Please back up your outrageous claims with some real data. I, for one, upgraded to XP only because the IT folks told me I had to. I'm not sure why they decided everyone needed to upgrade, but I'm pretty sure it wasn't because they wanted everyone to have more eye candy. And when I did finally upgrade, I immediately turned off all the silly PreSkool fluff and made it look like Win95/2000 again.
And besides, even if you're right, Linux is in a different place in the "hierarchy of needs" than Windows right now. Windows users haven't had to worry about lack of useful documentation or easy configuration for years, so they can afford to obsess over eye candy more than poor Linux users who have enough trouble just getting their printing to work
More effort should be going to providing more usable and better documented GUIs, not making more eye candy. But this is open source, and people work on what they want to work on. And people don't want to work on making things usable. There's no fun in that. You can't post to Slashdot saying "hey! i was just up programming all night -- check out my new highly usable CUPS setup wizard". No one will care, because the folks reading Slashdot all got CUPS working ages ago. What do they care if it's now easier to do? But you can post "check out these super new transparent menus!" and all the 14-year olds will gasp in awe and amazement and call you a h4X0r 60D.
If you RTFM it explains what RSS is pretty well. I wasn't too up on it either.
The thing that made me curious in the article is this quote:
"For an average Internet user who regularly visits about 50 Web sites, rather than have to go visit those 50 sites wouldn't it be cool if those sites could somehow visit you? "
Does the average internet user really visit 50 sites regularly? If so I'm way below average. I regularly visit maybe 2 or 3 sites, so I really don't see any reason to invest time in getting RSS set up. I find it hard to believe that an average user really visits so many sites. It seems to me RSS is really more of a niche thing that's really great for people who like to overload themselves with information, rather than a super-hot technology that's about to explode into widespread popularity.
It can't be too long before more third party suppliers spring up who do nothing but build game engines and design tools.
There already are quite a few 3rd party engines from tools only companies. I'm not sure why Mr. Blow only mentioned the Quake and Unreal engines. There are other engines like NetImmerse (or I guess it's called "Gamebryo" now), and, well, there are more links here. Go see for yourself.
I think The Unreal and Quake engines are actually some of the most expensive you can buy. From what I understand, you pay a premium for these engines because there's marketing value in being able to associate yourself with greatness by advertising "Built on the Unreal Engine". Whereas "Built on the LithTech engine" isn't going to impress anyone. So you pay a lot for those engines, when in fact, it's sort of akin to movies that advertise "From the producer of 'some good movie'". It's just not a very good predictor of whether it will or will not be a good movie.
If you can walk around the thing and the steroscopic pairs change accordingly, how is that different from reality?
The two main differences are latency and multi-viewer capability.
When you move around a true 3D image there is ZERO latency. You move your head back and forth, you always see the right view and it's perfectly in sync with your viewpoint. Any system that has to track your head, and then generate a stereoscopic pair based on that tracking result is going to have some latency. The result is that the image seems to swim a little bit. And it doesn't take much latency to make many people get a form of motion sickness. (Consider with 60fps display you have generally at least 16 msec of latency, and trackers usually pile on at least another 10-30msec or so, at best. That's plenty to induce motion sickness in many people, and in those that don't feel sick, at least it is enough that the swimming of the image is obvious when you make quick movements of your head.
Second, with stereo pairs, only one person can get the correct 3D view at a time. So it kind of cuts down on the potential for use in a group setting. Not to mention that you have to track the viewer and/or wear special eye gear. That cuts down on the potential uses also. You can't, say, have a 3D kiosk that people can just idly walk by and be wowed by if they have to line up one by one and put on some kind of tracking head gear to see the effect.
So there are a number of real reasons why you'd want to have a real 3D image-generation device instead of a device that's merely stereoscopic. If you just want to sit in front of your monitor and appreciate 3D porn, then there's nothing wrong with stereoscopic images, and at $150 or so (compared to $50K) the price is certainly right.
