Slashdot Mirror

I'd have one computer burning ROMs,
and another one dl'ing images,
and an even larger RAID drive just for show,
to show I'm a rich man now...

As it stands, though, downloading 650MB in one fell swoop and burning it to a CD is only for the internet "haves" in the world. (i.e. not me--I have ethernet, but not a CD burner, I guess I could mount it on loopback, but I don't have that much space. Aaggh! New computer, here I come...)

Otherwise, you'll get it in mangled, 657KB segments, copy it to floppies, and wonder why CDs aren't formatted as FAT...
---
pb Reply or e-mail; don't vaguely moderate.

I am working on a research project to use Linux to record and edit professional quality video such as S-VHS video cassettes. We are using frame accurate methods to record so editing is pretty minimal.

Each frame is rendered on a Linux Beowulf cluster and stored on a large hard drive in PPM format. Next each frame is loaded into a frame buffer that supports component video out. The image is then recorded onto a Sony LVR/LVS 5000. A Linux machine completely controls the LVR. (Code is GPL'd of course.) From there we use the rs-422 remote control interface of the LVR and a JVC S-VHS VCR BR-S822U to edit and make S-VHS and VHS recordings.

The problem that I have is finding a frame buffer card for Linux that supports component video out. There is lots of stuff for getting video in but that is not what I need at this point. Currently, I am using an SGI O2 for video out. Unfortunately, the video hardware is crippled forcing me to reduce the quality of video it will produce. Any one have any suggestions about a frame buffer?

With the proper frame buffer I will be able to quickly record with excellent quality each frame. (These frames make up a scientific animation of myoglobin.)

Please take a look at the website at http://prisant.ncsu.edu/~neely

The website is a bit out of date but will be updated soon. You can also e-mail me at jjneely@eos.ncsu.edu with any questions, comments or sugfgestions on video cards. You may be interested in the group of people I am researching with. Please see http://prisant.ncsu.edu

One of the outstanding goals of this project is to create a complete Linux and Open Source solution. We would also like to create DVDs of these animations but that's a completely different story.

Jack Neely

I am working on a research project to use Linux to record and edit professional quality video such as S-VHS video cassettes. We are using frame accurate methods to record so editing is pretty minimal.

Each frame is rendered on a Linux Beowulf cluster and stored on a large hard drive in PPM format. Next each frame is loaded into a frame buffer that supports component video out. The image is then recorded onto a Sony LVR/LVS 5000. A Linux machine completely controls the LVR. (Code is GPL'd of course.) From there we use the rs-422 remote control interface of the LVR and a JVC S-VHS VCR BR-S822U to edit and make S-VHS and VHS recordings.

The problem that I have is finding a frame buffer card for Linux that supports component video out. There is lots of stuff for getting video in but that is not what I need at this point. Currently, I am using an SGI O2 for video out. Unfortunately, the video hardware is crippled forcing me to reduce the quality of video it will produce. Any one have any suggestions about a frame buffer?

With the proper frame buffer I will be able to quickly record with excellent quality each frame. (These frames make up a scientific animation of myoglobin.)

Please take a look at the website at http://prisant.ncsu.edu/~neely

The website is a bit out of date but will be updated soon. You can also e-mail me at jjneely@eos.ncsu.edu with any questions, comments or sugfgestions on video cards. You may be interested in the group of people I am researching with. Please see http://prisant.ncsu.edu

One of the outstanding goals of this project is to create a complete Linux and Open Source solution. We would also like to create DVDs of these animations but that's a completely different story.

Jack Neely

I am working on a research project to use Linux to record and edit professional quality video such as S-VHS video cassettes. We are using frame accurate methods to record so editing is pretty minimal.

Each frame is rendered on a Linux Beowulf cluster and stored on a large hard drive in PPM format. Next each frame is loaded into a frame buffer that supports component video out. The image is then recorded onto a Sony LVR/LVS 5000. A Linux machine completely controls the LVR. (Code is GPL'd of course.) From there we use the rs-422 remote control interface of the LVR and a JVC S-VHS VCR BR-S822U to edit and make S-VHS and VHS recordings.

