Doubts on first time StaxRip converted Video

[src2206]

Hello

I have converted a DVD for the first time using StaxRip. Please find the report on the converted video as follows:



The source is ripped DVD folder on the HDD. It is progressive with 16:9 display ratio.

It would be nice if someone experienced could comment on the result.

I would also like to mention that in most of the WxH combinations were giving error.

There is one more doubt in my mind. I have taken it as a backup.

• Will it produce a better quality DVD in comparison with a XviD obtained from the same source of similar file sizes (if this at all can be converted)?
• Will this mp4 output file can be played in a MP4 capable Stand Alone DVD player?

Please suggest.

Thank you.

[guns1inger]

Don't confuse mp4 with mpeg4 when referring to stand alone players. Very few will play AVC encoded material at this time. Most of the players that refer to be mpeg-4 compatible mean Xvid/Divx encoded (mpeg-4 part 2), not H264/AVC (mpeg-4 part 10).

Yes, AVC will generally give a better quality output than Xvid at the same bitrate/resolution

I'm not exactly sure what your doubts are, however personally, I believe you are squeezing a two and a half hour movie too much, even with AVC. I would encode to 1.4 - 1.5 GB mark instead.

[src2206]

Thanks very much....

You have addressed the point of misunderstanding (on my part) very correctly, I was confusing MP4 with mpeg4..so this doubt is cleared for now. So if one needs to play this H.264 files in MP4 container, a player is required which is capable of playing AVC file.

I was just experimenting with different file sizes, so chose about 1GB size as the output. Though I was wondering whether all the other settings like WxH, Audio etc are optimally chosen. Though I understand that once I increase the output size, the video bit rate will also increase.

[src2206]

Hello guns1inger

Hope you are still watching this topic....

I have recently converted a source using AGK and StaxRip, in both cases I have used 80% Quality Setting in output. Here are the reports using MediaInfo for both the sources:

XviD

General #0
Complete name        : K:\Movies\Fistful of Dollars\Fistful of Dollars (AGK).avi
Format               : AVI
Format/Info          : Audio Video Interleave
Format/Family        : RIFF
File size            : 1.76 GiB
PlayTime             : 1h 40mn
Bit rate             : 2515 Kbps
Writing application  : VirtualDubMod 1.5.4.1 (build 2178/release)
Writing library      : VirtualDubMod build 2178/release

Video #0
Codec                : XviD
Codec/Family         : MPEG-4
Codec/Info           : XviD project
Codec profile        : Unknown
Codec settings/Packe : No
Codec settings/BVOP  : No
Codec settings/QPel  : No
Codec settings/GMC   : 0
Codec settings/Matri : Default
PlayTime             : 1h 40mn
Bit rate             : 4404 Kbps
Nominal bit rate     : 2023 Kbps
Width                : 720 pixels
Height               : 304 pixels
Display Aspect ratio : 2.35
Frame rate           : 23.976 fps
Resolution           : 8 bits
Chroma               : 4:2:0
Interlacement        : Progressive
StreamSize           : 3.08 GiB

Audio #0
Codec                : AC3
PlayTime             : 1h 40mn
Bit rate             : 4424 Kbps
Nominal bit rate     : 448 Kbps
Bit rate mode        : CBR
Channel(s)           : 6 channels
Channel positions    : Front: L C R, Rear: L R, Subwoofer
Sampling rate        : 48 KHz
StreamSize           : 3.09 GiB

X.264

General #0
Complete name        : K:\Movies\Fistful of Dollars\Fistful of Dollars (StaxRip).mp4
Format               : MPEG-4
Format/Info          : ISO 14496-1 Base Media
Format/Family        : MPEG-4
File size            : 909 MiB
PlayTime             : 1h 40mn
Bit rate             : 1271 Kbps
StreamSize           : 2.78 MiB
Encoded date         : UTC 2008-03-17 00:35:22
Tagged date          : UTC 2008-03-17 00:35:22

