How to automatically Match original video quality attribute?

I am a wildlife videographer. I may shoot an hour's worth of video for 30 seconds of footage I want. In my old video editor that I used before switching to DR (which I love) when I imported a clip, the program would ask me if I wanted the program to automatically adjust the project settings to match those of the imported video clip (i.e. Framerate, Resolution, Bitrate, etc.) Then I could trim out just the segment of the video I wanted and save it to the same exact setting as the original file. The file would not actually have to be rendered, the program would just trim it.

In DR, it does ask if I want the project to match the Framerate of the imported clip, but it matches nothing else. Is there a way to tell DR to match all of the original clip's settings?

Not yet.

viewtopic.php?f=33&t=97260

I see a lot of posts on this forum about this and many people requesting it. I also see a lot of people being very belligerent, demeaning, and insulting toward those people for some inexplicable reason. It seems that this is a known issue, and people have drawn battle lines about it.

To me, it is a ridiculous argument. Why not make things easier to use? Since this is a feature I have seen in just about every other video editor I have used, it is not an ODD or unusual request. It is something that a lot of other people and developers think is a good thing. Why is there such hatred on this forum towards anyone that suggests it? We are not talking about changing the whole program, just adding an option for those who want to use it. I just don't understand the mentality of those that seem to hate anyone who says DR should have this option.

This what you want to do is simple trim as you don't want to change source in any way.
There are tools for it (this is not a must have feature of an NLE) and it always have its limits.
If you shoot to I frame based codec the it's rather trivial and many tools support it including Resolve (for most common codecs). If you shoot to long GOP codec (h264/5, etc.) it's more difficult (or very difficult in some cases- eg. open GOP files) to trim your footage and you always will be limited to points where video can start/end. You can't set your in-point on any frame- it has to be on a key frame, so closest GOP start. Typical files have eg 15-150 frames GOPs so in worse case your in-point has to be shifted up to length of the GOP in your file. There're tools which allow to trim to any frame and then they re-encode just a single GOP and copy rest of the file. This is quite complex feature and those tools are missing many NLE features.
For what you trying to do it's enough to use eg. QT X, ffmpeg as it allows to trim h264/5 files. There is no need to use NLE if you want simple trim. NLE is about doing many cuts, adding effects, transitions, color adjustments etc. and then rendering new file to desired format. Trimming keeping original data is a quite specific feature, which is not supported by most NLEs or partially supported by some (eg. Edius does it quite well for mpeg2, but also has its limits). You think it's easy operation, but your are very wrong (it's only easy for an I frame based formats). If it would be trivial to implement then all NLEs would already have it.

Which NLEs support what you have described for long GOP formats?
Fact that tool matches project to your file parameters doesn't mean much at all when it comes to preserving original data during export.

It is not that complicated. You just want to make it so. Just another example of the unexplainable hatred on this forum for this common feature. It boggles the mind.

Which NLEs support what you have described for long GOP formats?

I think there are two issues here.

First is the ability to simply create a timeline that matches all the specs of the clip - resolution, frame rate, PAR - when those specs differ from the Master Settings. Premiere Pro had this feature.

I think he's baffled/frustrated by those who would argue against that feature, as happened in the Feature Request thread.


Issue two is Smart Rendering.

It is not that complicated. You just want to make it so. Just another example of the unexplainable hatred on this forum for this common feature. It boggles the mind.

How much do you know about codecs and programming, so you can claim it's not complicated?
It just sounds simple, but it's far from it.

Which NLEs which you used support this feature at full extent ? I don't know any and used about all of them.

I would be very happy for all NLEs to have smart-rendering support for all codecs. For now not all even support I frame based codecs. I use this feature a lot and thankfully Resolve/Premiere support it at least for main intermediate codecs.

"Which NLEs which you used support this feature "at full extent" ?"

Listen, you can try to sneak in your restrictive qualifiers so that you can claim no answer is sufficient all day long, but nobody is fooled by it. You are unnecessarily overcomplicating the issue so you can prove yourself right. If you claim have used all NLEs and have not found one that has this feature, you are lying. It is quite common. I am sure you are justifying your claim because they did not support it "At Full Extent" but that is just a scapegoat. That argument applies to ever feature ever invented when it comes to video. If that is what you are going with, then EVERYTHING is "Too Complicated".

