The PS1 already contained a cooperatively multi-tasking OS in its ROM. I don't think that many games used its tasking features, but the developer kit was based on functionality found in the OS (but also replaced many of the routines with faster, more versatile or simply bugfixed versions).
If there are enough cores to compute both branches from the conditional
I don't see how that could really be useful.
Doing it with multiple cores would probably be a waste, but isn't that what the IA64's predicated execution is all about? To avoid pipeline bubbles it executes both paths from the branch, and once the branch condition is known the results from the not-taken path are thrown away.
There is nothing new with Out of Order Execution. It's been implemented in all the Pentium cores as well as AMD chips from the K6 (I think) on up.
The original Pentium was superscalar, but in-order. The big "wow" thing about the P6 was that contrary to everyone's expectation, Intel had managed to make an out-of-order version of the x86.
It takes no longer to translate into one or fifteen languages, as it is work that can be done in parallel. However I find this piece of news unlikely, as that would mean Nintendo had given up their highly successful business practice of "fucking Europe up the ass, hard."
nVidia, IBM Sony Cell, and Altivec support only 32-bit floats.
The Cell SPEs do support double-precision floats, but IIRC you take a massive performance hit when using them. The PPE has a normal floating-point unit in addition to the Altivec instruction set.
The 6600GT can be passively cooled (eg. Gigabyte sell one such model, which I have), but the heatsink is much bigger than the one the 7600GS has in the picture. The performance increase wouldn't be that big however, and the 7600GS actually sports a lower memory clock than the 6600GT. It would be nice to get a power consumption figure as well - the 6600GT uses about 50 W and the linked item lists the 7600GT at 67 W.
CBS's "Without a Trace" drew the $3.6 million fine against 111 stations for an episode that showed no nudity, but featured scenes suggesting a teen orgy.
I don't think Apple's "pro" market would would be satisfied with a machine limited to 2GB of memory (which seems to be the limit of OSX on 32-bit hardware).
Your end user would not care about compile times, but they would care about the DVD they were watching skipping because Mail decided it was time to refresh their inbox. The title of my post should also give the hint that I don't give a hoot about your end user in this matter.
The only really important thing to me is that they improved on the atrociously poor HDD of the original Mini. About 80% of the beach-ball-spinning I've had to endure would just go away, and videos wouldn't immediately start stuttering whenever any other program at all touched the HDD. Compiling software should also be much quicker, which is another gripe I have.
Some DOS ANSI drivers were also vulnerable, and there were "ANSI bombs" that would delete your files or whatever the author could think of. Not supporting the keyboard redefinition feature was a big feature of some replacement drivers.
Weren't Nintendo quite keen on getting third party developers already for the Gamecube? Still, all their efforts will amount to nothing if they can't keep their platform commercially attractive (again, Gamecube).
My "weapon of choice" is the TAC-2, also from Suncom. It's got two buttons (mapped to the same pin), a slightly bigger ball handle and a lovely metal shaft. Very sturdy and had really precise control.
My least favourite controller would have to be Commodore's original 1311 joystick, which has been described as a "successful attempt at making an even worse joystick than Atari's". Absolutely horrible.
IIRC it was QuickShot who had something similar, but even worse in the late 80s. Not only was their half-sphere absolutely huge (not a chance of it fitting under your palm), but the two buttons were these big cassette-player-type things that had to be pressed really far down to register.
It's funny that QuickShot made some of the best Atari-type joysticks there was, but at some point they went absolutely batshit crazy and started producing a huge range of models, all of which were shit.
It really is rubbish. The +controller is also too far back, you have to bend your thumb uncomfortably backwards to reach it. The same goes for the C-stick, but since it's not flat it's not quite as bad.
Huh, didn't know there was Epyx-branded hardware. Hereabouts that one was sold as the Konix Speedking. Decent stick, but utterly useless if you're lefthanded. There was also a version with autofire and even an analog version for PCs.
Sony are specifying that the PS3 has only seven SPUs as a way to increase yield (and thus reduce costs) of the Cell processor. When processors are manufactured, a more or less constant percentage of the silicon wafer will be corrupted by impurities, specks of dust etc. With a chip as big as Cell there is thus a high risk that some part of it won't work correctly, and by using seven SPUs Sony can take all the chips where one SPU doesn't validate correctly and still use them.
