I'm not a lawyer, but I was around at Williams/Midway at the end of MK3 Wavenet and by my reading of the 523 patent claims, Wavenet is not prior art. The patent claims a set of computers pushing messages up to a server every so often, the server aggregating them and pushing a compound message back down to everyone who needs to know. It claims this running on a fixed, short time interval, over the Internet, with the groups being registered and subsequently joined by individual computers.
However, there is nothing in common with Wavenet. Wavenet was an asynchronous peer-to-peer arrangement which did no aggregation and worked over ISDN lines so that it had much higher bandwidth and lower latency than was available via the Internet for arcades at that time (28kbit modem).
With all that said, I don't see merit in this lawsuit. I've shipped Xbox games including online play and I know how Xbox Live works. I don't think there's anything I can publicly disclose, but suffice it to say there's none of it that I know of that works on this aggregated fixed-framerate multicast sort of model.
I'm not familiar with the Harrod's vans but electric vehicles have been used for milk delivery for many, many years in Britain. Milk floats don't go fast (there are multiple TV sitcom sketches involving milk float chases making light of this fact) or have much range but they are quiet and thus preferable for pre-dawn driving in residential areas.
Better graphics and optimised loading times mean more developers working for more time on the game. If publishers feel the game will sell X more units at retail because of the extra polish, they'll pay for it but usually it doesn't make a difference - especially now when PSP titles are thin on the ground. The system itself doesn't help this because, paradoxically, it's powerful enough that you need a larger, more expensive team to make a competitive game on PSP. Look at Wipeout Pure - 10 programmers and 13 artists.
If Sony make the business side easier you'll see more games and there'll be more competitive pressure to deal with loading times, to do ports properly, to make more original games and to build up a technology base to make it easier to make games on the system. Right now the lacklustre market is just dragging all of these issues.
I wouldn't count on the PS2 dying a premature death. Publishers feel that they "left money on the table" by moving away from PSone too quickly and they'll be looking to PS2 sales to help smooth the transition to PS3 and Xbox360. A Sony guy once told me they half of all the PSones after the Japanese PS2 launch and there's an enormous PS2 installed base (100 million, I believe) which won't be going away any time soon. I completely believe that people will be making PS2 games through 2008.
Now, it won't be at the same rate as now, and a significant percentage of those games will be budget titles, children's titles, highly established franchises and greatest hits. Retailers only have so much shelf space for videogames and they'll want to cut back how much PS2 product they carry so they wanted the least risky product. A good game can still punch through and be profitable despite the higher hurdle - there's so much existing code, art and knowledge that development costs should be lower. Marketing costs are also likely to be low, making the break-even point much more reachable.
The interesting thing to me is that most of the techniques described in this article have been known in the PS2 development community for several years. It's just that no-one working on a PS2-exclusive game has put them all together at once.
Mode 4 at the top (2 colour, 320 pixels wide) and mode 5 (4 colour, 160 pixels wide) at the bottom. They didn't actually use the full screen width so they took less than 10KB of memory (and correspondingly less resolution). The BBC Master version used mode 1/mode 2 for 4 colour and 8 colour respectively.
Other games had used the trick of changing the video registers partway through the frame (via an interrupt) although only to change the colour palette. Elite was the first to change the bit depth and so on as well, effectively changing the display mode. They also had the timing so rigid that there was no need for a black 'gutter' between the two states, as most other games needed.
I've no idea about the keyboard buffer, but the game used about every trick in the book so it wouldn't surprise me. Elite was so far ahead of its time I think cutting edge graphics and a deeper in-game story (ie. taking all that backstory from the novella and making it count in the game itself) would be enough to make it a viable product today. Convincing the retailers and money-men about a space game is a tougher proposition.
What you've described is the very bottom level of the sound system and is rather PC-centric. If your audio needs are extremely simple this is theoretically enough. However, next-generation console games (and even current-gen games) require much more sophisticated work than this.
