Editing parts of a video without re-encoding it

[Ferden]

I, I'm new at this, and I have a very trivial problem.

I want to cut the beginning of a video, but I can't. I thought that cutting was the dumbest thing ever, a matter of clicking this and that on a program, but apparently it's complicated. This is what I've done: so far:

- I tried simple programs like Ultra video splitter and Avs video remaker, but they left over a piece that I explicitly and repeatedly cut out. Avs video remaker even says "WARNING: splitting can be performed based on keyframes only. The split point will be moved to the left nearest keypoint.", which is precisely what the program is not supposed to do.

- I then tried more sophisticated programs, like Power director and Wondershare filmora, that, without even asking, casually start to reencode the video on stupid formats using time and making it ten times bigger.

- Then I googled it, and a video suggested me to try Avidemux. This program did cut the audio at the indicated place, but deleted a full "segment" of the video, so now the new file doesn't display the video's first seconds, and shows a still frame from the next "segment" instead.

How can I cut parts of a video without re-encoding it? Thanks

[smrpix]

You haven't said what kind of video it is (and I bet you don't know,) but based on your description it is long-GOP video of some sort. Your best solutions are VideoRedo and TMPGenc Smart renderer which will re-render only the GOPs your cuts interfere with. Neither are free.

It's not trivial, as you've discovered.

[Ferden]

You haven't said what kind of video it is (and I bet you don't know,) but based on your description it is long-GOP video of some sort.

I am the user. For me this is "A VIDEO". Any kind of technical shit is the program's problem, not mine. And yet, I have to deal with this. Anyway:

Video Format:
odivx
480 X 368
29.97 fps
1391 kbps
Audio Format:
mp3
2 Channels
44100 Hz, 16 Bits
128 kbps

By the way, I intend to make, in a foreseeable future, a documentary entirely with pasted scenes. Of all codecs. All of them. I don't even know which I will find, but the internet is big, so I really need some program that deals with all of this. Something that doesn't start to re-encode just because, and capable of dealing with everything.

[davejavu]

You had better get used the idea of dealing with the 'technical shit' as you are going to need to know it for your future project. You can't just throw random video clips together with multiple video and audio formats, video resolution and framerate difference and all the other 'technical shit', without proper editting and reencoding into a finished product.

Most video editting software will only let you cut on keyframes, which may not be where you actually want to cut, if you want to keep the existing video and audio without reencoding. Some don't allow that and will always reencode and others will do a combination of both, only reencoding the parts at the cut point, which lets you cut anywhere you want.

[usually_quiet]

SolveigMM Video Splitter claims to smart render and can work with more types of video and audio than VideoReDo TV Suite or TMPGEnc MPEG Smart Renderer 4, but not all. I'm not aware of any editing software that can edit all codecs and not many editors smart render/smart encode (only re-encode frames in incomplete GOPs adjacent to a cut). You will need more than one editor. It is possible that for some types of obscure video and audio formats, you will be forced to use an editor that cuts only on I frames to avoid re-encoding or use an editor that does re-encode everything.

However, as already stated, there is no container file format that supports joining random clips with varying resolutions, frame rates, and all types of video and audio together without re-encoding them first to be uniform. You will be unable to distribute your documentary in one media file.

[Ferden]

So, to resume:

- If I want to cut segments, I need a program with an hybrid approach, like SolveigMM Video Splitter or VideoReDo, 'cause most codes don't allow to split segments. Do you have a list of the codecs that have this segment mechanics and those that treat frames one by one?

- If I want to glue different scenes in a common video I'll have to re-encode everything on a common format with a heavy program. No re-encoding, obviously, will improve the quality of the videos, and even risk to deteriorate it. Do you have any suggestion for a 1080p codec to choose for the common standard?

P. S.: "hybrid" is a common word, it shouldn't be automatically hyperlinked with a program's page.

[manono]

Do you have a list of the codecs that have this segment mechanics and those that treat frames one by one?

What difference does it make? If you're going to join a bunch of videos with a bunch of different characteristics, they'll all have to be reencoded anyway. When reencoding to a common audio and video codec you can split them any way you like.

Do you have any suggestion for a 1080p codec to choose for the common standard?

h or x 264 video with AAC audio in MKV or MP4 container.

Any kind of technical shit is the program's problem, not mine.

Hardly.