You're not explaining yourself well. Or you may misperceive the needed processes. And none of us know which it is.

As noted by several others ... it would be sort of possible with interframe codecs like ProRes. Though any added effects would require processing of course.

Not really possible with long-GOP codecs.

ffmpeg and some other apps can mostly kind of do this, with long-GOP media, by starting with the iframe before the "cut" point and end with the iframe after the second "cut" point.

But again, that doesn't involve processing any added effects. Just trimming the file.

So to be clear, are you asking for a process with interframe codecs, or one with intraframe long-GOP codecs?

Or both?

Sent from my SM-S908U using Tapatalk

I think the goal is to remove the 55 minutes of redundant footage that it took to get the 5 worthy minutes?
Maybe best practise here would be to edit the footage to the length you want and render out to a 444 or uncompressed dpx sequence or something that will retain quality and discard the original gop codec?

It is important to understand codecs and how cameras can get high quality footage that is actually highly compressed. It does take a lot of processing to edit from. Resolve processing will not deteriorate the footage - assuming you render to something sturdy. MPG based codecs are sort of “dehydrated”: the cpu adds the water as you watch and removes it again after but keeping that codec will lead to deterioration - it is recompressing. Like having a photocopy and photocopying the copy … leads to poor resolution very quickly.

The dehydrated analogy fits well with the idea that once you add water it is going to be a much bigger file than you put in. GOP based footage records a full frame and then for a number of frames, only records the changes against the full frame. When you “unpack it” all the pixels that were recorded only in the full frame have to be added back into the frames in between.

That said, it might work to use the “Bypass re-encode when possible” under the “Advanced Settings” tab on the Deliver page but it would have to be rendering out to the same codec and would not work if you have applied any processing like grades or titles or resizes, fades …

If you're filming the clip yourself I'd assume you have a set resolution . No? Set the defaults to match.

Unless you don't have a standard clip.

If you're filming the clip yourself I'd assume you have a set resolution . No? Set the defaults to match.

Unless you don't have a standard clip.

Thanks for giving me some simple advice.

You're not explaining yourself well.

No, I am speaking in plain English, which is more than what I can say for some others here. Just because some people have lost the capacity to communicate normally, it is not my fault.

No, I am speaking in plain English, which is more than what I can say for some others here.

Cap in hand for some advice and a great display of gratitude to the people who have taken time to offer you answers to your questions. Delivered in plain English?

You're not explaining yourself well.

No, I am speaking in plain English, which is more than what I can say for some others here. Just because some people have lost the capacity to communicate normally, it is not my fault.

You still haven't answered very basic questions that cut to the heart of your point.

Until you do, no, you aren't (as far as this English major) communicating clearly.

Clarity please. Not open statements.

Sent from my SM-S908U using Tapatalk

No, I am speaking in plain English, which is more than what I can say for some others here.

Cap in hand for some advice and a great display of gratitude to the people who have taken time to offer you answers to your questions. Delivered in plain English?

Listen, I asked a simple question. I understand you were trying to be helpful, but what you said had absolutely nothing to do with what I am talking about. Why people are trying to overcomplicate what I said is mind-boggling to me. Not to mention frustrating and a little bit insulting. It's like we are speaking two different languages. I am asking how to peel an orange and people here are trying to describe the atomic structure of a citrus atom. I am sorry if I seem to be lacking in gratitude. I don't like it when people insult me by failing to read what I wrote and hijack my thread for their own agenda.

You're not explaining yourself well.

No, I am speaking in plain English, which is more than what I can say for some others here. Just because some people have lost the capacity to communicate normally, it is not my fault.

You still haven't answered very basic questions that cut to the heart of your point.

Until you do, no, you aren't (as far as this English major) communicating clearly.

Clarity please. Not open statements.

Sent from my SM-S908U using Tapatalk

You have everything you need, you just refuse to understand because you have a dead horse to beat.

Read the manual and work it out.