Music and speech is non-resident because there's just too big a variety in it to preload it all even on Xbox360 - besides, why spend time at a loading screen when we don't need to? A sports game will have literally thousands of speech calls and these days games have hours of background music. Music is streamed so there's a double-buffer in system RAM in addition to the double-buffer you describe. Speech is load-on-demand so we just slurp it all into memory, play it as if it was resident and then flush it. More sophisticated choreography of audio such as stitching together speech calls, transitioning from music intros to looping mainplays or changing hardware effects processing when sounds start or stop (eg. a sound might duck the volume on other voice channels) all require monitoring of the hardware state with fairly high precision and doing more work than is advisable from an interrupt callback.
Better yet, the next-gen consoles audio hardware is basically just a DAC. All the mixing and effects processing (reverb, 3d spatial, etc) is done in software. The CPUs are crazy-fast at math so why put dedicated hardware on the board for that, I guess?
RP is "rating pending". It's used as a placeholder on trailer videos for games and on game demos to indicate the game is still under development and hasn't been submitted to the ESRB yet. You'll see the same thing on movie trailers for films that are a long way from release; "This film is not yet rated".
I am a console game developer. 512x512 textures are pretty rare, even on Xbox. With a 640x448 (480 for Xbox/GCN) framebuffer it's rare you'd get close enough to see all the texels and the performance tends to be poor. 128x128 8bit paletted (on PS2) or DXT (on Xbox/GCN) is where it's at. Things like JPEG aren't that useful because you want to come off disc in the final format you'll use with the hardware. That way you can load-in-place in memory and fix up pointers. Anything else and you're spending realtime doing the decode and you need temporary buffer space in RAM, too.
Mark Turmell created Smash TV as his first game at Williams as an homage to Robotron (Robotron and Defender were Vid Kidz games, created by Eugene Jarvis and Larry DeMar). To this day Mark and Eugene are two of the best Robotron players I've ever met.
If you look at this picture you can clearly see two analog sticks on the controller. They replaced the 4 C buttons from the Nintendo 64 controller with a "C stick".
I stand corrected. I was looking through astronautix and must have not paid close enough attention to his leaving date. I didn't realise wikipedia had good info on Apollos 18, 19 and 20. Thanks for the update.
It was a stepping stone to being a moonwalker
on
Apollo 12 at 35
·
· Score: 3, Informative
Most things in the astronaut corps came through experience. You were backup crew on a mission and 3 missions later you were usally prime crew, for example. Being the command module pilot put you in good stead to be the mission commander on a later flight. Jim Lovell was CM pilot on Apollo 8 and commander on Apollo 13 (Frank Borman was commander of Apollo 8 and probably would have been commander of Apollo 11 if he'd not quit being an astronaut). Dave Scott was CM pilot of Apollo 9 and commander of Apollo 15. John Young was CM pilot of Apollo 10 and commander of Apollo 16.
As for the others, Apollo 7's crew was blacklisted because of their "grumpiness" in flight, Mike Collins quit being an astronaut after Apollo 11, Dick Gordon did the same after Apollo 12 and so did Jack Swigert after Apollo 13 (can't say I blame him). Stu Roosa was Apollo 14's CM pilot but his shot at commanding Apollo 17 was overtaken by Gene Cernan who had been LM pilot on Apollo 10. Apollo 18, 19 and 20 were cancelled and that was that.
There are several different meanings of "64-bit" and they all have differing impact on making videogames (and computing in general for that matter).
32bit vs 64bit address space: Currently most PCs and all game consoles can handle up to 4 gigabytes of memory. This is getting to be a problem on PC because games are using hundreds of megabytes of textures and because memory-mapped I/O for things like PCI cards eats into that total available memory. Going to 64bit addressing completely solves this problem. This is the "64bit" this article is about. The game in question doesn't really take advantage of this, however.
