Editing without Recoding

[criggs]

criggs-I deleted your other post,just continue here.

Thanks. Originally I assumed that folks would be finding my question on the Editing discussion page, which is why I posted it there. I had tentatively concluded that it was a mistake my posting this question in the newbie section. But I've been getting a perfectly good response here, and no response there, for whatever reason. So focusing on the one thread here will make my life easier.

[criggs]

If possible you should test TMPGEnc MPEG Smart Renderer for editing a recording with obvious glitches before making a decision.

Thanks for the suggestion. I think what I'll do is make a recording on the DVR+ when I get it, today (Friday) or tomorrow (Saturday) and, while the recording is happening, deliberately re-position the antenna away from the window and behind a solid object to provoke some breakup, and then see how SR 5 handles it. That should be a valid torture test.

I tried a few smart-rendering editors, including TMPGEnc MPEG Smart Renderer, before settling on VideoReDo Plus for editing recordings from over-the-air sources. With the other editors I tried, one small glitch in the recording either caused the audio and video to go out of sync in the edited file, or caused the editor to stop exporting the file at the glitch.

Well, just now, after installing the free trial, I pulled in a 29.97 fps 1920 x 1080 .mpeg stream capture, which ran five minutes. The video codec was h.264 and the audio was AAC.

I then pulled that into Smart Renderer 5, and immediately had a problem. SR 5 saw the clip as being a 59.94 fps capture rather than a 29.97 capture. I was sort of confused by that, but proceeded to pull out a ten-second chunk in the middle arbitrarily, making sure that I was NOT doing a clean cut in/cut out at an i-frame. Well, it played through fine, so I went ahead and outputted it.

The file was twice the size of the original! I pulled both into VLC in turn; they were completely different; the original was 29.97 fps and the SR 5 file was 59.94 fps. The SR 5 file was evidently a total re-encoding; none of it had been smart-rendered at all.

WTF?????????!!!!!!!!!!!

Then I pulled a completely different file into SR 5, an mp4 file at 1280 x 720 with a 25 fps frame rate. SR 5 saw it perfectly.

Another WTF???!!!

Did SR 5 have some stupid problem with .mpeg files???!!!

So I opened XMedia Recode and re-contained the original .mpeg file with which I'd had the problem. I re-contained it as a copy, no trans-coding, into an mp4 file, and pulled that mp4 file into SR 5.

PERFECT!!!

I pulled out a chunk of about a few seconds, and pushed the button.

Perfect result, the final SR 5 file being slightly shorter/smaller than the original, which made perfect sense because of the cut I'd made, and the rest read with stats identical to the original in VLC.

Does this mean that any and every time I want to do some smart rendering in SR 5, I have to re-contain all my source material into mp4? That's a pain! I can't believe that's the way it's supposed to work. Please tell me I'm simply doing something stupid, and that that isn't necessary. Thanks!

[usually_quiet]

criggs-I deleted your other post,just continue here.

Thanks. Originally I assumed that folks would be finding my question on the Editing discussion page, which is why I posted it there. I had tentatively concluded that it was a mistake my posting this question in the newbie section. But I've been getting a perfectly good response here, and no response there, for whatever reason. So focusing on the one thread here will make my life easier.

Forum rules prohibit members from creating multiple threads on essentially the same subject for that very reason. If in the future you feel that you have placed your thread in the wrong sub-forum, you can PM a moderator to have it moved.

[criggs]

Forum rules prohibit members from creating multiple threads on essentially the same subject for that very reason. If in the future you feel that you have placed your thread in the wrong sub-forum, you can PM a moderator to have it moved.

Sounds good, thanks.

[TimA-C]

1. VideoReDo is basically king in this regards.
2. TMPGEnc MPEG Smart Renderer 5 is very good and also has a commercial candidate detector optional plugin

I just went to the VideoReDo 90% convinced I would be trying that out first on a free trial basis.

Which is when I discovered that the version that can handle both h.264 and mpeg 1/2 is not $50, as I first thought, but actually $96.

So I probably owe it to myself to try the cheaper $70 Smart Renderer 5 ap first. Who knows? If I'm happy with the cheaper SR 5, there may be no need for me to try the VideoReDo ap. So I think that's what I'll do.

Oh for God's sake, just download both trial versions and try them yourself - why all the drama? Why don't you make this one an informed choice?

Just to echo another post here, the support for VideoRedo through their forum is way, way better than TMPGenc's - I have VideoRedo as well as a number of TMPGenc's products, and TMPGenc's support is virtually nil whilst VideoRedo's authors are very active on their forum and have sorted out a number of issues I've had with recordings from a number of devices over the years. (I'm not connected with either company, other than being a paying customer of both of them.)

Here's another thought, you couldn't find a direct comparison of these two products, so there probably are/will be others in the same position. Why don't you try them both and write that comparison yourself?

