can't we test for "realism" the same way as we (try) to test for AI? if you can't tell that it was CG, why would they bother telling you so?
I remember a quote that goes like "the holy grail of CG is that you don't even notice it." Example being in spiderman, where peter parker woke up all ripped and stuff - that was, IIRC, CG - but you don't even notice it.
haha, or go back to WINDOWS! X-mouse functionality has been part of the "power-toys" since windows 95; and I think it might be a standard feature in XP...
Funny thing is that when I use a X-server within windows, all the native windows apps conforms to the X-mouse behavior, while the X-server stuff don't. Irony, no?
I don't think the lack of realism really came from the redering technologies, but rather the modelling technologies.
Take the comparison between the real-actress and the CGI (Final Fantasy X-2 promotional video) for example, the lighting and such are all perfectly fine; but you can notice how "rigid" and un-natural the CG character's body moves.
I think, personally, that during movement, any fancy rendering effects are lost, but the actual movement themselves are the critical "realism" that needs to be addressed.
For one, human limbs move on a 1) feed-back system, which would be hard to simulate its complexities simply by dragging the block that says "arm" from here to there, 2) the feed-back is also has a lot to do with balance, another thing difficult to simulate properly, with such a complex system as the human body.
Interesting enough, Final Fantasy (the movie) is completely shot with the little humans too; I think it has to do with the fact that we cannot track the positions of the dots perfectly, though.
It should be possible eventually to do a GPS-esq system where the room has "location transponders" and each "dot" on the actor/actress's body would calculate it's location and send it out wirelessly to a computer somewhere nearby. I think after that, we can see some very good reproduction of human motions.
just my arm-chair thoughts after watching CG generated stuff for a long time.
Because we don't have anything better, if you don't count Macs.
I mean, I'd love for OSS community to develop something with no strings attached, like BeOS, say. but it's not about doing that - it's about getting things ported for it, and having COMPANIES BELIEVE IN IT - the latter takes a lot of time.
With the momentum behind linux (and possibly FreeBSD, to a lesser extent), what other OSS alternatives are there for a real desktop operating system?
As far as "company belief," many game makers are releasing linux native versions now - it took how many years? If we all made this whiz-bang new OS tomorrow, how many years do you think it would take for them to start releasing for it? before IBM spills the dough for a widget-OS consulting dept and runs commercials? when CIO maganize talk about your widget-OS strategy?
The fact is, all of this takes time, and by the time something is "ready" for the desktop, they would have been loaded down with baggages too. Revolution is nice and good - but their results have usually been less-than-extrodinary.
I remember reading somewhere that there is a chasm when the models gets too near "realistic." Like from no resemblance to real world (cartoon networks' toonami robot) to "100% human-like" models, the "realism" would scale linearly until you get to about 95%, which the realism just falls because the models simply "don't look right," or something. of course, eventually it would be possible to climb back the chasm to reach 100%, but I suppose that's really hard. =)
I read this in wired a while back, I believe; and no blinking is not a fault of the CG tech, but rather that of the guys doing this stuff - in better CG things (final fantasy for one - heck even in games they put in blinking characters now), they put in the details.
Btw, FF (movie) seemed weird and creepy but you CAN'T explain what's no "not real" about it - a good example of the "chasm."
And for another example - for anybody who seen the promotional video of Final Fantasy X-2 where there is a comparative track between the real singer doing her song and dance number, and cutting to the CG (Yuna) doing the same, you can see how strangely unreal the CG version is - though looking at CG version alone does not necessarily give you the idea.
Unless the female voice can render the below lines with feelings, I don't think it's a mature technology.
give me! give me! oh! I am coming!! OHHHH!
Actually I did try it. the result (of the above line) was not spectacular. I am impressed with the quality in general, though. Tried "Sticking feathers up your butt does not make you a chicken," but that needs to be said with feelings as well, I suppose.
Oh yeah, this kind of technology is excellent for a computer to read out the sites to you, if, say, your eyes are tired. It should work wonders for slashdot, even.
hmm. sounds like you are swapping out too much. windows tend to do that a lot, and worse if you don't have gobs of memory.