32bit vs 64bit precision for floating point math: Not really a big deal at all. You can do 64bit math already on all the systems, it's just not done in hardware so it's very, very slow by comparison. There's almost never a need for the extra precision anyway; things that lack precision at 32bit are usually flawed due to positive feedback or a lack of understanding of the math pipeline.
32bit vs 64bit data bus: We've already gone to 64bit data busses and beyond. PlayStation2 uses a 128bit wide data bus. Helps you feed data to the CPU (and other system devices) more quickly. Very useful but old technology these days.
32bit vs 64bit registers: Old news, we went to these with the original Pentium. Basically the same argument as for 64bit data bus.
32bit vs 64bit colour: Going from 8bit integer colour channels (ie. red, green and blue from 0-255 each) to 16bit floating point colour channels. This gives you a huge amount of dynamic range for colour and makes it easier to represent very subtle differences too. You need fairly complex pixel shaders for this to be worthwhile, but if you do have that capability it makes all the difference. The next generation of consoles will use this as will coming PC games - it will make their lighting feel much more realistic.
That maccer.net site is a sendup of The Happy Mondays and all the other Ecstasy-fueled bands of the Madchester genre from the early 1990s. I wouldn't be surprised if one of the radio stations played a lot of early Britpop and similar; Blur, Oasis, Happy Mondays, Stone Roses, Soup Dragons, Inspiral Carpets and so on. Perhaps mix it up with some eurobeat like The Shamen, 2 Unlimited, Utah Saints?
Personally I was psyched to hear Welcome To The Jungle on that trailer. Took me back to being 13 and hearing that song for the first time. It'd better be in the damn game now - it's such an LA song after all:-)
The Xbox division is losing a lot of money. Microsoft allocated seed capital to the division to get it going but it must have had the intent that the business would one day be profitable in its own right. Currently they're losing money on the hardware and they're not making nearly enough on game royalties to offset that. It looks like they need each Xbox owner to buy about twice as many games as they do now to hit break-even.
Even a company with as much cash as Microsoft doesn't keep throwing money down a black hole forever. If they decide they can't make a business segment profitable they get rid of it unless it's vital to their strategic future - and who's to say what that crystal ball may show?
Hard drives have a manufacturing cost floor. It's probably around $25-$30 to make the platters, the head, the voice coil, the casing and so on. Doesn't matter what the minimum capacity is, you can't beat those physical costs. Now, the hard drive manufacturer takes their profit and sells the device to the console maker. The console needs a few dollars more in parts (thermal, electrical, shock mounts etc) to integrate the drive. Altogether it's a significant amount of money against the wholesale price of the box.
Microsoft are hurting like hell right now because they have to pay that cost on every Xbox they sell and Sony doesn't. This is why Sony came up with that PSX console/PVR hybrid - they could offer a lot more functionality, charge much more for it and more than cover the extra manufacturing cost.
The 'sufficient' amount of RAM for a console is a function of many things. Sustained read speed from the optical disc, system bus bandwidth, processing power of the CPU and GPU. No point filling memory with data you can't express on-screen or that can't contribute to your simulation of the world. People also don't want to wait 2 minutes for memory to fill from the disc. If all of the other things could scale up, RAM would too - but then the system would be even more expensive.
As for the caching, Xbox already provides scratch space for games - 3 sets allocated on an LRU basis.
Frankly I'd be surprised if any developers were considering supporting it in more than a token sense even before this announcement. There are so few out there and the perceived benefit to customers is so low you'd never recoup the extra cost of testing, let alone development.
Sony needed to release the thing in 2002 and to ship the SCPH-5000x (the 2nd-rev PS2 with the built-in IR/DVD remote) with the hard drive pre-installed if they wanted anyone to pay attention to it. The game market is hit-and-miss enough these days without shackling yourself to some peripheral only a subset of your audience owns or cares about.
I called and got that message as well and thought "Damn! They punked some random chick." That's a valid San Fran area code by the way. But, the site lists Dana as Margaret's niece...