The problem that I have is finding a frame buffer card for Linux that supports component video out. There is lots of stuff for getting video in but that is not what I need at this point. Currently, I am using an SGI O2 for video out. Unfortunately, the video hardware is crippled forcing me to reduce the quality of video it will produce. Any one have any suggestions about a frame buffer?

With the proper frame buffer I will be able to quickly record with excellent quality each frame. (These frames make up a scientific animation of myoglobin.)

Please take a look at the website at http://prisant.ncsu.edu/~neely

The website is a bit out of date but will be updated soon. You can also e-mail me at jjneely@eos.ncsu.edu with any questions, comments or sugfgestions on video cards. You may be interested in the group of people I am researching with. Please see http://prisant.ncsu.edu

One of the outstanding goals of this project is to create a complete Linux and Open Source solution. We would also like to create DVDs of these animations but that's a completely different story.

Jack Neely

Thank you for the information. BTW, I've played around with evolving neural networks a little. Here is some Java source I wrote to a fully connected net that learns to count to 7 using only one input to tell it when to start counting. It is based on the technique that Foley and Chellapilla used to evolve a checkers playing network. The program is pretty simple but I found it interesting. I wrote it for a class I am currently taking.

Thank you for the information. BTW, I've played around with evolving neural networks a little. Here is some Java source I wrote to a fully connected net that learns to count to 7 using only one input to tell it when to start counting. It is based on the technique that Foley and Chellapilla used to evolve a checkers playing network. The program is pretty simple but I found it interesting. I wrote it for a class I am currently taking.

The Pinkerton Inc. Website

The Commission's Website

Their Final Report

The Center For the Prevention of Youth Violence, one of the brainchildren behind the idea

and of course

W.A.V.E.'s site.

It took me a while to track this down, and it leads me to suspect that our Mr Katz didn't do very good research either.

1) The name of the organization isn't W.A.V.E, it's S.A.V.E - Students Against Violence Everywhere. Its home page can be found here.

2) The program is not about a month old. This site lists the coordinator of the program from October of 1994. That makes it about 65 months old instead. Actually, according to the organization's history SAVE was begun in 1989.

3) What has happened is that the program - Pinkerton's toll free phone number - was included as part of the recommendations from the Task Force on Youth Violence and School Safety, which in April of 1999 was tasked with further reducing school violence. You can read a press release from August 11, 1999 here.

4) The official connection is here, the page of the School Programs of the Crime Prevention Unit of the NC Governor's Crime Commission. (I assume it's supposed to be the crime prevention commission, but we all have an opinion about government these days.)

5) Part of the information is loosely correct. There is a toll-free phone number. The reporting, however, is to notify law enforcement of school violence either in progress or anticipated soon. It is not a place to report students who might be potential "dangerous students" someday - the geeks, the depressed, the lonely, or whatever.

6) The program is run very much like the DARE program. This is intentional as the source material indicates. I am not discussing whether this is good or bad.

As with any such program there is a potential for abuse, and I'm not in a location to evaluate it. But the degree of error in Mr. Katz's report leaves him with a significant loss of credibility - which is too bad because I actually find most of his columns interesting. Now I'll have to verify his facts before I trust his opinions.

It took me a while to track this down, and it leads me to suspect that our Mr Katz didn't do very good research either.

1) The name of the organization isn't W.A.V.E, it's S.A.V.E - Students Against Violence Everywhere. Its home page can be found here.

2) The program is not about a month old. This site lists the coordinator of the program from October of 1994. That makes it about 65 months old instead. Actually, according to the organization's history SAVE was begun in 1989.

3) What has happened is that the program - Pinkerton's toll free phone number - was included as part of the recommendations from the Task Force on Youth Violence and School Safety, which in April of 1999 was tasked with further reducing school violence. You can read a press release from August 11, 1999 here.

4) The official connection is here, the page of the School Programs of the Crime Prevention Unit of the NC Governor's Crime Commission. (I assume it's supposed to be the crime prevention commission, but we all have an opinion about government these days.)

