In Arthur C Clarke's "The Songs of Distant Earth" set hundreds of years in the future, there's a nice throwaway line about no-one having come up with a better alternative to the keyboard in the intervening years.
I bought a Philips CDD3610. Installed under Win95 on a Winchip class box, wasted many evenings, it was a complete nightmare and just wouldn't work for me. That is what pushed me over the edge to go 100% Linux. Sure I had to do the SCSI emulation thing and a bit of RTFMing, but at the end of the day CD writing on a CDD3610 under Linux works and is rock solid.
Of course my counter example proves nothing, but CD burning worked for me. It even survived a SuSE upgrade from 6.0 to 6.2.
Keeping track of the kids is easy in this smart kidswear concept which incorporates GPS-driven locators and miniature camera's allowing parents to ensure they're safe, while a computer game console worn on the sleeve keeps the kids happy.
Get a life and stop posting that stupid statement on Slashdot. Why should Slashdot not discuss "things that don't exist" (yet)? We're looking to the future, we're not a hardware review site.
Also, X-Box does exist, inasmuch as there are developers writing for it. You can (and probably will) split hairs about "exists". It won't change from non-existence to existence overnight, hence
thre -is- an X-Box, it just isn't ready for market yet.
I agree, but why bother with the OS, which
has increasingly bad associations for consumers?
If I were MS I'd package Office 2002 as a standalone "thing" that takes over your machine just like a game. By definition it wouldn't have to worry about DLL hell or other applications so
would be more stable. It would do most of what most people want it to do. Swapping discs and restarting the box is just as intuitive to many as using the Start menu.
I don't like world domination either, but if they can make something that "just works" like a TV or a typewriter, give 'em credit.
It's grammar and the Nazis did evil stuff between 1933 and 1945 you've probably read about, in a completely different league from trivia that breeds stupid expressions like "drinks nazi" (I heard that one at Siggraph) or "grammar nazi".
Your argument about the ideas and the mold is a dangerous one. The point of using the eraser (as we all do) is to create a better X, as opposed to shipping a bad X and calling anyone who complains a nazi.
Sigh. Go read what I wrote, stop parroting figures you don't understand. System performance is what counts.
You haven't (can't) explain how 48 Gbytes/s
is sustained over busses that can't handle that
bandwidth. You said the 4 Mbyte in the GS was a cache to the 32 Mbyte main memory attached to the EE, remember?
I still don't see why Sony couldn't help your developer friends to program the chip to deliver what you claim it can.
Moderator, wake up! This isn't informative, it's mindless
repetition of marketing hype.
-PS2 CPU has a max. performance of 6.2GFLOPS (don't see XBox' PIII
doing that).
An X-Box 733 MHz Pentium III does 2.9 Gflops in the same vague way
that a PS2 does 6.2 Gflops, but a PS2 hasn't got hardware assist for
transformation and lighting. Compare sustained system performance
figures, discuss the pros and cons of the different architectures,
but mindless Gflop comparisons don't cut it..
-PS2 has the full screen anti-aliasing feature implemented on the
display hardware. This means there is absolutely NO performance hit
when using it. (developers didn't find this at first, that's why the
first ps2 games look jagged)
No, there's 4 Mbyte of RAM and they couldn't figure out how to
page textures so they shrank the frame and depth buffers. Why didn't
developers find it? Are Sony unable to tell them? Are developers
stupid? No, just working against time and an architectural limitation.
And the 4MB of VRAM that's implemented on the Graphics Synthesizer,
isn't really the VRAM. The VRAM is included in the 32MB of RAM on the
EE. That 4MB on the GS is just cache with a 47GB/sec bandwidth. This
means PS2 doesn't need external bandwidth for Z-buffering, rendering
and framebuffer.
There's excellent middleware that can page textures, but your cache
argument is wishful thinking, . How do you figure 47GB/s? Even if
you're talking bits not bytes, thats 6 GBytes/s which is twice the
published 3.2 Gbytes/s memory bandwidth, while the EE internal bus is
2.4 Gbytes/s.
Don't forget the giant inflatable lemons bouncing around the hall (think "The Prisoner"). The free drink beforehand was also appreciated. I felt sorry for the nervous speaker who ended every other sentence with "OK", only to find 1000 people shouting "OK" every time she said it. I sort of liked the man who gave as good as anyone got by saying we were all just geeks and proceeded to tell us how he was here to make money - that took guts (or implied insanity) because Web3D roundup was a bear pit.