Video #0
Codec                : H.264
Codec/Info           : H.264 (3GPP)
PlayTime             : 1h 40mn
Bit rate             : 984 Kbps
Nominal bit rate     : 965 Kbps
Width                : 704 pixels
Height               : 288 pixels
Display Aspect ratio : 2.444
Frame rate           : 23.976 fps
StreamSize           : 704 MiB
Title                : Video
Encoded date         : UTC 2008-03-17 00:35:22
Tagged date          : UTC 2008-03-17 00:36:35

Audio #0
Codec                : AAC LC
Codec/Info           : AAC Low Complexity
PlayTime             : 1h 40mn
Bit rate             : 283 Kbps
Nominal bit rate     : 288 Kbps
Bit rate mode        : VBR
Channel(s)           : 6 channels
Channel positions    : Front: L R,   Rear: L R, Subwoofer
Sampling rate        : 48 KHz
Resolution           : 16 bits
StreamSize           : 202 MiB
Title                : English (5.1)
Language             : English
Encoded date         : UTC 2008-03-17 00:36:16
Tagged date          : UTC 2008-03-17 00:36:35

Now what baffles is the Bit Rate. If I understand correctly, X.264 should produce better quality in a smaller file size, but in this case though same quality setting is applied for both; XviD seems to have produced better quality (higher bit rate) than X.264.

Or is there something tat I'm getting wrong...

Please suggest...

[Baldrick]

Try compare with same bitrate settings. H264 is not that much better than XviD and you might not even see the difference with same bitrate.

[src2206]

Hello Baldrick

Thank you very much for replying...?(I hope you have received my PM )

I have tried to set the same bit rate, and there is hardly much of a difference to the file size. And that also I think can be attributed to the use of AAC codec which in case of the XviD is AC3 (thus larger file size). At lower bit rate (the above mentioned present scenario), it appears that XviD looks more saturated than X.264, and that I think due to the higher bit rate of XviD.

I shall try to use the same bit rate as XviD and then encode, but that's gonna take some time as 2 Pass Slower quality does take a hell of a time.

[guns1inger]

If you use the same bitrate you should get the same file size, regardless of which codec you use. I have been encoding with Videora and the 3GP iPod Converter and I am not seeing any loss in saturation or similar changes.

Something else to consider when doing comparisons, especially if you are doing side by side comparisons. Most software video players will use the hardware overlay by default. Only one overlay can be active at a time, so if you have two players open at once, one will use the overlay, the second will use software rendering. The overlay controls can be set independently of all other playback settings. Unless you have both the overlay settings and software rendering settings identical, side by side comparison will produce very different results. The only way to truly do side by side comparison of contrast/colour etc if you load both clips via an avisynth script and stack them, then load the script into virtualdub. By doing this, both will use the overlay (because they are now a single display) and you will get a true feel for the differences.

[src2206]

Thank you for your reply guns1inger

If X.264 does not produce smaller file size while giving better quality, then why should it be used? Only to play in portable player like iPod? In judging 'quality' I hope bit rate should be the most important criteria, and for same bit rate the file sizes are almost identical! Except the little difference caused because the usage of Nero Audio Codec, which again is inherently VBR, hence must be lower in quality than its AC3 448 counterpart (the original DVD audio which is retained by the XviD), though both can be 5.1.

Regarding the method you describe, I must admit that I have no clue hoe exactly I can do this. What I have done is, I have played both the files but not simultaneously. When I played one, the other was paused. Though I did run two instances of VLC player. But thanks for the explanation as I never had any clear idea about this.

[poisondeathray]

file size = bitrate x running time

Like guns1inger said, if you use the same bitrate (video + audio) for a given movie, it will be the same size regardless of codec used.

x.264 will give you better quality at a given bitrate. The difference is profound at lower bitrates, mainly because of x.264's deblocking feature. XviD/DivX tend to appear blocky & pixellated. At high bitrates, you may not notice the difference.

Hope this helps
Cheers

[src2206]

Thanks poisondeathray

What is normally considered as a high bit rate? Is it about 40% of the original (as guns1inger mentioned)? But I have not seen any software converting to AVI or H264 with that high of a bit rate. For example, the average bit rate of the original DVD9 movie is about 8100 kbps, so the 40% of this is 3240 kbps. Even in 100% quality setting I do not think AGK will encode to XviD in that hig of a bit rate, or did I understand something wrong!

