Does Adobe Encore/DVDLab Re-Encode Video?

[EViS]

I'm capturing a 15000 bitrate .mpg file from old family VHS tapes. The next step will be to remove the audio hiss using Adobe Audition. Is then doing the following going to lose me noticeable quality;

Re-Mux the clean DD2.0 audio file back to the high 15000 bitrate mpg video file using TMPGENC Plus & lower the bitrate to a DVD standard... and then author the video with a menu/titles between clips...

Or am I doing things the wrong way around here?

[guns1inger]

Re-encoding should be avoided if possible, especially when working with temporal lossy compression schemes like mpeg1/2/4.

The short answer to your question is in two parts

a) there is no loss of quality from demuxing and remuxing the audio, and if the restoration is done right, the audio should be improved.

b) There is always the risk of quality loss when re-encoding, especially with temporal lossy compression. The question is - will you notice it ?

There is nothing inherently wrong with your process. However if your running time is in the 60 - 70 minute range, you can avoid the re-encoding of the video by capturing at a constant bitrate at the higher end of DVD compliant - say 8500 - 8800 kbps - and use the video as-is.

How are you capturing - DVDR or capture card ?

[EViS]

My runtime for some videos is in the 180minutes range.
I'm capturing using a Hauppauge PVR250 via a TBC using WinTV & some registry sharpness hacks.

The questio I guess I'm asking is whether I should re-encode to ~8000kbps in TMPGENC Plus before using the Adobe packages to author the DVD's, split them up, mix them around, and put them back together again in a different order? Or should I do the splitting and mixing the order around first prior to authoring menu's and title between clips?

[guns1inger]

You will have to re-encode, and depending on whether you are using DVD5 or DVD9 will determine the bitrate you need to re-encode to (DVD5 = approx 3600 kbps with 224 kbps audio, DVD9 = approx6550 kbps).

If you are editing (as opposed to just cutting it up to author in chunks) do it first, and do it in a good mpeg2 editor. This should include adding the titles between scenes. After you have finished the edit, restore the audio. Now you can re-encode the whole thing to the correct bitrate, and author.

I would not cut the clip into smaller parts and add separate title clips unless absolutely necessary. You have more freedom if you have a single title to work from in most cases.

[EViS]

And how do I go about taking chunks out of various .mpg files and then creating a new .mpg file with all the chunks placed in the correct order? I've got bits and pieces scattered across different VHS tapes (each VHS capture = new .mpg) and I would like to put them all in the same DATE order to create a "story of my life" type DVD Video...

I would not cut the clip into smaller parts and add separate title clips unless absolutely necessary. You have more freedom if you have a single title to work from in most cases.

By that do you mean cutting a 3hour .mpg into 10 .mpg's using a cutting tool and then importing all 10 .mpg's into my authoring program to create a new 3hour .mpg with the clips re-arranged? Meaning it's best to do all the cutting and moving in a quality authoring program?

[edDV]

During capture with PVR-250, you have two choices.

1. Capture only the clips you need for the DVD. Select a legal DVD MPeg2 bitrate and be done as far as encoding is concerned. Import the resulting clips into a DVD authoring program and specify no re-encode for legal files.

2. Capture at a high bitrate with the idea clips will be selected later. Use an editing tool to issolate the clips (Womble products are good for this since they only process the frames within the GOP). If filtering or effects are needed, use a normal editor.

Now you need to re-encode the MPeg to an accepable rate for DVD.
see https://www.videohelp.com/calc

Import the clips into the authoring program and author without re-encode. Many authoring programs (e.g Encore) include an MPeg2 encoder if you want to use that for final encode.

[EViS]

Point 2 is my preferred option .

Some slight filtering may be needed, and effects (each video clip fading in and fading out prior to the next clip) will be needed. Thus instead of using Womble can clips be issolated from each .mpg in Encore or Premiere, titles be added inbetween each clip, and the whole DVD then authored and re-encoded?