I like Chris Hecker's articles in Game Developer and on Gamasutra. I was disappointed in his talk, especially when he seemed to say that because game developers were so driven by commercial pressures all they could do was make incremental improvements rather than innovate. Fair enough, but I got the impression he expected someone else to pay for research while game development rakes in the $$$. Maybe I'm being harsh.
Robin Green of Bullfrog gave an excellent presentation about Steering Behaviours, clear diagrams, code fragments, a pleasant contrast to Chris' talk.
You're talking about nVIDIA and Auschwitz in the same context? Asking us to flame people, to scream kick, fuss and rant?
Grow up.
Read what u4eahh said. I agree with him, I know folks at nVIDIA, it won't help them or anyone to be unreasonable. Someone moderate u4eahh up, if only because he is advocating moderation.
Wireframe display on the Voodoo has been around for a while, and it lets you see what is really happening. I had a good look at Mario64 on UltraHLE on a PC/Voodoo, and my admiration for Nintendo increased when I saw how they did stuff, also I learned why some things that puzzled me had to happen as they did. (Our family owns an N64 btw, for me UltraHLE is just a curiosity.)
So one benefit of nVIDIA wireframe is that it might encourage a few more people to do 3D graphics. Another is that with hardware transform and lighting we can see whose drivers are culling more efficiently, the benefits of clever Z buffer algorithms and so on.
It's important to distinguish Talisman from tile based rendering - I'm an advocate of the latter and have issues with the former. I recall seeing a newsgroup post by Michael T Jones of SGI saying how the Talisman paper at Siggraph 96 didn't give credit to people who'd been working in the area for years, but I couldn't find it. (I did try, and I'd appreciate being given the reference if anyone else can find it). I didn't believe in the image warping aspects because if you turned you head or went round a corner the whole frame had to be re-rendered anyway. Tile based rendering gets you so much, in particular anti-aliasing, elegantly.
Life's too short to reverse engineer all this "I 4|v| 31337" stuff or to hang out with script kiddies, but if there's a non bogus Rosetta stone URL I could turn to in times of utter bafflement....
The clever idea used for Dreamcast GD-ROM is to include "trademark Sega" as human readable text that can also be read by the drive. Not the usual stuff you find on an audio CD, look about 3 cm in from the outside edge of the disk. Without this text disks should be rejected by GD drives.
I'm not sure how the BootCD gets around this, I'm guessing that applications forget to check frequently, allowing the disk swap trick, hence the concept should work, even if the execution is flawed. The clever bit is that if they want to lean on someone producing illegal disks (assuming they know who they are - people who pass off fakes as the real thing) then it's easier to explain "trademark infringement" to the legal profession - just point to the human readable text.
Anyway, thing is, would the Sega text work on the dumb copier - I wouldn't think so.
MS is using COTS hardware to speed R&D time and ultimately, time to market. For a company brand new to the gaming world, I'm worried. They're going to integrate existing hardware and software.
Granted the CPU and memory are COTS (I assume you mean Cheap Off The Shelf) hardware, but why did Microsoft pay nVIDIA $200 million - surely the GPU is a custom design, not just cobbled together?
Textures will need to be mip-mapped so as not to stall the cache.
Why don't you have a go at saying that again? It just makes no sense the way you wrote it.
Abrash explains the issues (concisely, clearly) in this link at the top of the original article referenced by Slashdot - it's easy to miss the link if you're skimming the article.
Slashdot Mirror
Insightful. +5. If not, why not?
retsina or ouzo usually ...
In Arthur C Clarke's "The Songs of Distant Earth" set hundreds of years in the future, there's a nice throwaway line about no-one having come up with a better alternative to the keyboard in the intervening years.
I bought a Philips CDD3610. Installed under Win95 on a Winchip class box, wasted many evenings, it was a complete nightmare and just wouldn't work for me. That is what pushed me over the edge to go 100% Linux. Sure I had to do the SCSI emulation thing and a bit of RTFMing, but at the end of the day CD writing on a CDD3610 under Linux works and is rock solid. Of course my counter example proves nothing, but CD burning worked for me. It even survived a SuSE upgrade from 6.0 to 6.2.
