You're missing an important point. As you say it is possible (not trivial) for a team of programmers to replicate functionality of another application. But often the differentiating factor is not the actual function but the quality of its implementation. For example, the principles of indexing data are well understood but what makes Oracle a leader is its speed, scalability and robustness. Photoshop is another example, its functions are very well defined yet no one has been able to challenge it as a best seller. I'll give one last example, many scientific packages implement a fast fourier transform, but the application I work on has a very fast (perhaps the fastest on x86) implementation. Opening our source would give away a huge competetive advantage.
You assume that all implementations are alike apart from a "tiny percentage". This is simply not the case, many leaders are leaders because of the quality of their code, along with its function.
I have hundreds of DVD discs lying around. For a while I've been pondering what it would take to build a networked video server, Audiotron for video, recently I finally got it working. The objective was to be able to play DVD's seamlessly from a server share, by seamlessly I mean complete with menus, extras etc., with high quality video and digital sound. Several STB DVD players play files over ethernet but this is usually mpeg4 or DivX, not DVD quality mpeg2. In theory a 100Mbit LAN should have enough bandwidth since DVD's video stream is at most 9Mbit, the most difficult part was putting together a quiet client machine capable of generating a good quality analog video signal. I ended up buying a cheap ($400) Dell Poweredge machine on ebay, it's practically silent, this is important since it sits beside the TV. For DVD decoding I'm using something called the XCard (http://www.sigmadesigns.com/products/xcard.htm), it decodes video and outputs digital audio. On the software side I am using TVedia (http://www.8dim.com/default.asp?linkid=vcdhelp), it generates menus on the TV to access media on the LAN. It also works quite well with the XCard, which comes with a serial port IR remote sensor. I can play DVD's by picking the title from a menu or search previously entered metadata such as director/genre etc. When I select a movie the DVD starts as if it were in the drive, the only real difference is it is much faster to jump between chapters and menus.
The system is limited only by storage. So far I have over 100 DVD's on the server. It's very cool to be able to pick a movie using the remote and play it instantly, no discs, no jewel cases and most importantly no exhausting roundtrips from the armchair to the DVD player. The system also decodes DivX (in hardware) and a few other formats, which means I can download clips/movies and display them on the big screen.
You're confusing exceptions with assertions. Exceptions can be used to elegantly handle boundary conditions. Assertions flag events that should never happen *if* the code logic is correct. Exceptions stay in both debug and release builds, assertions are not usually compiled into release binaries.
I have hundreds of DVD discs lying around. For a while I've been pondering what it would take to build a networked video server, recently I finally got it working. The objective was to be able to play DVD's seamlessly from a server share, by seamlessly I mean complete with menus, extras etc., with high quality video and digital sound. In theory a 100Mbit LAN should have enough bandwidth since DVD's video stream is at most 9Mbit, the most difficult part was putting together a quiet client machine capable of generating a good quality analog video signal. I ended up buying a cheap ($400) Dell Poweredge machine on ebay, it's practically silent, this is important since it sits beside the TV. For DVD decoding I'm using something called the XCard (http://www.sigmadesigns.com/products/xcard .htm), it decodes video and outputs digital audio. On the software side I am using TVedia (http://www.8dim.com/default.asp?linkid=vc dhelp), it generates menus on the TV to access media on the LAN. It also works quite well with the XCard, which comes with a serial port IR remote sensor. To make a DVD playable from the network I use DVDDecrypter to create an iso image and daemon tools to mount the image on a virtual drive, TVedia can then play the "disc" as if it was in the local machine's drive. The system is pretty neat if I do say so myself. I picked up a couple of Maxtor 250G drives from Frys, that's about 50-100 DVDs online. In theory I could stick another cheap PC in the bedroom and have access to the movie collection there also.
In the 80's when I was studying for my first degree in C.S. (yes, I'm an old fart), the fad of the time was formal methods. The basic idea is this: a problem is specified using a formal language, the specification is then translated into code automatically. This was supposed to herald the end of imperative programming languages and programmers. The major flaw was assuming anyone could put together a specification in the first place.
Hyperthreading does NOT equal parallel processing. Hyperthreading replicates the "architectural state" of the CPU, this is the registers and queues, the ALU would need to be replicated for parallel processing to happen. What it does do is make a context switch between threads faster in some circumstances. In Windows the speedup is realised because threads have affinity for processors i.e. they tend to be scheduled on the same physical CPU. Assuming two CPU bound threads in the ready state, hyperthreading should result in a speedup because less time will be spent in the context switch between scheduling quantums.
OK fuckwit, how would you record it if the only media available was watermark sensitive? The fact that a watermark can survive in the analog domain seems to be beyond you. That's why I suggested you google watermarking. Try engaging your brain before you respond.
Google watermarking. A good audio watermark would be difficult to remove and would prevent reencoding efforts *if* the right infrastructure was present. Bottom line, for this to be somewhat effective it would require support at the hardware level.
