I think you're underestimating ATI's and nVidia's latest generations. If you go and read this technical document (PDF, ~4 MB) on the nVidia GeForce FX, you'll see that it does, in fact, support full 128-bit floating point all the way up to, and including, the frame buffer. I'm not
certain why 12 bits of integer should be better than 32 bits of floating point, per component, maybe you could educate me a bit there? Also, I don't know what the actual analog output quality of the FX chipset is, maybe SGI has a (small) lead there, still.
"[...] the 48-bit RGBA provides the highest level of precision available on any desktop system today"
Oh? Quick, everyone with Radeon 9700 PRO graphics boards in your PCs, make sure you have them in tower cases, or something!
For reference, the ATI specs page states:
Pixel shaders up to 160 instructions with 128-bit floating point precision
I guess SGI might refer to actual output precision, i.e. the RAMDAC D/A-converters... In that case, it seems they still have the edge, since the ATI boards only have 10 bits per component. Still, I think that's of lesser value than the actual precision image operations are performed at.
Re:It's heavier, consumes more power, etc...
on
Rear View LCD?
·
· Score: 2
From the way I parsed it, you're missing the fact that they use a LCD+camera system instead of an external mirror, since the latter affects wind resistance. So by going with a power hungry but aerodynamically friendly LCD+camera system, and then just turning it off, they get the best of two worlds. Except for the added mass, of course. Does that make sense?
Probably true, yes. Unfortunately, we never felt we had even enough momentuum to motivate us to release binaries... Which possibly was a mistake, of course. If only N people had mailed and bitched that they wanted to try Verse out, but couldn't be bothered to compile it, that might have helped. It's all speculation now, of course. Plus the core architecture of Verse (very client-intensive client/server, requiring many clients to get anything done) makes it a bit involved to just try out, it needs packaging, which we also hardly had the time for... Oh well.
As someone who has spent almost two years working on building, from the ground up, the technology for something similar, but having had to kill the project mainly due to lack of external interest, I think the main advantage of DustCity is the target audience. The Counter-Strike user community seems to be very strong, and of course already centers around creating 3D content. Plus, the idea of integrating clans and giving them the ability to build their own "homes" in DustCity seems obvious, but is hopefully brilliant just as well. Now, if only my home PC was powerful enough to actually do 3D things on...;^)
Oh, and for the curious: check out the results of the above-mentioned effort, which was sponsored by one of Sweden's coolest research companies, at verse.sf.net. It's all Free Software, using a combination of GPL, LGPL and BSD licenses. Never mind the bitterness of the opening (final) diary entries.;^)
The reason that the instructions are "hard to follow", might be that the site the article is on, TechTV, is a network video broadcasting site... So there's a video show, "The Screen Savers" (Weekdays 7 p.m. Eastern) that's going to be showing more about the project, I assume. Further, I guess that's almost impossible to figure out if you're a Slashdot editor (or, now that I've started to bitch, a story submittor). Heh.
If you're into SF, then I simply have to recommend Stephen Baxter. I can't recall any of his books that doesn't involve, oh, the universe coming to an end in some spectacular way. He's a good writer, imo, but the crushing dystopicality (heh) of his novels often gets to me. For a place to start reading Baxter, check out the beginning of his Xeelee sequence, Raft.
Which, I see, is out of print. Hm. Typical.
Um, I see a lot of comments along the lines of "NVIDIA aren't including loops in the language because GPUs just can't do loops". It seems
NVIDIA aren't aware of that, the next-generation "NV3x" hardware supports loops up to 64 levels of nesting... They've also grown the maximum pixel shader program size by a nice little factor of 512 (65,536 rather than 128 instructions per program). Also, it says "dynamic flow control" in the chart, which sounds like maybe arbitrary branching (there's your GOTO right there) could be supported.
That said, it does seem a bit weird not to make Cg strong enough to include features that are obviously needed for their own next-generation of hardware... But all the conspiracy theories have already been used up, I'll just settle for introducing some facts in the discussion.;^)
One thing: Lian Li PC-9300, perhaps? No idea about availability, I sure haven't seen it around these parts yet, but... Check out the other similar case mentioned in the review, too, perhaps. And start reading Dan's stuff, and give him plenty of donations so I don't have to. Thank you.;^)
Oh. Then, next weekend, Chris is going to ask for references about how to program books! Whoa!! Well, I guess in that case, there can be only one...;^)
By their very definition, binaries kinda have to be "(pre)compiled", you know... You probably mean precompiled headers, at least that would make more sense.
That, and the fact that it was mainly backed by Intel. For some reason, the people at that company seem to like things that load down the main CPU. It's weird. But, actually, I'm not complaining all too much; I think USB's fairly neat. And there's a "movement" to develop host-less USB technology, which will very much mimic FireWire and allow e.g. a camcorder to talk to a VCR without the involvement of a "smart" host such as a PC. Can't seem to remember the exact name for this new standard at the moment, though.
