Regionals Not Archived

[Greg Marra]

Originally Posted by xcentrix

We've got pretty much the entire Hawaii regionals archived in windows media format. Windows media player is not the preferred format for TBA.

I haven't had a chance to send the archives to the NASA folks to have it converted. I have time tomorrow morning to try to track down my NASA contacts and/or TBA to see about getting the archives up.

Preferred, maybe not, acceptable, absolutely. Check your PMs.

[Uberbots]

for future years, you guys just need a peice of TBA-compatible software which sits at one of the hubs at each regional. this application simply listens to the signals from mission control (or whoever starts/stops the matches) and uses that data to record and encode the movie. files are saved based on match data from mission control.


its kinda forward, and maybe a little complex logistic-wise, but if it could be done then the organization would be increased, and man-hours and waiting time would be decreased.

[Vikesrock]

Originally Posted by Uberbots

for future years, you guys just need a peice of TBA-compatible software which sits at one of the hubs at each regional. this application simply listens to the signals from mission control (or whoever starts/stops the matches) and uses that data to record and encode the movie. files are saved based on match data from mission control.


its kinda forward, and maybe a little complex logistic-wise, but if it could be done then the organization would be increased, and man-hours and waiting time would be decreased.

Not sure FIRST would like this any more than they liked Sundial........

[David55]

Originally Posted by Uberbots

for future years, you guys just need a peice of TBA-compatible software which sits at one of the hubs at each regional. this application simply listens to the signals from mission control (or whoever starts/stops the matches) and uses that data to record and encode the movie. files are saved based on match data from mission control.


its kinda forward, and maybe a little complex logistic-wise, but if it could be done then the organization would be increased, and man-hours and waiting time would be decreased.

This idea has been thought of before and discussed with FIRST. There is no simple way to trigger the recording unless you receive a direct signal from the field computer. This is not a simple task and requires software modifications on the field computer, as well as a good amount of scripting on the recording side. There are current attempts being made to get this working and hopefully there will be more details later on...

[Bongle]

Originally Posted by David55

This idea has been thought of before and discussed with FIRST. There is no simple way to trigger the recording unless you receive a direct signal from the field computer. This is not a simple task and requires software modifications on the field computer, as well as a good amount of scripting on the recording side. There are current attempts being made to get this working and hopefully there will be more details later on...

I don't think the video capture side of it would be too difficult. Set up a Directshow video capture graph that spits the data to a file. As for the field communication part, that might be more difficult. Setting up the graph would probably an hour or two of work to research the filters you'd need.

When you get contacted by the field computer (or the user pushes a button to say that the match has started)
IMediaControl* pifController = NULL;
m_pifGraph->QueryInterface(IID_IMediaControl,(void**)&pifCont roller) // gets the media control pointer from the main graph
pifController->Play(); // starts the graph running
pifController->Release(); // releases our COM reference
// yay! now the graph is playing and saving to a file!

[Elgin Clock]

Originally Posted by Bongle

You can see the GTR regional at http://www.discoverychannel.ca/shows....aspx?sid=7826. It will not be on a publically available archive though because since Discovery holds the copyright on it, we can't redistribute it.

Error. Is this the link that can only be viewed if you are in Canada?
If not, then the link is now bad.

Originally Posted by Bongle

I don't think the video capture side of it would be too difficult. Set up a Directshow video capture graph that spits the data to a file. As for the field communication part, that might be more difficult. Setting up the graph would probably an hour or two of work to research the filters you'd need.

When you get contacted by the field computer (or the user pushes a button to say that the match has started)
IMediaControl* pifController = NULL;
m_pifGraph->QueryInterface(IID_IMediaControl,(void**)&pifCont roller) // gets the media control pointer from the main graph
pifController->Play(); // starts the graph running
pifController->Release(); // releases our COM reference
// yay! now the graph is playing and saving to a file!

Can you just base it off of a sound input? The match start sound for example? I know this a question coming from a non-programmer like me, but how hard would it be to set up something to record and be started somehow by the match start & end sounds??

Then again, someone can just sit there at the feed box at the regional and start and stop recording manually as well.

[Bongle]

Originally Posted by Elgin Clock

Error. Is this the link that can only be viewed if you are in Canada?
If not, then the link is now bad.

It should be viewable. Search for the GTR archive thread, maybe I fail at copy/pasting. Discovery had the geo-lock off for the regional itself, I would assume they'd have it off for these as well.