At high bitrates, you may not notice the difference.

If guns1inger and others agrees to this comment, and provided high bit rate means at least 40% of original; then I find no reason to encode in H.264 when the the same bit rate output can be obtained in much lesser time.

[poisondeathray]

The definition of "high" is relative, and depends on video dimensions as well. You need a higher bitrate for bigger dimensions. A 1080p video will require a higher bitrate than a 720p to "look the same quality"

Some static films, with simple scenes don't require high bitrates. Some action movies with fast sequences and complex pictures require high bitrates too look normal.

Bottom line: find out what your priorites are. If file size doesn't matter. Leave it as original. Don't re-encode. If you need to reduce to a particular size, choose one. You can be 100% certain at the same file size h.264 > XviD for quality. Do some tests if you don't believe me. Check out the codec shootouts at Doom9 forums. When you expand the video to fullscreen, the aberrations and blocking become more evident with XviD. The deblocking of h.264 smooths the video out. Unless you're happy playing it in a small window

Another thing to consider: If you want to play on a standalone DivX compatible player, h.264 does you no good - not compatible. And as you noticed, it takes a LOT longer to encode! You should consider these trade offs between quality, file size, CPU resources.

Test sample clips of what you want to encode, say 1 minute clips to reduce the testing time, and vary the bitrate, codec etc.... and figure out a happy balance.

Good Luck

[guns1inger]

It really depends on your threshold for pain/quality. I believe the argument has been muddied by the continued use of CD capacity as a measure, be-it 700 MB or 350 MB. I encode some TV shows to Xvid in order to get a series on as few discs as possible. I choose to encode Xvid through AutoGK (because this is what my players play) at around 10MB per minute. So a 50 - 55 minute program such as Top Gear, with lots of action and transitions, comes in at around 550 MB per episode.

Is this DVD quality ? Short answer : No.

Long answer : Of course not. The resolution has been lowered to 640 pixels across (down from 720), a lossy encode has been compounded by a lossy encode, and there are still small artifacts in some scenes that are not in the original digital stream. However on playback, this is acceptable to me. In order to maintain the same quality as the source, the size would have to be closer to 1 GB per episode. So I have compromised. 550 MB gives me a quality I can live with and gets a series on a single disc. To maintain the original quality would require 2 - 3 discs per series.

My (brief) experiments with H264 show me that I can get comparable quality in as little as 420 MB per episode.

An ideal solution would be to encode with H264 at 550 MB an episode and still be able to play the files back on my DVD player (Apple TV be damned forever).

You ave to find a balance that suits you that covers you quality requirements, playback needs and the practicality of encoding. I have a quad core machine that can 2 pass Xvid encode 8 x 50 minute episodes in under 4 hours with AutoGK, and only slightly longer for H264

[src2206]

Thank you very much posidondeathray...

I am encoding at present with the same bit rate as the XviD. I shall try and post the feedback here.

BTW, my main aim is to take back up of the original movie to MPEG4, so that if by any chance I loose the original DVD I still have something to fall back on, and I can convert the MPEG4 file to DVD. Though I understand that this way I can never get the original DVD quality back, but it will be still better than nothing. Now through this discussion I am trying to zero on the codec that I would like to use for this job (either XviD or H.264).

[poisondeathray]

src2206 - you don't have to test an entire movie, just 1000 frames or so. This will limit time time wasted on testing

From your original post, the dimension errors (w x h) for 16 x 9, might be because they were not multiples of 16. So called "mod16" dimensions are required for compatibility. Your original screenshot gspot showed 608x256. The correct dimension would be 608x336.

Your video bitrate in that encode was 608kbps. That is quite low for that dimension, and I am 100% certain if you used the same dimensions and video bitrate with XviD, it would be noticably inferior than a video encoded with h.264.

Your original post mentioned a DivX/XviD stand alone player, so h.264 is probably not an option then.

