temporal:/cdrom/video_ts# cat vts_01_1.vob cat: vts_01_1.vob: Input/output error temporal:/cdrom/video_ts# well fuck
Like I said, no encrypted data for you. I can read everything else on the disk just fine, but the videos are not available until the drive has been unlocked. Try it yourself. Pop in The Matrix, and try to cat any of the vobs in video_ts. If you get anything other than a plain old Access-fucking-DENIED (aka: I/O error, read error, etc.), then I guess I'm just stupid.
Once again, someone has replied to me by incorrectly telling me that I'm wrong, and been moderated up for it. Damn, not being part of the collective sucks. At least I posted without my +1, and managed to avoid getting modded down for being right.
Sounds nice, but what are the performance costs? Seems to me that even detecting circularly linked stuff would be expensive, but then I don't know too much about the details of how that is done, so I could be wrong.
Right. Having a smart pointer class in C++ gives you the advantages without the disadvantages. If you want to allocate on the stack, you just don't use the smart pointer wrapper. (I don't like Java either.:)
Really, a C++ smart pointer class is a trivial thing to write and gives you all the same advantages. I wrote one in 100 lines a few weeks ago and it works just as well as Java's garbage collection. The article seems to suggest that this is one of C#'s greatest features. Hmm, kinda like W2k having that "symbolic linking" thing.
Err, just so there is no confusion here... Most hardware that can access DVDs will not allow you to even read the *encrypted* data unless you unlock it first. Try it with linux and your DVD-ROM drive. I'd imagine that hardware companies agree to do that in order to get a license to use DVD technology... and I'd imagine that means that you would not be able to copy a DVD without DeCSS. Sorry. (I hope I'm wrong)
Hey, I am a proud owner of one of those t-shirts with the DeCSS source code on the back, so don't get me wrong or anything. Die, MPAA.
------
Re:Can someone give 1 good reason to use C++ over
on
Who's Afraid Of C++?
·
· Score: 2
There are tons of reasons. Here's an interesting one:
In the program I am writing, dynamically allocated data deletes itself when you are done with it, the same way java does. You just have to use my smart pointer class in the place of regular pointers. They work exactly the same syntactically, but they keep a reference count for whatever they are pointing to and free it once nothing is pointing to it anymore. No more memory leaks!
Now, this is not ideal for all situations, but it is something you just can't do easily and transparently in C. In my project, it has been incredibly useful, as alot of the code involves distributing and re-distributing resources to many distant places, and it is not always clear who's job it is to free some memory. But with smart pointers, it is not a problem.
Also, when using smart arrays, I can stick in bounds-checking for debug builds, which is pretty useful.
Several cheater bots were entered in the last tournament. They were disqualified, of course, but here are the funniest ones:
Fork Bot: Every move, this bot would fork itself into 3 processes and make a different move in each one. Any process that lost would be killed off in the next round, with the winning process continuing the tournament. Thus, you would think that it would never lose. However, when playing against the Psychic Friends Network, all three moves resulted in a loss, causing the Fork Bot to kill itself off, ending in a forfeit.
The Psychic Friends Network: This program won 998/1000 rounds against any opponent other than The Matrix. No one really knows how it works, being incredibly obfusicated, but it appears to mess with the stack directly, among other things.
The Matrix: Based on the simple premise "There is no spoon", this program won every single round of every match it was in. (Being written by the author of the tournament software, this was not very hard)
A smart program will NOT start out random
on
Rock-Paper-Scissors
·
· Score: 2
A smart program will use all kinds of techniques to figure out its opponent's prediction algorithm. I would imagine that such a program would start out with some carefully planned tricks to get the other program to reveal itself.
Of course, I don't know for sure, since I have not tried it. But, if you look at the results from the last tournament, you see that one program, called "Iocaine Powder", won every single round my a significant margine. The second-place program was based on an earlier version of Iocaine Powder, and was as far ahead of third place as it was behind first. Clearly, this is more complex than it sounds.:) Perfect name for the winner, too.
Doesn't the portability of such a program depend alot more on the GUI toolkit used? I mean, the level of back-end API that an office program would need is pretty trivial (and mostly provided by ANSI-C anyway). It's the front-end that counts. AFAIK, OSX uses its own GUI stuff, not X11 or gtk.
Err... did you bother to read the link he gave to magram? No? Well, apparently, magram is faster than DRAM, non-volotile, and not much more expensive that hard drive space... or something... read the article. What konstant proposed is perfectly plausable.
The official AOL documentation for TOC is included in the GAIM distrobution. I doubt that they are just re-releasing that, but I suppose it is possible.