No, the Actuality display (Perspecta) uses projectors too. But the difference is that in Perspecta parts of the optics (some mirrors) move with the screen. Effectively it's like the projector is whizzing around as fast as the screen is, so that the projector is always at the perfect angle to project on the screen.
With the Hitachi display, there appear to be 24 fixed optical paths rather than one continuously varying path that tracks with the screen. So most of the time the Hitachi's spinning screen will be projected onto by a projector that's at a slightly wrong angle. As the screen sweeps through the 15 degree arc that corresponds to one optical path, the image will be first have a slight perspective/keystone distortion in one direction, then it will be correct for one slight fraction of a section, then it will proceed to become keystoned in the opposite direction. The result: smeary 3D images, and more smeary the further you get from the central axis of the volume.
That is unfortunate for this kind of display because the central axis is bad for viewing to begin with. The center axis is always dim because a significant fraction of the sweep time you're looking at the screen nearly edge-on, so the optimal part of the viewing volume is the off-axis part. But with Hitachi's display it seems like you will have dimness in the center axis, and smearing off axis. No sweet spots at all.
But I could be wrong, the Hitachi folk could possibly be pre-warping the images to correct for the keystoning as the screen sweeps past, but I doubt it, based on the video, and based on the update rate required of the projector to do that. Maybe it's on their to-do list, though.
I think that only projects a 2D image into 3D space. It looks like they are basically doing the trick of projecting on a thin laminar stream of fog. Although they don't call it fog, they call it "transformed air" because that sounds a lot cooler and more mysterious.
It does look pretty neat. But it's not 3D imagery.
It does look similar to Actuality's system. But it seems to have much lower resolution. Take a look at the video someone posted in this message. It's really really smeary, almost to the point that the subject is unrecognizable.
From the spec sheet you can see Actuality's display does 198 slices of the volume compared to Hitachi's 24, and each slice is 768x768 resolution, compared to whatever Hitachi does. Just guessing, but assuming they Hitachi splits one projector frame up into 24 subframes (which it looks like they do because the schematics seem to indicate fixed optics), and generously assuming no wasted pixels, that comes to something like 213x256 resoultion per view, assuming they start with a 1280x1024 projector. So the frame resoulution is also a good bit lower than Actuality's.
Also looking at the vid of the Hitachi, and how smeary the images are, it almost makes me think they are projecting ALL 24 images ALL the time rather than blanking all but the two projecting most perpendicular to the screen. Or maybe it's smeary because they're using the same image for 15 degree chunks (360/24), compared to Actuality's 1.8 degrees (360/198). Or it could just be an artifact introduced by the video camera.
The other big difference is you can actually buy a display from Actuality today -- if you have $39,995.:-)
The search box.:-) Search for "non-commercial". I think you're right, though, they have basically removed all the direct links to that page. I certainly couldn't find one anywhere. Maybe they don't even mean for it to be findable via search.
Is GTK for Win32 actually viable now? All the rumors I've heard up till now is that it doesn't really work quite right on Windows yet.... Oh and yeh, I can't stand writing GUIs in C.
Well anyway there's no one-size-fits-all toolkit. You've gotta pick what's right for your requirements. They all have pros and cons. FOX has a good mix of what I want: flexibility, ease of coding, cross-platform support, attractive GUI, OpenGL support, and freeness. It doesn't have quite the wide support of advanced features that wxWindows or Qt have, but I find it easier to work with than wxWindows, and it's definitely got fewer licensing issues than Qt.
It's not libre free, but beer free. And it's old as hell. But the poster was correct that there is a "non-commercial" version for windows that's freely available. Be warned though that Trolltech's definition of "non-commercial" is rather peculiar.
I'm wondering if they'll make this new non-comm 3.2 edition similarly available.
The old 2.3 non-com version for windows is still on Trolltech.com. Just search for "non-commercial" in the little search box at the top of the page, And you'll get this link as the top hit.