CodeWarrior is what most people in North America use. Sony supply a Linux-based port of gcc and gdb but it's really only Japanese who use that. Lastly, SN Systems' ProDG is popular in Europe. That's the Sony ee-gcc and ee-gdb backend ported to Win32 and integrated with Visual Studio.NET via makefile projects.
None of these is available to you without a Sony developer agreement. Note that the same is true for the Microsoft XDK. Taking an Xbox game to completion requires knowledge of a whole bunch of Xbox-specific APIs you won't find (legally) documented anywhere outside the licensed developer arena. Most people license either Renderware or Gamebyro or another 3rd party cross-platform API or use another Microsoft API; learning DirectX on PC will not give you more than a small headstart on writing an Xbox game.
Personally there are things I like and dislike about CW and VS.NET - neither is my ideal although both are definitely good enough.
The game biz is more like the music biz than the film biz. Glut of titles in the 4th quarter, only big established artists do well then and the rest wither on the vine. New artists that get launched in the 1st quarter benefit from the breathing space.
We just followed this model by releasing The Suffering in March, NBA Ballers in April and Psi-Ops in June. Mortal Kombat is coming towards the end of the year. The results seem to be paying off since those games have sold decently.
None of this is rocket science, as you point out, and I'd be surprised if other companies aren't working towards it as well. It depends where your games are in their development schedules as to when you can start adjusting their release dates.
I would take slight issue with your use of Warcraft as an example. There are a very few games released each year that will generate monster sales no matter when you release them. Halo 2 and The Sims 2, Gran Turismo 4... no-one wants to go up against them with anything. I'd say the Warcraft games fell into that category. Far Cry might be a better PC example - it wasn't already a guaranteed home run.
Many things affect the size of game saves on the memory card. Games aren't allowed to vary the size of their saves so they claim enough space for their worst case scenario. Beyond the game's own data, the icon, optional banner and description are counted in that block total. A block is only 8KB so a game with an 8-frame animated icon and with a banner could well use 6-7 blocks just for that.
It's much worse on PS2 where the (admittedly usually very cool looking) 3D models take anything from 50KB to 150KB each. Since a game can have 3 models (display, copy and delete) that could be almost half a megabyte just for icons!
PS2 actually is the first - it launched in March 2000 in Japan while GBA launched in the USA on June 2000.
GBC is not a different console. Same CPU and memory map, just with some upgraded functionality mostly in the display system. Games can determine the existence of the extra hardware at runtime and tailor their behaviour - hence the GB/GBC combo games in the black cartridges. A better analogy would be a PC with a 3D accelerator.
Hopefully the goal for all 3 next-gen systems is to allow 1080i with multisampling antialiasing and 64bit pixels (16bit floating point colour channels for high dynamic range). Beyond the inherent coolness of this for those of us with HDTVs there is the side benefit that, because 1080i is widescreen only, more games will support 16:9 widescreen.
Xbox is the only current system with enough memory to go over 480p and it doesn't really have the fillrate to make 720p or 1080i worthwhile unless you have simplistic scenes and turn off antialiasing or drop framerate to 30Hz. Personally I'd rather spend the fillrate on antialiasing and/or nicer scenes at 480p.
Games that don't connect to Live themselves but use the bundled content and/or stats downloader to support non-realtime updates are "Assisted Download". Games that do connect to the Live service but just to allow friends list presence and cross-game invites are called "Xbox Live Aware". Games that support full online play are "Xbox Live Multiplayer".
In my personal opinion, and this is only anecdotal because I don't have hard data to back it up, network support is one of those loss-leader things you do for strategic reasons much like supporting widescreen or HDTV 480p. Only a small minority of gamers (15% at most for any of these in the USA) will ever benefit from the feature directly but the extra mindshare it earns you from the media, influential hardcore gamers and the retail channel makes it a good business decision. There's also the experience you're building up doing that sort of development.