The ILS (instrument landing system) allows very low visibility (zero-zero) approaches using a glidesope indicator for height and localizer for direction, however, often flights are cancelled because fog prevents safe manouvering on the ground. What is really needed is a way to see static and moving objects through the fog. The visualization technology is cute and would be especially useful for training.
This would be pretty difficult to implement efficiently since a single instruction has the potential to make massive state changes, however, even a limited implementation would be extremely useful and would provide an alternative to restarting the application.
Build a custom quiet PC with a video card with onboard mpeg2 decoding and component out. Connect it to a gigabit over copper LAN (very affordable) and play the disks using PowerDVD by browsing to the network VIDEO_TS folder. To control the PC I use my laptop and PCAnywhere.
I looked into using one of Kiss tech's players (http://www.kiss-technology.com/) but they do not yet support streaming of ripped DVD's. Their tech support suggested ripping to mpeg4, apparently DVD streaming may be supported in a future firmware release.
The ideal solution would be an audiotron like device for video, Kiss seems the closest yet, but for now a custom solution is the best. The PC I spec'd out will run about $1400.
It's a common fallacy that teaching someone how to race will magically make them safer drivers on the open road. Checkout the Institute of Advanced Motorists and learn that becoming a good (safe) driver takes much more than a couple of hours pretending you are Michael Schumaker.
You don't know what you're talking about. Flight plans are filed for flight through many kinds of airspace. For a VFR flight the plan is mainly used as a search & rescue aid when the pilot does not close out the plan within a certain period. For IFR (instrument flight) the flight plan helps controllers give clearances during the flight, otherwise the pilot would have to make explicit requests en route.
I'm a matrix fan, but this movie was the biggest piece of dung I'm seen in a while. It's as if they ran out of money and had an amateur write and direct. It's just one big, boring shoot-em-up- machines spew through the dome at 10,000 per second and the dozen humans somehow manage to shoot every last one. Whatever.
Oh for Christ sakes, all Amdahl's law says is that the maximum parallel speedup is limited by the inherently serial portion of the algorithm. It's friggin obvious, fair play to him for sticking his name on it, but we're not talking e=mc^2 here.
Slashdot Mirror
Man, I don't no where to start. FFT's are used all over the place, our FFT is not specicialized, it is just FAST.
You're missing an important point. As you say it is possible (not trivial) for a team of programmers to replicate functionality of another application. But often the differentiating factor is not the actual function but the quality of its implementation. For example, the principles of indexing data are well understood but what makes Oracle a leader is its speed, scalability and robustness. Photoshop is another example, its functions are very well defined yet no one has been able to challenge it as a best seller. I'll give one last example, many scientific packages implement a fast fourier transform, but the application I work on has a very fast (perhaps the fastest on x86) implementation. Opening our source would give away a huge competetive advantage.
You assume that all implementations are alike apart from a "tiny percentage". This is simply not the case, many leaders are leaders because of the quality of their code, along with its function.
I have hundreds of DVD discs lying around. For a while I've been pondering what it would take to build a networked video server, Audiotron for video, recently I finally got it working. The objective was to be able to play DVD's seamlessly from a server share, by seamlessly I mean complete with menus, extras etc., with high quality video and digital sound. Several STB DVD players play files over ethernet but this is usually mpeg4 or DivX, not DVD quality mpeg2. In theory a 100Mbit LAN should have enough bandwidth since DVD's video stream is at most 9Mbit, the most difficult part was putting together a quiet client machine capable of generating a good quality analog video signal. I ended up buying a cheap ($400) Dell Poweredge machine on ebay, it's practically silent, this is important since it sits beside the TV. For DVD decoding I'm using something called the XCard (http://www.sigmadesigns.com/products/xcard.htm), it decodes video and outputs digital audio. On the software side I am using TVedia (http://www.8dim.com/default.asp?linkid=vcdhelp), it generates menus on the TV to access media on the LAN. It also works quite well with the XCard, which comes with a serial port IR remote sensor. I can play DVD's by picking the title from a menu or search previously entered metadata such as director/genre etc. When I select a movie the DVD starts as if it were in the drive, the only real difference is it is much faster to jump between chapters and menus.
The system is limited only by storage. So far I have over 100 DVD's on the server. It's very cool to be able to pick a movie using the remote and play it instantly, no discs, no jewel cases and most importantly no exhausting roundtrips from the armchair to the DVD player. The system also decodes DivX (in hardware) and a few other formats, which means I can download clips/movies and display them on the big screen.
Let's hope nothing goes wrong. As soon as the system fuck's up it'll be your fault and you'll be standing in the dole queue.
You're confusing exceptions with assertions. Exceptions can be used to elegantly handle boundary conditions. Assertions flag events that should never happen *if* the code logic is correct. Exceptions stay in both debug and release builds, assertions are not usually compiled into release binaries.