[src2206]

Thank you guns1inger for your reply, and I do understand your point.

Sorry, I did not see your last post before posting my last one...

Hello posiondeathray,

for XviD, I chose 80% quality output which gave a avg bit rate of about 2100 kbps.

[src2206]

Ok, I have encoded the video in X264. But to be very honest I could hardly see any difference in quality- very little indeed. Now it may be because I have a old CRT 15'' (Flat) monitor and my stand alone DVD player can not play MPEG4. That's why I mentioned about my intention to convert the file back to DVD in case the original DVD does not work.

[guns1inger]

Then the real test should be to take your Xvid and H264 videos, convert them back to DVD, and compare them to he original DVD. That will be the true test of the quality of the process.

[src2206]

Hello guns1inger...

Thanks for replying.

I have earlier used HQ Fast profile, I am encoding again using HQ Slowest profile. Lets see if I can spot any difference this time.

In the meantime, could you please be kind enough to let me know how to do the following?

The only way to truly do side by side comparison of contrast/colour etc if you load both clips via an avisynth script and stack them, then load the script into virtualdub.

Thanks again.

[guns1inger]

https://forum.videohelp.com/topic348086.html#1827547

It is actually for a 2 x 2 grid, but you get the idea.

However I stand by my last post. If you want this as a backup for a failed DVD, then need to convert them back again and compare the quality of the original DVD with the two multi-encoded, hyper compressed DVDs.

[src2206]

However I stand by my last post. If you want this as a backup for a failed DVD, then need to convert them back again and compare the quality of the original DVD with the two multi-encoded, hyper compressed DVDs.

Not that I disagree with you, and I would love to do that. The only thing that deters me is Time. But still I shall surely try once I get a little holiday and let you know either through PM or posting in this thread. Not to mention that I shall work on the AviSynth to play both converted files side by side by the method you have kindly mentioned.

BTW, to convert beck to DVD and check, is it necessary that I convert the whole movie or a small 10-15 clip will be enough? If clip will be sufficient then I can surely squeeze it in my schedule and do the tests ASAP. But please let me know how to cut a part of VOB file, because VirtualDub does not support MPEG2 and I am not sure whether VirtualDubMod can be used for this.

Thank you for your reply.

[xXNateXx]

hey im kinda new to this converting stuff lol ive been trying to convert mostly dvd to avi. i came across a program called dr.divx. it was doing very good quality somewhat like axxos but now its been acting up(freezes 1/3 on the second pass). i seen a mp4 video with great quality and in the range of axxos video size 700mbs (and for the ppl who dont know who axxo is check out mininova.org) question is does anyone know of a program that would do that? im fairly new to the ripping and converting of movies so i dont know to much. auto gordian knot has been good to me. so i tryed just plan old gordian knot to profect my skills but get some errors.. well not errors really but the file size comes out to be in the 300mb range and no picture but sound. so now im interested in this mp4 codec. oh also i have adobe premier. what does anyone think about using that for converting movies and what not? prob the ppl that are taking video editing in school could come up with some good quality stuff. well hope this getting some answers......

[src2206]

There is a guide on using AGK here: http://sroyc.blogspot.com/2008/02/dvd-backup-complete-guide.html

See if this helps..

BTW, I think all here will agree that Dr DivX is not as good as AGK...so give it a try

[xXNateXx]

thats a well written guide. i been trying to get quality like axxo video and size of file wise. i think he made his own software or something cuz theres nothing that matches it out there. dr divx though came quite close to the results i was looking for. why is no one trying to fix up dr divx? its been forever at version 2.0.

[src2206]

thats a well written guide. i been trying to get quality like axxo video and size of file wise. i think he made his own software or something cuz theres nothing that matches it out there. dr divx though came quite close to the results i was looking for. why is no one trying to fix up dr divx? its been forever at version 2.0.

Please steer clear off illegal P2P download and its reference in this site. This is not allowed....and Mods are watchin'... (first I thought this is some kind tool, )

BTW, thanks very much for appreciating my guide...