into memory more dynamically, so over the course of generating a frame of graphics on screen the whole 4MB of video ram may be used several times over, instead of trying to do everything in it.
The bottom line is that programming for the PS2 requires much more dynamic memory managment, whereas programming on the PC utilizes a more static style.
Ars Technica recently had a very interesting article on the two completely different architectures somewhere (I can't seem to find it).