Can you just base it off of a sound input? The match start sound for example? I know this a question coming from a non-programmer like me, but how hard would it be to set up something to record and be started somehow by the match start & end sounds??

Then again, someone can just sit there at the feed box at the regional and start and stop recording manually as well.

That might work, but you'd have to train it to recognize sounds, and that (might be) a tall order. You could assume sounds follow the match start -> autonomous end -> teleop start -> end of match order and just change states as you detect loudness coming out of the field input without bothering to recognize each sound, but then that wouldn't account for them testing the field or having field errors. Having someone sit there would probably work best.

[Joe Ross]

Originally Posted by Elgin Clock

Error. Is this the link that can only be viewed if you are in Canada?
If not, then the link is now bad.

I was able to view the videos in California.

[Greg Marra]

Originally Posted by Elgin Clock

Then again, someone can just sit there at the feed box at the regional and start and stop recording manually as well.

This is possible now, but there are still dozens of unarchived regionals.

[Uberbots]

Originally Posted by Greg Marra

This is possible now, but there are still dozens of unarchived regionals.

Thats why the 'archiving team' needs a bit more collaboration, communication, and fallback.
What you guys do is amazing, and im guessing a couple tiny improvements will bring you from 25% to 100% easily.

[Tom Bottiglieri]

The problem with setting up a "drop in" solution (Believe me, this is not a new idea..) is differences in hardware between users. There's no guarantee that different video capture devices and their drivers will be compatible with recording software packages.

We have looked into DirectShow pumping video into VLC, which seems like a decent solution. (This is how we recorded the Boston regional). You can set the video recording settings once, then press play and stop to stop and start recording. I haven't looked into the VLC documentation enough though, but if it has OLE-esque controls it may be a feasible implementation.

[Madison]

We have Portland and Seattle archived, but the files are very large -- into the dozens of gigabytes each -- and we haven't had time to get someone to sit down and encode them into a more manageable format. Someday soon, perhaps.

Sorry, everyone.

[Bongle]

Originally Posted by Tom Bottiglieri

The problem with setting up a "drop in" solution (Believe me, this is not a new idea..) is differences in hardware between users. There's no guarantee that different video capture devices and their drivers will be compatible with recording software packages.

We have looked into DirectShow pumping video into VLC, which seems like a decent solution. (This is how we recorded the Boston regional). You can set the video recording settings once, then press play and stop to stop and start recording. I haven't looked into the VLC documentation enough though, but if it has OLE-esque controls it may be a feasible implementation.

You shouldn't need to pump it into VLC. It is possible to set up a filter graph so it goes:
Capture Source -> Compressor -> File

You can have the user select their capture source and codec at run-time (thus diminishing problems with hardware compatibility), and you could distribute some directshow compatible codec (preferably one that allows for quick seeking) with the software package to ensure uniformity of output files.

Granted, all this would require a custom app, but even that isn't that hard to create. I worked with directshow for 4 months my previous co-op term, it's really a nice technology.

[Greg Marra]

Originally Posted by Bongle

Granted, all this would require a custom app, but even that isn't that hard to create. I worked with directshow for 4 months my previous co-op term, it's really a nice technology.

I started poking at building a custom app, but ran into roadblocks trying to find out information about how to do this in Windows or, preferably, something cross platform.

If there is interest among the community, do you think some people would want to band together to create a FIRST Robotics Video Archiving Application? If we want to get really fancy, we can even have it push video up to the Internet as matches happen, so archiving is "5 minutes behind live."

[Jonathan Norris]

As mentioned earlier the most difficult aspect of creating an application to record the matches automatically is the different hardware configurations. The only way to really solve this is to have the recording equipment come with every field (I believe there are about 9??). For that we would need either convince FIRST that a computer with the necessary video capture card, and adequate performance is a good investment. Or some how get a specific sponsor to donate the equipment for all the field setups.

On top of that it would crate a new volunteer position, and the person running the recording would need to be very adept in using computers and solving problems that arise... (I can tell you as much as you plan these things out, capturing video with a computer isn't always a reliable system... computers like to crash). So doing all this isn't as easy as it sounds, but if the community creates a application to automatically record, split, encode, and upload all the matches we would have a strong case to bring to FIRST to get a system like this installed for every field setup.