I have way over a thousand messages in my inbox and it hardly slows down for anything (the only time it would hiccup is when checking / downloading new messages). Do yourself a favor, get some RAM upgrades - they are not expensive, and turn off swap altogether (registry hack for win2k). should speed up things a bit. (Don't do this unless you have half gig or more, though)
Somebody with a high-gain antenna will just immediately "see" what you are seeing. fuck spy-satellites; just let everyone do the spying for you (well, sell a lot of those glasses first, anyhow)!
Stalkers are gonna have such a blast with all these.
On the other hand, I predict that shit like this will be embedded into our bodies within no time.
From what I have read so far is that they put voltage combinations into the hippocampus (I will type HC from here on) and observes its output - much like a combinational logic circuit; which is assuming that one input is exactly correspondent to one output.
Being neurons, I'd assume that the HC portions at the "input" area would also "fire-back," as well, which makes it a little more complicated.
Worse yet, if I do a CPU analogy - load Reg A with 0 and Reg B with 1; keep clocking in commands of "A=A+B" (and output Reg A every step). Of course, while your 1000 commands looks exactly the same, the results of "A=A+B" is different. Similarly, I would expect something similar from any part of the brain - as a signal would take multiple paths to get from one end to the other, and more likely than not the order of pulse inputs are very important as well.
so two things, mainly:
1) the bidirectional-ness of the HC is apparently not duplicated
2) It is impossible to even check for a unknown CPU's full range of functions (let's say, you get a 8086 and no manuals or knowledge of internal structure (or experience in semiconductors) - but you do get to slam different combinations of "stuff" into it - how much of the testing would it take to duplicate the CPU completely? - now, try a 386; and now try a PIV with its huge cache and massive pipelines.
I dunno; I can just imagine something will be lost with the current implementation - though I hope otherwise.
Obviously you have not been through many wars... I had the unfortunate opportunity of browsing through some books about WW2 era media, and you cannot possibly believe the size and energy of the propaganda machine.
To be honest, it continues to this day - winners write the history books, friend.
OTOH, besides the money-earning aspect, I really don't think oscars are all that, it's like a bunch of movie people get together and give themselves an award. how about something we get to vote on?
first of all, I don't think it's true that you can completely disable paging in windows, or AFAIK in win2k anyhow. you can disable "data paging" by setting a registry key, but deleting the page file or setting it to 0 will make windows bitch at you to the extent of "you NEED page file. I create for you." bastards.
WinXP has a "disable paging" button. with that said - it might be that windows just go do something fishy behind your back. but I am giving MS a little trust right now and I decided that for this alone I am willing to stick with XP, even though minimal config it takes 20MB extra than 2k
anyway - that's the thing though; with, say, moz, I tell it to stay in memory (leave that quick-start portion) - and I think it should be an option to do this to all of the programs. I mean, there are just these few things I use, and instead of keeping a "portion" in memory - I'd rather it would just kept all of it in memory, all the time.
I am willing to buy a half gig extra, JUST so that I can keep all these things in memory. and I think both of us know that if it was done, even 15krpm drives can't beat my load speed, which would be near instantaneous. averaging out over the long run, I think the few times I do load infrequently-used-stuff will be more than balanced out by my super-fast other stuff.
It's like people who chokes out the extra for a huge HD so they can store CD images on their HDs (and just mount them) instead of flipping through CDs. same thing here; I believe that a such a way of decreasing wait-time should be an option, and I think it's a good option too, being the cheap price of DRAM now-a-days - but very disapponited that it's not quite possible, or at least on a large scale.
btw - on laptops, which is mainly what I use now, 10k drives are not an option, so maybe that's where I came up with my little optimization idea.
btw if you translate the japanese title directly i think it's "heaven's door" - I thought the "knocking no heaven's door" would be a good english interpretation (yes I watched the japanese one)... probably have to do with copyrighted song titles? though AFAIK you can't copyright song titles at all.
ack. that is just whack.
btw, no nudity as far as I remembered (though there was points where Faye was pretty close), so I don't think much will be cut. Violence was not extreme either as well, IIRC;
btw, anybody (who saw it in japanese) knows who sang the opening song? it was in english, but I can't figure it out.
that wasn't my point, and you don't have to explain memory hierachy to me, I know it well.
The point is - hard drives with high transfer rate (okay, so 10k will afford you a few microseconds of access time too) have very few benefits, and only in a very few areas (that *I* can think of, anyway):
now, with a large enough memory, you shouldn't ever NEED to swap, or worry about using massive space for kernel compile (and really now, you gentoo kids need to chill out a little), etc etc.
