Pixar: Good movies, suck-ass company
on
A Tour of Pixar
·
· Score: 2, Interesting
Pixar is the Microsoft of the computer graphics world. They have created some good stuff, yes, but they have not given much of anything back and often hurt the field.
They do things like produce a suposedly open standard like Renderman, then sue anyone who uses it into oblivion. Most small projects get by without them batting an eye, but if competition rises up they are quick to lay the smack down.
They have all these secrets, and keep everything to themselves. Just read some of their licensing agreements (read the agreement for the "open" Renderman spec.. haha). And they provide no public inovation in the field. Rather than help the community at large, they stifle it.
Have you ever seen Pixar release anything like Massive (Weta)? I didn't think so.
Again, they usually produce good stuff, but they also are incredibly self-serving. I just have a bad feeling about the company. They're scary in a Microsoft and Disney (duh) way.
This is just another one of those "team training" ideas that was popular in the 90's.
I can't count how many of these dumbass ideas I've had to live through and how many team building sessons, getaways, and classes I've endured. It's a total waste of money and employees time.
It doesn't work, it didn't do a damn thing to change anything. Good companies with good people were always successful even with no special team crap. Bad companies with pathetic management and/or moronic employees always failed, no matter what "team building" exercises they put us through.
I also have had this drive for about 6 months and have had no problems at all and would recommend it to anyone. I have the external (firewire/usb) version.
A great drive that works really good with the Linux DVD tools.
I know of quite a few others that have had very good success with these drives.
I never said Linux wasn't ready for the desktop, I just said that it needs some work to catch up with Windows. I never mentioned the install process either.
It's not the speed of my machine (2.2 Ghz, 1GB RAM). But would that matter? I'm comparing Windows and Linux on the same machine.
I disagree with Microsoft's licensing scheme and Microsoft's business practices.
However, the main reason is security. I'm a consultant and I have to plug into different networks all the time. Linux provides me with more control than Windows can. Sure, there are firewall products for Windows but I've never been convinced those truely work (eg. there have been exploits that go right past some Windows firewalls).
I'll start off by saying I use Linux (Debian) as my primary OS, and usually only run Windows in VMware.
However, Windows still feels better than anything on Linux (KDE 3.1, GNOME 2, or whatever). Windows is more consistant (all the widgets look the same), cut/copy/paste works better, and the UI just feels snappy and nice (file manager, desktop, start menu). Contrast that with Linux. The UI is all over the place, KDE is extremely slow to start up and just generally feels somewhat funky, the file managers are all slow and tend to crash a lot or just not work, there are not many good GUI administration tools. I could go on. Please note that I am way past the initial "neato" Linux phases. I've been using Linux since the pre-1.0 days. A lot of people who say Linux is so much better, faster, or whatever are actually just excited about using something different. Linux is getting better, but there is quite a ways to go before it's better than Windows.
Really, the Linux and *BSD communities should take note of Mac OS X. That is UNIX with a nice (although slow) user experience. Now, Apple has it easy because the hardware configurations is extremely limited, but the friendlyness could be copied. I'm not talking about copying OS X in the strict sense, I'm just saying that OS X is an example of something complicated and text based (UNIX) that was made friendly.
(***cringe*** at the thought of implementing the FFT in VB).
It not really that bad. Implement it in C/C++ as an ActiveX control (or I guess the young'uns are using.NET these days). In fact, you don't even need to implement the whole thing as there's tons of FFT C/C++ source code out there.
I was wondering the exact same thing when I looked at the shots.
I also wonder what else isn't rendering properly on the ATI. There could be other things left out because they don't have the source to the demo. Who knows.
I would say that it's impossible to say which card is faster from using this demo.
Basically, I would like to make some minor updates to C that add minimal high-level modularity (similar to Erlang) and some generic programming capability (eg. a better preprocessor).
Our current computer hardware sucks, that's why
on
AI Going Nowhere?
·
· Score: 1
I mean, we can't even do "basic" stuff for computer/human interaction, let alone full blown AI.
Realtime voice recognition is a problem that has an easily verifiable solution. We have made a lot of progress but I've never seen a system that is usable like talking to another person. Note that I'm just talking about the speech recognition, not language processing which would be the next step (another area we still can't do very well).
Realtime image/face/whatever recognition falls into the same catagory of easily testable systems that we have yet to make anything even remotely close to what a human can do.
And there are many more areas like this. You don't need AI to do those things, but those things are the baby steps towards real AI. And we've barely scratched the surface of making those "simple" systems usable. Real AI is a ways off, for sure.