I intend to make, in a foreseeable future, a documentary entirely with pasted scenes. Of all codecs. All of them.

Right. Let us know how that turns out.

[smrpix]

Your final encode specifications would be determined by your distributor's delivery requirements. Ask them.

[blud7]

I, I'm new at this, and I have a very trivial problem.

...
How can I cut parts of a video without re-encoding it? Thanks

You can't.

[Ferden]

Do you have a list of the codecs that have this segment mechanics and those that treat frames one by one?

What difference does it make? If you're going to join a bunch of videos with a bunch of different characteristics, they'll all have to be reencoded anyway. When reencoding to a common audio and video codec you can split them any way you like.

Knowledge is always useful, so it does makes a difference. For example: yesterday I downloaded from youtube something called "Dexter early cuts". I did it from youtube, 'cause I didn't find a torrent. As usual, it wasn't simple. Although composed by just five eight-minutes episodes, I didn't find a single file: the first four episodes were in individual pages, the fifth was split in seven parts, simply because, for commercial reasons, it was released in seven different moments. You know, this way people get addicted. On top of that, I don't know if the original distributor or the one that uploaded the video on youtube, had the perfectly fine reflex of adding at the beginning as well as in the end of each and every segment of episode an annoying "Machimina" logo. Most annoying. So what I did was, first, to glue the seven parts in one with one of the many programs I tried these days, Ultra video joiner. A program that failed to properly edit other stuff, even when the different parts were encoded the same way, but inexplicably did it right now. Why it didn't work with other projects but did now, it's something probably we'll never know, but I guess it has something to do with the codecs. Hence my curiosity of knowing if there is somewhere a list of codecs that behaves this way, or rather that way. But then I wanted to edit out the fourteen (I repeat: fourteen) ugly logos embedded in the fourteen-minutes video. Again, my gut feelings, developed in a few days' experience, told me one of the programs I tried, Avs video remaker, wouldn't do it. I don't know why, again I can't tell, but maybe it's because it's an mildly sophisticated software, and something tells me that it will surely leave untold blunders behind every cut. So I used yet another program, a big one, called Power Director. I knew it would have re-encode and resize the file, but it's not like I had a choice. Curiously enough, it worked. After carefully selecting the end format, and explicitly instructing the program to use the same definition the original file had, I managed to have an end result that, although it looks the same, it's three times bigger. Who knows why. Why an mp4 14:33 1280x720 268.184.268 bytes long becomes an mp4 13:38 1280x720 848.560.297 bytes long, it's simply beyond me. But maybe it's because my narrow minded mentality doesn't grasp the depths of codecs behavior, or the philosophy of programs automatically choosing different, heavy formats so as to not to have to auto-detect and then handle the limitations and of all the possible input formats. Silly me. But the bottom line is, as you yourself noted, the technical shit is actually my problem.
And funny enough, editing this file as well as writing this post took me more than watching all seven the segments of that episode.
So, I want to know if you have some section or link to a page that explains the different codecs, their unsplittable "segments", or even what h264 or x264 are (they are codecs, I know).

[Cornucopia]

You have big ideas & ambitions ( you don't even realize yet how big), yet your approach to them utterly lacks the drive/effort and openness/willingness to learn about the fundamentals of constructing them puts you in a quandary where you are doomed to not realize those ambitions until you change your attitude.

Tell me, how would you cut up a finished cake so that you remove the sugar (for a diabetic's needs, etc), without ruining the cake? "People cut up cakes all the time, it ought to be simple", you say.

Scott

[smrpix]

Per your request, here is a list of many of the video and audio codecs you are likely to encounter. It contains individual links with information about whether they are GOP or i-frame based.
https://en.wikipedia.org/wiki/List_of_codecs#Video_compression_formats

[Ferden]

You have big ideas & ambitions (you don't even realize yet how big), yet your approach to them utterly lacks the drive/effort and openness/willingness to learn about the fundamentals of constructing them puts you in a quandary where you are doomed to not realize those ambitions until you change your attitude.

I agree: I don't have the willingness to learn the technical intricacies of video formats. Then again, people don't really have this kind of choices, do they?