Slashdot Mirror
I'm not a lawyer, but I was around at Williams/Midway at the end of MK3 Wavenet and by my reading of the 523 patent claims, Wavenet is not prior art. The patent claims a set of computers pushing messages up to a server every so often, the server aggregating them and pushing a compound message back down to everyone who needs to know. It claims this running on a fixed, short time interval, over the Internet, with the groups being registered and subsequently joined by individual computers.
However, there is nothing in common with Wavenet. Wavenet was an asynchronous peer-to-peer arrangement which did no aggregation and worked over ISDN lines so that it had much higher bandwidth and lower latency than was available via the Internet for arcades at that time (28kbit modem).
With all that said, I don't see merit in this lawsuit. I've shipped Xbox games including online play and I know how Xbox Live works. I don't think there's anything I can publicly disclose, but suffice it to say there's none of it that I know of that works on this aggregated fixed-framerate multicast sort of model.
I'm not familiar with the Harrod's vans but electric vehicles have been used for milk delivery for many, many years in Britain. Milk floats don't go fast (there are multiple TV sitcom sketches involving milk float chases making light of this fact) or have much range but they are quiet and thus preferable for pre-dawn driving in residential areas.
Better graphics and optimised loading times mean more developers working for more time on the game. If publishers feel the game will sell X more units at retail because of the extra polish, they'll pay for it but usually it doesn't make a difference - especially now when PSP titles are thin on the ground. The system itself doesn't help this because, paradoxically, it's powerful enough that you need a larger, more expensive team to make a competitive game on PSP. Look at Wipeout Pure - 10 programmers and 13 artists.
If Sony make the business side easier you'll see more games and there'll be more competitive pressure to deal with loading times, to do ports properly, to make more original games and to build up a technology base to make it easier to make games on the system. Right now the lacklustre market is just dragging all of these issues.
I wouldn't count on the PS2 dying a premature death. Publishers feel that they "left money on the table" by moving away from PSone too quickly and they'll be looking to PS2 sales to help smooth the transition to PS3 and Xbox360. A Sony guy once told me they half of all the PSones after the Japanese PS2 launch and there's an enormous PS2 installed base (100 million, I believe) which won't be going away any time soon. I completely believe that people will be making PS2 games through 2008.
Now, it won't be at the same rate as now, and a significant percentage of those games will be budget titles, children's titles, highly established franchises and greatest hits. Retailers only have so much shelf space for videogames and they'll want to cut back how much PS2 product they carry so they wanted the least risky product. A good game can still punch through and be profitable despite the higher hurdle - there's so much existing code, art and knowledge that development costs should be lower. Marketing costs are also likely to be low, making the break-even point much more reachable.
The interesting thing to me is that most of the techniques described in this article have been known in the PS2 development community for several years. It's just that no-one working on a PS2-exclusive game has put them all together at once.
Mode 4 at the top (2 colour, 320 pixels wide) and mode 5 (4 colour, 160 pixels wide) at the bottom. They didn't actually use the full screen width so they took less than 10KB of memory (and correspondingly less resolution). The BBC Master version used mode 1/mode 2 for 4 colour and 8 colour respectively.
Other games had used the trick of changing the video registers partway through the frame (via an interrupt) although only to change the colour palette. Elite was the first to change the bit depth and so on as well, effectively changing the display mode. They also had the timing so rigid that there was no need for a black 'gutter' between the two states, as most other games needed.
I've no idea about the keyboard buffer, but the game used about every trick in the book so it wouldn't surprise me. Elite was so far ahead of its time I think cutting edge graphics and a deeper in-game story (ie. taking all that backstory from the novella and making it count in the game itself) would be enough to make it a viable product today. Convincing the retailers and money-men about a space game is a tougher proposition.
The Xbox360 core does have a dot product instruction with 1 cycle of throughput so technically the article is correct about that.
Your wider point is well made, however.