That is "non-commercial", though, and not "free". Which means according to Trolltech you can't use it for "any activities for which you are compensated," and that covers academic research too.
I don't think they've touched the code in over a year. Winamp is dead. Kiss it goodbye. The "new! Winamp 3!" button on the web page is just a joke now.
The "new and improved" Winamp3 really sucked the llama's cock. The video features are crap, all the neat plugins that used to work with winamp2 broke, and display of non-Roman character sets (the only thing I really felt lacking in Winamp2) was still not implemented, and actually was made more difficult to get working by their great bitmap GUI and fonts idea. The big new feature out of the year or so of development on Winamp3 was, what, a UI that could be scripted and bitmapped? Whee. Oh boy. Look -- we just want to access our music with minimal hassle, people. Most of us couldn't care less whether the buttons on the UI get jiggy on mouseover, or whether we can make it look like a 6-year drew it, as long as it can play the music and display the song titles.
I'll just stick with iTunes for Windows, thanks, which at least can display song titles properly, even if they're Japanese. And which integrates very nicely with my iPod as a bonus.
People have mostly been harping on how baseless this guy's criticisms of Qt are. But it's not just Qt where the facts are wrong. The whole thing is a pile of uninformed troll droppings. He criticizes FOX for requiring the FX_IMPLEMENT macro, but completely ignores the very similar macros that wxWindows uses, like DECLARE_EVENT_TABLE()and IMPLEMENT_APP(), and Qt's QOBJECT macro. Then there's a criticism of FOX requiring you to fiddle with message id enums, but this is also applicable to wxWindows. It's just a side effect of using enums and message tables for identifying messages.
Then there's the RAD Y/N column. FLTK's FLUID is a very basic code generator for FLTK dialogs. It's got nowhere near the functionality (or usability) of Qt's QDesigner, or from what I can see GTK's GLADE. FLUID hardly qualifies as a RAD tool.
Anyway, this article is a very shallow analysis of GUI toolkits that seems to be culled mostly from reading the web pages of the projects involved. This guy doesn't seem to have spent any significant amount of time actually *using* these toolkits, or talking to people who have.
Just to get the facts right, the 2.3.0 version of Qt for Win32 is most definitely NOT GPL. The source code is not available and the use of the binaries is restricted to "non-commercial" activities. Incidentally this apparently also excludes the use of that version of Qt for academic research, since Trolltech considers research to be "commercial activity". See the FAQ.
IAOAP (i am only a programmer) but "highly repetitive filler/rubbish/crap sounds" sounds like a pretty good description of a lot of C code I've seen. Especially that written by higher organisms such as math and engineering Ph.D.'s.
I don't think you have it quite right there. ATI always uses 24-bit floating point math internally per component, while NVIDIA uses either 16 or 32-bit floating point, depending on the size of the operands. So ATI isn't really using "full floating point" all the time, it's more like "3/4 full floating point" all the time, whereas NVIDIA uses the precision you ask for which is either "half floating point" or "full floating point".
[4]: NVIDIA GeForce FX supports two floating-point data formats: 16-bit per component and 32-bit per component. GeForce FX performs 32-bit floating-point calculations twice as slow as 16-bit ones: its 16-bit ALUs have to get in pairs for 32-bit calculations.
ATI RADEON 9700 PRO supports both 16-bit and 32-bit data precision, but performs all floating-point calculations with 24-bit precision. The result can be then translated into the 16-bit format, or expanded to the 32-bit one.
The article claims the image quality was improved, but the first thing I noticed when I opened up the one of the 4xAA screen shots was that the antialiasing on Dawn's hair isn't working properly in the ATI version. The NVIDIA demo uses the GL_SAMPLE_ALPHA_TO_COVERAGE_ARB feature to feather out the hair at the ends. (This makes the card use the alpha value of a fragment to determine the number of covered samples when multisample antialiasing is on.)