Let me tell you a metaphor. French grammar it's impossible. Up until the sixties, elementary school teachers used the stick to teach it, and a certain percentage of students learnt. Afterwards, the stick was forbidden, and this percentage dropped. But this doesn't change anything, because those who still learn simply use mental sticks on themselves. Either way, those who know proper grammar refuse any reform, and are even proud the language it's so complex. They say it's "rich". That's because now it's their turn to make fun of others. Meanwhile, in Italian you write things as they are pronounced, and in Chinese you don't have such a thing as syntax. Not to mention that SMSs are erasing the very concept of grammar. You see, if digital video evolved in a stupid way, you should say it's a flawed system, not cherish every trick you learnt and consider impatient those who don't know yet or never will. If you sweated to learn every trick on your bag, then you should be angry against those who put obstacles in front of you, not resell them. It's precisely the humble, open minded attitude of the user that allowed the market to develop different codecs, DRMs, regional codes, advertisement... and software without an undo button. Why don't you have a single video program, that reads everything, writes everything, convert everything, does it wright, does it fast, doesn't bug, and it's free? Something with a fast-forward button, that handle subtitles, who knows the name of that actor and doesn't start videos too loud? Because you have a thousand codecs, and million companies who want to speculate, and a billion users who sigh in frustration and say "that's just the way it is, it's my fault because I don't have enough money".

As per my ambitions, I just want to copy and paste video, man. That doesn't sound extreme to me.

[TreeTops]

"I just want to copy and paste video"

I could have swore that you mentioned cutting out parts of videos also. That's the part that requires some knowledge.

[usually_quiet]

Why don't you have a single video program, that reads everything, writes everything, convert everything, does it wright, does it fast, doesn't bug, and it's free? Something with a fast-forward button, that handle subtitles, who knows the name of that actor and doesn't start videos too loud? Because you have a thousand codecs, and million companies who want to speculate, and a billion users who sigh in frustration and say "that's just the way it is, it's my fault because I don't have enough money".

As per my ambitions, I just want to copy and paste video, man. That doesn't sound extreme to me.

If those are your expectations regarding editing software, you will need to delay your project until someone designs editing software with true artificial intelligence that can also read minds. Until then you'll need to learn a fair amount about video and audio as well as learn how to operate multiple programs.

[hech54]

Is this going to be one of those threads where the OP gives it about a week, comes back saying that he's found the holy grail of softwares that does exactly what he wants.....but refuses to say what it is?

[davejavu]

You see, if digital video evolved in a stupid way, you should say it's a flawed system, not cherish every trick you learnt and consider impatient those who don't know yet or never will. If you sweated to learn every trick on your bag, then you should be angry against those who put obstacles in front of you, not resell them.

What is stupid about it? It's called development. The technology has evolved over several decades and as new technology has evolved so have new codecs. Some codecs couldn't exist before the technology existed that was capable of using the more advanced, and more demanding, compression techniques.

Nobody is putting obstacles in your path, you just don't have the necessary knowledge and don't seem to be willing to learn it. Would you prefer we were all still watching black and white video on our 8" cathode ray tube screens?

The fact that something is more complicated than you are capable of understanding doesn't make it stupid, it just means you need to learn more about it or forget about it and do something else instead.

Why don't you have a single video program, that reads everything, writes everything, convert everything, does it wright, does it fast, doesn't bug, and it's free? Something with a fast-forward button, that handle subtitles, who knows the name of that actor and doesn't start videos too loud? Because you have a thousand codecs, and million companies who want to speculate, and a billion users who sigh in frustration and say "that's just the way it is, it's my fault because I don't have enough money".

One size doesn't fit all and those billion users would also sigh in frustration if they were told there was only one thing they were allowe to have even if it wasn't what they wanted. Software can only do what it is told to do, within the limitations of it's programming, and any program that could do all the things you suggest would be immensely complicated and have a very steep learning curve. I'm pretty sure you would be at the front of the queue of people complaining that it was a stupid complicated program!

[Ferden]

any program that could do all the things you suggest would be immensely complicated

No, actually, if a program handles everything on its own, and does it right, as they all should, it would be way easier than a tape recorder, with less than 10 button including volume:

https://upload.wikimedia.org/wikipedia/commons/2/29/RadioShack-ctr-119.jpg

Remember those? What an utopia...