What you've described is the very bottom level of the sound system and is rather PC-centric. If your audio needs are extremely simple this is theoretically enough. However, next-generation console games (and even current-gen games) require much more sophisticated work than this.
Music and speech is non-resident because there's just too big a variety in it to preload it all even on Xbox360 - besides, why spend time at a loading screen when we don't need to? A sports game will have literally thousands of speech calls and these days games have hours of background music. Music is streamed so there's a double-buffer in system RAM in addition to the double-buffer you describe. Speech is load-on-demand so we just slurp it all into memory, play it as if it was resident and then flush it. More sophisticated choreography of audio such as stitching together speech calls, transitioning from music intros to looping mainplays or changing hardware effects processing when sounds start or stop (eg. a sound might duck the volume on other voice channels) all require monitoring of the hardware state with fairly high precision and doing more work than is advisable from an interrupt callback.
Better yet, the next-gen consoles audio hardware is basically just a DAC. All the mixing and effects processing (reverb, 3d spatial, etc) is done in software. The CPUs are crazy-fast at math so why put dedicated hardware on the board for that, I guess?
RP is "rating pending". It's used as a placeholder on trailer videos for games and on game demos to indicate the game is still under development and hasn't been submitted to the ESRB yet. You'll see the same thing on movie trailers for films that are a long way from release; "This film is not yet rated".
I am a console game developer. 512x512 textures are pretty rare, even on Xbox. With a 640x448 (480 for Xbox/GCN) framebuffer it's rare you'd get close enough to see all the texels and the performance tends to be poor. 128x128 8bit paletted (on PS2) or DXT (on Xbox/GCN) is where it's at. Things like JPEG aren't that useful because you want to come off disc in the final format you'll use with the hardware. That way you can load-in-place in memory and fix up pointers. Anything else and you're spending realtime doing the decode and you need temporary buffer space in RAM, too.
Mark Turmell created Smash TV as his first game at Williams as an homage to Robotron (Robotron and Defender were Vid Kidz games, created by Eugene Jarvis and Larry DeMar). To this day Mark and Eugene are two of the best Robotron players I've ever met.
If you look at this picture you can clearly see two analog sticks on the controller. They replaced the 4 C buttons from the Nintendo 64 controller with a "C stick".
I stand corrected. I was looking through astronautix and must have not paid close enough attention to his leaving date. I didn't realise wikipedia had good info on Apollos 18, 19 and 20. Thanks for the update.
Most things in the astronaut corps came through experience. You were backup crew on a mission and 3 missions later you were usally prime crew, for example. Being the command module pilot put you in good stead to be the mission commander on a later flight. Jim Lovell was CM pilot on Apollo 8 and commander on Apollo 13 (Frank Borman was commander of Apollo 8 and probably would have been commander of Apollo 11 if he'd not quit being an astronaut). Dave Scott was CM pilot of Apollo 9 and commander of Apollo 15. John Young was CM pilot of Apollo 10 and commander of Apollo 16.
As for the others, Apollo 7's crew was blacklisted because of their "grumpiness" in flight, Mike Collins quit being an astronaut after Apollo 11, Dick Gordon did the same after Apollo 12 and so did Jack Swigert after Apollo 13 (can't say I blame him). Stu Roosa was Apollo 14's CM pilot but his shot at commanding Apollo 17 was overtaken by Gene Cernan who had been LM pilot on Apollo 10. Apollo 18, 19 and 20 were cancelled and that was that.
There are several different meanings of "64-bit" and they all have differing impact on making videogames (and computing in general for that matter).
32bit vs 64bit address space: Currently most PCs and all game consoles can handle up to 4 gigabytes of memory. This is getting to be a problem on PC because games are using hundreds of megabytes of textures and because memory-mapped I/O for things like PCI cards eats into that total available memory. Going to 64bit addressing completely solves this problem. This is the "64bit" this article is about. The game in question doesn't really take advantage of this, however.