For video edit, you can use the extra space anyhow so for similar price a RAID 7.2k drive array would work out better price/perf wise, I'd think (and sorry but a raided 7.2k would get better rates than 10k single, while probably not costing much more). (with raid card, you can get probably three 7200 drives while only two 10k drives)
so, besides boot-time, WHY would you need a faster hard-drive; or the question being, why invest the money into a faster drive, instead of a LOT of memory?
can you imagine how your system will scream if it never have to page, ever again? (technically, you can't really "page" anyhow since you already filled all 32 bits - that's AFAIK, correct me if you know better; it's been a while)
so, it's more like a economics question.
I am not really suggesting RAMDISK, btw - I just think that you can compile your OS / programs with option like "I have massive memory so use it lavishly and don't touch that drive."
on the second page of the eweekarticle, it seems that SCO may have pretended to be a market-research company that called around and asked how would the companies feel if SCO had sued them on intellectual property rights. including SONY and Ford, which also runs linux on... stuff.
my immediate question is, Ford runs linux? on what?
and second question is, isn't SCO like... Caldera?
aaaanyway; I guess linux is taking away the marked share of UNIX boxes a lot more than it's headway into the desktop (windows) arena. fighting the wrong crowd, I'd say...
maybe microsoft will file a suit that says "we want damages because linux makes our business model un-profitable."
Have anybody ever actually thought about it? For the amount of extra money to blow, why not spend more for memory and have EVERYTHING run from there? what, 4G is not enough for your desktop system? x86 only addresses that much right now, ya know...
Sure, your's might boot faster, but I still don't understand why we can't have linux just run off memory and don't even touch the hard-drives. How much trouble is to cache (even, user selectively) some directory of common-used code into memory? ls / ifconfig / vi or emacs depending on your flavor?
Anyway, I know there are situations in which a fast harddrive is warrented, but I still think that the money is better spent on more memory; a lot more memory. or a RAID controller.
One word: cellphone Actually that might be two words.
Anyways, in Japan they ALREADY are taking advantage of this technology - you can take 3D pictures on your photo-capable cellphones, print them out, etc etc. I don't know how well it works because I havn't seen it in action yet, but it has sure been in commercials a lot lately.
Don't think of 3D as a real 3D like "volumetric" but more like those magic-eye things - where it's an illusion of 3D, in the other words, you don't get more data (i.e. you never see more of the sides of the 3D thing by changing your perspective, trying to look at the display from the sides), but the object appears 3D, fooling your eyes.
So assuming that, ahem, the growth is linear (it won't be) and has been keeping at it for there years (probably more), that's 1.1 _____bytes! so, erm, when do we expect the database to achieve consiousness again?
And, dosn't the gigabit speed seems kinda trivial, compared to the massive amounts of data stored there?
Heck how do they manage corrupt bits? the chance of random bits failing here and there is just too high to ignore, no?
reminds me of a dilbert strip where he was in his gf's place (on the computer, no less) and she asks "you've been here quite often lately, wouldn't have anything to do with the fact that interent at your house is broken, right?"
well, something to that extent.
speaking of which, though - Scott Adams was quite skethy on *how* did Dilbert get a gf...
The same that was said about the benefit of AC posts still applies even in the TMF window - people posting valuable information but fear of prosecution or whatnot. I mean, to access the TMF window you have to me subscribing anyhow, why take away freedom to AC post? Could you explain your reasoning?
Here is an inquire article about it too
on
ATi Radeon 9800 Pro
·
· Score: 3, Interesting
Quoting: ATI will call the extended set of DX9 features the DX9++, although we suppose it could add just as many ++++++ as it wanted to....... Nvidia should perhaps call its own DX9 extensions DX9## or DX9.NET.
the sad thing is, though - I would not be surprised if Nvidia did release a DX9# or something stupid like that. I mean, look at Athlons naming themselves AthlonXP. ack
Why does Nvidia's demo with vid-card shows you this and this, but ATI shows you this? I think we should buy Nvidia based on their sense of ascthetics alone.
seriously though - was it like last week 9700PRO became available? what's up with this break-neck card-releasing? I didn't think it was christmas yet...
Slashdot Mirror
that gives me an interesting idea:
can't we test for "realism" the same way as we (try) to test for AI? if you can't tell that it was CG, why would they bother telling you so?