Anyway, good luck with your deliberations. Let us know how you get on.

[criggs]

Oh for God's sake, just download both trial versions and try them yourself - why all the drama? Why don't you make this one an informed choice?

I don't feel I'm in a position to make an informed choice. Just spending a day with one free trial of one ap, and then a day with another free trial of another ap, really doesn't cut it in my book. It doesn't compare to the experience and knowledge of those who've been using one or both of these aps for an extended period of time, sometimes years.

The fact is I'm reasonably proficient at learning stuff like this, so I'm fairly sure I can get both to work for me. The problems associated with a particular ap frequently don't emerge right away. Quirks that directly impinge on one's ability to get a specific job done are frequently not discovered without extended and constant use. Many folks here have already had those experiences, and have walked that path; I haven't, which is why I put much store in the reports of those in this forum.

Just to echo another post here, the support for VideoRedo through their forum is way, way better than TMPGenc's - I have VideoRedo as well as a number of TMPGenc's products, and TMPGenc's support is virtually nil whilst VideoRedo's authors are very active

Now that's good to know, thanks.

Here's another thought, you couldn't find a direct comparison of these two products, so there probably are/will be others in the same position. Why don't you try them both and write that comparison yourself?

Sorry, I don't have that kind of time; I need to make the decision now.

Anyway, good luck with your deliberations. Let us know how you get on.

Well, other than the mp4 quirk, Smart Render 5 seems to be working for me so far. If I have a choice between taking out a minute to make an mp4 copy of anything I wish to edit and smart render or of paying an additional $26, there's no question in my mind I'll accept the mp4 two-step. If that's the only problem with SR5, I'm good to go and I'm going ahead with the free trial on that ap. VideoReDo will have to wait unless I bump into something on SR5 that's a deal-killer. After all, I need to get to work here.

[TimA-C]

Oh for God's sake, just download both trial versions and try them yourself - why all the drama? Why don't you make this one an informed choice?

I don't feel I'm in a position to make an informed choice. Just spending a day with one free trial of one ap, and then a day with another free trial of another ap, really doesn't cut it in my book. It doesn't compare to the experience and knowledge of those who've been using one or both of these aps for an extended period of time, sometimes years.

WHAT? You're the ONLY one in a position to make an informed choice FOR YOU. You try an app, does it do what YOU want it to do with YOUR files? Do YOU find it easy to use & understand? Are the instructions & help useful and make sense to YOU? Yes, fine, read the reviews and ask for comments and recommendations, and Absolutely you take them on-board, but ultimately it's got to work for YOU, otherwise it's a waste of YOUR time and/or money. A lot of people rate Sony Vegas as a excellent video editor, I've used it on occasions and I hate it. I'm not saying that it's not a very good program but I struggle with it because I don't find it remotely intuitive. I don't have those issues with the various versions of Adobe Premiere or Corel VideoStudio etc. that I've used. Back in the day I got on quite happily with the then Ulead Video Studio programs but struggled with Pinnacle's alternative. I'm sure there are many who would say the exact opposite.

Here's another thought, you couldn't find a direct comparison of these two products, so there probably are/will be others in the same position. Why don't you try them both and write that comparison yourself?

Sorry, I don't have that kind of time; I need to make the decision now.

The time you've wasted wringing your hands and proclaiming yourself unworthy to have an opinion, as well as writing that last post, you could have tried the two programs and written just a few lines describing your initial impressions of each program, listing the things you liked and disliked about both of them, and then giving a single paragraph conclusion. I wasn't expecting a 300 page step-by-intricate-step manual with diagrams, screen caps and links to a comprehensive video tutorial in 50 easy-to-follow hour long clips!

Try both of them. You may find that the decision is REALLY easy for you as a result. Or you may find that they're both adequate for what you want and both equally easy to use, in which case go for the cheaper one, but at least YOU'LL know and you won't be wondering if the few dollars you saved weren't a massive false economy.

[johnmeyer]

You can lead a horse to water ...

[criggs]

Well, other than the mp4 quirk, Smart Render 5 seems to be working for me so far. If I have a choice between taking out a minute to make an mp4 copy of anything I wish to edit and smart render or of paying an additional $26, there's no question in my mind I'll accept the mp4 two-step. If that's the only problem with SR5, I'm good to go and I'm going ahead with the free trial on that ap. VideoReDo will have to wait unless I bump into something on SR5 that's a deal-killer.

Well, I did just bump into something that may be a deal-killer with Smart Renderer 5. I bumped into a clip, an flv file, which, for some mysterious reason, Smart Renderer 5 cannot smart-render. Accordingly I downloaded the free trial of VideoReDo. Sure enough, VRD can handle the flv file, no problem.