32bit vs 64bit precision for floating point math: Not really a big deal at all. You can do 64bit math already on all the systems, it's just not done in hardware so it's very, very slow by comparison. There's almost never a need for the extra precision anyway; things that lack precision at 32bit are usually flawed due to positive feedback or a lack of understanding of the math pipeline.
32bit vs 64bit data bus: We've already gone to 64bit data busses and beyond. PlayStation2 uses a 128bit wide data bus. Helps you feed data to the CPU (and other system devices) more quickly. Very useful but old technology these days.
32bit vs 64bit registers: Old news, we went to these with the original Pentium. Basically the same argument as for 64bit data bus.
32bit vs 64bit colour: Going from 8bit integer colour channels (ie. red, green and blue from 0-255 each) to 16bit floating point colour channels. This gives you a huge amount of dynamic range for colour and makes it easier to represent very subtle differences too. You need fairly complex pixel shaders for this to be worthwhile, but if you do have that capability it makes all the difference. The next generation of consoles will use this as will coming PC games - it will make their lighting feel much more realistic.
That maccer.net site is a sendup of The Happy Mondays and all the other Ecstasy-fueled bands of the Madchester genre from the early 1990s. I wouldn't be surprised if one of the radio stations played a lot of early Britpop and similar; Blur, Oasis, Happy Mondays, Stone Roses, Soup Dragons, Inspiral Carpets and so on. Perhaps mix it up with some eurobeat like The Shamen, 2 Unlimited, Utah Saints?
:-)
Personally I was psyched to hear Welcome To The Jungle on that trailer. Took me back to being 13 and hearing that song for the first time. It'd better be in the damn game now - it's such an LA song after all
The Xbox division is losing a lot of money. Microsoft allocated seed capital to the division to get it going but it must have had the intent that the business would one day be profitable in its own right. Currently they're losing money on the hardware and they're not making nearly enough on game royalties to offset that. It looks like they need each Xbox owner to buy about twice as many games as they do now to hit break-even.
Even a company with as much cash as Microsoft doesn't keep throwing money down a black hole forever. If they decide they can't make a business segment profitable they get rid of it unless it's vital to their strategic future - and who's to say what that crystal ball may show?
Hard drives have a manufacturing cost floor. It's probably around $25-$30 to make the platters, the head, the voice coil, the casing and so on. Doesn't matter what the minimum capacity is, you can't beat those physical costs. Now, the hard drive manufacturer takes their profit and sells the device to the console maker. The console needs a few dollars more in parts (thermal, electrical, shock mounts etc) to integrate the drive. Altogether it's a significant amount of money against the wholesale price of the box.
Microsoft are hurting like hell right now because they have to pay that cost on every Xbox they sell and Sony doesn't. This is why Sony came up with that PSX console/PVR hybrid - they could offer a lot more functionality, charge much more for it and more than cover the extra manufacturing cost.
The 'sufficient' amount of RAM for a console is a function of many things. Sustained read speed from the optical disc, system bus bandwidth, processing power of the CPU and GPU. No point filling memory with data you can't express on-screen or that can't contribute to your simulation of the world. People also don't want to wait 2 minutes for memory to fill from the disc. If all of the other things could scale up, RAM would too - but then the system would be even more expensive.
As for the caching, Xbox already provides scratch space for games - 3 sets allocated on an LRU basis.
This is solely my personal opinion.
Frankly I'd be surprised if any developers were considering supporting it in more than a token sense even before this announcement. There are so few out there and the perceived benefit to customers is so low you'd never recoup the extra cost of testing, let alone development.
Sony needed to release the thing in 2002 and to ship the SCPH-5000x (the 2nd-rev PS2 with the built-in IR/DVD remote) with the hard drive pre-installed if they wanted anyone to pay attention to it. The game market is hit-and-miss enough these days without shackling yourself to some peripheral only a subset of your audience owns or cares about.
I called and got that message as well and thought "Damn! They punked some random chick." That's a valid San Fran area code by the way. But, the site lists Dana as Margaret's niece...