Listen, I asked a simple question. I understand you were trying to be helpful, but what you said had absolutely nothing to do with what I am talking about. Why people are trying to overcomplicate what I said is mind-boggling to me. Not to mention frustrating and a little bit insulting. It's like we are speaking two different languages. I am asking how to peel an orange and people here are trying to describe the atomic structure of a citrus atom. I am sorry if I seem to be lacking in gratitude. I don't like it when people insult me by failing to read what I wrote and hijack my thread for their own agenda.

You want to extract a small part of a gop based codec to reduce your storage needs?

You were told that gop based codecs are problematic.

You said your old system (which you might use just for the purpose of reducing your storage needs) did this by matching the timeline codec to the camera codec. In DR we don’t want that because it degrades the media too much.

DR isn’t for people who don’t want to learn how it works.

In DR, it does ask if I want the project to match the Framerate of the imported clip, but it matches nothing else. Is there a way to tell DR to match all of the original clip's settings?

As a standard workflow practice, before we start a project, we sit down, take a look at all the material being used, and make a judgement call as to what's the best choice to make on Frame Rate, Image Size, Aspect Ratio, Bit-Depth, and all the other factors relating to final delivery. Sometimes, the Delivery dictates everything else: if it's going to wind up as 29.97fps for U.S. broadcast, that might change our decision to work in 23.98 or 24.00fps.

Choosing the right workflow is a pivotal decision, and I don't expect Resolve to do it for me because there's lots and lots of consideration involved. I might have 25fps PAL material, 24fps HD material, film material transferred to 29.97 standard def, film material transferred to 23.98 high-def, still frames, graphics, slo-mo shots... there's a lot of permutations and complications.

Two books I would recommend on workflow:

"Modern Post: Workflow & Techniques for Digital Filmmakers"
by Scott Arundale
https://www.amazon.com/Modern-Post-Work ... 0415747023

"The Guide to Managing Postproduction for Film, TV, and Digital Distribution"
by Susan Spohr & Barbara Clark
https://www.amazon.com/Guide-Managing-P ... 1138482811

Adobe has the free "Best Practices and Workflow Guide," and while it's designed around Premiere, the basic concepts also apply to Resolve:

https://helpx.adobe.com/premiere-pro/us ... tices.html

Read at least one of those, then armed with that knowledge, you can sit down and figure out where's the best path to follow. If it was whittling down a 4-hour clip to :30 seconds (or something), I'd render the sub-clip in place to a high-res format, then archive the original clip "just in case" after the fact. Problem solved. The alternative would be to Media Manage and either transcode to a high-res format as you go, or live with the limitations of the original format. I'm not sure that H.264 can be Media Managed (as one potential issue), but maybe it works now in 18.1/18.5. For me, we usually transcoded all those to ProRes 422HQ before media managing and archiving the show.

If you have a massive documentary with huge amounts of footage, my only advice is to buy a spitload of drives. I've worked on documentaries that literally had 400-500-600 hours of footage (easily), and that was just the stuff culled from thousands of hours of source material. Eventually it was cut down to 4-5 hours and then tweaked over time.

You want to extract a small part of a gop based codec to reduce your storage needs?

Completely and utterly wrong. Not even close. Go back and read my OP.

Is there a way to tell DR to match all of the original clip's settings?

No.

In DR, it does ask if I want the project to match the Framerate of the imported clip, but it matches nothing else. Is there a way to tell DR to match all of the original clip's settings?

As a standard workflow practice, before we start a project, we sit down, take a look at all the material being used, and make a judgement call as to what's the best choice to make on Frame Rate, Image Size, Aspect Ratio, Bit-Depth, and all the other factors relating to final delivery. Sometimes, the Delivery dictates everything else: if it's going to wind up as 29.97fps for U.S. broadcast, that might change our decision to work in 23.98 or 24.00fps.

Choosing the right workflow is a pivotal decision, and I don't expect Resolve to do it for me because there's lots and lots of consideration involved. I might have 25fps PAL material, 24fps HD material, film material transferred to 29.97 standard def, film material transferred to 23.98 high-def, still frames, graphics, slo-mo shots... there's a lot of permutations and complications.