However there is an important strength of Smart Renderer 5 which, apparently, VRD does not have, and that is the ability to mix and match video clips of different resolutions and utilize, as its resolution, the size presented by the clip one designates as Master. When I tried to do that on VRD, I was baffled. A search in their help file uncovered a paragraph which specifically stated it won't mix and match video clips of different sizes and will not normalize/standardize them to a particular standard/clip.

I had a vague recollection that I had read somewhere that VRD CAN do that, so this sorta surprised me.

However I'd rather solve the problem with SR5 than with VRD, because of the former's more reasonable price.

Here are the specifics of the problem I've bumped into with SR5.

When I go to Format this particular flv file, I get the following message: "There is no Smart Render-able clip in the Source. It is necessary to add one clip that can be smart rendered at least."

Here were the video stats that I got on this particular flv file in the VLC player:

Codec: H264 - MPEG-4 AVC (part 10) (avc1)
Resolution: 1280x738
Display Resolution: 1280x720
Frame Rate: 29.970389
Decoded format: Planar 4:2:0 YUV

And here were the audio stats from VLC:

Codec: MPEG AAC Audio (mp4a)
Channels: Stereo
Sample rate: 48000 HZ

When I recontain as a copy in mp4 with XMedia Recode, Windows Media Player reports that the video plays back at 1937 Kbps and, according to XMedia Recode, the audio plays back at 108 Kbps, for a combined playback speed of 2045 Kbps.

There was another peculiarity with the file. While it played back fine in VLC, it played back at double-speed in the GOM player.

I've been using Smart Renderer 5 for similar projects for about two weeks and this is the first time I've encountered this specific problem.

But I have encountered one other Smart Renderer 5 problem that may be related. I've bumped into one or two .ts files where pulling them into Smart Renderer would result in altered stats. Specifically, these were files that played back at 29.97 fps but, when pulled into Smart Renderer 5, the ap would see the files as playing back at 59.94 fps. Then, when I went to render it in Smart Renderer 5, and requested a Smart Rendering analysis, SR5 displayed the whole timeline in red, meaning it was going to re-render the whole thing, no smart render at all. ??!! That happened twice to me and I figured out a workaround. I recontained the ts files in XMedia Recode as copies in mp4 files and Smart Renderer 5 then handled the files fine.

I tried the same thing here, recontaining the original flv file in XMedia Recode as copies in an mp4 file but no dice: Still the same "No Smart Render-able clip" message in Smart Renderer 5 (and still the double-speed playback in the GOM player).

I then tried demuxing into two separate audio and video mp4 files in XMedia Recode. When I did that, GOM played back the video file at the correct speed, but was still playing back the audio file at double-speed. Pulling the video only file into SR5 got the same No Smart Render-able clip message.

I then remuxed the two files in tsMuxer; well, that fixed the double-speed playback problem in GOM (go figure!), but still no dice with SR5. I then tried recontaining the tsMuxer result as an MP4 copy in XMedia Recode. That accomplished nothing; still no dice in SR5 (and the double-speed problem then returned in the GOM player!).

Any ideas? Thanks!

The bottom line: At the moment, I'm bumping into this "No Smart Render-able clip" problem with SR5, and I'm bumping into this mix-and-match problem with VRD. Does anyone here know a solution for either of these problems? My preference is to solve the SR5 problem because of SR5's lower price tag, but I'll take anything at this point! Because if I can't solve them, I'll need to buy BOTH, which is a $166 outlay, which is way North of what I was hoping to spend.

[TimA-C]

It's easy enough to encode all the clips to the same settings in Vrd and then join them (Not entirely sure what the method is now if you add a number of mismatched clips to the joiner and try to create a new clip as it's not something I do.)

Have you tried looking at the file in MediaInfo or loading it into Vrd and pressing {Ctrl}{L} to view the stats?

[criggs]

It's easy enough to encode all the clips to the same settings in Vrd

Here's the specific situation which I would like to make work in VRD.

I have two clips. The first clip is already encoded in a form which I like, say 1920 x 1080, 29.97 fps, AAC 48k sampling, 384k. The second clip is something else (say 720 x 404, 25 fps, mp3 128k).

I don't want to take the time to go into the VRD encoding settings and laboriously and specifically decide what codec, what i-frame rate, variable rate or constant rate, etc. etc. and all that nonsense. Instead, I just want VRD to see the settings in clip 1, take a snapshot of them in their entirety, and then plug those parameters into any and all other clips being joined that do not match that set of parameters; this is, in fact, what I've figured out how to make SR5 do right now, but of course SR5 now has this other problem which has led me to experiment with VRD in the first place.

The bottom line is that I want to end up with a joined clip where the portion that is my original master clip is smart-rendered, and not encoded AT ALL (other than stuff preceding the first i-frame from the original master clip and stuff following the last i-frame in the original master clip, which, as I understand it, is the precise definition of smart-rendering).