Several days ago, the TOC server went down for some time. I re-compiled GAIM to use the "experimental" OSCAR support, and I've been online ever since.
Ah, the joys of open source software. My GAIM conversation windows have a big toggle button on the bottom marked "Sveedish Cheff". Turn it on, and everything I say gets converted before being sent. Bork Bork Bork!
Granted they may not have the best Linux drivers in the world (so I hear), but no company is perfect.
In fact, they DO have the best Linux drivers in the world. According to LinuxGames' benchmarks, nVidia's Linux drivers come closer to matching the performance of thier corresponding Windows drivers that any other 3D drivers available on Linux.
And 3dfx has launched plenty of FUD in the past. For instance, in the FAQ for the Bashee, one of the questions was "With your competitors supporting textures 2k x 2k in size, don't you think that your 256x256 limit will hurt your products?" The answer, "Because we are the market leaders in 3D graphics hardware, developers will write their games with our hardware in mind. So, they will not exceed the 256x256 texture limit anyway."
800MPixel with two texels per pipeline. Since almost every game uses multitexturing, that means 1600MTexels per second, compared to the V5-6000 1200.
Quake 3 is a T&L game, and depending on the map you play, T&L can make a big difference. Many more T&L games will be out in a few months. And I'm a game developer, so I sure as hell am seeing serious improvement NOW on my GF2 with T&L. Thank you.
Err, apparently you did not read yesterday's article on driver performance. The current drivers from nVidia perfrom at betweed 85% and 99% of the speed of their Windows counterparts, whereas the 3dfx drivers perform at 75% - 84%. Please go and download the XFree 4.0 drivers from nVidia.
The TNT2 is most definately faster than the banshee, especially in Linux.
Rendering at a higher resolution and interpolating down would be the same speed as 3dfx's 4xFSAA. 3dfx's implementation cuts performance quite a bit as well.
I find it sort of odd that nVidia's method, as you describe it, does not produce the exact same image. It is essentially the same operation. 3dfx renders the same frame 4 times, with pixel offsets of (0,0), (0,0.5), (0.5,0), and (0.5,0.5). After averaging, that should look the same as just rendering at a higher res and then averaging every 4 pixels into one, right? What am I missing here? Can you point me to a comparison, preferably with screenshots?
Perhaps it is a matter of opinion, but I think that the high polygon counts provided by the GeForce and GF2's T&L engines look much better than FSAA. example
Here is an article which compares the raw power of the cards. As you can see, the V5 5500 has nothing over the GeForce 2, but the V5 6000 might have a chance, but only when T&L is not a factor.
BTW, "128MB" on a V5 6000 is no better than 32 on a GeForce 2 due to the multiprocessor design. But the GF2 can have up to 128MB, which would be like a V5 with 512MB on-board. heh.
I wonder how much of this was a quality concern, and how much was the sudden realization that the nVidia GeForce 2 is faster, cheaper, available NOW,and doesn't require a freeking AC adapter to be plugged into its rear-end?
From what I've read, the V5 6000 is the only card from 3dfx that has any chance of beating the GF2, but only in a few select situations, such as running older games at super-high-res with 4xFSAA. And then it is only a little bit faster. (yes, the GeForce 2 does FSAA.) Add to that the fact that the 6000 will cost US$600 (when it finally comes out) as opposed to the GF2's current price of $300, and you have a sorry situation indeed...
Linux's default sound driver allows direct access. mmap("dev/dsp"). There is no need for DirectX.
OK, imagine Windows is a thick forest. You are a programmer, and somewhere in this thick forest is your destination. There are several winding dirt pathways that you could take to get there, but you obviously can't go very fast on such pathways. So, the government of this forest (Microsoft) realizes that they need to improve the trasportation around here. So, they build a bunch of multi-lane super-highways that are perfectly stait and had no speed limit (DirectX). Now you can get where you need to go fast and efficiently. Everyone is happy.
What about Linux? Well, with Linux, there is no forest, and you don't have to go anywhere. You are already there.
Disclaimer: This basically applies to everything except for X. To fix the X problem, we have the new direct-rendering stuff. As you can see, nVidia's beta Linux drivers were able to almost match their Windows' couterparts' performance, so it is clearly quite possible for Linux to match Windows' speed. For everything non-graphical, Linux does not need a "DirectX" because it already gives you direct access to the hardware.
In the article, they complained about the nVidia drivers being unstable, and they suggested that this bug would be fixed if the drivers were open. But then they said that they were not using the latest drivers due to "issues" with Q3A.
I am using the latest drivers. There are no "issues". They work perfectly. What's more, the instability they mentioned WAS FIXED (mostly) in the newer drivers. Supposedly, there is still an occasional crash, but I have tried to make it happen with no success.