Duh, because Quake3 is one of the very few engines that scales well with faster processors. It pretty much runs as fast as your CPU and graphics can handle because it's written down at the bare metal level.
It's a more "raw" benchmark than the rest of the games out there. Other games often have many strange and bizarre bottlenecks due to the way there were written.
The Q3 benchmark is not about showing how fast the game is. It's about showing how fast the processor is compared to older versions.
LOL, I understand what you are saying, but that's not what I was talking about. I am speaking from a professional standpoint. I'm assuming the artist is talented and that isn't an issue. We are past that point and now talking tools.
Good tools help an artist and don't force you into a mold. They allow you to work freely, fast, and efficient. They allow you to really express yourself.
Yes, it's true a good artist can create good art no matter the tool, but the question is how quickly and easily.
This mimics the same debate about programming languages. A good programmer could create something like MSWord using binary 1's and 0's, but would you want to?
Each to his own, but I would argue that Mirai has a much better interface.
In fact, XSI has been slowly becoming more Mirai-like over the years (same with Maya). Mirai had it first though. Both Maya and XSI come from some really bad user interfaces (Alias|Wavefront, and the original Softimage 3D) and they have been improving the interface over the years. But Mirai comes from N-World which was arguably the first commercial-grade graphics and animation package, and quickly developed the interface that everyone else now realizes they should have had all along.
If you haven't used Mirai, then you're missing out. It's a shame the program is so hard to get though. No demo and it's been almost impossible to buy for the last few years (management problems in the company). It hardly gets any publicity because so many graphics shops use it as a "secret weapon".
At first, most old-time 3D artists don't realize how good Mirai is because they've been stuck in the land of crappy software for so long that they have a skewed view of what's good. Once you use it though, the world looks different.
Frankly, I don't think so. I think the limitation is all the crappy software out there. Maya isn't too bad, but the other software you mentioned have horrible interfaces.
The "talent" is learning to use some piece of shit user interface effectively. That seems like a waste of time to me. There is some software out there that is better than the most popular packages but they are not anywhere near perfect. The packages I'm speaking of are Mirai, Nendo, Wings3D, etc. They are much easier to get prefessional results without spending years learning the damn UI. However, Mirai is limited compared Maya and Nendo/Wings are currently only modelers (and they can't handle high poly models well).
It's just a sad state of affairs because there is no complete, powerful, fast, easy to use graphics and animation package (I'm currently working to fix this by the way:).
I haven't tried using the Knoppix hardware stuff with regular Debian. Would be nice but I just haven't gotten around to trying it.
You might try running Knoppix since it can run right off the CD and see if it's any good at detecting your particular hardware. It has worked really well on all the machines I've tried, including my accessory laden laptop.
And another thing, KDE 3.1 and newer kernels have been in the regular Debian unstable for quite a while now. I'm running KDE 3.1 under unstable right now in fact.
Uh, Knoppix has had that latest stuff (KDE 3.1, new kernel, etc.) for some time now.
In case you didn't know, Knoppix is Debian based and has some awesome hardware auto-detection utilities.
Re:See, the Internet is good for something
on
SARS and the Internet
·
· Score: 4, Interesting
Meh, they have no idea. The Internet is an extremely useful source of information. Any time I need to know something, I just head to the 'Net.
Here are some examples of what I've used the Internet for (besides all the programming and normal computer junk I do):
- I wanted to know how to make those hollow egg shell decorations for Easter. Looked it up on the 'Net.
- Someone asked me if I knew how to spell chlamydia (I could hardly guess how to spell it; thanks Google) and if it was a bacterial infection. Looked it up on the 'Net.
- I needed the blue book value of several cars I was helping a friend look for. Head to the 'Net.
- I needed some tax forms from the IRS and my state. They're all on the 'Net.
- Tons of DMV stuff can be done on the 'Net. license renewalls have never been so quick and easy.
- I needed some hummingbird pictures for a nephew who is learning to paint and needed some color samples.
- I do almost all my shopping online (especially for gifts). No more travelling 20 miles to find they don't even stock what I was looking for.
- I got a weird engine check code on my car and needed to know how to fix it. Online forums are great.
- I got more information on the Pony Express Rally that I plan on competing in next year.
- I wanted to set up an official sized vollyball court in my backyard and got the dimensions on the 'Net.
And that's just in the past few days/weeks. I could go on and on.
No, you have to look at "f encrypted with g" as a whole new algorithm h. It's a complete system. In a sense it's really just one algorithm (with more steps) not two.
