AVInaptic. Free utility which analyzes multimedia files (Linux and Windows)

[rberger]

It's apparently his own compiler

Kind of unfortunate. Obviously, this little tool is very useful, but as the whole application logic code base depends on that compiler, which itself implements all the relevant external bindings, there's likeliness this will turn into a maintenance nightmare and eventually die.

E.g. for the switch to GTK2, he has to revamp that very compiler code and - very understandably - already shown some reluctance. It'll be interesting to see if this keeps going anyway.

[QUEENFAN]

Hello,

great Tool.
I just began to encode some DVD-Rip with AutoMKV and now have two samples, one XVID-AVI and another x264-MKV.
Now I wanted to discover via AVInaptic which encode has better quality.

On DRF you say: So the higher the DRF the more compression/blurriness the video has? The quantizer is involved in the quantization of "DCT coefficients". A typical effect of an excessive quantization is blockiness.

To my surprise the DRF-analysis of the XVid-File gives me an averadge DRF of 3.24 while the x264 has 21.39 ???
I thought x264 would bring much better quality than xvid, the encode took about 3 times of the xvid-encoding-time.

[LoRd_MuldeR]

@QUEENFAN

1. Neither Quantizers (that's what DRF refers to) nor PSNR or SSIM can give you information about the actual subjective quality!
2. H.264 and MPEG-4 ASP use completely different Quantizer scales, that's why the values from x264 cannot be compared with Xvid or DivX.

I think you can say: The higher the quantizer (DRF value), the more information was removed from the frame. But this says absolutely nothing about how "visible" that information was. So judging the quality only from quantizer values is pure nonsense. Metrics like "PSNR" and "SSIM" should give some more hint, but they still don't match the human perception. For example: The latest psycho-visual improvements in x264 can improve the subjective quality drastically, although the metrics actually show "worse" values...

[QUEENFAN]

@LoRd_MuldeR

Thanks for your quick reply.

I'm just trying to find out the best solution for me for encoding DVD/HD-/BR-DVD in the future regarding settings/profiles for AutoMKV which seems very promissing to me being able to encode XVID as well as x264.
I want to keep my originals in a safe place thus watching the encoded files in the near future optimized for LCD-HD-Ready over HDMI and a HTPC with HD-DVD/BR-Playback.

So to follow your suggestion I can't use a tool like AVInapotic but have to take a look at the destination myself?

[LoRd_MuldeR]

So to follow your suggestion I can't use a tool like AVInapotic but have to take a look at the destination myself?
AVInaptic is a really useful tool, which can give a lot of information. But expressing a subjective thing like "Quality" in numbers simply is not possible :p
You will have to do a few encoding tests and decide what looks best for your eyes...

Nevertheless x264 should give much better quality at a given bitrate, compared to Xvid. That's because H.264 is much more efficient than MPEG-4 ASP.

[telmoMRC]

yei have made one report whit the AVInaptic i select drf analys and is missing the [ About quality ] !!

i whant to make one report like this one but how ???

[ About file ]

Name: xxxxxxxxxx
Date: 31/05/2008 21:01:47
Size: 2,341,105,664 bytes (2232.652 MB)

[ Generic info ]

Play duration: 01:37:14 (5833.625291 s)
Container type: AVI OpenDML indexes multi-chunks (*)
Number of streams: 2
Type of stream nr. 0: video
Type of stream nr. 1: audio
Audio streams: 1
ISFT: VirtualDubMod 1.5.10.2 (build 2540/release)
JUNK: VirtualDubMod build 2540/release

[ About quality ]

Resolution: HIGH (704 x 384)
Width: multiple of 16 (GOOD)
Height: multiple of 16 (GOOD)
Average DRF quality: HIGH (2.873419)
Standard deviation quality: HIGH (0.682232)
Std. dev. weighted mean: HIGH (0.451463)

[ Video track ]

FourCC: xvid/XVID
Resolution: 704 x 384
Frame aspect ratio: 11:6 = 1.833333 (~1.85:1)
Pixel aspect ratio: 1:1 = 1
Display aspect ratio: 11:6 = 1.833333 (~1.85:1)
Framerate: 23.976 fps
Number of frames: 139867 (126206)
Stream size: 2,008,658,415 bytes
Bitrate: 2754.593673 kbps
Qf: 0.424988
Key frames: 2535 (0; 240; 480; 697; 937; ... 139825)
Null frames: 0
Min key int: 1
Max key int: 240
Avg key int: 55.174358
Delay: 0 ms

[ Audio track ]