Can I interpret this to mean that a large part of an nVidia card's performance comes from the software in the driver?
No. In fact, nVidia does have the best and fastest x86-based graphics hardware available today. The thing is, 3D graphics is extremely broad. There are an almost infinite number of features that could be implemented in hardware. The specification for OpenGL is several hundred pages thick, despite being almost too brief to understand. No hardware ever made implements every OpenGL call on-board. However, the hardware vendor must supply a complete OpenGL implementation. Otherwise, the application programmer would have to check every feature to make sure it is supported, and implement it themselves if not.
Of 3dfx, ATI, Matrox, and nVidia, nVidia is the ONLY company that has written a complete implementation of OpenGL for their cards (I don't know about S3, but I'd guess not). Their hardware accellerates more OpenGL calls than anyone else's, but it is still far from doing everything, and they have to support their older hardware, which does alot less. Thus, they have to write software to do the rest. Obviously, this software would be very useful to other graphics card companies wanting to write drivers for their cards.
On Linux, OpenGL is much better. 3dfx, Matrox, and ATI (I think) have based their Linux drivers off of Mesa, which is a more or less complete OpenGL implementation. However, I am programming a 3D game engine (GPL'd), and I see a distinct difference between running it on Mesa-based drivers and running it on nVidia's new drivers. The nVidia drivers look much better, especially with lighting. I would *like* to see them release their souce code so that everyone would get these superior drivers, but I don't expect them to do it because then customers of other companies would get the advantage that only nVidia customers have now (this is how capitolism works, BTW).
You may be wondering why I am so eager to support nVidia. Well, like I said, I am writing a 3d game engine. nVidia has traditionally gone to great lengths to support new features in their cards even when games don't actually use them yet. Developers love them for this, because nVidia is rapidly expanding the amount of cool stuff that can be done in games. But if people go out and buy, say, a 3dfx Voodoo 5, it is like taking a step back. If too many people end up using V5's, then I won't be able to make games that use these advanced features, because no one with a V5's will be able to use them, and they will all blame ME for their buying mistake.
So, the choice seems to be between nVidia's high image quality and high driver quality and everyone else's medium image quality, medium to mediocre driver quality, and open source code. (John Carmack himself said that he believes nVidia's drivers are better than anything he and UtahGLX have come up with.)
Go ahead, take the mediocre solution with the open drivers. But don't complain to ME when the software I write doesn't work on it.
Slashdot Mirror
Err, YES.
temporal:/cdrom/video_ts# cat vts_01_1.vob
cat: vts_01_1.vob: Input/output error
temporal:/cdrom/video_ts# well fuck
Like I said, no encrypted data for you. I can read everything else on the disk just fine, but the videos are not available until the drive has been unlocked. Try it yourself. Pop in The Matrix, and try to cat any of the vobs in video_ts. If you get anything other than a plain old Access-fucking-DENIED (aka: I/O error, read error, etc.), then I guess I'm just stupid.
Once again, someone has replied to me by incorrectly telling me that I'm wrong, and been moderated up for it. Damn, not being part of the collective sucks. At least I posted without my +1, and managed to avoid getting modded down for being right.
------
Sounds nice, but what are the performance costs? Seems to me that even detecting circularly linked stuff would be expensive, but then I don't know too much about the details of how that is done, so I could be wrong.
------
Right. Having a smart pointer class in C++ gives you the advantages without the disadvantages. If you want to allocate on the stack, you just don't use the smart pointer wrapper. (I don't like Java either. :)
------
Really, a C++ smart pointer class is a trivial thing to write and gives you all the same advantages. I wrote one in 100 lines a few weeks ago and it works just as well as Java's garbage collection. The article seems to suggest that this is one of C#'s greatest features. Hmm, kinda like W2k having that "symbolic linking" thing.
------
Err, just so there is no confusion here... Most hardware that can access DVDs will not allow you to even read the *encrypted* data unless you unlock it first. Try it with linux and your DVD-ROM drive. I'd imagine that hardware companies agree to do that in order to get a license to use DVD technology... and I'd imagine that means that you would not be able to copy a DVD without DeCSS. Sorry. (I hope I'm wrong)
Hey, I am a proud owner of one of those t-shirts with the DeCSS source code on the back, so don't get me wrong or anything. Die, MPAA.
------
There are tons of reasons. Here's an interesting one:
In the program I am writing, dynamically allocated data deletes itself when you are done with it, the same way java does. You just have to use my smart pointer class in the place of regular pointers. They work exactly the same syntactically, but they keep a reference count for whatever they are pointing to and free it once nothing is pointing to it anymore. No more memory leaks!