I remember a quote that goes like "the holy grail of CG is that you don't even notice it." Example being in spiderman, where peter parker woke up all ripped and stuff - that was, IIRC, CG - but you don't even notice it.
haha, or go back to WINDOWS! X-mouse functionality has been part of the "power-toys" since windows 95; and I think it might be a standard feature in XP...
Funny thing is that when I use a X-server within windows, all the native windows apps conforms to the X-mouse behavior, while the X-server stuff don't. Irony, no?
I don't know what you refered to by "it," but:
I don't think the lack of realism really came from the redering technologies, but rather the modelling technologies.
Take the comparison between the real-actress and the CGI (Final Fantasy X-2 promotional video) for example, the lighting and such are all perfectly fine; but you can notice how "rigid" and un-natural the CG character's body moves.
I think, personally, that during movement, any fancy rendering effects are lost, but the actual movement themselves are the critical "realism" that needs to be addressed.
For one, human limbs move on a
1) feed-back system, which would be hard to simulate its complexities simply by dragging the block that says "arm" from here to there,
2) the feed-back is also has a lot to do with balance, another thing difficult to simulate properly, with such a complex system as the human body.
Interesting enough, Final Fantasy (the movie) is completely shot with the little humans too; I think it has to do with the fact that we cannot track the positions of the dots perfectly, though.
It should be possible eventually to do a GPS-esq system where the room has "location transponders" and each "dot" on the actor/actress's body would calculate it's location and send it out wirelessly to a computer somewhere nearby. I think after that, we can see some very good reproduction of human motions.
just my arm-chair thoughts after watching CG generated stuff for a long time.
Because we don't have anything better, if you don't count Macs.
I mean, I'd love for OSS community to develop something with no strings attached, like BeOS, say. but it's not about doing that - it's about getting things ported for it, and having COMPANIES BELIEVE IN IT - the latter takes a lot of time.
With the momentum behind linux (and possibly FreeBSD, to a lesser extent), what other OSS alternatives are there for a real desktop operating system?
As far as "company belief," many game makers are releasing linux native versions now - it took how many years? If we all made this whiz-bang new OS tomorrow, how many years do you think it would take for them to start releasing for it? before IBM spills the dough for a widget-OS consulting dept and runs commercials? when CIO maganize talk about your widget-OS strategy?
The fact is, all of this takes time, and by the time something is "ready" for the desktop, they would have been loaded down with baggages too. Revolution is nice and good - but their results have usually been less-than-extrodinary.
I remember reading somewhere that there is a chasm when the models gets too near "realistic." Like from no resemblance to real world (cartoon networks' toonami robot) to "100% human-like" models, the "realism" would scale linearly until you get to about 95%, which the realism just falls because the models simply "don't look right," or something. of course, eventually it would be possible to climb back the chasm to reach 100%, but I suppose that's really hard. =)
I read this in wired a while back, I believe; and no blinking is not a fault of the CG tech, but rather that of the guys doing this stuff - in better CG things (final fantasy for one - heck even in games they put in blinking characters now), they put in the details.
Btw, FF (movie) seemed weird and creepy but you CAN'T explain what's no "not real" about it - a good example of the "chasm."
And for another example - for anybody who seen the promotional video of Final Fantasy X-2 where there is a comparative track between the real singer doing her song and dance number, and cutting to the CG (Yuna) doing the same, you can see how strangely unreal the CG version is - though looking at CG version alone does not necessarily give you the idea.
give me! give me! oh! I am coming!! OHHHH!
Actually I did try it. the result (of the above line) was not spectacular. I am impressed with the quality in general, though. Tried "Sticking feathers up your butt does not make you a chicken," but that needs to be said with feelings as well, I suppose.
Oh yeah, this kind of technology is excellent for a computer to read out the sites to you, if, say, your eyes are tired. It should work wonders for slashdot, even.
though I have second thoughts about MY TAX DOLLARS (erm, euros) funding it.
whatever floats your boat, though.
darn, eh?
hmm. sounds like you are swapping out too much. windows tend to do that a lot, and worse if you don't have gobs of memory.