Anyway, coming down to Earth, I tried the following programs:

Avidemux: Displays the freezed first frame of the next gop and plays audio
Avs video remaker: Leaves behind blunders but at least warns you and has an average interface
Easy video splitter: Total crap
Power director: Re-encodes everything every time and has fancy options, to use only for making films
Solveigmm video splitter: Hybrid tool that works (so far)
Ultra video splitter: Total crap (although Ultra video joiner works fine for files with the same size and codec)
Video redo: Allegedly an hybrid tool but actually re-encodes everything like Power director (so far)
Wondershare filmora: Re-encodes everything and has a bad interface

All the mentioned programs have heavy UI flaws: they don't remind their window size, they reset the volume to maximum every time (that one is on purpose, so as to imprint the mind of the user), they don't remember the default directory for saves, they don't allow to select the file's original directory as default, many of them don't even have a next frame button or display on a line the audio wave. But hey, that's not stupid,

It's called development.

[usually_quiet]

Don't blame the tool just because you don't understand its purpose. That's like blaming a screwdriver for not being any good at driving nails.

I told you VideoReDo TV Suite isn't designed to work with as many types of video and audio as SolveigMM Video Splitter. VideoReDo TV Suite is a specialty tool for MPEG-2 and H.264, used mainly for working with recorded TV shows captured with various devices. It can smart encode MPEG-2 and H.264 along with the audio formats commonly associated with them. However, it isn't designed to smart encode any video formats other than MPEG-2 and H.264. It is designed to convert any other video format the user imports to either MPEG-2 or H.264 on export.

I have been using VideoReDo TV Suite for a few years. The user interface is very efficient once someone understands how to use it.

[Edit]VideoReDo TV Suite does remember the previous folders I used for both import and export, as well as the last file that I imported and the last file I exported. It remembers the previous volume setting and the previous window size too.

[davejavu]

No, actually, if a program handles everything on its own, and does it right, as they all should, it would be way easier than a tape recorder, with less than 10 button including volume:

https://upload.wikimedia.org/wikipedia/commons/2/29/RadioShack-ctr-119.jpg

The software can't read your mind. Not everybody wants the same type of video, audio, subtitles, chapters, metadata, tags, etc., that you want, and many are constrained by the intended use for the video. A video format that is good for Bluray would be bad for streaming, and can't be used for DVD. People have many different needs and preferences for the video they create, that is why there are so many different ways to do it.

You will need to find a piece of software that can output the final product that you want and LEARN how to use it properly to achieve your goal. Unless you intend to write you own omnicient, omnipresent, psychic software that knows everything and does everything you are out of luck.

Remember those? What an utopia...

Really, I seem to remember things like that being a POS compared to the technology available today.

[Cornucopia]

Also, it was one single (audio only) format that could do just a few things well. Ever try to do an edit of the type you mentioned in the OP on one of those things? Nope, I didn't think so. Well I can tell you from firsthand experience it was a major PITA. And you would be re-recording (aka recompressing).

The world has greatly expanded beyond your narrow scope, and with it are options. With those options, however, comes complexity.

It appears at first simple (or as if it should be simple) to us at our elevated technological spot until we look down and around and acknowledge that we are standing on the shoulders of giants.

Re: your language argument: not everything in a language mixes down into LCD internet discussion and memes. The same language that handles those uses must also be able to handle the intricacies of concepts in science and medicine and law, as well as evoke strong feelings in novels and poetry and propaganda. This requires versatility, and thus, complexity.
Same too with multimedia uses, tools & techniques.

Still haven't unmixed the cake?

Scott

[Ferden]

Look, I was about to answer to all of your provocations one by one, but I guess it's for no use. You guys got territorial defending something (the programs, formats and duelling companies) that actually steal your time and money with crappy, expensive and conflicting products. If you learnt, with pain and time, to do this and that, it doesn't mean a complex system is defendable. If the market decided to invent a thousand formats, it should also invent a program that handles all of them, and keep the interface with less than ten buttons, 'cause you actually don't need more than that. Provided the software works, of course. Not to mention that the more their products break and are incompatible, the more money they make. I don't like it. I don't like companies to make a profit, let alone create gratuitous problems to make even more profits.

That said, I have a few questions:

-A program that smart re-encodes, re-encodes the split GOP into a different format from the original, or simply creates a new, smaller GOP with the same standard?

-For the time being, I had a number of videos to polish individually, so I used Solveigmm and Videoredo. Solveigmm have this ugly habit of crashing with half of the videos (mp4 and flv) I tried. It doesn't crash when I order to save one of those files without having made any modification, it doesn't crash when I try to make a cut on a random place, it does crash when I cut on the exact frame I want. This means I can't even test if a given file will cause a crash or not, and I will discover it only after the work at the moment of saving. Do you know why? Do you know how to prevent it?