In other words, my concern is that if I wind up giving VRD encoding instructions of any sort at all, rather than relying on VRD's snapshot ability, it will end up also encoding those portions of the final clip (drawn from my original master clip) which I don't want to be re-encoded at all, but just smart-rendered.

I hope all of that makes sense. As I've said before, I'm new to this, and I'm working somewhat in the dark at the moment.

Have you tried looking at the file in MediaInfo or loading it into Vrd and pressing {Ctrl}{L} to view the stats?

Sure, I've looked at the stats in Control L; in fact, I've noted that there is even a copy to Clipboard option in that menu. The nice thing would be if I could simply copy those parameters and paste them into my VRD encoding settings for the non-master clips that don't match the parameters of my master clip. But that's where I'm stymied; I've been unable to figure out how to do that.

[johnmeyer]

You cannot join, into a single video clip, video that is different pixel size, frame rate, or encoding container. You either have to re-encode some of your clips so that they are in a format that matches your other video, or you just keep them as separate clips, and play them one at a time.

So there is no software that will do what you are asking because the various video container formats do not permit shifting from one encoding format to another; from one pixel size to another; or from one fps to another (well, that actually can be done in some cases).

[criggs]

there is no software that will do what you are asking because the various video container formats do not permit shifting from one encoding format to another; from one pixel size to another; or from one fps to another (well, that actually can be done in some cases).

Well, actually, there is such software, TMPGEnc MPEG Smart Renderer 5, and I've been using it to do just that for a few weeks now. My understanding was that there was a way, in VRD, to do the same thing using the Joiner function. The trick seems to be to copy and paste the precise parameters of my master file into VRD's encoding instructions, and convert the mismatched clips in my project to those parameters. Well, I've gotten as far as copying the master clip's parameters. The problem is that, while I've been hearing repeated rumors to the effect that one can do such copy/pasting with VRD, I have yet to figure out how that VRD process works in terms of pasting those copied parameters into VRD's encoding instructions for the other mismatched clips. My working presumption, for now, is that someone here will be familiar with how that VRD function works.

Of course, my preference is to fix this #$%^& problem I'm bumping into with Smart Renderer 5, where a few clips, the latest being an flv file, refuse to smart render at all in SR5. If I could figure out how to correct this intermittent "No Smart Render-able clip" problem in SR5, I would love to tell VRD to go take a hike, since the Smart Renderer 5 is not only cheaper but strikes me as the smoother more user-friendly product. Of course, it strikes me as being so on the basis of a grand total of less than two days working with VRD, which is why it continues to make such gigantic sense to me to continue to solicit the opinions and feedback on these matters from the infinitely more experienced participants in this forum.

[johnmeyer]

there is no software that will do what you are asking because the various video container formats do not permit shifting from one encoding format to another; from one pixel size to another; or from one fps to another (well, that actually can be done in some cases).

Well, actually, there is such software, TMPGEnc MPEG Smart Renderer 5, and I've been using it to do just that for a few weeks now.

That, of course, is software that works with one format: MPEG-2 (and maybe MPEG-1, for anyone who cares). Each codec, and each container format are designed in ways that sometimes permit some of the things you talked about. As the simplest example, MPEG-2 permits variable bitrate, whereas a codec like DV does not. I don't fully understand what you are trying to do because I don't have access to each and every clip you are attempting to join. However, my original statements still stand when it comes to the general discussion of all video containers and codecs, especially H.264 implemented as an MP4: you may very well not be able to accomplish your goal no matter what software you choose.

[criggs]

That, of course, is software that works with one format: MPEG-2 (and maybe MPEG-1, for anyone who cares). Each codec, and each container format are designed in ways that sometimes permit some of the things you talked about. As the simplest example, MPEG-2 permits variable bitrate, whereas a codec like DV does not. I don't fully understand what you are trying to do because I don't have access to each and every clip you are attempting to join. However, my original statements still stand when it comes to the general discussion of all video containers and codecs, especially H.264 implemented as an MP4.

Hmmm. My current experience would tend to suggest otherwise, at least as far as Smart Renderer 5 goes.

My first clip, my master clip, was an h.264 mp4, with a resolution of 1920 x 1080 and a frame rate of 29.97. My second clip was an MPEG-2 ts file, with a resolution of 1280 x 720 and a frame rate of 59.94. I put them both into SR5, assigning the 1080 clip as the master. Smart Renderer 5 had no problem rendering the two together as one mp4 clip and re-encoding the second clip to match the first. If you're curious, I'm happy to post each file's full stats here if you wish to examine them closely, but so far, I've had no problem mixing and matching any combination you can think of into one clip using Smart Renderer 5.

Except when I can't even get Smart Renderer 5 to get to first base with a file, before I even try to mix and match, like this flv file which works fine in VRD but where I get this "No Smart Render-able clip" message on Smart Renderer 5. So, instead of trying to get VRD to mix and match with the same power as SR5 can do, does anyone have any ideas about how to get around this "No Smart Render-able clip" error message in SR5?