Thus everything is only re-encoded once as opposed to twice - the first time being with the software used to issolate each .mpg clip (even if it is the few frams in the GOP).

Or do I have no choice but to use Womble to split the clips and then import them into Encoe or Premiere?

I guess the question I'm asking here is, can Encore (or another authoring software?) do everything that I need (summarised below) & only re-encode once?

Issolate a number of clips out of a number of separate large .mpg captures,
Put the clips in order with title screens inbetween,
Add video effects (fading video & sound in/out),
Author with a DVD Menu.

Note. By normal editor I believe you mean an authoring package like Encore or Premiere, correct?

[edDV]

Womble type editors re-encode the affected GOP (between I frames) only. If the captured bitrate is within DVD spec, the resulting edit can be authoed to DVD without further recode.

Premiere or Virtualdub type editors will decode all MPeg on import. For Premiere, while decompressed you can apply filters and titles to the timeline and play with it until you are happy. Then at the end you re-encode back to MPeg2 at the desired DVD bitrate. Then you pass that MPeg2 to Encore (and specify no recode).

You don't have to create the entire timeline before encoding. You can encode separate clips then collect the clips into the authoring program and author without recode. The authoring program may want to recode if the DVD capacity is exceeded or may try to recode unless you control it. Some authoring programs are difficult to control.

I haven't used recent versions of Encore. It does have an internal MPeg2 encoder that you want to avoid using unless necessary. Authoring programs like TMPEGEnc use an external MPeg2 encoder.

[edDV]

By normal editor I mean all that don't edit native MPeg. They will decode all MPeg to the internal processing format (usually RGB or YCbCr).

Some editors have a "smart render" feature that monitors which frames have changed during editing/filtering. When done editing, and when the timeline is encoded to the same MPeg2 spec as the input (e.g. bitrate, GOP length, etc.), then the software copies unedited GOPs from the input MPeg file to the output file for zero generation loss. Only GOPS that contain edited/filtered frames are recoded.

[EViS]

Right, I think I'm getting confused again now... Could you answer my question (in bold) with a simply yes/no in the above post please?

And why should I refrain from using Encore's internal encoder? Is it not based on the same one (forgot the name) that TMPGEnc Plus uses?

[edDV]

I guess the question I'm asking here is, can Encore (or another authoring software?) do everything that I need (summarised below) & only re-encode once?

Issolate a number of clips out of a number of separate large .mpg captures,
Put the clips in order with title screens inbetween,
Add video effects (fading video & sound in/out),
Author with a DVD Menu.

Basic answer is No. You are asking Encore to do the work of Premiere (or other editor) and Encore. The basic project flow assumes external creation of "assets" such as edited video clips with video filtering, transitions and titles in Premiere and normal practice is to encode to MPeg2(+ AC3 audio) in Premiere (or other video editor+encoder).

That said, Encore will import raw clips and allow heads and tails cuts and encode to MPeg2. It may do other things but these would be outside the normal job of a DVD authoring program.

The key concept to understand is project flow and what program does what in sequential order.

Encore's main job is to collect assets, create links, "Author with a DVD Menu" and then prepare the Video_TS folder.

Here is a free set of tutorials for Encore. Other authoring programs are similar in concept but not all include an MPeg2 encoder.
http://www.vtc.com/products/encore.htm

Encore and Premiere use the Mainconcept based "Adobe MPeg Encoder". TMPGEnc is a different encoder. TMPGEnc DVD Author is a different DVD authoring program.

[EViS]

Right, I'm confused between the role of Premiere and Encore...

I was under the impression that Encore can do all the editing, effects, titles, etc. too. However this is all to be done in Premiere, and Encore is only used to create DVD Menu's. Correct? (I sincerely do hope for both our sakes I've understood this alll now ).

[guns1inger]

Encore is an authoring tool. It can transcode, but you are usually better off doing this prior to authoring where possible. You use this to create menus and navigations structures for your disc.