So, if you've written a piece of software, all you have to do to get an article about it on Slashdot is wait until an article about a similar program is published. Then, you
bomb
the
thread
with
links
to
your
project, wait a while, submit a story about your project, and Bob's your uncle. Easy. *Crawls back under a rock, waiting for the next article even remotely touching file managers*.;^)
Right, but that makes the total dimension mm^9. The guy who said it was mm^27 is wrong, but mm is just as wrong, too. The proper unit is the (plain, one-dimensional) millimeter, since it is then taken to apply to each of the three dimensions listed.
Hrm. I actually think you're giving them too much of a hard time. I hadn't seen the company before just now, and am certainly not overly impressed by their product or their attitute (too much marketing speech!). Still, I think most people in this thread are mis-reading the tech specs. At least you didn't interpret it to have a pair of 512x768 pixel displays, like someone above, but I think it has dual VGA controllers, too. One per display: the tech specs, under "VGA" (item #8) say "SMI 721 * 2". If so, then that should be adequate for 1024x768. That said, I agree that it's heavy, fragile-looking, and expensive.
No, it is not. The original article really has mmas the unit after the measurements, and that is incorrect (notice the little digit 3 flying there). Your example, with only mm, is of course just fine. It's just another case of lousy proofreading of a (semi-scientific) article.
(OK, totally off topic now, but whattheheck). I just tried mucking with the prefs some, and actually got it into a shape which compiles. There's an extraneous space in there, which is very annoying in a sig designed to be compact, but... I feel better. Thanks for poking me.
I *know*, but thanks for pointing it out. You're not the first, either, which says a lot about the crowd here. I fit right in, I guess. Anyway, the reason the sig doesn't compile is because Slashcode mangles it. There's very little I can do about it.
Matrix Orbital. Many of their displays support keypads for input, one even goes all the way up to a full PS/2 keyboard. I haven't tried these for myself, but from reading their site some, I get the impression the keypads run over the same serial link as is used to control the display. Makes sense to me that it should, too.
Smaller? Hello... You need a bigger controller, of course!:D It does seem that the promised console adapters for this lovely-looking piece of joystickery still aren't available, but if they through some fluke of nature actually appear, this looks like a solid enough stick. I know I want one. And a suitable TV-ized box to MAME on. Oh, and a living room to put it all in. And, perhaps, a job with enough of a salary to pay for it, too. Darn.
Then stuff like quake, rtcw, half-life, etc would be a bit more submersive.
Um, I don't know how you prefer to play those games, and it sure has been hot where I am today, so the idea of submerging oneself while playing action games has some appeal... Still, I do think the word you're looking for is immersive. Could just be me, though.;^)
Slashdot Mirror
Is it, really? :)
Thank you. I was not being clear enough, it seems.
I think you're underestimating ATI's and nVidia's latest generations. If you go and read this technical document (PDF, ~4 MB) on the nVidia GeForce FX, you'll see that it does, in fact, support full 128-bit floating point all the way up to, and including, the frame buffer. I'm not certain why 12 bits of integer should be better than 32 bits of floating point, per component, maybe you could educate me a bit there? Also, I don't know what the actual analog output quality of the FX chipset is, maybe SGI has a (small) lead there, still.
Oh? Quick, everyone with Radeon 9700 PRO graphics boards in your PCs, make sure you have them in tower cases, or something!
For reference, the ATI specs page states:
I guess SGI might refer to actual output precision, i.e. the RAMDAC D/A-converters... In that case, it seems they still have the edge, since the ATI boards only have 10 bits per component. Still, I think that's of lesser value than the actual precision image operations are performed at.
From the way I parsed it, you're missing the fact that they use a LCD+camera system instead of an external mirror, since the latter affects wind resistance. So by going with a power hungry but aerodynamically friendly LCD+camera system, and then just turning it off, they get the best of two worlds. Except for the added mass, of course. Does that make sense?
Probably true, yes. Unfortunately, we never felt we had even enough momentuum to motivate us to release binaries... Which possibly was a mistake, of course. If only N people had mailed and bitched that they wanted to try Verse out, but couldn't be bothered to compile it, that might have helped. It's all speculation now, of course. Plus the core architecture of Verse (very client-intensive client/server, requiring many clients to get anything done) makes it a bit involved to just try out, it needs packaging, which we also hardly had the time for... Oh well.
As someone who has spent almost two years working on building, from the ground up, the technology for something similar, but having had to kill the project mainly due to lack of external interest, I think the main advantage of DustCity is the target audience. The Counter-Strike user community seems to be very strong, and of course already centers around creating 3D content. Plus, the idea of integrating clans and giving them the ability to build their own "homes" in DustCity seems obvious, but is hopefully brilliant just as well. Now, if only my home PC was powerful enough to actually do 3D things on... ;^)
;^)
Oh, and for the curious: check out the results of the above-mentioned effort, which was sponsored by one of Sweden's coolest research companies, at verse.sf.net. It's all Free Software, using a combination of GPL, LGPL and BSD licenses. Never mind the bitterness of the opening (final) diary entries.