Now, this is not ideal for all situations, but it is something you just can't do easily and transparently in C. In my project, it has been incredibly useful, as alot of the code involves distributing and re-distributing resources to many distant places, and it is not always clear who's job it is to free some memory. But with smart pointers, it is not a problem.
Also, when using smart arrays, I can stick in bounds-checking for debug builds, which is pretty useful.
------
Several cheater bots were entered in the last tournament. They were disqualified, of course, but here are the funniest ones:
For more info, see this page (near the bottom).
------
A smart program will use all kinds of techniques to figure out its opponent's prediction algorithm. I would imagine that such a program would start out with some carefully planned tricks to get the other program to reveal itself.
Of course, I don't know for sure, since I have not tried it. But, if you look at the results from the last tournament, you see that one program, called "Iocaine Powder", won every single round my a significant margine. The second-place program was based on an earlier version of Iocaine Powder, and was as far ahead of third place as it was behind first. Clearly, this is more complex than it sounds. :) Perfect name for the winner, too.
------
Doesn't the portability of such a program depend alot more on the GUI toolkit used? I mean, the level of back-end API that an office program would need is pretty trivial (and mostly provided by ANSI-C anyway). It's the front-end that counts. AFAIK, OSX uses its own GUI stuff, not X11 or gtk.
------
Err... did you bother to read the link he gave to magram? No? Well, apparently, magram is faster than DRAM, non-volotile, and not much more expensive that hard drive space... or something... read the article. What konstant proposed is perfectly plausable.
------
The official AOL documentation for TOC is included in the GAIM distrobution. I doubt that they are just re-releasing that, but I suppose it is possible.
Several days ago, the TOC server went down for some time. I re-compiled GAIM to use the "experimental" OSCAR support, and I've been online ever since.
Ah, the joys of open source software. My GAIM conversation windows have a big toggle button on the bottom marked "Sveedish Cheff". Turn it on, and everything I say gets converted before being sent. Bork Bork Bork!
------
Oh, yes, I understand. I'd much rather have them take their time than rush and create a lower-quality product. We can still joke about it, though. :)
------
Diablo 2 goes gold
Oh, come on! It's not even April anymore!
------
Granted they may not have the best Linux drivers in the world (so I hear), but no company is perfect.
In fact, they DO have the best Linux drivers in the world. According to LinuxGames' benchmarks, nVidia's Linux drivers come closer to matching the performance of thier corresponding Windows drivers that any other 3D drivers available on Linux.
And 3dfx has launched plenty of FUD in the past. For instance, in the FAQ for the Bashee, one of the questions was "With your competitors supporting textures 2k x 2k in size, don't you think that your 256x256 limit will hurt your products?" The answer, "Because we are the market leaders in 3D graphics hardware, developers will write their games with our hardware in mind. So, they will not exceed the 256x256 texture limit anyway."
------
800MPixel with two texels per pipeline. Since almost every game uses multitexturing, that means 1600MTexels per second, compared to the V5-6000 1200.
Quake 3 is a T&L game, and depending on the map you play, T&L can make a big difference. Many more T&L games will be out in a few months. And I'm a game developer, so I sure as hell am seeing serious improvement NOW on my GF2 with T&L. Thank you.
------
Err, apparently you did not read yesterday's article on driver performance. The current drivers from nVidia perfrom at betweed 85% and 99% of the speed of their Windows counterparts, whereas the 3dfx drivers perform at 75% - 84%. Please go and download the XFree 4.0 drivers from nVidia.
The TNT2 is most definately faster than the banshee, especially in Linux.
------
Rendering at a higher resolution and interpolating down would be the same speed as 3dfx's 4xFSAA. 3dfx's implementation cuts performance quite a bit as well.
I find it sort of odd that nVidia's method, as you describe it, does not produce the exact same image. It is essentially the same operation. 3dfx renders the same frame 4 times, with pixel offsets of (0,0), (0,0.5), (0.5,0), and (0.5,0.5). After averaging, that should look the same as just rendering at a higher res and then averaging every 4 pixels into one, right? What am I missing here? Can you point me to a comparison, preferably with screenshots?
------
Perhaps it is a matter of opinion, but I think that the high polygon counts provided by the GeForce and GF2's T&L engines look much better than FSAA. example
The GF2's pixel shaders are pretty nice as well.
------
BTW, "128MB" on a V5 6000 is no better than 32 on a GeForce 2 due to the multiprocessor design. But the GF2 can have up to 128MB, which would be like a V5 with 512MB on-board. heh.