Mathematically it's possible that h is weaker than either f or g alone. A very simplistic example:
- f encrypts by adding 5 to every binary value. - g encrypts by substracting 5 from every binary value.
Use the two together and, oops, no encryption. Of course a real algorithm is more complicated, but it's all just math.
I think you're basically agreeing with me in principal. That is that using different algorithms to encrypt multiple times is mostly safe.
However, your example is missing a possibility. It is completely possible that the interaction between algorithm f and g could produce a weaker result M". Of course it would be easier for anyone trying to break it to know which algorithms you used, but we have to accept that there is a chance they would have this information.
The attack would not be to try and break each algorithm separately (eg. break g, then to try and break f). Instead the attack would be based on breaking some weaker result of the two algorithms. You can look at "f encrypted with g" as a new algorithm "h" which is possibly weaker than f or g alone.
Not necessarily. Especially not when encrypting multiple times using the same algorithm. Read Bruce Schneier's "Applied Cryptography" book. Good stuff. He covers this question much better than I can answer here.
Even when using multiple different algorithms there is a chance of weaking the whole thing. Depends on which algorithms you're using and how you're using them. I think you are generally safe using different known-good algorithms though (say 3DES then AES). I would not encrypt multiple times with the same algorithm unless it has been mostly proven to be more secure.
What systems must you support?
on
SSH or IPSec?
·
· Score: 1
- SSH only tunnels specified ports which is a pain if you need to tunnel several services or complicated protocols. Plus it makes things difficult to configure if you're constantly switching between wireless and ethernet on the same system.
- PPPoE (or similar) over SSH can be incredibly unreliable. It's the TCP-over-TCP problem and you will sometimes see connections just freeze up.
- IPsec is a pain in the ass to configure. I've noticed most implementations are pretty crappy. I've seen similar problems that you get with PPPoE over SSH. Connections freezing and such.
- stunnel can't do UDP connections (so no Quake over wireless) among other problems.
I use OpenVPN. It uses UDP and tunnels all traffic like IPsec except it's super easy to configure. In my experience it has also been much more reliable than IPsec. It has several connection and server recovery mechanisms that work very well. I don't think there is a Windows client though so that might be a problem. I really like OpenVPN though.
Slashdot Mirror
Pixar is the Microsoft of the computer graphics world. They have created some good stuff, yes, but they have not given much of anything back and often hurt the field.
They do things like produce a suposedly open standard like Renderman, then sue anyone who uses it into oblivion. Most small projects get by without them batting an eye, but if competition rises up they are quick to lay the smack down.
They have all these secrets, and keep everything to themselves. Just read some of their licensing agreements (read the agreement for the "open" Renderman spec.. haha). And they provide no public inovation in the field. Rather than help the community at large, they stifle it.
Have you ever seen Pixar release anything like Massive (Weta)? I didn't think so.
Again, they usually produce good stuff, but they also are incredibly self-serving. I just have a bad feeling about the company. They're scary in a Microsoft and Disney (duh) way.
This is just another one of those "team training" ideas that was popular in the 90's.
I can't count how many of these dumbass ideas I've had to live through and how many team building sessons, getaways, and classes I've endured. It's a total waste of money and employees time.
It doesn't work, it didn't do a damn thing to change anything. Good companies with good people were always successful even with no special team crap. Bad companies with pathetic management and/or moronic employees always failed, no matter what "team building" exercises they put us through.
I also have had this drive for about 6 months and have had no problems at all and would recommend it to anyone. I have the external (firewire/usb) version.
A great drive that works really good with the Linux DVD tools.
I know of quite a few others that have had very good success with these drives.
I never said Linux wasn't ready for the desktop, I just said that it needs some work to catch up with Windows. I never mentioned the install process either.
It's not the speed of my machine (2.2 Ghz, 1GB RAM). But would that matter? I'm comparing Windows and Linux on the same machine.
I disagree with Microsoft's licensing scheme and Microsoft's business practices.
However, the main reason is security. I'm a consultant and I have to plug into different networks all the time. Linux provides me with more control than Windows can. Sure, there are firewall products for Windows but I've never been convinced those truely work (eg. there have been exploits that go right past some Windows firewalls).
I'll start off by saying I use Linux (Debian) as my primary OS, and usually only run Windows in VMware.