The reason that the instructions are "hard to follow", might be that the site the article is on, TechTV, is a network video broadcasting site... So there's a video show, "The Screen Savers" (Weekdays 7 p.m. Eastern) that's going to be showing more about the project, I assume. Further, I guess that's almost impossible to figure out if you're a Slashdot editor (or, now that I've started to bitch, a story submittor). Heh.
If you're into SF, then I simply have to recommend Stephen Baxter. I can't recall any of his books that doesn't involve, oh, the universe coming to an end in some spectacular way. He's a good writer, imo, but the crushing dystopicality (heh) of his novels often gets to me. For a place to start reading Baxter, check out the beginning of his Xeelee sequence, Raft. Which, I see, is out of print. Hm. Typical.
Um, I see a lot of comments along the lines of "NVIDIA aren't including loops in the language because GPUs just can't do loops". It seems NVIDIA aren't aware of that, the next-generation "NV3x" hardware supports loops up to 64 levels of nesting... They've also grown the maximum pixel shader program size by a nice little factor of 512 (65,536 rather than 128 instructions per program). Also, it says "dynamic flow control" in the chart, which sounds like maybe arbitrary branching (there's your GOTO right there) could be supported.
;^)
That said, it does seem a bit weird not to make Cg strong enough to include features that are obviously needed for their own next-generation of hardware... But all the conspiracy theories have already been used up, I'll just settle for introducing some facts in the discussion.
Sounds like a problem with an easy enough fix...
One thing: Lian Li PC-9300, perhaps? No idea about availability, I sure haven't seen it around these parts yet, but... Check out the other similar case mentioned in the review, too, perhaps. And start reading Dan's stuff, and give him plenty of donations so I don't have to. Thank you. ;^)
Oh. Then, next weekend, Chris is going to ask for references about how to program books! Whoa!! Well, I guess in that case, there can be only one... ;^)
By their very definition, binaries kinda have to be "(pre)compiled", you know... You probably mean precompiled headers, at least that would make more sense.
That, and the fact that it was mainly backed by Intel. For some reason, the people at that company seem to like things that load down the main CPU. It's weird. But, actually, I'm not complaining all too much; I think USB's fairly neat. And there's a "movement" to develop host-less USB technology, which will very much mimic FireWire and allow e.g. a camcorder to talk to a VCR without the involvement of a "smart" host such as a PC. Can't seem to remember the exact name for this new standard at the moment, though.
Me, too! (There, did you get it now?)
So, if you've written a piece of software, all you have to do to get an article about it on Slashdot is wait until an article about a similar program is published. Then, you bomb the thread with links to your project, wait a while, submit a story about your project, and Bob's your uncle. Easy. *Crawls back under a rock, waiting for the next article even remotely touching file managers*. ;^)
Right, but that makes the total dimension mm^9. The guy who said it was mm^27 is wrong, but mm is just as wrong, too. The proper unit is the (plain, one-dimensional) millimeter, since it is then taken to apply to each of the three dimensions listed.
Hrm. I actually think you're giving them too much of a hard time. I hadn't seen the company before just now, and am certainly not overly impressed by their product or their attitute (too much marketing speech!). Still, I think most people in this thread are mis-reading the tech specs. At least you didn't interpret it to have a pair of 512x768 pixel displays, like someone above, but I think it has dual VGA controllers, too. One per display: the tech specs, under "VGA" (item #8) say "SMI 721 * 2". If so, then that should be adequate for 1024x768. That said, I agree that it's heavy, fragile-looking, and expensive.
No, it is not. The original article really has mmas the unit after the measurements, and that is incorrect (notice the little digit 3 flying there). Your example, with only mm, is of course just fine. It's just another case of lousy proofreading of a (semi-scientific) article.
(OK, totally off topic now, but whattheheck). I just tried mucking with the prefs some, and actually got it into a shape which compiles. There's an extraneous space in there, which is very annoying in a sig designed to be compact, but... I feel better. Thanks for poking me.
I *know*, but thanks for pointing it out. You're not the first, either, which says a lot about the crowd here. I fit right in, I guess. Anyway, the reason the sig doesn't compile is because Slashcode mangles it. There's very little I can do about it.
Matrix Orbital. Many of their displays support keypads for input, one even goes all the way up to a full PS/2 keyboard. I haven't tried these for myself, but from reading their site some, I get the impression the keypads run over the same serial link as is used to control the display. Makes sense to me that it should, too.
Smaller? Hello... You need a bigger controller, of course! :D It does seem that the promised console adapters for this lovely-looking piece of joystickery still aren't available, but if they through some fluke of nature actually appear, this looks like a solid enough stick. I know I want one. And a suitable TV-ized box to MAME on. Oh, and a living room to put it all in. And, perhaps, a job with enough of a salary to pay for it, too. Darn.
Then stuff like quake, rtcw, half-life, etc would be a bit more submersive. ;^)
Um, I don't know how you prefer to play those games, and it sure has been hot where I am today, so the idea of submerging oneself while playing action games has some appeal... Still, I do think the word you're looking for is immersive. Could just be me, though.