Back 2 back recordings

[zehd]

Just a thought, how about splitting the paddings by dividing into 2 the greater amount of padding.
ie.
First Recording post-padding = 2 minutes
Second Recording pre-padding = 1 minute

therefore 2 > 1 then 2 / 2 = 1 minute

so give 1 minute of padding for each show.

What you're dealing with is, in this case, there is no padding to play with. The post padding from the first recording is discarded, as with the prepadding of the second...

If two recordings are back to back, then neither can have padding, and the end/start point has to be the one scheduled. Hopefully, the network adjusts their clocks to your PC...

I'm currently quite well on the way to a program that will steal time from the end of the first show, and glue it on to the beginning of the second. It will also steal time from the beginning of the second show and glue it on to the end of the first.

IMHO, it's the only way to reclaim snippets of a show that have gone wandering...

(I've been able to get it working already, but it's quite manual at the moment. I'm, automating it so it will take care of the padding reclamation in the PostProcessing.bat (or in ZProcess)...

[zehd]

Just a thought, how about splitting the paddings by dividing into 2 the greater amount of padding.
ie.
First Recording post-padding = 2 minutes
Second Recording pre-padding = 1 minute

therefore 2 > 1 then 2 / 2 = 1 minute

so give 1 minute of padding for each show.

I think what you're trying to do here is somehow, get some post padding....

When there is back to back recordings on one tuner, it's impossible to have realtime Post and Pre Paddings...

[nia]

I guess it's just me, but most of my half hour shows follow each other on the same channel...

Well - at least for the last few versions, I have PERFECT back-to-back recordings on same channel. I mean - I loose about <½ second. It's very impressive.

I usually get this with the weekend-morning cartoon bonanza, where I record different shows for junior. In the case the ending of one is at the beginning of the next I simply use mpeg2cut2 to create the full show. It is usually results in seamless assemblies with no or unnoticeable loss.

I guess this is due to intelligent use of caching while creating the new file. No matter how it's implemented - please don't change it

[fhmanas]

What you're dealing with is, in this case, there is no padding to play with. The post padding from the first recording is discarded, as with the prepadding of the second...

If two recordings are back to back, then neither can have padding, and the end/start point has to be the one scheduled. Hopefully, the network adjusts their clocks to your PC...

I'm currently quite well on the way to a program that will steal time from the end of the first show, and glue it on to the beginning of the second. It will also steal time from the beginning of the second show and glue it on to the end of the first.

IMHO, it's the only way to reclaim snippets of a show that have gone wandering...

(I've been able to get it working already, but it's quite manual at the moment. I'm, automating it so it will take care of the padding reclamation in the PostProcessing.bat (or in ZProcess)...

I like that idea of getting 2 minutes (or whatever post-padding of first recording) from first part of 2nd recording and transferring it to the 1st recording and transferring 2 minutest (or whatever pre-padding of 2nd Recording) from last part of 1st Recording and transferring it to 2nd Recording. Hopefully, post-processing would know that this should only be done with back-to-back processing.

What I was proposing was a sort of averaging out the paddings to give partial paddings to both recordings. This way no post-processing is needed but would need the system to know when to actual stop the 1st recording and start of the 2nd Recording.

[zehd]

I like that idea of getting 2 minutes (or whatever post-padding of first recording) from first part of 2nd recording and transferring it to the 1st recording and transferring 2 minutest (or whatever pre-padding of 2nd Recording) from last part of 1st Recording and transferring it to 2nd Recording. Hopefully, post-processing would know that this should only be done with back-to-back processing.

What I was proposing was a sort of averaging out the paddings to give partial paddings to both recordings. This way no post-processing is needed but would need the system to know when to actual stop the 1st recording and start of the 2nd Recording.

I hope you don't think me rude to try to point out why this idea won't work.