The zoom on the NV example isn't quite as close-up, but you can still see how the hair feathers away at the tips, while the hair on the ATI is all jaggy and uniform thickness right out to the tips. Like some kind of Raggedy Anne yarn hair.
So does the ATI not support alpha-to-coverage? Or did these guys making the wrapper just not translate it properly? The hair looks pretty bad without it.
I seem to remember when I saw the NV guys give the demo that they made a point of how expensive the hair was to render. Each hair is a separate GL_LINE_STRIP, there are thousands of them, and if you turn off the complicated blending and smoothing on each one of those little hairs, that could maybe add up to a large savings. I don't know, but I wonder if that would account for most of the speed-up they see running it on ATI.
Still a neat hack. It would really rock to have complete NV<->ATI compatibility dlls that would work for all apps and not just this demo.
While they may have made some slight improvement to one aspect of image quality by improving normalization, which I guess makes the lighting a little more accurate, I really doubt the improvement is all that noticeable. Maybe I could tell given a side-by-side comparison, but I doubt I would notice if only shown a version normalized one way or the other. On the other hand, that ugly hair is pretty obvious. To me that makes the NVIDIA sreen shots look better.
As it is, it doesn't even deal with differences between C++ implementations - it doesn't even list GCC in the index.
That's too bad, because 9 out of 10 times when I've had troubles with templates it's because of differences between C++ implementations. Beautiful, well thought out, intricate standards-compliant examples are useless if I can't actually get them to compile with my real-world compiler!
The book I'm looking for is the one that gives me real-world recipes for getting around bone-headded compilers. For example, there are at least 3 different ways to declare templatized friend functions depending on the compiler. Only one is correct according to the standard, but the standard isn't worth a whole lot to me today if the compilers I'm stuck using don't follow it. And likewise, an advanced templates book isn't of much use to me today if the examples won't compile on my compilers.
But the 2 minute rule gets in the way of real back and forth discussion too often, and also makes it a pain to post corrections if you realize a mistake just after submitting.
A less restrictive, but equivalently troll-subduing rule would be to allow, say, only 4 posts per 8 minutes. You could post 'em all in the first minute if you want, but then you've got to wait out 7. Trolls would still only be able to get in 720 posts/day, but average users shouldn't ever notice the limit.
Slashdot Mirror
Please back up your outrageous claims with some real data. I, for one, upgraded to XP only because the IT folks told me I had to. I'm not sure why they decided everyone needed to upgrade, but I'm pretty sure it wasn't because they wanted everyone to have more eye candy. And when I did finally upgrade, I immediately turned off all the silly PreSkool fluff and made it look like Win95/2000 again.
And besides, even if you're right, Linux is in a different place in the "hierarchy of needs" than Windows right now. Windows users haven't had to worry about lack of useful documentation or easy configuration for years, so they can afford to obsess over eye candy more than poor Linux users who have enough trouble just getting their printing to work
More effort should be going to providing more usable and better documented GUIs, not making more eye candy. But this is open source, and people work on what they want to work on. And people don't want to work on making things usable. There's no fun in that. You can't post to Slashdot saying "hey! i was just up programming all night -- check out my new highly usable CUPS setup wizard". No one will care, because the folks reading Slashdot all got CUPS working ages ago. What do they care if it's now easier to do? But you can post "check out these super new transparent menus!" and all the 14-year olds will gasp in awe and amazement and call you a h4X0r 60D.
The thing that made me curious in the article is this quote:
Does the average internet user really visit 50 sites regularly? If so I'm way below average.
I regularly visit maybe 2 or 3 sites, so I really don't see any reason to invest time in getting RSS set up. I find it hard to believe that an average user really visits so many sites. It seems to me RSS is really more of a niche thing that's really great for people who like to overload themselves with information, rather than a super-hot technology that's about to explode into widespread popularity.