CodeWarrior is what most people in North America use. Sony supply a Linux-based port of gcc and gdb but it's really only Japanese who use that. Lastly, SN Systems' ProDG is popular in Europe. That's the Sony ee-gcc and ee-gdb backend ported to Win32 and integrated with Visual Studio .NET via makefile projects.
None of these is available to you without a Sony developer agreement. Note that the same is true for the Microsoft XDK. Taking an Xbox game to completion requires knowledge of a whole bunch of Xbox-specific APIs you won't find (legally) documented anywhere outside the licensed developer arena. Most people license either Renderware or Gamebyro or another 3rd party cross-platform API or use another Microsoft API; learning DirectX on PC will not give you more than a small headstart on writing an Xbox game.
Personally there are things I like and dislike about CW and VS.NET - neither is my ideal although both are definitely good enough.
The game biz is more like the music biz than the film biz. Glut of titles in the 4th quarter, only big established artists do well then and the rest wither on the vine. New artists that get launched in the 1st quarter benefit from the breathing space.
We just followed this model by releasing The Suffering in March, NBA Ballers in April and Psi-Ops in June. Mortal Kombat is coming towards the end of the year. The results seem to be paying off since those games have sold decently.
None of this is rocket science, as you point out, and I'd be surprised if other companies aren't working towards it as well. It depends where your games are in their development schedules as to when you can start adjusting their release dates.
I would take slight issue with your use of Warcraft as an example. There are a very few games released each year that will generate monster sales no matter when you release them. Halo 2 and The Sims 2, Gran Turismo 4... no-one wants to go up against them with anything. I'd say the Warcraft games fell into that category. Far Cry might be a better PC example - it wasn't already a guaranteed home run.
Many things affect the size of game saves on the memory card. Games aren't allowed to vary the size of their saves so they claim enough space for their worst case scenario. Beyond the game's own data, the icon, optional banner and description are counted in that block total. A block is only 8KB so a game with an 8-frame animated icon and with a banner could well use 6-7 blocks just for that.
It's much worse on PS2 where the (admittedly usually very cool looking) 3D models take anything from 50KB to 150KB each. Since a game can have 3 models (display, copy and delete) that could be almost half a megabyte just for icons!
PS2 actually is the first - it launched in March 2000 in Japan while GBA launched in the USA on June 2000.
GBC is not a different console. Same CPU and memory map, just with some upgraded functionality mostly in the display system. Games can determine the existence of the extra hardware at runtime and tailor their behaviour - hence the GB/GBC combo games in the black cartridges. A better analogy would be a PC with a 3D accelerator.
Hopefully the goal for all 3 next-gen systems is to allow 1080i with multisampling antialiasing and 64bit pixels (16bit floating point colour channels for high dynamic range). Beyond the inherent coolness of this for those of us with HDTVs there is the side benefit that, because 1080i is widescreen only, more games will support 16:9 widescreen.
Xbox is the only current system with enough memory to go over 480p and it doesn't really have the fillrate to make 720p or 1080i worthwhile unless you have simplistic scenes and turn off antialiasing or drop framerate to 30Hz. Personally I'd rather spend the fillrate on antialiasing and/or nicer scenes at 480p.
Games that don't connect to Live themselves but use the bundled content and/or stats downloader to support non-realtime updates are "Assisted Download". Games that do connect to the Live service but just to allow friends list presence and cross-game invites are called "Xbox Live Aware". Games that support full online play are "Xbox Live Multiplayer".
In my personal opinion, and this is only anecdotal because I don't have hard data to back it up, network support is one of those loss-leader things you do for strategic reasons much like supporting widescreen or HDTV 480p. Only a small minority of gamers (15% at most for any of these in the USA) will ever benefit from the feature directly but the extra mindshare it earns you from the media, influential hardcore gamers and the retail channel makes it a good business decision. There's also the experience you're building up doing that sort of development.