Audio tag: 0x2000 (AC3)
Bitrate: 448 kbps CBR
Channels: 6
Sample rate: 48000 Hz
Chunks: 60688
Stream size: 326,258,688 bytes
Delay: 0 ms

[ About MPEG4 encoding ]

User data: XviD0047
Packed bitstream: No
QPel: No
GMC: No
Interlaced: No
Aspect ratio: Square pixels
Quant type: MPEG custom (*)
Custom intra quant matrix:
8 8 10 12 14 16 18 20
8 10 12 14 16 18 20 22
10 12 14 16 18 20 22 24
12 14 16 18 20 22 24 26
14 16 18 20 22 24 26 28
16 18 20 22 24 26 28 30
18 20 22 24 26 28 30 32
20 22 24 26 28 30 32 32
Custom inter quant matrix:
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
Number of frames: 139867
Drop/delay frames: 0
Corrupted frames: 0

I-VOPs: 2535 ( 1.812 %)
P-VOPs: 56556 ( 40.436 %) ##########
B-VOPs: 80741 ( 57.727 %) ##############
S-VOPs: 0 ( 0.000 %)
N-VOPs: 35 ( 0.025 %)

Max consecutive B-VOPs: 2
1 consec: 18117 ( 36.653 %) #########
2 consec: 31312 ( 63.347 %) ################

[ DRF analysis ]

Average DRF: 2.873419
Standard deviation: 0.682232
Max DRF: 4

DRF<2: 0 ( 0.000 %)
DRF=2: 42512 ( 30.402 %) ########
DRF=3: 72508 ( 51.854 %) #############
DRF=4: 24812 ( 17.744 %) ####
DRF>4: 0 ( 0.000 %)

I-VOPs average DRF: 2.001972
I-VOPs std. deviation: 0.044367
I-VOPs max DRF: 3

P-VOPs average DRF: 2.293231
P-VOPs std. deviation: 0.455631
P-VOPs max DRF: 4

B-VOPs average DRF: 3.307179
B-VOPs std. deviation: 0.461324
B-VOPs max DRF: 4

[elguaxo]

Misc -> Preferences -> Judgments about quality -> enabled

but you should trust your own eyes more than any program. ;)

[telmoMRC]

Misc -> Preferences -> Judgments about quality -> enabled

but you should trust your own eyes more than any program. ;)

thx gx :thanks::thanks::thanks:

[telmoMRC]

i try to do that and doesn´t work

i give up

one morre time thx for the help

[rkalwaitis]

Hello all,

I hope a savy person can help me to understand some of the user entries in my report on a mkv file. Let me tell you quickly what I did. I noticed a film that had been compressed to a really small file size, but still had good quality. It was encoded with megui and used x264. I thought I could use AVInaptic to see what the settings were and if a custom matrix was used. Any ways here is where is stuck.

User data: x264
User data: core 54 svn-654
User data: H.264/MPEG-4 AVC codec
User data: Copyleft 2005
User data: http://www.videolan.org/x264.html
User data: cabac=1 I assume 1 is activated in megui's case
User data: ref=16
User data: deblock=1:-2:-1
User data: analyse=0x3:0x133
User data: me=umh
User data: subme=7
User data: brdo=1
User data: mixed_ref=1
User data: me_range=16
User data: chroma_me=1
User data: trellis=2
User data: 8x8dct=1
User data: cqm=0
User data: deadzone=21,11 (I do not understand how to change this or what it is)
User data: chroma_qp_offset=0 (I do not understand how to change this or what it is)
User data: threads=3
User data: nr=0
User data: decimate=1(I do not understand how to change this or what it is)
User data: mbaff=0(I do not understand how to change this or what it is)
User data: bframes=3
User data: b_pyramid=1
User data: b_adapt=1(I do not understand how to change this or what it is)
User data: b_bias=0(I do not understand how to change this or what it is)
User data: direct=3
User data: wpredb=1(I do not understand how to change this or what it is)
User data: bime=1(I do not understand how to change this or what it is)
User data: keyint=250
User data: keyint_min=25
User data: scenecut=40(pre)
User data: rc=2pass
User data: bitrate=450
User data: ratetol=1.0(I do not understand how to change this or what it is)
User data: rceq='blurCplx^(1-qComp)'(I do not understand how to change this or what it is)
User data: qcomp=0.60
User data: qpmin=10
User data: qpmax=51
User data: qpstep=4(I do not understand how to change this or what it is)
User data: cplxblur=20.0
User data: qblur=0.5
User data: ip_ratio=1.40
User data: pb_ratio=1.30

I think I found all of the other things, maybe quess ill see. But if someone can help me figure out what the red items are and how to change them or how they or changed I will be forever greatful.