It can't be too long before more third party suppliers spring up who do nothing but build game engines and design tools.
There already are quite a few 3rd party engines from tools only companies. I'm not sure why Mr. Blow only mentioned the Quake and Unreal engines. There are other engines like
NetImmerse (or I guess it's called "Gamebryo" now), and, well, there are more links here. Go see for yourself.
I think The Unreal and Quake engines are actually some of the most expensive you can buy. From what I understand, you pay a premium for these engines because there's marketing value in being able to associate yourself with greatness by advertising "Built on the Unreal Engine". Whereas "Built on the LithTech engine" isn't going to impress anyone. So you pay a lot for those engines, when in fact, it's sort of akin to movies that advertise "From the producer of 'some good movie'". It's just not a very good predictor of whether it will or will not be a good movie.
If you can walk around the thing and the steroscopic pairs change accordingly, how is that different from reality?
The two main differences are latency and multi-viewer capability.
When you move around a true 3D image there is ZERO latency. You move your head back and forth, you always see the right view and it's perfectly in sync with your viewpoint. Any system that has to track your head, and then generate a stereoscopic pair based on that tracking result is going to have some latency. The result is that the image seems to swim a little bit. And it doesn't take much latency to make many people get a form of motion sickness. (Consider with 60fps display you have generally at least 16 msec of latency, and trackers usually pile on at least another 10-30msec or so, at best. That's plenty to induce motion sickness in many people, and in those that don't feel sick, at least it is enough that the swimming of the image is obvious when you make quick movements of your head.
Second, with stereo pairs, only one person can get the correct 3D view at a time. So it kind of cuts down on the potential for use in a group setting. Not to mention that you have to track the viewer and/or wear special eye gear. That cuts down on the potential uses also. You can't, say, have a 3D kiosk that people can just idly walk by and be wowed by if they have to line up one by one and put on some kind of tracking head gear to see the effect.
So there are a number of real reasons why you'd want to have a real 3D image-generation device instead of a device that's merely stereoscopic. If you just want to sit in front of your monitor and appreciate 3D porn, then there's nothing wrong with stereoscopic images, and at $150 or so (compared to $50K) the price is certainly right.
No, the Actuality display (Perspecta) uses projectors too. But the difference is that in Perspecta parts of the optics (some mirrors) move with the screen. Effectively it's like the projector is whizzing around as fast as the screen is, so that the projector is always at the perfect angle to project on the screen.
With the Hitachi display, there appear to be 24 fixed optical paths rather than one continuously varying path that tracks with the screen. So most of the time the Hitachi's spinning screen will be projected onto by a projector that's at a slightly wrong angle. As the screen sweeps through the 15 degree arc that corresponds to one optical path, the image will be first have a slight perspective/keystone distortion in one direction, then it will be correct for one slight fraction of a section, then it will proceed to become keystoned in the opposite direction. The result: smeary 3D images, and more smeary the further you get from the central axis of the volume.
That is unfortunate for this kind of display because the central axis is bad for viewing to begin with. The center axis is always dim because a significant fraction of the sweep time you're looking at the screen nearly edge-on, so the optimal part of the viewing volume is the off-axis part. But with Hitachi's display it seems like you will have dimness in the center axis, and smearing off axis. No sweet spots at all.
But I could be wrong, the Hitachi folk could possibly be pre-warping the images to correct for the keystoning as the screen sweeps past, but I doubt it, based on the video, and based on the update rate required of the projector to do that. Maybe it's on their to-do list, though.
I just wonder how they're going to avoid all of Actuality's patents.
I think that only projects a 2D image into 3D space. It looks like they are basically doing the trick of projecting on a thin laminar stream of fog. Although they don't call it fog, they call it "transformed air" because that sounds a lot cooler and more mysterious.
It does look pretty neat. But it's not 3D imagery.
It does look similar to Actuality's system. But it seems to have much lower resolution. Take a look at the video someone posted in this message.
