Mpeg vs AVI Resolution

[jchunter_2]

Coalman,
re your post in the top sticky:

See http://phpbb.ulead.com.tw/EN/viewtopic.php?t=861 for some tests that show that AVI and Mpeg2 video files show no degredation in resolution after 8 edits and re-renderings.
John

[thecoalman]

Hello JC,

I'm not sure if you missed my point and I'm certainly not putting down your method. Whatever works for you is how to do it. I had been meaning to create some examples of exactly what I mean in the future, it just got bumped up a few days. This is an extreme example considering the bitrates and the source I'm using. The lights are moving so it best illustrates my point. Also note tha I'm using JPEG images for my examples so they will not be entirely accurate since they themselves are lossy but should be good enough for these examples.

My source video is a DV-AVI it's a very high quality source from a Canon GL2. Results from a lower quality source may mask the results. I encoded the AVI to both 8000CBR and 3000CBR. I also reencoded the 8000 CBR to 3000CBR. I'm using MSP not VS8.

Here'a a screenshot of the AVI:


The 8000CBR MPEG, notice the edges of the center light produce a little macroblocking. Very little and would be imperceptible when it's playing:


The 3000CBR MPEG encoded from the AVI, Lot's of macroblocking:


The 3000CBR MPEG encoded from the 8000CBR MPEG, a real lot of macroblocking:


As you can see the the mpeg reencoded from the mpeg shows signifgantly more macroblocking compared to the one encoded directly from the AVI. Encoding from the best source always produces a better result, if your just cutting and trimming the ends, adding a few transitions and most of the video doesn't have to be reencoded that just may be the way to do it. It's definitley faster. Personally I'm a perfectionist and prefer to have the highest quality video I can get.

Here's links to the videos (10 Secs.) the example frame above is at 7 secs. 2nd frame:

Right Click and select "Save As"

4MB 3000CBR from AVI
4 MB 3000CBR from MPEG
10MB 8000CBR
36MB AVI

[rwindeyer]

Personally I'm a perfectionist and prefer to have the highest quality video I can get.

I agree totally. As you say, whatever works for the person concerned is fine, and I don't want to put anyone down or say that one method is the only way. However from an instinctive and purist viewpoint I always capture to DV, edit, render to DV, then export to another package for the final transition to mpeg and burning. Never had a problem.

[GeorgeW]

Personally I'm a perfectionist and prefer to have the highest quality video I can get.

I agree totally. As you say, whatever works for the person concerned is fine, and I don't want to put anyone down or say that one method is the only way. However from an instinctive and purist viewpoint I always capture to DV, edit, render to DV, then export to another package for the final transition to mpeg and burning. Never had a problem.

I also take the DV .avi approach for capture/edit. Depending on the amount of editing, you might be able to improve your encoding results on transitions/effect by going from your editing timeline straight to an mpeg file (that would also depend on the quality of the mpeg encoder in your editing package). The reason is to save the encoding of the transitions/effects to dv .avi, then re-encoding that to mpeg.

Alternatively, you could export the timeline to a "lossless" codec like Huffy -- but that takes up alot more space.

Of course, exporting to dv .avi allows you to backup the entire video to tape (something I do most of the time -- I still have a few videos I need to get off my computer...).

[jchunter_2]

There is no question that any compression algorithm will display defects if pushed to the limit (e.g., too low a bitrate) and Mpeg2 is no exception.

However, my interest lies in whether video compression will affect me in a practical way when editing or archiving and reconstructing my projects. If I can't see any difference after 8 or 10 edits and re-renderings (with sufficient bitrate), then that is good enough for me.

Users with slower computers should definitely capture and edit in AVI because of the extra stress that mpeg2 encoding places on the computer system. Users with faster computers have a choice.

Here is another test using moving video - I was unable to find it in this forum using our "crack" forum Search function. I hope that you will download this video - it is really a lot of fun and has extremely high resolution.

MOVING VIDEO TEST

Conclusion: There is no visible difference in video resolution when capturing direct to mpeg2 or AVI Type I (DV).

Test Procedure:
I downloaded the a moving test video from http://www.streamcrest.com/mp4.shtml and created the uncompressed AVI file. This is a very jazzy movie that really tests any thing that encodes, decodes, and displays video. I encourage everyone to try this at home.

I used VS8.01 to convert the uncompressed file to “DV format” (A.K.A. Type 1 AVI), which I exported to the camcorder and then captured back to the computer, first as AVI Type 1, then as Mpeg2. Then I compared the two capture files using Windows Media Player.

The results: After watching these two files about 20 times, I see no measurable difference between the Mpeg and the AVI capture files. If there are differences, they are smaller than can be detected using this test video.

Some Statistics:
Uncompressed AVI file: 264 MB, Bitrate: 30,345 Kbps.
NTSC drop frame (29.97 fps)
Microsoft AVI files
24 Bits, 720 x 480, 29.97 fps
Lower Field First
Uncompressed
Interleave audio for every 15 frames
DV[Moving Video Test]
Audio -- NTSC, 48.000 kHz, 16 Bit, Stereo

“DV” format - Type 1 AVI: 30.5 MB
NTSC drop frame (29.97 fps)
Microsoft AVI files
24 Bits, 720 x 480, 4:3, 29.97 fps
Lower Field First
DV Video Encoder -- type 1
DV Audio -- NTSC, 48.000 kHz, 16 Bit, Stereo

Captured Type 1 AVI: 33.5 MB
NTSC drop frame (29.97 fps)
Microsoft AVI files
24 Bits, 720 x 480, 4:3, 29.97 fps
Lower Field First
DV Video Encoder -- type 1
DV Audio -- NTSC, 48.000 kHz, 16 Bit, Stereo

Captured Mpeg2: 8.1 MB
NTSC drop frame (29.97 fps)
MPEG files
24 Bits, 720 x 480, 29.97 fps
Lower Field First
(DVD-NTSC), 4:3
Video data rate: Variable (Max. 8000 kbps)
Audio data rate: 224 kbps
MPEG audio layer 2, 48 KHz, Stereo

[thecoalman]

Conclusion: There is no visible difference in video resolution when capturing direct to mpeg2 or AVI Type I (DV).