[johnmeyer]

Hmmm. My current experience would tend to suggest otherwise, at least as far as Smart Renderer 5 goes.

My first clip, my master clip, was an h.264 mp4, with a resolution of 1920 x 1080 and a frame rate of 29.97. My second clip was an MPEG-2 ts file, with a resolution of 1280 x 720 and a frame rate of 59.94. I put them both into SR5, assigning the 1080 clip as the master. Smart Renderer 5 had no problem rendering the two together as one mp4 clip and re-encoding the second clip to match the first.

Well, if you re-encode, then you can certainly combine any type of video clips, without any restriction whatsoever.

However ...

... the title of this thread is " Editing without Recoding" (my emphasis). Not only is that the title, that has been the main topic of conversation as the thread progressed. So, if you are now content to be rendering (and there is nothing wrong with that, other than the loss of quality which sometimes can be minor, but other times can be huge), then everything I have said, and everything most other people have said in every post in this thread is utterly pointless and useless.

If you are willing to re-encode clips, then you can use almost any competent NLE, and I would suggest you use one of them rather than the two programs you have been using. The two programs you have been using (VideoRedo and Smart Renderer 5) exist mainly to do what you originally said you want to do (edit without re-rendering) but, because of the restrictions that imposes, and also because of the nature of the situations where most people want to do this cutting without re-encoding (mostly for removing commercials from DVR captures) they have a fairly specific and somewhat unusual feature set. You can do a LOT more with a competent and fully-featured NLE.

So, I would recommend Vegas (the cheap version), Studio, or some other low-end NLE.

[usually_quiet]

@criggs Initially you asked for a smart encoding editor to work with files recorded by an ATSC DVR. Now you have added the requirement that it must also be able handle combining random video files with different resolutions and frame rates all by itself.

VRD is a product used primarily for editing files created by devices which capture the transport stream output from digital TV tuners. It was recommended to you because it does a great job with those sources. It also retains closed captions in parts of the video that are simply copied, which is a nice feature if someone wants or needs them. It can work with other kinds of files as input, but most here would not recommend it as an all-purpose editor.

Get something else if you intend to use the editor primarily to combine all kinds of media files containing any kind of video and audio and don't want to make any decisions about encoding. ...but such editors don't always do good a job editing glitchy recordings from over-the-air sources and are likely to remove closed captions. I tried TMGEnc Smart Renderer 5 today. It handled the minor glitches in the recording that I edited (an H.264 TS file from my CableCARD tuner) well enough, but stripped the closed captions.

[criggs]

Well, if you re-encode, then you can certainly combine any type of video clips, without any restriction whatsoever.

Unfortunately you've misunderstood. THE FIRST CLIP WAS NOT RE-ENCODED AT ALL. THAT'S THE WHOLE POINT. THE CLIP ASSIGNED AS THE MASTER CLIP IN THE SR5 DIALOGUE IS THE REFERENCE FOR THE ENCODING OF THE SECOND CLIP. BY DEFINITION, THE MASTER CLIP IS THE ONE THAT IS SMART-RENDERED.

A picture is worth a thousand words. Here's the screen grab of the SR5 Smart Rendering Analyzer, which tells the user which portions of the final clip being requested will be smart-rendered (i.e. not encoded but merely copied, for the most part) and which portions will be completely re-encoded.

Note that those portions which will be smart-rendered are shown in grey and those portions which will be re-encoded are shown in red. (Note that the second clip had two audio tracks while the master clip had only one. SR5 resolved that by leaving the audio track of the master alone (of course) but creating a second track from scratch for the master clip which was an encoding match for the second track of the second non-master clip, which it left alone and smart-rendered. Meanwhile it did the opposite for the master clip's single audio track, smart-rendering it but then re-encoding the first track of the second non-master clip to match the parameters of the one and only audio track of the master clip.)

The bottom line here: There is no confusion here. I wish to use smart-rendering in order to preserve quality and downgrade the quality of the original video as little as humanly possible. That is precisely what SR5 is doing for me in this example. The problem is not what SR5 is doing; it is doing exactly what I want it to be doing in this example. The problem is that it sometimes intermittently gets super-finicky about certain files for reasons which totally escape me, as, for example, it did with that flv file which triggered the current exchange. When it gets a finicky attack, it screams that there is "No Smart Render-able clip", and when that happens I have yet to come up with a solution. That is, hopefully, where folks in this forum can come in.

Or, at least, that's my first choice. If there is no solution to this SR5 problem, that's O.K. if we can figure out how to encode non-matching clips to match my master clip with VRD. If we can do that, then I'm happy to pay the extra $26 for VRD. But my first choice is to solve this "No Smart Render-able clip" problem, if we can, because SR5 is cheaper.