Two books I would recommend on workflow:

"Modern Post: Workflow & Techniques for Digital Filmmakers"
by Scott Arundale
https://www.amazon.com/Modern-Post-Work ... 0415747023

"The Guide to Managing Postproduction for Film, TV, and Digital Distribution"
by Susan Spohr & Barbara Clark
https://www.amazon.com/Guide-Managing-P ... 1138482811

Adobe has the free "Best Practices and Workflow Guide," and while it's designed around Premiere, the basic concepts also apply to Resolve:

https://helpx.adobe.com/premiere-pro/us ... tices.html

Read at least one of those, then armed with that knowledge, you can sit down and figure out where's the best path to follow. If it was whittling down a 4-hour clip to :30 seconds (or something), I'd render the sub-clip in place to a high-res format, then archive the original clip "just in case" after the fact. Problem solved. The alternative would be to Media Manage and either transcode to a high-res format as you go, or live with the limitations of the original format. I'm not sure that H.264 can be Media Managed (as one potential issue), but maybe it works now in 18.1/18.5. For me, we usually transcoded all those to ProRes 422HQ before media managing and archiving the show.

If you have a massive documentary with huge amounts of footage, my only advice is to buy a spitload of drives. I've worked on documentaries that literally had 400-500-600 hours of footage (easily), and that was just the stuff culled from thousands of hours of source material. Eventually it was cut down to 4-5 hours and then tweaked over time.

Again with the college lecture. What I am saying is not this complicated. It is a feature that other NLEs have had for a LONG time. It exists because SOME people want it. Your unique, individual, particular, needs, likes, and desires are just that. I am not saying that EVERYONE has a need for this or that it should be made MANDATORY for everyone to use it. I am saying it would be a good OPTION for those that want it. Not everyone works the way you do. Why is it so hard for you guys to get that through your head?

Is there a way to tell DR to match all of the original clip's settings?

No.

Okay. See that was not so hard.

I guess the friendly, helpful forum folk just wanted you to understand.

I rarely find the forum to be combative.

"Which NLEs which you used support this feature "at full extent" ?"

Listen, you can try to sneak in your restrictive qualifiers so that you can claim no answer is sufficient all day long, but nobody is fooled by it. You are unnecessarily overcomplicating the issue so you can prove yourself right. If you claim have used all NLEs and have not found one that has this feature, you are lying. It is quite common. I am sure you are justifying your claim because they did not support it "At Full Extent" but that is just a scapegoat. That argument applies to ever feature ever invented when it comes to video. If that is what you are going with, then EVERYTHING is "Too Complicated".

I’m still waiting for that list of NLEs which support smart-rendering at full extent ( check dictionary if you don’t know what it means).
Setting project to a source parameters (fps, size etc.) is one thing (very easy), but smart-rendering is a different story. You want both and there is no single NLE which does it completely.

I think you're looking at an incidental issue here, Andrew. I think the heart of Roen's question is the Feature Request.

viewtopic.php?f=33&t=97260

I'm a wildlife videographer too. If I understand the posts correctly, this is how I trim and save clips in the same format as my camcorder in version 18.5 Using the dropdown menu on Edit page for the timeline I want to save the trimmed clips from:

<File>
<Media Management>
<Copy>
- Used media and trim keeping <30> handle frames
<Start>

I'm on Windows Pro 10. Checking the file properties in Explorer, everything is the same in the trimmed files as in the source files except for the length of the file. File size, fps, bitrate, etc.

I'm using a JVC HM170U shooting 4K 30P Bit Rate 150M

Note: if you don't check the Used media and trim keeping... selection, the entire clip is saved rather than the trimmed clips for all the clips on the timeline. Saving just the trimmed clips allows me to save just the keepers from a day of shooting, the trimmed keepers and discard all the rest of those large 4K video files.

This is the way in Resolve to make project smaller. It's not exactly what original question is about, but good 'workaround'.

Jim, your feature request is almost four years old. I can see why these issues bleed into other threads. Feature requests often get lost in the crowd. By the way, I say +1, I hope some day we get it.