------
Hell, they don't even have FSAA over nVidia now. The GeForce and GeForce 2 can do FSAA with the 5.x drivers.
------
I wonder how much of this was a quality concern, and how much was the sudden realization that the nVidia GeForce 2 is faster, cheaper, available NOW,and doesn't require a freeking AC adapter to be plugged into its rear-end?
From what I've read, the V5 6000 is the only card from 3dfx that has any chance of beating the GF2, but only in a few select situations, such as running older games at super-high-res with 4xFSAA. And then it is only a little bit faster. (yes, the GeForce 2 does FSAA.) Add to that the fact that the 6000 will cost US$600 (when it finally comes out) as opposed to the GF2's current price of $300, and you have a sorry situation indeed...
------
Well, for one thing, make sure you do have that 'Load "glx"' line in XF86Config. (I think I made that mistake, and got that error)
------
Linux's default sound driver allows direct access. mmap("dev/dsp"). There is no need for DirectX.
OK, imagine Windows is a thick forest. You are a programmer, and somewhere in this thick forest is your destination. There are several winding dirt pathways that you could take to get there, but you obviously can't go very fast on such pathways. So, the government of this forest (Microsoft) realizes that they need to improve the trasportation around here. So, they build a bunch of multi-lane super-highways that are perfectly stait and had no speed limit (DirectX). Now you can get where you need to go fast and efficiently. Everyone is happy.
What about Linux? Well, with Linux, there is no forest, and you don't have to go anywhere. You are already there.
Disclaimer: This basically applies to everything except for X. To fix the X problem, we have the new direct-rendering stuff. As you can see, nVidia's beta Linux drivers were able to almost match their Windows' couterparts' performance, so it is clearly quite possible for Linux to match Windows' speed. For everything non-graphical, Linux does not need a "DirectX" because it already gives you direct access to the hardware.
------
In the article, they complained about the nVidia drivers being unstable, and they suggested that this bug would be fixed if the drivers were open. But then they said that they were not using the latest drivers due to "issues" with Q3A.
I am using the latest drivers. There are no "issues". They work perfectly. What's more, the instability they mentioned WAS FIXED (mostly) in the newer drivers. Supposedly, there is still an occasional crash, but I have tried to make it happen with no success.
------
Can I interpret this to mean that a large part of an nVidia card's performance comes from the software in the driver?
No. In fact, nVidia does have the best and fastest x86-based graphics hardware available today. The thing is, 3D graphics is extremely broad. There are an almost infinite number of features that could be implemented in hardware. The specification for OpenGL is several hundred pages thick, despite being almost too brief to understand. No hardware ever made implements every OpenGL call on-board. However, the hardware vendor must supply a complete OpenGL implementation. Otherwise, the application programmer would have to check every feature to make sure it is supported, and implement it themselves if not.
Of 3dfx, ATI, Matrox, and nVidia, nVidia is the ONLY company that has written a complete implementation of OpenGL for their cards (I don't know about S3, but I'd guess not). Their hardware accellerates more OpenGL calls than anyone else's, but it is still far from doing everything, and they have to support their older hardware, which does alot less. Thus, they have to write software to do the rest. Obviously, this software would be very useful to other graphics card companies wanting to write drivers for their cards.
On Linux, OpenGL is much better. 3dfx, Matrox, and ATI (I think) have based their Linux drivers off of Mesa, which is a more or less complete OpenGL implementation. However, I am programming a 3D game engine (GPL'd), and I see a distinct difference between running it on Mesa-based drivers and running it on nVidia's new drivers. The nVidia drivers look much better, especially with lighting. I would *like* to see them release their souce code so that everyone would get these superior drivers, but I don't expect them to do it because then customers of other companies would get the advantage that only nVidia customers have now (this is how capitolism works, BTW).
You may be wondering why I am so eager to support nVidia. Well, like I said, I am writing a 3d game engine. nVidia has traditionally gone to great lengths to support new features in their cards even when games don't actually use them yet. Developers love them for this, because nVidia is rapidly expanding the amount of cool stuff that can be done in games. But if people go out and buy, say, a 3dfx Voodoo 5, it is like taking a step back. If too many people end up using V5's, then I won't be able to make games that use these advanced features, because no one with a V5's will be able to use them, and they will all blame ME for their buying mistake.
So, the choice seems to be between nVidia's high image quality and high driver quality and everyone else's medium image quality, medium to mediocre driver quality, and open source code. (John Carmack himself said that he believes nVidia's drivers are better than anything he and UtahGLX have come up with.)
Go ahead, take the mediocre solution with the open drivers. But don't complain to ME when the software I write doesn't work on it.
------