Keeping track of the kids is easy in this smart kidswear concept which incorporates GPS-driven locators and miniature camera's allowing parents to ensure they're safe, while a computer game console worn on the sleeve keeps the kids happy.
That's my daughter in the silver jacket :-)
Get a life and stop posting that stupid statement on Slashdot. Why should Slashdot not discuss "things that don't exist" (yet)? We're looking to the future, we're not a hardware review site.
Also, X-Box does exist, inasmuch as there are developers writing for it. You can (and probably will) split hairs about "exists". It won't change from non-existence to existence overnight, hence thre -is- an X-Box, it just isn't ready for market yet.
Duh.
I agree, but why bother with the OS, which has increasingly bad associations for consumers?
If I were MS I'd package Office 2002 as a standalone "thing" that takes over your machine just like a game. By definition it wouldn't have to worry about DLL hell or other applications so would be more stable. It would do most of what most people want it to do. Swapping discs and restarting the box is just as intuitive to many as using the Start menu.
I don't like world domination either, but if they can make something that "just works" like a TV or a typewriter, give 'em credit.
People have no responsibility to the console manufacturer to fit into their revenue projections
Sure, but try making a phrase from these words: kill, goose, egg, golden.
grammer nazi
It's grammar and the Nazis did evil stuff between 1933 and 1945 you've probably read about, in a completely different league from trivia that breeds stupid expressions like "drinks nazi" (I heard that one at Siggraph) or "grammar nazi".
Your argument about the ideas and the mold is a dangerous one. The point of using the eraser (as we all do) is to create a better X, as opposed to shipping a bad X and calling anyone who complains a nazi.
Sigh. Go read what I wrote, stop parroting figures you don't understand. System performance is what counts.
You haven't (can't) explain how 48 Gbytes/s is sustained over busses that can't handle that bandwidth. You said the 4 Mbyte in the GS was a cache to the 32 Mbyte main memory attached to the EE, remember?
I still don't see why Sony couldn't help your developer friends to program the chip to deliver what you claim it can.
Moderator, wake up! This isn't informative, it's mindless repetition of marketing hype.
-PS2 CPU has a max. performance of 6.2GFLOPS (don't see XBox' PIII doing that).
An X-Box 733 MHz Pentium III does 2.9 Gflops in the same vague way that a PS2 does 6.2 Gflops, but a PS2 hasn't got hardware assist for transformation and lighting. Compare sustained system performance figures, discuss the pros and cons of the different architectures, but mindless Gflop comparisons don't cut it..
-PS2 has the full screen anti-aliasing feature implemented on the display hardware. This means there is absolutely NO performance hit when using it. (developers didn't find this at first, that's why the first ps2 games look jagged)
No, there's 4 Mbyte of RAM and they couldn't figure out how to page textures so they shrank the frame and depth buffers. Why didn't developers find it? Are Sony unable to tell them? Are developers stupid? No, just working against time and an architectural limitation.
And the 4MB of VRAM that's implemented on the Graphics Synthesizer, isn't really the VRAM. The VRAM is included in the 32MB of RAM on the EE. That 4MB on the GS is just cache with a 47GB/sec bandwidth. This means PS2 doesn't need external bandwidth for Z-buffering, rendering and framebuffer.
There's excellent middleware that can page textures, but your cache argument is wishful thinking, . How do you figure 47GB/s? Even if you're talking bits not bytes, thats 6 GBytes/s which is twice the published 3.2 Gbytes/s memory bandwidth, while the EE internal bus is 2.4 Gbytes/s.
16 (not 8) mobos apparently - and the -32- Mbyte GS
There's a Quicktime movie of the wooden mirror on one of the Siggraph CD-ROMs, you can hear the nice sound it makes
Thanks for that reply. Someone should moderate it up.
I was at Siggraph when this Slashdot article was current. Too busy to surf. No-one's going to read this, but here goes anyway.
It's the PS3. Moore's Law. Just slap down 16 PS2 in parallel. Inherent back compatibility with PS2 and original PS.