-Since it's clear you need dozens of programs to edit video, do you know other software, a part those I mentioned here plus TMPGEnc I should try? So far, I don't have a way to join files (with the same codec, obviously) whenever Ultra video joiner fails.

As for the cake, I still don't get it.

[usually_quiet]

Look, I was about to answer to all of your provocations one by one, but I guess it's for no use. You guys got territorial defending something (the programs, formats and duelling companies) that actually steal your time and money with crappy, expensive and conflicting products. If you learnt, with pain and time, to do this and that, it doesn't mean a complex system is defendable. If the market decided to invent a thousand formats, it should also invent a program that handles all of them, and keep the interface with less than ten buttons, 'cause you actually don't need more than that. Provided the software works, of course. Not to mention that the more their products break and are incompatible, the more money they make. I don't like it. I don't like companies to make a profit, let alone create gratuitous problems to make even more profits.

That said, I have a few questions:

-A program that smart re-encodes, re-encodes the split GOP into a different format from the original, or simply creates a new, smaller GOP with the same standard?

-For the time being, I had a number of videos to polish individually, so I used Solveigmm and Videoredo. Solveigmm have this ugly habit of crashing with half of the videos (mp4 and flv) I tried. It doesn't crash when I order to save one of those files without having made any modification, it doesn't crash when I try to make a cut on a random place, it does crash when I cut on the exact frame I want. This means I can't even test if a given file will cause a crash or not, and I will discover it only after the work at the moment of saving. Do you know why? Do you know how to prevent it?

-Since it's clear you need dozens of programs to edit video, do you know other software, a part those I mentioned here plus TMPGEnc I should try? So far, I don't have a way to join files (with the same codec, obviously) whenever Ultra video joiner fails.

As for the cake, I still don't get it.

It is becoming clear that your reason for creating this thread has little to do with getting technical help. You are here mainly to use this forum as a soapbox to express your philosophical views regarding technology/software development and your personal economic opinions. VideoHelp exists to solve real-world problems. I'm not interested in playing your game anymore.

To smart encode, a program has to be able to figure out and replicate the settings used to encode the original video and audio. It isn't always easy to figure out what settings were used, and not all of the possible settings are available in every encoder. Defects in the re-encoded video/audio and crashes might result in when the smart encoding can't figure out or replicate the original encoding settings. If all the editor needs to do is copy and join clips composed entirely of the original video and audio, then no re-encoding is needed.

[PuzZLeR]

Interesting how a simple thread about linear cuts and joins got so philosophical.

It's really simple, and has already been discussed in several forms already.

You can do-it-all in one step if you like. I'm sure a simple tool like VideoStudio can input all your clips into one timeline, and output with the edits you want. And you're done.

Or, are you?

Yes, your done, but this simple solution, depending on format(s), will likely require re-encoding, and would destroy alot of quality along the way to the finished product. Knowledge of formats is essential if you want better quality.

The key here is to "smart render" as much as possible - to not re-encode common formats, and each format seems to require a separate dedicated editor better designed for it.

Sure you can probably do-it-all in one step. Or, you can take the time to learn about formats and their requirements and save yourself tons of time in processing, and tons of quality in the result.

Yeah, I know it sucks to have many different formats. I get it. But each was designed to better suit a technological era, and/or a specific function (such as higher file size for more quality, better speed, better capturing, better editing - whatever the objective is). The only way around this is knowledge, maintaining source, and having a clear objective for your result.

[Ferden]

Yeah, I know it sucks to have many different formats. I get it. But each was designed to better suit a technological era, and/or a specific function

You see: it's not difficult to understand that it sucks to have different formats, many programs, conflicts, and so on. I'm afraid is pretty consensual. If you are smart you can understand the path that lead there, but understanding and having an accurate map in your head of what happened doesn't make it good.

The only way around this is knowledge, maintaining source, and having a clear objective for your result.