However, Windows still feels better than anything on Linux (KDE 3.1, GNOME 2, or whatever). Windows is more consistant (all the widgets look the same), cut/copy/paste works better, and the UI just feels snappy and nice (file manager, desktop, start menu). Contrast that with Linux. The UI is all over the place, KDE is extremely slow to start up and just generally feels somewhat funky, the file managers are all slow and tend to crash a lot or just not work, there are not many good GUI administration tools. I could go on. Please note that I am way past the initial "neato" Linux phases. I've been using Linux since the pre-1.0 days. A lot of people who say Linux is so much better, faster, or whatever are actually just excited about using something different. Linux is getting better, but there is quite a ways to go before it's better than Windows.
Really, the Linux and *BSD communities should take note of Mac OS X. That is UNIX with a nice (although slow) user experience. Now, Apple has it easy because the hardware configurations is extremely limited, but the friendlyness could be copied. I'm not talking about copying OS X in the strict sense, I'm just saying that OS X is an example of something complicated and text based (UNIX) that was made friendly.
Yes, thanks for the tips. I've been using Debian for years and probably don't even know but 1% of these nice features.
(***cringe*** at the thought of implementing the FFT in VB).
.NET these days). In fact, you don't even need to implement the whole thing as there's tons of FFT C/C++ source code out there.
It not really that bad. Implement it in C/C++ as an ActiveX control (or I guess the young'uns are using
Then use the control from VB... simple.
I was wondering the exact same thing when I looked at the shots.
I also wonder what else isn't rendering properly on the ATI. There could be other things left out because they don't have the source to the demo. Who knows.
I would say that it's impossible to say which card is faster from using this demo.
The $100 standard edition is the normal compiler without the optimizer (and a few other minor tools).
So your apps run slower. No pop-ups or distribution issues, that is specific to the academic/learning version.
I've posted on this before, here it is again.
Basically, I would like to make some minor updates to C that add minimal high-level modularity (similar to Erlang) and some generic programming capability (eg. a better preprocessor).
I mean, we can't even do "basic" stuff for computer/human interaction, let alone full blown AI.
Realtime voice recognition is a problem that has an easily verifiable solution. We have made a lot of progress but I've never seen a system that is usable like talking to another person. Note that I'm just talking about the speech recognition, not language processing which would be the next step (another area we still can't do very well).
Realtime image/face/whatever recognition falls into the same catagory of easily testable systems that we have yet to make anything even remotely close to what a human can do.
And there are many more areas like this. You don't need AI to do those things, but those things are the baby steps towards real AI. And we've barely scratched the surface of making those "simple" systems usable. Real AI is a ways off, for sure.
Duh, because Quake3 is one of the very few engines that scales well with faster processors. It pretty much runs as fast as your CPU and graphics can handle because it's written down at the bare metal level.
It's a more "raw" benchmark than the rest of the games out there. Other games often have many strange and bizarre bottlenecks due to the way there were written.
The Q3 benchmark is not about showing how fast the game is. It's about showing how fast the processor is compared to older versions.
LOL, I understand what you are saying, but that's not what I was talking about. I am speaking from a professional standpoint. I'm assuming the artist is talented and that isn't an issue. We are past that point and now talking tools.
Good tools help an artist and don't force you into a mold. They allow you to work freely, fast, and efficient. They allow you to really express yourself.
Yes, it's true a good artist can create good art no matter the tool, but the question is how quickly and easily.
This mimics the same debate about programming languages. A good programmer could create something like MSWord using binary 1's and 0's, but would you want to?
Each to his own, but I would argue that Mirai has a much better interface.
In fact, XSI has been slowly becoming more Mirai-like over the years (same with Maya). Mirai had it first though. Both Maya and XSI come from some really bad user interfaces (Alias|Wavefront, and the original Softimage 3D) and they have been improving the interface over the years. But Mirai comes from N-World which was arguably the first commercial-grade graphics and animation package, and quickly developed the interface that everyone else now realizes they should have had all along.
If you haven't used Mirai, then you're missing out. It's a shame the program is so hard to get though. No demo and it's been almost impossible to buy for the last few years (management problems in the company). It hardly gets any publicity because so many graphics shops use it as a "secret weapon".
At first, most old-time 3D artists don't realize how good Mirai is because they've been stuck in the land of crappy software for so long that they have a skewed view of what's good. Once you use it though, the world looks different.
The limitation is talent.
:).
Frankly, I don't think so. I think the limitation is all the crappy software out there. Maya isn't too bad, but the other software you mentioned have horrible interfaces.
The "talent" is learning to use some piece of shit user interface effectively. That seems like a waste of time to me. There is some software out there that is better than the most popular packages but they are not anywhere near perfect. The packages I'm speaking of are Mirai, Nendo, Wings3D, etc. They are much easier to get prefessional results without spending years learning the damn UI. However, Mirai is limited compared Maya and Nendo/Wings are currently only modelers (and they can't handle high poly models well).