:-)
It's really really smeary, almost to the point that the subject is unrecognizable.
From the spec sheet you can see Actuality's display does 198 slices of the volume compared to Hitachi's 24, and each slice is 768x768 resolution, compared to whatever Hitachi does. Just guessing, but assuming they Hitachi splits one projector frame up into 24 subframes (which it looks like they do because the schematics seem to indicate fixed optics), and generously assuming no wasted pixels, that comes to something like 213x256 resoultion per view, assuming they start with a 1280x1024 projector. So the frame resoulution is also a good bit lower than Actuality's.
Also looking at the vid of the Hitachi, and how smeary the images are, it almost makes me think they are projecting ALL 24 images ALL the time rather than blanking all but the two projecting most perpendicular to the screen. Or maybe it's smeary because they're using the same image for 15 degree chunks (360/24), compared to Actuality's 1.8 degrees (360/198). Or it could just be an artifact introduced by the video camera.
The other big difference is you can actually buy a display from Actuality today -- if you have $39,995.
I think he's talking about saccades.
How did you find it?
:-) Search for "non-commercial".
The search box.
I think you're right, though, they have basically removed all the direct links to that page. I certainly couldn't find one anywhere. Maybe they don't even mean for it to be findable via search.
Is GTK for Win32 actually viable now? All the rumors I've heard up till now is that it doesn't really work quite right on Windows yet. ... Oh and yeh, I can't stand writing GUIs in C.
Well anyway there's no one-size-fits-all toolkit. You've gotta pick what's right for your requirements. They all have pros and cons. FOX has a good mix of what I want: flexibility, ease of coding, cross-platform support, attractive GUI, OpenGL support, and freeness. It doesn't have quite the wide support of advanced features that wxWindows or Qt have, but I find it easier to work with than wxWindows, and it's definitely got fewer licensing issues than Qt.
Then what do you call this ?
It's not libre free, but beer free. And it's old as hell. But the poster was correct that there is a "non-commercial" version for windows that's freely available. Be warned though that Trolltech's definition of "non-commercial" is rather peculiar.
I'm wondering if they'll make this new non-comm 3.2 edition similarly available.
In the meantime I continue to happily use FOX.
The old 2.3 non-com version for windows is still on Trolltech.com. Just search for "non-commercial" in the little search box at the top of the page,
And you'll get this link as the top hit.
That is "non-commercial", though, and not "free". Which means according to Trolltech you can't use it for "any activities for which you are compensated," and that covers academic research too.
So is Non-commercial Windows Qt 3.2 going to be available for free download eventually, too (like the old one is)?
Right on brutha!
I don't think they've touched the code in over a year. Winamp is dead. Kiss it goodbye. The "new! Winamp 3!" button on the web page is just a joke now.
The "new and improved" Winamp3 really sucked the llama's cock. The video features are crap, all the neat plugins that used to work with winamp2 broke, and display of non-Roman character sets (the only thing I really felt lacking in Winamp2) was still not implemented, and actually was made more difficult to get working by their great bitmap GUI and fonts idea. The big new feature out of the year or so of development on Winamp3 was, what, a UI that could be scripted and bitmapped? Whee. Oh boy. Look -- we just want to access our music with minimal hassle, people. Most of us couldn't care less whether the buttons on the UI get jiggy on mouseover, or whether we can make it look like a 6-year drew it, as long as it can play the music and display the song titles.
I'll just stick with iTunes for Windows, thanks, which at least can display song titles properly, even if they're Japanese. And which integrates very nicely with my iPod as a bonus.
People have mostly been harping on how baseless this guy's criticisms of Qt are. But it's not just Qt where the facts are wrong. The whole thing is a pile of uninformed troll droppings. He criticizes FOX for requiring the FX_IMPLEMENT macro, but completely ignores the very similar macros that wxWindows uses, like DECLARE_EVENT_TABLE()and IMPLEMENT_APP(), and Qt's QOBJECT macro. Then there's a criticism of FOX requiring you to fiddle with message id enums, but this is also applicable to wxWindows. It's just a side effect of using enums and message tables for identifying messages.