[Brother John]

Most of the red items have a more or less identically named CLI option for x264.exe. And most of them were left at their defaults anyway.

cabac=1 --> CABAC is used (it’s on by default)
deadzone=21,11 --> Refers to --deadzone-inter and --deadzone-intra respectively (default settings).
chroma_qp_offset=0 --> --chroma-qp-offset 0 (default setting)
decimate=1 --> --no-dct-decimate was not used (default setting)
mbaff=0 --> MBAFF is a type of interlacing. 0 means source was not MBAFF interlaced.
b_adapt=1 --> --no-b-adapt was not used (default setting)
b_bias=0 --> --b-bias 0 (default setting)
wpred=1 --> --weightb was used.
bime=1 --> --bime was used.
ratetol=1.0 --> --ratetol 1.0 (default setting)
rceq='blurCplx^(1-qComp)' --> default setting for --rceq
qpstep=4 --> --qpstep 4 (default setting)

[rkalwaitis]

Brother John,

I guess that is why I am unable to change them or find some of them to change. Thanks for helping me out, I really appreciate it.

:thanks:

[foxyshadis]

That's HQ-Insane, btw.

[rkalwaitis]

You could tell that just from looking at what I posted! I'm amazed. When I did the avinaptic on the file it did not have a custom matrix, but the bitrate was really low and the film looked great. I was trying to see how it was made. Is there a standard for what they call "HC", Im assuming that is "Heavy Compression". I would of thought that superlow bitrates would have used one.

Thanks Foxyshadis :thanks:

[nm]

When I did the avinaptic on the file it did not have a custom matrix, but the bitrate was really low and the film looked great. I was trying to see how it was made.
Clean, not too complex source video and good preprocessing (denoising, for example) are probably more significant than the x264 encoding parameters.

[rkalwaitis]

NM
I learning that simple usually gets the best results.

[Jorge_O]

Hello,
I've installed AVInaptic (Dl from the author's site) but...

I cannot change directories!
I click/double click in a directory name in the left side window, in the small upper square with a drive letter in it - but files in the right window won't change.
Uninstall/reinstall didn't help.

Otherwise the program seems ok - an AVI file at the driver's root gives lots of data, graphics, etc.

Win XP SP3.

Thanks for your help!

[LoRd_MuldeR]

I cannot change directories!
I click/double click in a directory name in the left side window, in the small upper square with a drive letter in it - but files in the right window won't change.
Uninstall/reinstall didn't help.

Yes, the tool is great, but the "open" dialog sucks!

However it does work here. Double-click on directory names in the left view, will change the directory and show files in the right view...

(WindowsXP x64 Edition, SP-2)

[buzzqw]

i added to SendTo ... very easy! (but more easy on konqueror)

BHH

[OrionZodiac]

How do you get this thing to work? I have spent over an hour figuring this out. I downloaded avinaptic-20071118-full-fixed and extracted it. There is no gui in there. There is only magic and avinaptic.exe along with about 14 .dll files. Am I missing something? I have never had any program give me as much trouble as this. Any help would be nice at this point in time.

Thank You

[LoRd_MuldeR]

Uhm, avinaptic.exe is an GUI application :confused:

[OrionZodiac]

Whenever I try to open it or run it nothing happens. Whenever I drag and drop a video onto the .exe nothing happens whenever I do anything with the .exe nothing happens. Maybe my computer is being a piece of crap?

[LoRd_MuldeR]

Whenever I try to open it or run it nothing happens. Whenever I drag and drop a video onto the .exe nothing happens whenever I do anything with the .exe nothing happens. Maybe my computer is being a piece of crap?

When I run "avinaptic.exe" on my system (Windows XP) then I see this:

http://img294.imageshack.us/img294/5117/avinapticcmdkr9.png


Sure your download is okay?

[OrionZodiac]

Well it looks like it is my computer. I am unable to get this running. And yes, I have downloaded the correct file from the authors website. I have tried all download links, but none of then yield a working program. However, when I run the aspect ratio calculator which itself is just a .exe I get a working gui. Ill use something else. Thanks for your help.

[LoRd_MuldeR]

Well it looks like it is my computer. I am unable to get this running. And yes, I have downloaded the correct file from the authors website. I have tried all download links, but none of then yield a working program. Ill use something else. Thanks for your help.

