Made my first DVD. What do you think?

[44echo]

I wanted to get your opinion on my process. I'm new at this and still don't know if I'm heading in the right direction. It worked, but I'm looking for suggestions.

I transferred my video via Firewire from my Sony DCR-HC30 to my PC using WinDV. The video was from a musical performance so I transferred and saved one song at a time. Currently I'm still using a 550 MHz processor, so I thought a smaller file size to work with would be better. I took each file and cropped the beginning and end to my liking with AVIEdit. The video was taken at an informal setting so there was a lot of "down time" in between songs. At this point, I had 18 avi files, each containing one song. I then encoded them to MPEG2 with CinemaCraft Encoder. I selected MPEG-2 (PS: Video + Audio) with a 8000 bitrate. Then I took the resulting mpg files and authored/burned my DVD using Nero Vision Express.

I plan to record a lot of musical performances. The one song per clip seems like a logical method, from my inexperienced view. This also makes it easy to plan how much will fit on a DVD. If I add too much, I take off the last song. Otherwise, how would I do it with one big file? Doesn't it become a guessing game as to where to split the file?

In the end, I want AC3 audio. How do I know what audio format I have? As far as size goes, how does AC3 compare to MPEG-1?

Please comment on ANY of the settings of ANY of the software titles I've used. I welcome ideas for a whole new process, as well as other software titles to try.

Thank you.

[guns1inger]

I would probably ditch Nero Vision Express in favour of something like Tmpgenc DVD Author or DVD Lab. If you go down the Tmpgenc route, consider getting the AC3 plugin as well. It only encodes two channel AC3, however if your source if DV, this is more than adequate. The main difference between AC3 and Mpeg1-Layer2 audio. AC3 is mandated in the DVD spec as compliant in all regions. Mpeg1-Layer2 is only mandated in some areas of PAL land. Use AC3 in preference wherever possible.

Look into Variable Bit Encoding (VBR) as it is a more efficient way of encoding video. CCE Basic supports 2 pass VBR. If you get DVD Lab, you will be better off producing seperate streams for video and audio. Tmpgenc will take PS or ES.

As for breaking them down - that is a matter of preference. If you have a lot of dead time in between, then this is probably the most efficient way of capturing the video. You also have the advantage that the first edit pass has been done early.

As for fitting material on a disk, where possible use bitrates to make your titles fit. Only cull if there is not other way. With the correct bitrates, you should be able to fit up to 2 hous of good quality footage on a disk.

[redwudz]

I'm guessing you mean MPEG-1, Layer 2 sound. AC3 is more DVD compliant, but 'Layer 2' is fine. AC3 is a little more compact and somewhat better quality for the size.

You can check the audio format with AVIcodec or Gspot.

I use TMPGEnc DVD Author for authoring, but if Nero works for you, good. Using smaller files and encoding individually is probably a good choice with your present computer. I assume you a not joining the files, just letting the authoring program combine them?

Yes, a large single file can be harder to calculate the end size, depending on the encode method you use.

You could also edit with Virtualdub, then frameserve the results to your encoder. Saves a few steps and some HD space.

All in all, I'd say you are doing fine with your setup.

[44echo]

You also have the advantage that the first edit pass has been done early.

Sorry, but I don't understand what that means.

As for fitting material on a disk, where possible use bitrates to make your titles fit.

Is that where bitrate calculators come in?

At what point would I notice a difference in quality? 8000 down to 7000?...6500? I realize it's subjective.


Along the lines of quality, won't more passes (multi-pass) produce better quality?

Thank you for your reply.

44E

[Mr_khyron]

Is that where bitrate calculators come in?

At what point would I notice a difference in quality? 8000 down to 7000?...6500? I realize it's subjective.


Along the lines of quality, won't more passes (multi-pass) produce better quality?

Thank you for your reply.

44E

http://www.mediachance.com/dvdlab/tutorial/bitrate.html

[guns1inger]

Mr_khyron - good call on the bitrate tutorial.