Then there's the RAD Y/N column. FLTK's FLUID is a very basic code generator for FLTK dialogs. It's got nowhere near the functionality (or usability) of Qt's QDesigner, or from what I can see GTK's GLADE. FLUID hardly qualifies as a RAD tool.
Anyway, this article is a very shallow analysis of GUI toolkits that seems to be culled mostly from reading the web pages of the projects involved. This guy doesn't seem to have spent any significant amount of time actually *using* these toolkits, or talking to people who have.
Just to get the facts right, the 2.3.0 version of Qt for Win32 is most definitely NOT GPL. The source code is not available and the use of the binaries is restricted to "non-commercial" activities. Incidentally this apparently also excludes the use of that version of Qt for academic research, since Trolltech considers research to be "commercial activity". See the FAQ.
Heck, with all those PCs lined up on the floor it I bet it sounds pretty much like an airplane engine in there even without speakers.
IAOAP (i am only a programmer) but "highly repetitive filler/rubbish/crap sounds" sounds like a pretty good description of a lot of C code I've seen. Especially that written by higher organisms such as math and engineering Ph.D.'s.
Echorun, this is God. Stop playing with yourself.
Here's a quote from a review on xbitlabs
Check out the hair on this NVIDIA shot
and compare with the hair in this shot off the ATI.
The zoom on the NV example isn't quite as close-up, but you can still see how the hair feathers away at the tips, while the hair on the ATI is all jaggy and uniform thickness right out to the tips. Like some kind of Raggedy Anne yarn hair.
So does the ATI not support alpha-to-coverage? Or did these guys making the wrapper just not translate it properly? The hair looks pretty bad without it.
I seem to remember when I saw the NV guys give the demo that they made a point of how expensive the hair was to render. Each hair is a separate GL_LINE_STRIP, there are thousands of them, and if you turn off the complicated blending and smoothing on each one of those little hairs, that could maybe add up to a large savings. I don't know, but I wonder if that would account for most of the speed-up they see running it on ATI.
Still a neat hack. It would really rock to have complete NV<->ATI compatibility dlls that would work for all apps and not just this demo.
While they may have made some slight improvement to one aspect of image quality by improving normalization, which I guess makes the lighting a little more accurate, I really doubt the improvement is all that noticeable. Maybe I could tell given a side-by-side comparison, but I doubt I would notice if only shown a version normalized one way or the other. On the other hand, that ugly hair is pretty obvious. To me that makes the NVIDIA sreen shots look better.
That's too bad, because 9 out of 10 times when I've had troubles with templates it's because of differences between C++ implementations. Beautiful, well thought out, intricate standards-compliant examples are useless if I can't actually get them to compile with my real-world compiler!
The book I'm looking for is the one that gives me real-world recipes for getting around bone-headded compilers. For example, there are at least 3 different ways to declare templatized friend functions depending on the compiler. Only one is correct according to the standard, but the standard isn't worth a whole lot to me today if the compilers I'm stuck using don't follow it. And likewise, an advanced templates book isn't of much use to me today if the examples won't compile on my compilers.
Read the article. They've made it possible to cram 2 cards in one slot by making a half-width card. Standard size cards aren't being eliminated.
I saw that happening on a machine with Hyperthreading turned on in the bios. Problem went away after we turned hyperthreading off.
But the 2 minute rule gets in the way of real back and forth discussion too often, and also makes it a pain to post corrections if you realize a mistake just after submitting.
A less restrictive, but equivalently troll-subduing rule would be to allow, say, only 4 posts per 8 minutes. You could post 'em all in the first minute if you want, but then you've got to wait out 7. Trolls would still only be able to get in 720 posts/day, but average users shouldn't ever notice the limit.