Once the Cell goes through a process shrink it will become physically smaller and as a result more will have eight fully-working SPUs, but commercial games will of course still not be able to use all of them.
Nitpicking, but it wasn't Irem who sued Rainbow Arts but Activision, who held the home computer license for R-Type. It was also Activision who then hired RA for the official port after they ran into problems with their own port.
The "Classification/impact" column is not from the original document. I don't know who wrote the comments or what their technical background is, but they seem a little alarmist to me.
Slashdot Mirror
The PS1 already contained a cooperatively multi-tasking OS in its ROM. I don't think that many games used its tasking features, but the developer kit was based on functionality found in the OS (but also replaced many of the routines with faster, more versatile or simply bugfixed versions).
It takes no longer to translate into one or fifteen languages, as it is work that can be done in parallel. However I find this piece of news unlikely, as that would mean Nintendo had given up their highly successful business practice of "fucking Europe up the ass, hard."
The 6600GT can be passively cooled (eg. Gigabyte sell one such model, which I have), but the heatsink is much bigger than the one the 7600GS has in the picture. The performance increase wouldn't be that big however, and the 7600GS actually sports a lower memory clock than the 6600GT. It would be nice to get a power consumption figure as well - the 6600GT uses about 50 W and the linked item lists the 7600GT at 67 W.
I don't think Apple's "pro" market would would be satisfied with a machine limited to 2GB of memory (which seems to be the limit of OSX on 32-bit hardware).
Your end user would not care about compile times, but they would care about the DVD they were watching skipping because Mail decided it was time to refresh their inbox. The title of my post should also give the hint that I don't give a hoot about your end user in this matter.
The only really important thing to me is that they improved on the atrociously poor HDD of the original Mini. About 80% of the beach-ball-spinning I've had to endure would just go away, and videos wouldn't immediately start stuttering whenever any other program at all touched the HDD. Compiling software should also be much quicker, which is another gripe I have.
Some DOS ANSI drivers were also vulnerable, and there were "ANSI bombs" that would delete your files or whatever the author could think of. Not supporting the keyboard redefinition feature was a big feature of some replacement drivers.
Me neither. Meanwhile, the original link is working just fine.
Weren't Nintendo quite keen on getting third party developers already for the Gamecube? Still, all their efforts will amount to nothing if they can't keep their platform commercially attractive (again, Gamecube).
My least favourite controller would have to be Commodore's original 1311 joystick, which has been described as a "successful attempt at making an even worse joystick than Atari's". Absolutely horrible.
IIRC it was QuickShot who had something similar, but even worse in the late 80s. Not only was their half-sphere absolutely huge (not a chance of it fitting under your palm), but the two buttons were these big cassette-player-type things that had to be pressed really far down to register. It's funny that QuickShot made some of the best Atari-type joysticks there was, but at some point they went absolutely batshit crazy and started producing a huge range of models, all of which were shit.
It really is rubbish. The +controller is also too far back, you have to bend your thumb uncomfortably backwards to reach it. The same goes for the C-stick, but since it's not flat it's not quite as bad.
Huh, didn't know there was Epyx-branded hardware. Hereabouts that one was sold as the Konix Speedking. Decent stick, but utterly useless if you're lefthanded. There was also a version with autofire and even an analog version for PCs.
Oh you know, they're probably just delaying it so it doesn't have to compete against their own titles in the crowded GC market.
Once the Cell goes through a process shrink it will become physically smaller and as a result more will have eight fully-working SPUs, but commercial games will of course still not be able to use all of them.
Nitpicking, but it wasn't Irem who sued Rainbow Arts but Activision, who held the home computer license for R-Type. It was also Activision who then hired RA for the official port after they ran into problems with their own port.
At the time they wrote GGS they were, though. It was one of their first (if not the first) retail games.
The "Classification/impact" column is not from the original document. I don't know who wrote the comments or what their technical background is, but they seem a little alarmist to me.