44echo - yes, quality is subjective. However if the audience is wider than yourself, try to get the best possible. The idea behind variable bitrate encoding is to get the best use of the available bitrate/space. The first pass analises the video to work out where higher or lower bitrates can/should be applied. This entails looking at the size of change between frames (as mpeg compression is primarily temporal). The second pass allocates the bitrate as it encodes. You can usually set a minimum bitrate (don't encode lower than this), a maximum bitrate (don't encode higher than this) and an average bitrate - the target bitrate for the size of disk. Some encoders (CCE SP is one) allow for more passes. Each eaxtra pass refines the analysis for allocation of bitrates. You need to remember that there is a rapid rate of diminishing returns to extra passes. Unless you have some particularly difficult footage, you are unlikely in most cases to see any difference after 3 passes, and in 98% of cases, 2 passes is sufficient. Remember - each pass takes the same approx. amount of time, so generally 2 passes takes twice as long, 3 passes three times as long etc. I have read posts from people who apply 6 passes in DVD Rebuilder, yet I have seen no proof that this produces better results than 2 or 3 passes - it just takes longer.

What I meant about editing was that by choosing not to capture certain footage at the outset, you have removed the need to edit it out later. Some people capture everything, and make all the decisions when sitting at the editor. This has the advantage of giving you more to work with, but can make editing a much longer process. If you can cull early, this is no longer an issue.

[44echo]

Now I'm confused. Here is part of the log file from Nero.

[14:40:56] ExpressUI  ------<NeroVision Project Information>--------
[14:40:56] ExpressUI  --- Project type: DVD-Video
[14:40:56] ExpressUI  --- Project video options ---
[14:40:56] ExpressUI  Video mode: NTSC
[14:40:56] ExpressUI  Aspect ratio: Automatic
[14:40:56] ExpressUI  Quality: Automatic
[14:40:56] ExpressUI  Resolution: Automatic
[14:40:56] ExpressUI  File type: MPEG-2
[14:40:56] ExpressUI  Audio format: Dolby Digital (AC-3) 2.0
[14:40:56] ExpressUI  Encoding mode: High Quality (2-Pass VBR)

The last line tells me that the video was encoded by Nero, when it wasn't, AFAIK. Only the audio was transcoded. It started with MPEG-1 Layer 1 audio (thanks to redwudz link to AVIcodec ) and is now AC-3. Is anyone familiar with these Nero log files?

Thank you all for your help.

44E

[44echo]

I assume you a not joining the files, just letting the authoring program combine them?

That is correct.

Redwudz, what do you think of TMPEGEnc DVD Author? I'm using Nero just because it came with my burner. I don't care for the menus in NVE.

Do you use AVIcodec? Does it ever tell you that your video is 16:9 when you encoded it to 4:3? I thought that was kind of weird.

[lordsmurf]

Your process will work.
I would replace Nero Vision with a better authoring software. An easy one to start with is TMPGENC DVD AUTHOR ($99 or less). But give a careful look over to Ulead DVD Workshop. The Express version is very affordable ($199). As well as Sonic DVDit! PE ($149).

[funkguy4]

along the lines of frameserving from virtualdub:
you may want to consider using avisynth if all you're doing to the video is trimming the ends, or trimming at all, since it will speed up your encode times significantly (my CCE went from ~.84 speed using VDub to ~1.05 speed using avisynth)

i'm pretty sure avisynth also has noise cleanup filters and such, and pretty much anything a DV user would need. i know it's not graphical though there are some GUIs for it (which confuse me more than just learning the commandline ). After you gain some experience or just feel comfortable moving on to something like avisynth, consider giving it a try. on the AC3 audio: since you're making an AVI from DV source (which is WAV, I believe), you may want to consider using BeSweet w/ BeSweetGUI or AC3machine (which IS easier than BeSweet commandline . ). Also, I'm pretty sure Pegasys (makers of TMPGEnc stuff) also offer a Dolby Digital Plugin for use w/ their authoring program that outputs 2-channel AC3 for yah. So, if BeSweet gives you problems w/ silent AC3s (which has happened to many), you can try this.