It's just a sad state of affairs because there is no complete, powerful, fast, easy to use graphics and animation package (I'm currently working to fix this by the way
I haven't tried using the Knoppix hardware stuff with regular Debian. Would be nice but I just haven't gotten around to trying it.
You might try running Knoppix since it can run right off the CD and see if it's any good at detecting your particular hardware. It has worked really well on all the machines I've tried, including my accessory laden laptop.
Oops, hit post too fast.
And another thing, KDE 3.1 and newer kernels have been in the regular Debian unstable for quite a while now. I'm running KDE 3.1 under unstable right now in fact.
Uh, Knoppix has had that latest stuff (KDE 3.1, new kernel, etc.) for some time now.
In case you didn't know, Knoppix is Debian based and has some awesome hardware auto-detection utilities.
Meh, they have no idea. The Internet is an extremely useful source of information. Any time I need to know something, I just head to the 'Net.
Here are some examples of what I've used the Internet for (besides all the programming and normal computer junk I do):
- I wanted to know how to make those hollow egg shell decorations for Easter. Looked it up on the 'Net.
- Someone asked me if I knew how to spell chlamydia (I could hardly guess how to spell it; thanks Google) and if it was a bacterial infection. Looked it up on the 'Net.
- I needed the blue book value of several cars I was helping a friend look for. Head to the 'Net.
- I needed some tax forms from the IRS and my state. They're all on the 'Net.
- Tons of DMV stuff can be done on the 'Net. license renewalls have never been so quick and easy.
- I needed some hummingbird pictures for a nephew who is learning to paint and needed some color samples.
- I do almost all my shopping online (especially for gifts). No more travelling 20 miles to find they don't even stock what I was looking for.
- I got a weird engine check code on my car and needed to know how to fix it. Online forums are great.
- I got more information on the Pony Express Rally that I plan on competing in next year.
- I wanted to set up an official sized vollyball court in my backyard and got the dimensions on the 'Net.
And that's just in the past few days/weeks. I could go on and on.
I find the 'Net invaluable.
You are completely missing what I am saying. I see no point in continuing this.
Go read a good book on this subject.
No, you have to look at "f encrypted with g" as a whole new algorithm h. It's a complete system. In a sense it's really just one algorithm (with more steps) not two.
Mathematically it's possible that h is weaker than either f or g alone. A very simplistic example:
- f encrypts by adding 5 to every binary value.
- g encrypts by substracting 5 from every binary value.
Use the two together and, oops, no encryption. Of course a real algorithm is more complicated, but it's all just math.
I think you're basically agreeing with me in principal. That is that using different algorithms to encrypt multiple times is mostly safe.
However, your example is missing a possibility. It is completely possible that the interaction between algorithm f and g could produce a weaker result M". Of course it would be easier for anyone trying to break it to know which algorithms you used, but we have to accept that there is a chance they would have this information.
The attack would not be to try and break each algorithm separately (eg. break g, then to try and break f). Instead the attack would be based on breaking some weaker result of the two algorithms. You can look at "f encrypted with g" as a new algorithm "h" which is possibly weaker than f or g alone.
Not necessarily. Especially not when encrypting multiple times using the same algorithm. Read Bruce Schneier's "Applied Cryptography" book. Good stuff. He covers this question much better than I can answer here.
Even when using multiple different algorithms there is a chance of weaking the whole thing. Depends on which algorithms you're using and how you're using them. I think you are generally safe using different known-good algorithms though (say 3DES then AES). I would not encrypt multiple times with the same algorithm unless it has been mostly proven to be more secure.
- SSH only tunnels specified ports which is a pain if you need to tunnel several services or complicated protocols. Plus it makes things difficult to configure if you're constantly switching between wireless and ethernet on the same system.
- PPPoE (or similar) over SSH can be incredibly unreliable. It's the TCP-over-TCP problem and you will sometimes see connections just freeze up.
- IPsec is a pain in the ass to configure. I've noticed most implementations are pretty crappy. I've seen similar problems that you get with PPPoE over SSH. Connections freezing and such.
- stunnel can't do UDP connections (so no Quake over wireless) among other problems.
I use OpenVPN. It uses UDP and tunnels all traffic like IPsec except it's super easy to configure. In my experience it has also been much more reliable than IPsec. It has several connection and server recovery mechanisms that work very well. I don't think there is a Windows client though so that might be a problem. I really like OpenVPN though.