The only solution that I've come up with so far is to use both, meaning I have to spend $166. Well, before I do that, I'd really like to hear from some more experienced people here about other possible cheaper solutions to these two problems, preferably one for the SR5 glitch but, if not, then for the VRD encoding conundrum.

I hope that all makes sense, and thanks for any suggestions anyone may have to solve one or both of these problems.

[johnmeyer]

+1 usually_quiet.

As for the OP, I've tried to help, but I guess it hasn't been useful. Perhaps I have misunderstood, but even after reading the last post, I really think it is the OP who is confused. However, if he can get everything to work the way he wants, that is all that matters.

[criggs]

I really think it is the OP who is confused.

Where is my confusion? I want this stuff to be smart-rendered, insofar as possible. And that is happening with both VRD and SR5. This thread is focused on the problems I'm encountering while performing that process, specifically problem A with SR5 and problem B with VRD. Problems A and B do not specifically relate to the smart-rendering process itself. In the case of both SR5 and VRD, with most files I am achieving my goal of smart render.

Problems A and B instead relate to problematic situations that don't directly impact the smart-render process, namely, in the one instance, the outright rejection of certain files which is preventing me from getting even to first base and, in the other instance, the inability to properly encode files which I do NOT want to smart-render so that they match files I DO want to smart-render.

I ask again, where is this confusion on my part to which you refer, please??

However, if he can get everything to work the way he wants, that is all that matters.

Well, if I was getting everything to work the way I wanted, I'm guessing I would probably not be posting these questions and requests for help! You think?

[criggs]

@criggs Initially you asked for a smart encoding editor to work with files recorded by an ATSC DVR. Now you have added the requirement that it must also be able handle combining random video files with different resolutions and frame rates all by itself.

Guilty. I guess I got spoiled by Smart Renderer 5's ability to do that, and thought that VRD was an ap that supposedly could do everything that Smart Renderer 5 could do.

VRD is a product used primarily for editing files created by devices which capture the transport stream output from digital TV tuners. It was recommended to you because it does a great job with those sources.

And, as far as that goes, I have found that it indeed performs very well, yes.

It also retains closed captions in parts of the video that are simply copied, which is a nice feature if someone wants or needs them.

Actually, where my work is concerned, the matter of closed captions is not important, fortunately.

Get something else if you intend to use the editor primarily to combine all kinds of media files containing any kind of video and audio and don't want to make any decisions about encoding.

Just to clarify, it is not just that I don't want to make decisions about encoding, although that is certainly a time-saver. Primarily this is about my wanting to retain video in its first-generation state as much as possible, which is why I want to be able to mix and match, on the one hand, master video clips where I would like to retain the original video and audio with, on the other hand, video clips where I want to alter the original audio and video to match those master clips.

I tried TMGEnc Smart Renderer 5 today. It handled the minor glitches in the recording that I edited (an H.264 TS file from my CableCARD tuner) well enough, but stripped the closed captions.

Fortunately that's not a deal-killer for me. What IS a deal-killer is this intermittent "No Smart Render-able clip" message, for which I'm still hoping there's a solution out there.

Alternately, I hope there's a means out there, other than SR5 (because of this intermittent SR5 problem), of re-encoding certain clips so they will match master clips that I do NOT want to re-encode.

[poisondeathray]

Supposedly cyberlink powerdirector. Basically it's a simple NLE with a timeline which can smart render AVC. I've never tried it

But I know for 100% certain you will find clips that won't work in some or all programs. Even if you bought all of them . They don't support all AVC profiles or settings, nor do they claim to. For example 10bit AVC, different chroma subsampling variants. And all programs will have problems with editing VFR . Those programs are more likely to work for the generic "garden variety AVC" . But even in that subset I have examples which cause videoredo and solviegmm problems. Pros/cons for each of them

Earlier you said you wanted "So I am looking for the best Smart Rendering-capable editor that can handle the largest possible number of video codecs" - is that no longer a criteria ? So far you're looking at 2 video codecs. SR5 supports HEVC, but videoredo / solviegmm do not. Want to smart render WMV/VC-1 . Nope, videoredo and tmpg can't, only solveigmm can... and so forth ....

Another problem is "smart rendering" can sometimes be "bad" - in the sense those few re-encoded frames in the GOP will be noticably lower in quality. Especially in content that has a lot of noise or grain. So you can get this fluttering in quality. If you do it manually you have more control. Yes videoredo has intelligent re-encode options and can use x264, but people can struggle with trying to match the quality on some sources

[TimA-C]

I think Criggs is saying that he would like Vrd to pick up the settings of, say, the first clip added to the joiner, and for Vrd (or whichever program) to then encode any subsequent clips that don't match that first clip using the settings of the first clip. Or even just some easy way of copying the stats of one clip and then pasting those stats into a profile so that they're used to encode subsequent clips so that they match the first clip.