I have way over a thousand messages in my inbox and it hardly slows down for anything (the only time it would hiccup is when checking / downloading new messages). Do yourself a favor, get some RAM upgrades - they are not expensive, and turn off swap altogether (registry hack for win2k). should speed up things a bit. (Don't do this unless you have half gig or more, though)
Somebody with a high-gain antenna will just immediately "see" what you are seeing. fuck spy-satellites; just let everyone do the spying for you (well, sell a lot of those glasses first, anyhow)!
Stalkers are gonna have such a blast with all these.
On the other hand, I predict that shit like this will be embedded into our bodies within no time.
hah, pun in the title, but seriously:
From what I have read so far is that they put voltage combinations into the hippocampus (I will type HC from here on) and observes its output - much like a combinational logic circuit; which is assuming that one input is exactly correspondent to one output.
Being neurons, I'd assume that the HC portions at the "input" area would also "fire-back," as well, which makes it a little more complicated.
Worse yet, if I do a CPU analogy - load Reg A with 0 and Reg B with 1; keep clocking in commands of "A=A+B" (and output Reg A every step). Of course, while your 1000 commands looks exactly the same, the results of "A=A+B" is different. Similarly, I would expect something similar from any part of the brain - as a signal would take multiple paths to get from one end to the other, and more likely than not the order of pulse inputs are very important as well.
so two things, mainly:
1) the bidirectional-ness of the HC is apparently not duplicated
2) It is impossible to even check for a unknown CPU's full range of functions (let's say, you get a 8086 and no manuals or knowledge of internal structure (or experience in semiconductors) - but you do get to slam different combinations of "stuff" into it - how much of the testing would it take to duplicate the CPU completely? - now, try a 386; and now try a PIV with its huge cache and massive pipelines.
I dunno; I can just imagine something will be lost with the current implementation - though I hope otherwise.
Obviously you have not been through many wars... I had the unfortunate opportunity of browsing through some books about WW2 era media, and you cannot possibly believe the size and energy of the propaganda machine.
To be honest, it continues to this day - winners write the history books, friend.
OTOH, besides the money-earning aspect, I really don't think oscars are all that, it's like a bunch of movie people get together and give themselves an award. how about something we get to vote on?
first of all, I don't think it's true that you can completely disable paging in windows, or AFAIK in win2k anyhow. you can disable "data paging" by setting a registry key, but deleting the page file or setting it to 0 will make windows bitch at you to the extent of "you NEED page file. I create for you." bastards.
WinXP has a "disable paging" button. with that said - it might be that windows just go do something fishy behind your back. but I am giving MS a little trust right now and I decided that for this alone I am willing to stick with XP, even though minimal config it takes 20MB extra than 2k
anyway - that's the thing though; with, say, moz, I tell it to stay in memory (leave that quick-start portion) - and I think it should be an option to do this to all of the programs. I mean, there are just these few things I use, and instead of keeping a "portion" in memory - I'd rather it would just kept all of it in memory, all the time.
I am willing to buy a half gig extra, JUST so that I can keep all these things in memory. and I think both of us know that if it was done, even 15krpm drives can't beat my load speed, which would be near instantaneous. averaging out over the long run, I think the few times I do load infrequently-used-stuff will be more than balanced out by my super-fast other stuff.
It's like people who chokes out the extra for a huge HD so they can store CD images on their HDs (and just mount them) instead of flipping through CDs. same thing here; I believe that a such a way of decreasing wait-time should be an option, and I think it's a good option too, being the cheap price of DRAM now-a-days - but very disapponited that it's not quite possible, or at least on a large scale.
btw - on laptops, which is mainly what I use now, 10k drives are not an option, so maybe that's where I came up with my little optimization idea.
that title is just... AWFUL.
btw if you translate the japanese title directly i think it's "heaven's door" - I thought the "knocking no heaven's door" would be a good english interpretation (yes I watched the japanese one)... probably have to do with copyrighted song titles? though AFAIK you can't copyright song titles at all.
ack. that is just whack.
btw, no nudity as far as I remembered (though there was points where Faye was pretty close), so I don't think much will be cut. Violence was not extreme either as well, IIRC;
btw, anybody (who saw it in japanese) knows who sang the opening song? it was in english, but I can't figure it out.
that wasn't my point, and you don't have to explain memory hierachy to me, I know it well.
The point is - hard drives with high transfer rate (okay, so 10k will afford you a few microseconds of access time too) have very few benefits, and only in a very few areas (that *I* can think of, anyway):
1) video-edit
2) system boot
3) kernel compile; maybe
4) swap
now, with a large enough memory, you shouldn't ever NEED to swap, or worry about using massive space for kernel compile (and really now, you gentoo kids need to chill out a little), etc etc.