There is a difference, although my second example wasn't captured directly to MPEG there is definitely a degredatrion in the quality of the video. This is from a miniDV cam so it wasn't actually captured but transferred, it's lossless going from cam to computer.

Closeups:
AVI:

MPEG:


This is irrelevant because the AVI footage has to be encoded to the MPEG which will give us the same results. So if we were to take this footage either encoded from the AVI or captured directly to MPEG we would have identical quality if burned directly to disc providing you use the same method.

Back to my point though.... Any parts of the MPEG will be degraded where you make any edits, transitions, etc. Let's take for example if you adjust the brightness of the clip, the entire clip has to be reencoded. Same goes for any other filters you apply.

JC, I suggest you try this: take a short high motion AVI and apply a transition where there is a lot of movement. Do the same with a MPEG, I'll bet you see a difference. Signifigant difference? Probably not but you will see a difference.

I'd like to make one other point, I see a lot of posts about A/V sync here and elsewhere. In most cases the source footage is MPEG, just something to ponder.

[Steve__A]

It has been said that MPEG2 is a format for distributing video, not editing it. I do not believe that you can do sharp frame-by-frame editing with an MPEG2 file because you are restricted by the GOP structure (with the I frames, the P frames and the B frames). The I frames contain the most information but it is not like the AVI format where each frame contains all of the information you need to draw the screen. The compression scheme used for MPEG2 records the differences in a sequence of frames so it is really like a moving target which will not stand still at all.

That being said there are many advantages to capturing video to MPEG2 format. If you are just recording shows from the TV you may not have to do any editing at all except for removing commercial breaks so the AVI format might be overkill. By capturing to MPEG2 format you use much less disk space and you can also avoid the rather lengthy process in rendering the video to MPEG2 format to create a DVD.

I consider a burned DVD to be a success if I can avoid any visual or audio artifacts on splice points, and if the audio and video still in sync enough that it doesn't look like a poorly dubbed movie from Hong Kong... The subtleties in the video examples posted in this thread are beyond my frame of reference. (I think I'll pass on the Miller Lite...)

Steve A.

[thecoalman]

I Did a little more experimenting. I took my AVI source and created a 6000VBR MPEG to be used as a source file. I split the AVI source in half and did a 4 second crossfade, so we have 1 second on either side with no cross fade. I then created a 6000VBR MPEG from the AVI.


I then did the same thing to the MPEG that I'm using as a source file. Here's the results.

The MPEG encoded from the AVI:


The MPEG encoded from the MPEG:


If you focus your attention on the light to the right you can see a slight difference. Is it huge difference? No not really, but with the combination of high quality 3CCD cams coming into the consumer market and TV's becoming larger and larger any quality loss will be noticeable.

Here's some close-ups, notice the right edge of the blue light:

The MPEG encoded from the AVI: ---- The MPEG encoded from the MPEG:







The clips themselves (4MB each):

MPEG encoded from AVI
MPEG encoded from MPEG

[jchunter]

Coalman,
Your tests constitute valuable input to the Mpeg2 vs. AVI debate. You have shown that Mpeg2 compression artifacts are visible at a bit rate of 3 Mbps and slightly visible at 6 Mbps.

Your findings are not inconsistent with my earlier tests, which were run at 8Mbps and showed no loss of resolution after 8 edits and re-renderings.

Bottom line – General guidelines for users:

1) If you have a slow machine, you must capture to AVI.
2) If you have a fast machine, you have a choice of capturing to Mpeg2 or AVI.
3) If you capture to AVI, you will need significantly more disk space and you will have to re-render to Mpeg2 before you burn.
4) If you capture Mpeg2 use a high video bitrate and maintain consistent property settings through the Edit and Burn phases.
5) When burning a DVD, if you lower the bitrate (e.g., to fit a project onto the disk) you may see more compression artifacts with Mpeg2 than with AVI.

Does this fairly summarize what we have learned from the factual evidence so far?
John

[thecoalman]

Does this fairly summarize what we have learned from the factual evidence so far?
John

Yes I think so. One thing I'd like to point out though is there is a lot of factors to consider. When I do a project each one is different. I consider the source and what I want to do with it. If for example you want to fit a lot on one disc when considering a bitrate you have to consider your source. You can get sufficient quality from a regular D8 cam at about 6000VBR, this is what I have always used. You can lower it even further for VHS source (excluding S-VHS), I've had sufficent results from as low as 4000VBR ,this is about as low as you want to go for 720x480.

Also note if your doing a lot of editing particularly adding global filters to an entire clip, overlays, etc. use AVI. Editors love AVI it's meant for editing. For cutting and trimming MPEG is fine providing the software your using doesn't reencode the untouched video and works well enough to not make a huge mess out of it.

All these points I'm making pertain to me, the DV-AVI footage I have remains on tape and a second copy on disc for archiving so at any point I can retrieve the original, If the user is using there DVD MPEG to store there footage I'd reccommend the highest bitrate possible.

All of these points are going to become moot shortly once the next generation of drives come on the market. You can pretty much bet they'll be able to play DV-AVI natively without the need to encode to MPEG.