When there's a one hour recording from 1-2 (usually with two minutes extra at the end)

back 2 back with another one hour recording from 2-3 (usually with an extra two minutes at the beginning),

There is NO way to have any padding at all in between hour 1-2 and hour 2-3. You can't fudge what you don't have. The first recording stops at 'exactly' 2, and the second recording starts....

The only way to have any form of padding without any post processing, is to have two tuners, and have the end time of first recording On tuner 1, 'overlap' the start of recording two on tuner 2...

That's why, if you want it, there's post process involved...

[fhmanas]

I hope you don't think me rude to try to point out why this idea won't work.

Nope, not at all. I was just contemplating the idea that overlapping paddings could probably be detected, combined, then divided to the two recordings prior to it happening. This would actually entail more on pre-processing rather than post-processing. Of course, if the code is too much of a hassle, it's really not to much of a biggie. I would instead go for my original idea of cutting the second rather than the first as I would rather see a conclusion rather than an introduction.

[zehd]

Nope, not at all. I was just contemplating the idea that overlapping paddings could probably be detected, combined, then divided to the two recordings prior to it happening. This would actually entail more on pre-processing rather than post-processing. Of course, if the code is too much of a hassle, it's really not to much of a biggie. I would instead go for my original idea of cutting the second rather than the first as I would rather see a conclusion rather than an introduction.

Hmmm. Maybe you were ahead of me, and helping me with and idea for the Back2Back program. (which is doing well)

I was trying to make clearer the problem that existed and needed to be solved. Perhaps you were already ahead trying to solve it.

Anyway... I way I see it 'Back2Back' (Which I think I'm going to call it, cause ZBack2Back doesn't ring right)... Will run after every program, called from the PostProcessing.bat. When it runs, it checks the database to see if there is an adjacent program currently recording. If there is, It zonks off the last (however many) minutes and then exits. It may be told to cancel comskip, and wait for later.

When Back2Back is run on the next recording, it will know that there was a previous adjacent recording, zonk the first minute. It will then immediately glue on the last minute of the first show, and then proceed straight to Comskip...

While Comskip is comskipping, I think I will have Back2Back also glue the first minute, second recording to the first recording... It goes quickly, but I will be careful not to hog resources as having a recording, comskip and join operations at the same time, may be processor intensive...

There will be an option to have comskip run on the first show again, just to clean up the end of the show... But again, that's processor intensive, and we might leave that alone...

What do you think?

[nia]

To me it sounds like a very good idea!

It's basically automating what i often do manually...

One additional suggestion: Make the perod you add from the first to the secomd recording be configurable. E.g. I would maybe prefer 4 or 5 minutes being added, others may have different prefs.' Come to think of it, why not just use the normal padding prefs for the specific recording...

Just to be clear/check if I can read: After the completion af recording 1 and 2, the results are:

• x minutes of the beginning of recording 2 has been glued to the end of recording 1 => Recording 1 has a "padding-end".
  
• x minutes of the end of recording 1 has been glued to the beginning of recording 2 => Recoding 2 has a "padding beginning".
  

Correct?

/Niels

[Kasper]

Just to add my thoughts to this.

Why not where both recordings are on the same channel, at the start padding point of the second show open up a second file on the drive named as the second show and record into two files in parallel (from the same tuner) until the first recording is dropped off at its end pad point.

Then you would have each show complete in its own file with only 1 tuner. Seems a neat idea to me.

Wayne

[zehd]

To me it sounds like a very good idea!

It's basically automating what i often do manually...

One additional suggestion: Make the perod you add from the first to the secomd recording be configurable. E.g. I would maybe prefer 4 or 5 minutes being added, others may have different prefs.' Come to think of it, why not just use the normal padding prefs for the specific recording...

Just to be clear/check if I can read: After the completion af recording 1 and 2, the results are:

• x minutes of the beginning of recording 2 has been glued to the end of recording 1 => Recording 1 has a "padding-end".
  
• x minutes of the end of recording 1 has been glued to the beginning of recording 2 => Recoding 2 has a "padding beginning".
  

Correct?

/Niels

Correct.