I have hundreds of DVD discs lying around. For a while I've been pondering what it would take to build ad .htm), it decodes video andc dhelp), it generates menus on the
networked video server, recently I finally got it working. The objective
was to be able to play DVD's seamlessly from a server share, by seamlessly I
mean complete with menus, extras etc., with high quality video and digital
sound. In theory a 100Mbit LAN should have enough bandwidth since DVD's
video stream is at most 9Mbit, the most difficult part was putting together
a quiet client machine capable of generating a good quality analog video
signal. I ended up buying a cheap ($400) Dell Poweredge machine on ebay,
it's practically silent, this is important since it sits beside the TV. For
DVD decoding I'm using something called the XCard
(http://www.sigmadesigns.com/products/xcar
outputs digital audio. On the software side I am using TVedia
(http://www.8dim.com/default.asp?linkid=v
TV to access media on the LAN. It also works quite well with the XCard,
which comes with a serial port IR remote sensor. To make a DVD playable from
the network I use DVDDecrypter to create an iso image and daemon tools to
mount the image on a virtual drive, TVedia can then play the "disc" as if it
was in the local machine's drive. The system is pretty neat if I do say so
myself. I picked up a couple of Maxtor 250G drives from Frys, that's about
50-100 DVDs online. In theory I could stick another cheap PC in the bedroom
and have access to the movie collection there also.
In the 80's when I was studying for my first degree in C.S. (yes, I'm an old fart), the fad of the time was formal methods. The basic idea is this: a problem is specified using a formal language, the specification is then translated into code automatically. This was supposed to herald the end of imperative programming languages and programmers. The major flaw was assuming anyone could put together a specification in the first place.
The double-blind test involved one group getting banged in the ass.
A leaked FBI document confirmed that the walrus was indeed Paul.
Hyperthreading does NOT equal parallel processing. Hyperthreading replicates the "architectural state" of the CPU, this is the registers and queues, the ALU would need to be replicated for parallel processing to happen. What it does do is make a context switch between threads faster in some circumstances. In Windows the speedup is realised because threads have affinity for processors i.e. they tend to be scheduled on the same physical CPU. Assuming two CPU bound threads in the ready state, hyperthreading should result in a speedup because less time will be spent in the context switch between scheduling quantums.
Yes dear, I know that. The point is analog recording is not a generalized way to defeat copy protection, so your assertion is incorrect.
OK fuckwit, how would you record it if the only media available was watermark sensitive? The fact that a watermark can survive in the analog domain seems to be beyond you. That's why I suggested you google watermarking. Try engaging your brain before you respond.
Google watermarking. A good audio watermark would be difficult to remove and would prevent reencoding efforts *if* the right infrastructure was present. Bottom line, for this to be somewhat effective it would require support at the hardware level.
The ILS (instrument landing system) allows very low visibility (zero-zero) approaches using a glidesope indicator for height and localizer for direction, however, often flights are cancelled because fog prevents safe manouvering on the ground. What is really needed is a way to see static and moving objects through the fog. The visualization technology is cute and would be especially useful for training.
I'll be here all week.
Score:4, Interesting ??
Pray tell us how it would use flaps to stay parallel to the runway? Word to the wise: if you're going to spout off, try getting your facts straight.
This would be pretty difficult to implement efficiently since a single instruction has the potential to make massive state changes, however, even a limited implementation would be extremely useful and would provide an alternative to restarting the application.
Build a custom quiet PC with a video card with onboard mpeg2 decoding and component out. Connect it to a gigabit over copper LAN (very affordable) and play the disks using PowerDVD by browsing to the network VIDEO_TS folder. To control the PC I use my laptop and PCAnywhere.
I looked into using one of Kiss tech's players (http://www.kiss-technology.com/) but they do not yet support streaming of ripped DVD's. Their tech support suggested ripping to mpeg4, apparently DVD streaming may be supported in a future firmware release.
The ideal solution would be an audiotron like device for video, Kiss seems the closest yet, but for now a custom solution is the best. The PC I spec'd out will run about $1400.
Well they didn't do such a great job. I checked several of the c source files and found gotos. It don't get much ruder than "goto".
It's a common fallacy that teaching someone how to race will magically make them safer drivers on the open road. Checkout the Institute of Advanced Motorists and learn that becoming a good (safe) driver takes much more than a couple of hours pretending you are Michael Schumaker.
You don't know what you're talking about. Flight plans are filed for flight through many kinds of airspace. For a VFR flight the plan is mainly used as a search & rescue aid when the pilot does not close out the plan within a certain period. For IFR (instrument flight) the flight plan helps controllers give clearances during the flight, otherwise the pilot would have to make explicit requests en route.
I'm a matrix fan, but this movie was the biggest piece of dung I'm seen in a while. It's as if they ran out of money and had an amateur write and direct. It's just one big, boring shoot-em-up- machines spew through the dome at 10,000 per second and the dozen humans somehow manage to shoot every last one. Whatever.
nm
He said the ceremony was totally lacking in atmosphere.
Oh for Christ sakes, all Amdahl's law says is that the maximum parallel speedup is limited by the inherently serial portion of the algorithm. It's friggin obvious, fair play to him for sticking his name on it, but we're not talking e=mc^2 here.