5) Part of the information is loosely correct. There is a toll-free phone number. The reporting, however, is to notify law enforcement of school violence either in progress or anticipated soon. It is not a place to report students who might be potential "dangerous students" someday - the geeks, the depressed, the lonely, or whatever.

6) The program is run very much like the DARE program. This is intentional as the source material indicates. I am not discussing whether this is good or bad.

As with any such program there is a potential for abuse, and I'm not in a location to evaluate it. But the degree of error in Mr. Katz's report leaves him with a significant loss of credibility - which is too bad because I actually find most of his columns interesting. Now I'll have to verify his facts before I trust his opinions.

And I'd hardly call the people who code open-source "wannabe programmers" - I'd imagine they're far better coders than you are, seeing as how you're so closed-minded I doubt you ever try and learn new techniques.

This was not an attack at _all_ open source coders. That would be insane. It's simply a comment on that fact that we have hundreds (!) of Instant Messaging clients, and only three Instant Messaging _Servers_ (Jabber, gicqd and KiT) that I know of.

Linux is an excellent server platform, but has not made the leap in _current_ server-side apps, at least for the ones I'm looking for :)

And I do not profess to be a programmer, I know my strengths and weaknesses, and programming is one of my weaknesses.

Oh man, the government always does that crap.

I sent an e-mail to the FCC about this ridiculous policy, but I never got a real answer.

I guess I should have sent them 18 diskettes and cover letters and stuff in every format known to man so they could read it. :|
---
pb Reply or e-mail; don't vaguely moderate.

I am proud to be attending North Carolina State University and studying computer science here. Why? This college supports Open Source. We have our own linux distribution called EOS Linux that has tools for connecting to the campus AFS and Kerberos network. Most all of the computer labs are Solaris, Of course, we do have a few NT labs. C++ is taught with GNU tools. In one of my programming classes we were told that if we wanted to work at home on our own computer to install linux.

I really could not ask for a better computing environment.
Jack Neely

BTW, the URL for that Distribution is http://www.linux.ncsu.edu/eos-linux .

There are versions of Stella for Acorn, Irix, Linux, Solaris, Amiga, FBSD, Java, Mac, OS/2[my port :-)], Windows, and others.

I give them credit they at least wrote a unbiased article. But I am more than certain they didn't actually run the tests, but instead summed up the results from the student employee.

No, no, no, this is the government. And, believe it or not, it's gotten better .

But yes, I agree with you. If I have to e-mail them *at all*, it will be in ASCII. And the last time I did, all I got is an automated reply. Maybe they'll get back to me when I start collecting social security...
---
pb Reply or e-mail; don't vaguely moderate.

The Retinal Prosthesis Project has a lot of info about the technology. The doctor and university which are mentioned in the Stevie Wonder news reports are mentioned several places there.

BTW the English translation of SD3 is very cool.

The GNU ICQ-compatible Server

You can see their documents here

The GNU ICQ-compatible Server

You can see their documents here

I've got an ATI Rage 128 chip set, and it works very well for a desktop under the beta XFCom driver by Suse. 1600X1200X16M resolution works fast and well.

I can't run 3D games on it in Linux yet, but it runs Half Life with noticeably better framerate than my Nvidia TNT in Win'98 used to. I've also upgraded from K6/300 to P3/450, but I don't think that's the whole story.

I've had trouble with ATI in the distant past, and am not a big supporter of theirs. But it seems like they are improving and maybe I'll give 'em one more chance after this news.

Case in point:
my FCC rant from the last time I tried to check on this stuff... I never got a real reply from them, either. And we pay money for this?

Here's an old screenshot.

I messed up the link to Thought Stream. It's at http://www4.ncsu.edu/~bgdarnel/though tstream/

raj.phys.sfu.ca/~giles/lego-probe/

only partial at the moment, but more going up as it comes across. poor thing.

Now if you'd said it was a lego scanning probe microscope, I would have been there alot sooner!

Oh my god! A sound file available for public download! I never did that over ZMODEM before!