As I said earlier, I'm not sure what the Joiner in VRD does these days with non-matching clips as it's not something I do. Criggs, you'd probably be better off asking in the VRD forum. Even if none of us can help you directly, I'm sure that Dan or Dan (members of the team who wrote VRD) can give advice.

[criggs]

Supposedly cyberlink powerdirector. Basically it's a simple NLE with a timeline which can smart render AVC. I've never tried it

Sounds interesting, thanks. I took a quick look at their videohelp page. They have a rating of 6.1 out of 10, based on 45 votes. The four comments unfortunately are not very encouraging.

Earlier you said you wanted "So I am looking for the best Smart Rendering-capable editor that can handle the largest possible number of video codecs" - is that no longer a criteria ? So far you're looking at 2 video codecs.

Actually you just helped make my earlier point, which is that one day or one week of a free trial simply can't compare, in terms of valuable knowledge gained, with the feedback and information the far more experienced members of this forum can provide me in a discussion thread. Just because I've only mentioned two codecs, h.264 and mpeg-2, should not in any way be taken as a basis for concluding that those are the only codecs with which I expect to work. On the contrary, I want to be prepared for pretty much anything, which is why, at the moment, strictly as a smart-rendering ap, VRD is doing slightly better for me than SR5. It's a pity that this capability of mixing and matching clips is also something that would be helpful for me or I would probably have pretty much decided to go with VRD.

Another problem is "smart rendering" can sometimes be "bad"...Yes videoredo has intelligent re-encode options and can use x264, but people can struggle with trying to match the quality on some sources

I'd settle for any kind of matching at this point, as long as I don't have to sit there and laboriously feed in thirty or forty parameters for my encoding settings.

[usually_quiet]

As I said earlier, I'm not sure what the Joiner in VRD does these days with non-matching clips as it's not something I do. Criggs, you'd probably be better off asking in the VRD forum. Even if none of us can help you directly, I'm sure that Dan or Dan (members of the team who wrote VRD) can give advice.

I just tried adding dissimilar clips to the joiner list, so I know what the joiner does with them. It refuses to add anything which isn't similar enough to the clips that are already in the list. The video codec must match (i.e. all clips must be H.264, or all must be MPEG-2 video, etc.). The video resolution must match (i.e. all must be 720x480, or all must be 1980x1020, etc.). The audio sampling rate must match (i.e. the audio must be sampled at 48.0 KHz for all clips). Apparently bit rates can be different for both audio and video. ...and I know GOP length can be different, because it varies for many of the files I edit.

This stuff is in VideoReDo's Manual (Help->Contents->Contents Tab->Getting Started -> Joining Programs).

[criggs]

I think Criggs is saying that he would like Vrd to pick up the settings of, say, the first clip added to the joiner, and for Vrd (or whichever program) to then encode any subsequent clips that don't match that first clip using the settings of the first clip. Or even just some easy way of copying the stats of one clip and then pasting those stats into a profile so that they're used to encode subsequent clips so that they match the first clip.

PRECISELY!!! THANK YOU!!! By George, you've got it!

As I said earlier, I'm not sure what the Joiner in VRD does these days with non-matching clips as it's not something I do. Criggs, you'd probably be better off asking in the VRD forum. Even if none of us can help you directly, I'm sure that Dan or Dan (members of the team who wrote VRD) can give advice.

Thank you for the suggestion. I have followed up on it. I have posted my question in the VRD forum at http://www.videoredo.net/msgBoard/showthread.php?36042-Re-Encoding-Files-to-Match-Prim...-File&p=124342 .

In addition, for good measure, I've posted my other question in the Smart Renderer 5 forum at http://bbs.pegasys-inc.com/bbs/list/lang/en/board/TMSR5/messid/70555 .

Hopefully, between those two messages, someone will have an answer for one or the other problem. Hoping for the best.

[criggs]

I just tried adding dissimilar clips to the joiner list...It refuses to add anything which isn't similar enough to the clips that are already in the list.

Yep, that's what I found out also. The way to solve this in VRD, apparently, is to somehow first pre-encode the unmatching files before doing your joining. The goal is to pre-encode them with parameters which match those files that are already in the Joiner. In my case, I would first put into the Joiner those files which have parameters of which I approve. Once I've done that, the trick then is to pre-encode new copies of the other files so they match the files already in the Joiner.

And that's where I'm stuck. I've yet to figure out how to grab a file's parameters and then tell the VRD encoder to use those parameters and encode the other files to match. I have a hard time believing there's no way to do that. And considering that SR5 does precisely that (albeit with an overly limited selection of files), I'm also in the hunt for another piece of software that also does that, even if VRD is not capable of doing that.