Also each GS has 8 times as much embedded RAM as PS2 version of GS, hence much higher resolution, less jaggies, more textures on line.
Don't forget the giant inflatable lemons bouncing around the hall (think "The Prisoner"). The free drink beforehand was also appreciated. I felt sorry for the nervous speaker who ended every other sentence with "OK", only to find 1000 people shouting "OK" every time she said it. I sort of liked the man who gave as good as anyone got by saying we were all just geeks and proceeded to tell us how he was here to make money - that took guts (or implied insanity) because Web3D roundup was a bear pit.
Timothy Childs certainly rules
I like Chris Hecker's articles in Game Developer and on Gamasutra. I was disappointed in his talk, especially when he seemed to say that because game developers were so driven by commercial pressures all they could do was make incremental improvements rather than innovate. Fair enough, but I got the impression he expected someone else to pay for research while game development rakes in the $$$. Maybe I'm being harsh.
Robin Green of Bullfrog gave an excellent presentation about Steering Behaviours, clear diagrams, code fragments, a pleasant contrast to Chris' talk.
You're talking about nVIDIA and Auschwitz in the same context? Asking us to flame people, to scream kick, fuss and rant?
Grow up.
Read what u4eahh said. I agree with him, I know folks at nVIDIA, it won't help them or anyone to be unreasonable. Someone moderate u4eahh up, if only because he is advocating moderation.
Wireframe display on the Voodoo has been around for a while, and it lets you see what is really happening. I had a good look at Mario64 on UltraHLE on a PC/Voodoo, and my admiration for Nintendo increased when I saw how they did stuff, also I learned why some things that puzzled me had to happen as they did. (Our family owns an N64 btw, for me UltraHLE is just a curiosity.)
So one benefit of nVIDIA wireframe is that it might encourage a few more people to do 3D graphics. Another is that with hardware transform and lighting we can see whose drivers are culling more efficiently, the benefits of clever Z buffer algorithms and so on.
It's important to distinguish Talisman from tile based rendering - I'm an advocate of the latter and have issues with the former. I recall seeing a newsgroup post by Michael T Jones of SGI saying how the Talisman paper at Siggraph 96 didn't give credit to people who'd been working in the area for years, but I couldn't find it. (I did try, and I'd appreciate being given the reference if anyone else can find it). I didn't believe in the image warping aspects because if you turned you head or went round a corner the whole frame had to be re-rendered anyway. Tile based rendering gets you so much, in particular anti-aliasing, elegantly.
Life's too short to reverse engineer all this "I 4|v| 31337" stuff or to hang out with script kiddies, but if there's a non bogus Rosetta stone URL I could turn to in times of utter bafflement ....
Philips patented it around 1963, with a license policy that was basically "free to anyone but you must conform to the spec".
Hmm, sounds like a pragmatic use of the patent system where everyone wins. Heresy on /. ?
The clever idea used for Dreamcast GD-ROM is to include "trademark Sega" as human readable text that can also be read by the drive. Not the usual stuff you find on an audio CD, look about 3 cm in from the outside edge of the disk. Without this text disks should be rejected by GD drives.
I'm not sure how the BootCD gets around this, I'm guessing that applications forget to check frequently, allowing the disk swap trick, hence the concept should work, even if the execution is flawed. The clever bit is that if they want to lean on someone producing illegal disks (assuming they know who they are - people who pass off fakes as the real thing) then it's easier to explain "trademark infringement" to the legal profession - just point to the human readable text.
Anyway, thing is, would the Sega text work on the dumb copier - I wouldn't think so.
MS is using COTS hardware to speed R&D time and ultimately, time to market. For a company brand new to the gaming world, I'm worried. They're going to integrate existing hardware and software.
Granted the CPU and memory are COTS (I assume you mean Cheap Off The Shelf) hardware, but why did Microsoft pay nVIDIA $200 million - surely the GPU is a custom design, not just cobbled together?
Textures will need to be mip-mapped so as not to stall the cache.
Why don't you have a go at saying that again? It just makes no sense the way you wrote it.
Abrash explains the issues (concisely, clearly) in this link at the top of the original article referenced by Slashdot - it's easy to miss the link if you're skimming the article.
"All these worlds are yours except Europa. Attempt no landing there"
- 2010 Arthur C Clarke