By "maintaining source" you mean keeping the original video fragments that you want to join in the final version in their original form and codec? This is what I intend to do, but as you know it's trick. In my optimist vision, editing a video and cutting out the part I need does not require re-encode, but I'm afraid this is not true. So I have a couple of questions:

1 - A part the failures usually_quiet describes:

Defects in the re-encoded video/audio and crashes might result in when the smart encoding can't figure out or replicate the original encoding settings.

smart re-encoding/smart rendering means a program charged with splitting a GOP will replace it with a new, shorter GOP with the same codec and settings, or it will rather insert the retained part from the split GOP in a different codec, leaving to the player (or the second editing software) the task to the understand this and play the different parts?

2 - I noticed it’s perfectly possible that software re-encodes a video, makes it bigger, and the end result looks way worse than the original. How is it possible? Is there a way to predict the cases when this will happen? I mean, there are some specific reasons for this that I can avoid?

3 - Do you know a program capable of joining video without re-encoding other than Easy video joiner and Ultra video joiner?

[PuzZLeR]

Yeah, I know it sucks to have many different formats. I get it. But each was designed to better suit a technological era, and/or a specific function

You see: it's not difficult to understand that it sucks to have different formats, many programs, conflicts, and so on. I'm afraid is pretty consensual. If you are smart you can understand the path that lead there, but understanding and having an accurate map in your head of what happened doesn't make it good.

Again, I too agree this sucks having all these bloody formats all over the place.

But, also, I appreciate why: technological progress and/or function.

As per progress, I'm sure something like H.264 video was a plan, and algorithm, from, say, 1986, but the processing then would've taken like, a month for one movie on the home machines of that era. Even playback would've been choppy at best. We needed something reasonable for a given era otherwise nothing would've been practical.

As per function, we have formats best suited for capture. Others for edits. Others for low bitrate playback. Others for speed. It all depends on the application and objective. For example, a format that can be completely lossless will be bad for file size, however we use this format for flexibility in processing, and retaining quality to the final objective, and deal with the file size later, and even playback objective, later, such as with a format best suited for such. This is why we have Forums like this.

Which also leads me to my next point...

The only way around this is knowledge, maintaining source, and having a clear objective for your result.

By "maintaining source" you mean keeping the original video fragments that you want to join in the final version in their original form and codec? This is what I intend to do, but as you know it's trick. In my optimist vision, editing a video and cutting out the part I need does not require re-encode, but I'm afraid this is not true.

Yes indeed - another thing that has had huge benefits on this hobby for me. I keep all Source, which is common practice among many here.

You can see how things change, and formats come and go, playback needs change all the time, but keeping the Source always gives you the best options (which included highest quality possible, and recovery from any "mistakes" or selection of "wrong format", etc) at all times, now, and in the future. This is the best way you can deal with this and your problem as well.

And yes, encoding different formats, using lossy codecs, into one production will result in some re-encoding and some quality loss, along the way. You can learn techniques to minimize this loss, and you can also retain Source in case you want to do something different. Again, keeping the Source demands more storage, but that's the only disadvantage long-term.

So I have a couple of questions:

1 - A part the failures usually_quiet describes:

Defects in the re-encoded video/audio and crashes might result in when the smart encoding can't figure out or replicate the original encoding settings.

smart re-encoding/smart rendering means a program charged with splitting a GOP will replace it with a new, shorter GOP with the same codec and settings, or it will rather insert the retained part from the split GOP in a different codec, leaving to the player (or the second editing software) the task to the understand this and play the different parts?

2 - I noticed it’s perfectly possible that software re-encodes a video, makes it bigger, and the end result looks way worse than the original. How is it possible? Is there a way to predict the cases when this will happen? I mean, there are some specific reasons for this that I can avoid?

3 - Do you know a program capable of joining video without re-encoding other than Easy video joiner and Ultra video joiner?

I can't answer this without inviting you to do further research, or, better yet, tell you to list details of each video in your collection, and tell us exactly what you want it to play it on.

1. This is tricky. If you're changing formats, that's a complete re-encoding, and a different set of GOPs particularly suited to the new format. But, if a program "smart renders", such as Womble does for MPEG-2, then is should optimize the GOPs for the cuts and edits. If it doesn't, then it's not "smart rendering" very well.

2. Yes, this is true. Maybe it's the settings and a much bigger bitrate. Maybe it's encoding to another format that is less efficient in bitrate. Also, you can encode from lossy->lossy, and no matter how much bitrate you use, even 1,000,000mbps, you will still (at minimum in theory) lose quality. Or, maybe it's a bad, bad encoder doing the work. There are lots of reasons why this is possible.