Premiere is the Adobe editor. You use this to cut up and re-arrange your video, add titles, and make basic (and sometimes not so basic) adjustments to your clip, such as colour, brightness etc. It is also where you put in your transitions between clips.

[EViS]

Thankyou very much guns1inger and edDV for all your patience with me !!

Is there anything better to than premiere as an editing tool to move video clips around, etc?

[guns1inger]

Better . . . on par with . . .

It depends on what you want to do, and how you like to work. Personally, I prefer Vegas to Premiere. However if your needs are simple, both are probably overkill. Both Premiere and Vegas have more modest versions - Premiere Elements and Vegas Movie Studio respectively - that you might find easier to work with.

[GeorgeW]

If your source is mpeg (which it looks like based on your initial posts), then you might want to consider packages that can "Smart Render" unedited sections -- like Womble, and even Ulead's packages VideoStudio and MovieFactory. Vegas always re-renders your sd mpeg2 timeline -- I don't know how Premiere handles mpegs on the timeline.

Regards,
George

[EViS]

When you say re-renders, do you mean it re-encodes what-ever I import into the timeline, and everytime I cut clips up and move them around it will yet again re-encode the whole video?

I would have thought that once I import a .mpg into the timeline I can do as much editing and cutting as I desire, and only once I finish the editing tool will simply re-encode the whole thing, at which point I may as well tell it to re-encode at a lower DVD compliant bitrate... re-encoding everytime something is cut and moved around would take a VERY long time to complete a DVD video, no?

[edDV]

As said above, MPeg is decompressed for the major editors but if managed correctly, it will only recode once.

Since you are capturing at greater than DVD maximum bitrate, you will still need to recode even after using Womble or "Smart-Render" editors. High bitrate MPeg capture will achieve higher quality to the uncompressed timeline and is a good strategy if using Premeire or Vegas type editors.

The Womble/Smart-Edit techniques work well if you capture to DVD bitrate. They assure minimal frames are recoded during edits and transitions. If filters are used, all affected frames need recode anyway. Better to do that in Premiere, Vegas, Virtualdub, AVIsynth, etc.

[EViS]

Again, thanks edDV ! So renderring does not mean losing quality (i.e. re-encoding), correct?

[edDV]

Again, thanks edDV ! So renderring does not mean losing quality (i.e. re-encoding), correct?

Yes it does. MPeg has loss with every decompression and recompression. The higher the bitrate used during capture, the lower the losses will be.

Since you are capturing low quality VHS, it will be difficult to see these losses. If it was a higher quality source like consumer DVD camcorder MPeg2 or AVCHD MPeg4, the losses would be substantial.

Pro's work with minimal compressed source (e.g. DV, DVCPro, Digital Betacam, HDCAM, etc.) then edit and process as uncompressed and only encode MPeg2 for distribution. The closest consumer equivalent path is DV format acquisition and DV or uncompressed editing/processing.

[EViS]

Right... everytime I cut a video and move a section around, will this get recompressed? If I do it again to the same clip, further compression will have? Thus once I then come to re-encode the whole .mpg to a lower bitrate, one small clip in the whole .mpg would have been re-encoded a total of 3 times?

[edDV]

Right... everytime I cut a video and move a section around, will this get recompressed? If I do it again to the same clip, further compression will have? Thus once I then come to re-encode the whole .mpg to a lower bitrate, one small clip in the whole .mpg would have been re-encoded a total of 3 times?

That is why you decompress once, then edit uncompressed (or DV) and play around to your hearts content. Then when finished you encode again to DVD MPeg2. We all know done is not done so best to save a low compressed version of the timeline for future rework*. That would be ~130-160 GB/hr uncompressed (ouch), ~25-5 0GB/hr Huffyuv, 13GB/hr DV or ~7 GB/hr for 15Mb/s MPeg2.


* known as the "edit master"

[EViS]

What exactly does decompressing to a .mpg video do? And "when finished" is the video compressed again - aka. re-encoded?

If captured in say DV, does a video too need to be decompressed?