Either something on your system is seriously borked or you download is broken. Even when you downloaded the correct file, the download itself might have failed...

[talen9]

Be sure that you downloaded the right package ... this (http://fsinapsi.altervista.org/code/avinaptic/avinaptic-20071118-full-fixed.zip), for a clean install.

Drag&Drop, as far as i recall, doesn't work (it's a limitation of the graphic library - GTK 1.2 - not of AVInaptic itself); you have to double-click on avinaptic.exe and then you open the file you want to get the info from.

[Quark.Fusion]

>4GB mkv in avinaptic:
Size: #undef# bytes (#undef# MB)

[ Generic infos ]

File type: Matroska data

[ Hex dump ]

00000000: 1a 45 df a3 93 42 82 88 6d 61 74 72 6f 73 6b 61 |.E▀г.B..matroska|

[littleD]

Funny thing but actually drag&drop feature works well on my xp sp3 :) But only when drop on shortcut. I've got avinaptic-20071118-full-fixed.zip. No cmd messages.
One thing. Do u have directory "C:\Program Files\Common Files\GTK\2.0\..... " with some gtk libraries? I forgot when/with what this came along.:confused: Looks like Gimp.
Avidemux is based on gtk, but it seems installing gtk lib in its own folder.

[talen9]

@Quark.Fusion: I did succeed in analyzing >4GB MKV files with AVInaptic, it *is* supported. Make sure you have the latest version of AVInaptic (marked 18Nov2007). The MKV file could be corrupted, too.

@littleD: IIRC, the parts of GTK libs needed by AVInaptic are statically linked into avinaptic.exe, so to ease the distribution/installation of the product. BTW, AVInaptic is based on GTK1.2, so it's definitely not the responsible for your installation of GTK2.0, The Gimp may be though.

[Quark.Fusion]

I doubt that mkv is corrupt as it plays just fine and I tried 2 files >4GB, all files <2GB opens without problems. I don't have more big mkv files now — will try mux something.

[SquallMX]

>4GB mkv in avinaptic:
Size: #undef# bytes (#undef# MB)

[ Generic infos ]

File type: Matroska data

[ Hex dump ]

00000000: 1a 45 df a3 93 42 82 88 6d 61 74 72 6f 73 6b 61 |.E▀г.B..matroska|


If you´re using Vista use "XP Service Pack 2" compatibility mode, that fixed the problem for me.

[Quark.Fusion]

confirm, thanks.

[Quark.Fusion]

Can I request addition to Avinaptic following analysis?
1) consecutive b-frames number
2) average size of diffirent slices
3) number of b-pyramids, separate b-pyramid analysis (like total number of b-frames, number of non-pyramids, number of pyramids, same for avg QP and others)
4) macroblocks analysis as in x264 output
5) DPB usage.

[doragasu]

Anyone tried to compile this program in 64bit Linux?

I downloaded the sources, but they are in treepc, a language implemented by the creator of avinaptic. I downloaded the souces of the treepc compiler an runtime support and was able to compile and install them. Then I tried again to compile avinaptic and the compiler thrown an error:

doragasu@doragasu-C2D:~/Escritorio/avinaptic$ make
treepc -O2 avinaptic.trp
treepc: pass 1 (syntax analysis)...
undeclared
in /home/doragasu/Escritorio/avinaptic/init.tin
line 18
make: *** [avinaptic] Error 255

Line 18 in init.tin is:
(set _nl (if (cygwin) crlf nl))

It looks like this line is cygwin related and maybe it's only needed to generate the windows binary, so I tried commenting it and launched make again, to face a new 'undeclared' error:

doragasu@doragasu-C2D:~/Escritorio/avinaptic$ make
treepc -O2 avinaptic.trp
treepc: pass 1 (syntax analysis)...
undeclared
in /home/doragasu/Escritorio/avinaptic/init.tin
line 40
make: *** [avinaptic] Error 255

Line 40 in init.tin is:
(if (not gtk-connected)

There I stopped trying things. GTK libraries are installed and also are the dev headers...

I was able to run the windows version using Wine, but it would be great if I could run a native binary.

Any suggestion?

Also is there other program similar to avinaptic (better if it runs in Linux)?

[buzzqw]

Hi doragasu

i know the developer of Avinaptic

i will ping to your post

BHH

[nm]

Also is there other program similar to avinaptic (better if it runs in Linux)?
Depends on what you need. There's MediaInfo, and for retrieving basic information, mplayer -identify (see midentify script for usage). If you need a DRF analysis feature, I'm not sure if there are better alternatives.