3. If you're working with several different lossy formats, then no. And even with one common format, there is no one editing program that can do it all with all formats optimally. There are some that can do it, with multiple formats, more or less, kind of like the VideoStudio example I gave earlier, but will only give an easy result, not particularly the best one.

There are editing programs for MPEG-2, H.264, AVI formats, even WMV, etc. Each handles the format it's designed for well, but may really suck at something else.

That's just the way it is. Again, that's why we have Forums like this. You let us know what format your files are in, and what you want them to play back on, and someone here can help you with some of the better techniques. That's how I learned, and that's how you can too.

[usually_quiet]

smart re-encoding/smart rendering means a program charged with splitting a GOP will replace it with a new, shorter GOP with the same codec and settings, or it will rather insert the retained part from the split GOP in a different codec, leaving to the player (or the second editing software) the task to the understand this and play the different parts?

A smart encoding editor uses the same type of video and audio that the original input uses to re-encode partial GOPs for those types of video and audio that it is designed to smart encode . It does not encode partial GOPs as a different type of video. That would result in an unplayable file. No existing smart encoding editor can smart encode all possible video formats.

2 - I noticed it’s perfectly possible that software re-encodes a video, makes it bigger, and the end result looks way worse than the original. How is it possible? Is there a way to predict the cases when this will happen? I mean, there are some specific reasons for this that I can avoid?

Using different settings for encoding as the same type of video, or re-encoding as a different type of video can make the exported video file either much larger or much smaller. The user must to be able to control what video format will be used for output as well as the encoder settings to control the files size and reduce quality loss. (Some quality loss is inevitable when lossy video is re-encoded with a lossy encoder.) Enlarging a low-resolution video to high-definition, or using an overly low bit-rate are examples of things which make the resulting video look much worse than the original.

3 - Do you know a program capable of joining video without re-encoding other than Easy video joiner and Ultra video joiner?

No. ...and if someone is joining different types of video, then the video must be re-encoded for the file to be playable.

[Cornucopia]

It does suck to have different formats, but it would also suck to only have one format that didn't cover everyone's needs, and it would also suck to have one huge-a$$ complex superformat that covered everyone's needs but included variables (that one has to navigate) which only got used 5% of the time.
THERE IS NO EASY SOLUTION.
THERE IS NO FREE LUNCH.
All real-world options are flawed.
What we have can be improved upon (incl. smaller # of universally agreed-upon open codecs based around workflow stage: capture, edit/process, storage/archive, distribution), but there are plenty of people who appreciate having the options available to us.

Also, it seems you're stuck just seeing this from the point of view of a casual, instant-gratification-expecting consumer, not a producer. But it strikes me as odd that you won't consider giving the same respect to the engineering feats of creating/supporting/maintaining/expanding multimedia architecture that you likely would to civil engineers and physical architecture. "Only need 10 buttons" is beyond a joke, more like an insult. If you think it's so easy, why haven't you or anybody gone & created it? Seems that one could make a ton of money if you could pull it off.
So you wonder why it hasn't been done yet, but instead of going for the obvious answer (occam's razor) of "it can't be done, because things are much more complex than that", you jump to a ridiculous conspiracy conclusion of "the evil corporations are just trying to maintain their hold over everyone". Yeh, yeh, we've heard that one before, @newpball, et al.

What is there to not get about "unmixing" a cake? Can you do it? Why or why not? The analogy is closer to the truth that you might think.

...too busy to continue for the moment, but I should have a couple of informative responses to your questions (particularly about smart-rendering) in a future post.

Scott

[Ferden]

Hi guys,

sorry I didn't wrote anything before, but for a while I was waiting for Cornucopia's informative responses, then I was without internet, then I was busy, then a locust plague got in the way... and finally I forgot. Thank you for the advises.

As for Occam's shaving habits... greed and laziness are actually pretty parsimonious activities. Duh. Computer programs are universally bad precisely because companies go cheap about it. I never saw simple products like a hammer, a fork or, guess what, a cake, accumulate the number of defects a tiny program is capable of piling up precisely because they are simple.

[usually_quiet]

Apparently you have not baked many cakes from scratch (no mix). Many things can go wrong during preparation, mixing and baking. Even a box cake can turn out badly if someone doesn't follow the instructions, or the cake mix is too old.