[poisondeathray]

Supposedly cyberlink powerdirector. Basically it's a simple NLE with a timeline which can smart render AVC. I've never tried it

Sounds interesting, thanks. I took a quick look at their videohelp page. They have a rating of 6.1 out of 10, based on 45 votes. The four comments unfortunately are not very encouraging.

Those ratings are probably dealing with the editing /NLE part, not the actual smart rendering

But that would be the ideal thing that everyone wants - A NLE that smart renders everything and is stable. This means you could do real editing - overlays, effects, transitions, and only the parts that were untouched get passed through. But if you look at their forums, it seems very picky about what it seems to do successfully in terms of smart rendering . The fact is there are way too many configurations for software to support. MPEG2 is mostly "doable" because fewer varieties. AVC is a different ballgame with much higher complexity . Even mature programs like videoredo can still have problems with AVC types

Earlier you said you wanted "So I am looking for the best Smart Rendering-capable editor that can handle the largest possible number of video codecs" - is that no longer a criteria ? So far you're looking at 2 video codecs.

Actually you just helped make my earlier point, which is that one day or one week of a free trial simply can't compare, in terms of valuable knowledge gained, with the feedback and information the far more experienced members of this forum can provide me in a discussion thread. Just because I've only mentioned two codecs, h.264 and mpeg-2, should not in any way be taken as a basis for concluding that those are the only codecs with which I expect to work. On the contrary, I want to be prepared for pretty much anything, which is why, at the moment, strictly as a smart-rendering ap, VRD is doing slightly better for me than SR5. It's a pity that this capability of mixing and matching clips is also something that would be helpful for me or I would probably have pretty much decided to go with VRD.

Well videoredo only supports MPEG2 (and maybe MPEG1) , and AVC. If they do decide to eventually support newer formats (e.g. HEVC is a big one coming up, as is VP9/10/AV1 ), it will be a paid upgrade.

Also, in the context of mixing an matching clips with different characteristics such as resolution, frame rate etc... - although it might be "easier" doing it "automatically", letting a program decide for you is the surest way for lower quality . Scaling algorithms , FPS changes for example are usually not ideally done . They usually use the fast, non ideal methods. There can be problems with interlaced vs. progressive etc.. chroma issues etc...as well. Dedicated NLE's can be guilty of this too. If the entire exercise was to retain original quality of the main clip - you likely just took a dive on the other sections. If quality was the most important factor, IMO it would have been better not to use the smart renderer and specify your own encoding , manipulations and joining. You still pass through the main segment, but you have more control over the fluttering quality of the GOP re-encode, and importantly of the completely re-encoded sections as well.

I'd settle for any kind of matching at this point, as long as I don't have to sit there and laboriously feed in thirty or forty parameters for my encoding settings.

It depends what you deal with. For most generic typical source , videoredo will work with 99.9% of AVC videos. But there is that 0.1% where it will either outright fail, or have problems like glitches at the seams . It's very reliable for MPEG2 sources, but not perfect for AVC . I just wanted you to be aware just because tmpgenc "failed" on a certain source, be prepared that videoredo can and willl as well.

[poisondeathray]

For your FLV that "failed" in TMPGEnc Smart Renderer, can you post the mediainfo report (view=>text)
https://www.videohelp.com/software/MediaInfo

You said videoredo handled it, but what operations did you do in videoredo ?

[criggs]

it might be "easier" doing it "automatically", letting a program decide for you is the surest way for lower quality ...If the entire exercise was to retain original quality of the main clip - you likely just took a dive on the other sections. If quality was the most important factor, IMO it would have been better not to use the smart renderer and specify your own encoding , manipulations and joining.

Well, I do have Adobe Premiere Elements, and I'm very happy with it for what it does (no smart rendering as far as I can tell, plus a full complement of transitions (dissolves, wipes, text, etc. etc.)). So perhaps my mixing and matching should dump the concept of smart rendering completely.

videoredo will work with 99.9% of AVC videos. But there is that 0.1% where it will either outright fail, or have problems like glitches at the seams . It's very reliable for MPEG2 sources, but not perfect for AVC . I just wanted you to be aware just because tmpgenc "failed" on a certain source, be prepared that videoredo can and willl as well.

So I guess the question is whether it's just a random accident that I bumped into this file which Smart Renderer 5 can't handle but have not yet bumped into a similar problem with VRD. Am I A) likely to bump into just as many rogue files, over the long-term, with VRD as with SR5? Or will B) VRD always have a slight edge over SR5 in that regard?

If the former, A), then that is a strong argument for SR5, inasmuch as it is both cheaper than VRD and has this mix-and-match capability which VRD apparently lacks.

If the latter, B), then I might be better off dumping the mix-and-match capability of SR5, going with VRD, and simply accepting the fact that going forward my mixing and matching will be done Premiere Elements-style without smart-rendering.