For video edit, you can use the extra space anyhow so for similar price a RAID 7.2k drive array would work out better price/perf wise, I'd think (and sorry but a raided 7.2k would get better rates than 10k single, while probably not costing much more). (with raid card, you can get probably three 7200 drives while only two 10k drives)
so, besides boot-time, WHY would you need a faster hard-drive; or the question being, why invest the money into a faster drive, instead of a LOT of memory?
can you imagine how your system will scream if it never have to page, ever again? (technically, you can't really "page" anyhow since you already filled all 32 bits - that's AFAIK, correct me if you know better; it's been a while)
so, it's more like a economics question.
I am not really suggesting RAMDISK, btw - I just think that you can compile your OS / programs with option like "I have massive memory so use it lavishly and don't touch that drive."
my immediate question is, Ford runs linux? on what?
and second question is, isn't SCO like... Caldera?
aaaanyway; I guess linux is taking away the marked share of UNIX boxes a lot more than it's headway into the desktop (windows) arena. fighting the wrong crowd, I'd say...
maybe microsoft will file a suit that says "we want damages because linux makes our business model un-profitable."
WHY do we need fast hard-disks?
Have anybody ever actually thought about it? For the amount of extra money to blow, why not spend more for memory and have EVERYTHING run from there? what, 4G is not enough for your desktop system? x86 only addresses that much right now, ya know...
Sure, your's might boot faster, but I still don't understand why we can't have linux just run off memory and don't even touch the hard-drives. How much trouble is to cache (even, user selectively) some directory of common-used code into memory? ls / ifconfig / vi or emacs depending on your flavor?
Anyway, I know there are situations in which a fast harddrive is warrented, but I still think that the money is better spent on more memory; a lot more memory. or a RAID controller.
One word: cellphone
Actually that might be two words.
Anyways, in Japan they ALREADY are taking advantage of this technology - you can take 3D pictures on your photo-capable cellphones, print them out, etc etc. I don't know how well it works because I havn't seen it in action yet, but it has sure been in commercials a lot lately.
Don't think of 3D as a real 3D like "volumetric" but more like those magic-eye things - where it's an illusion of 3D, in the other words, you don't get more data (i.e. you never see more of the sides of the 3D thing by changing your perspective, trying to look at the display from the sides), but the object appears 3D, fooling your eyes.
Editors might want to get this straight too
1.1 ____ bytes = 1.1*(2^50) = 1238489897526886 bytes
Today's record was about 59.5 seconds for 6.7GB, so rounding off to 6.7GB/min: 7194070221 bytes / minute (roughly)
that would take 172154.27 minutes to get everything out on this fat pipe
equating to about 119.55 days, or roughly 4 monthes. That's keeping at the maximum record set today all day everyday for that duration.
Ouch...
So assuming that, ahem, the growth is linear (it won't be) and has been keeping at it for there years (probably more), that's 1.1 _____bytes! so, erm, when do we expect the database to achieve consiousness again?
And, dosn't the gigabit speed seems kinda trivial, compared to the massive amounts of data stored there?
Heck how do they manage corrupt bits? the chance of random bits failing here and there is just too high to ignore, no?
reminds me of a dilbert strip where he was in his gf's place (on the computer, no less) and she asks "you've been here quite often lately, wouldn't have anything to do with the fact that interent at your house is broken, right?"
well, something to that extent.
speaking of which, though - Scott Adams was quite skethy on *how* did Dilbert get a gf...
The same that was said about the benefit of AC posts still applies even in the TMF window - people posting valuable information but fear of prosecution or whatnot. I mean, to access the TMF window you have to me subscribing anyhow, why take away freedom to AC post? Could you explain your reasoning?
Quoting: ATI will call the extended set of DX9 features the DX9++, although we suppose it could add just as many ++++++ as it wanted to. ... ... Nvidia should perhaps call its own DX9 extensions DX9## or DX9.NET.
the sad thing is, though - I would not be surprised if Nvidia did release a DX9# or something stupid like that. I mean, look at Athlons naming themselves AthlonXP. ack
seriously though - was it like last week 9700PRO became available? what's up with this break-neck card-releasing? I didn't think it was christmas yet...