[doragasu]

Thanks a lot for the replies.

@buzzqw: I sent an email to the author just before writing to these thread. If he replies me, I'll post here his suggestions.

@nm: I don't need DRF analysin (I don't even know what DRF analysis is :p). I only need data of the h264 stream like ref frames, b frames, if b_pyramid is used, etc... I'll have a look to MediaInfo.

EDIT: Tried MediaInfo, and this is exactly what I need, but I'll try to compile avinaptic anyway. Maybe someday I'll learn what DRF is and what the extra parameters provided by avinaptic are for.

[talen9]

avinaptic gives a deep analysis of the audio streams and of the way they're muxed in the AVI/MKV container, too.

Regarding using it in a 64bit linux o.s. a viable way, as of now, is using the 32bit binary, having installed on the system the 32bit libraries it needs ...

[doragasu]

Regarding using it in a 64bit linux o.s. a viable way, as of now, is using the 32bit binary, having installed on the system the 32bit libraries it needs ...

Yes, I know that option, but I tend to avoid doing that kind of things (and having duplicate libraries).

[stanjr]

I'd love to try this, but I'd prefer to compile it from source on my AMD64 Ubuntu system, instead of installing all the 32bit libs.

What language is the source code in? I've never seen .trp and .tin files, but the code in them looks like lisp. Googling for the compiler, treepc, didn't turn up anything.

I'm a veteran Unix geek, so just point me at a web site for the compiler and I'll figure out if there are Ubuntu packages, or whatever.

Thanks,How I got it to work on my 64 bit Ubuntu system (Jaunty), I pretty much followed the instructions I originally found here (http://whynotwiki.com/How_I_installed_N_on_64-bit_Linux). I went through several .so files that needed to be in the /usr/lib32/ directory, starting with libgtk-1.2.so.0, then libgmp.so.3, then libgmodule-1.2.so.0, and libmagic.so.1. All packaged all the files into one easy download here (http://www.sendspace.com/file/nt7q3s) that you should copy into your /usr/lib32/ directory. Put those in and you should be fine running avinaptic. I have still come across, when running the program through commandline anyway (when not using a menu item to do it for me) that pops up the message "Gtk-WARNING **: Failed to load module "libcanberra-gtk-module.so": libcanberra-gtk-module.so: cannot open shared object file: No such file or directory" that I haven't been able to figure out since that is already on my system (I thought correctly, but if I put libcanberra-gtk-module.so into my /usr/lib32/ directory, I get a segmentation fault, so I just took it out. Everything seems to work fine, though, and if you run it through a created menu item, you don't even see that warning pop up.

EDIT: I don't know if you use all the dev versions of these things (and probably put them in the /usr/lib32/ directory), if that would get it to compile from scratch or not. That treepie custom compiler is not something that I've gotten used to, especially working on a 64 bit ubutntu distro.

[popper]

it would seem there's a new effort to update AVInaptic looking at the translated page
http://fsinapsi.altervista.org/code/avinaptic/index.html

"....
avinaptic2.tar.bz2 Updated on February 3, 2010. The new version of AVInaptic (rewritten in the new language). Note: it is still incomplete. To compile it you need to compile and install TreeP.
..."

anyone managed to make a win32 binary of this avinaptic2 yet?

[talen9]

There's actually nothing new, as far as I know ... I followed fsinapsi in his work of rewriting TreeP (the programming language which AVInaptic is written into) and AVInaptic itself is being rewritten, some months ago it was still missing many of the features which were in the 2007 version.

Of the silent update of February 3rd, I know nothing about, I suppose it was just due to some minor bugfix of which fsinapsi has become aware while working on the rewrite .... so, nothing to write home about, I think ;)

[Biggiesized]

Very cool program.

[Anacletus]

updated version

http://fsinapsi.altervista.org/ :)

[b66pak]

thanks a lot
_

[HoP]

new version released...

http://fsinapsi.altervista.org/code/avinaptic/index.html

[vrpatilisl]

what are the advantages of the avinaptic over mediainfo.

[nm]

what are the advantages of the avinaptic over mediainfo.

Avinaptic can draw bitrate graphs.

[movmasty]

what are the advantages of the avinaptic over mediainfo.
Does mediainfo gives you any clue on which quantizer is used on every frame?

[poisondeathray]

Does mediainfo gives you any clue on which quantizer is used on every frame?

No , mediainfo doesn't . You can use ffdshow's OSD, for the mean frame quantizer, and print out a .csv text file (you can also specify other things like frame number, frametype etc...)