Weird judder in encoded mp4?

[The.King]

Hi all,

I've been encoding for years with Staxrip without getting into the hugely technical stuff. Load a source, make a couple of standard codec tweaks for better results and let it do it's thing. I've always gotten good results with Staxrip, except for when I come across a funny source which is blended or something (according to the good folk here).

Anyway, aside from trying to get a bit of help with integrating QTGMC into Staxrip and/or doing similar with MeGUI - without having to write command lines every time (see other posts) I have a question about a few encodes I've recently done in the interim.

I've made an .mkv file from an original .vob using makemkv. Upon running it through Mediainfo, I know that it's interlaced TFF(), etc. I've loaded it into Staxrip which it indexed, then applied the usual deinterlace [currently yadif()] and made the normal codec tweaks such as ref frames, trellis, subpixel refinement, M.E. Range, like I normally would, etc.

The resulting encode looks fine on my PC. The resulting encode when played on my TV via a WDTV Live is also fine EXCEPT, every couple of seconds - about 3-4 secs, the picture jumps a little. It happens so fast, I can't tell if it jumps forwards or backwards? Almost like (in my VERY limited knowledge) the field order was switched every couple of seconds? So I have a good result for 3-4 seconds, then a second of judder, then it's good again for 3-4 secs, then judder - and so on.??

I've re-encoded it again and although all programs tell me it's interlaced TFF() , I've tried IVTC instead, which was worse. I remember someone saying years ago something about Bob() or something, but I can't remember and don't know what that is or if it would help?

Can anyone suggest what might be happening? It's not just with one isolated encode, it's been with 4-5 encodes all coming from PAL or NTSC vobs - except for one, which was a 1080p BRRIP x264. That was already encoded, but was a massive 14gb in size even with a standard AAC soundtrack taking only 6%. I ran it through Staxrip and kept AR the same, it was progressive so no deinterlacing required, no cropping, no resizing, etc - but shrunk it down to 4.5gb which still gave me a good bit rate. But it does the same thing - ok for about 4 secs, then a little judder for a split second, then good again for 4 secs, etc. The quality is great, but something's gone wrong?

I can upload samples of source or encode if it helps - just tell me what you need in order to help me and what to use to cut a short clip. (I was just using Allok Video Splitter - a simple little program, but the short clip it gives me is nothing like the quality of the encode?)

Thank you.

[sum_guy]

So it sounds like your saying that the problem lies between the WDTV and the TV, are you streaming it over a WIFI connection or ethernet? If it plays fine on the PC and stutters on the WDTV you may be exceeding the bandwidth of your LAN? Your bitrate could be too high for the encoded file.

[The.King]

So it sounds like your saying that the problem lies between the WDTV and the TV

Thanks for your reply.

While it could be the case, I'm not necessarily saying that. The more I look at it on my PC, the more I can notice the judder - it's just MUCH easier to see it on a 75" LED than it is on a PC screen 1/3 of the size. the WDTV has never had a problem playing anything to date and I've throwing practically everything at it.

are you streaming it over a WIFI connection or ethernet?

No, everything is wired Ethernet over Cat6.

you may be exceeding the bandwidth of your LAN?

Broadband coming in is 240MB not contended. LAN within network is G-Bit, so it's definitely not that.

It's an encoding issue for sure, but whether it's from an unusually blended source (which can't be the case for the 1080p but might be for the others) or because something is going on with the fields - I don't know? Hence the post here for some help. There's obviously something 'off', but I'm no expert when it comes to the inner workings of encoding.

[sum_guy]

Ok, if the LAN is good to go, lets assume an encoding issue. Maybe try Vidcoder or Handbrake instead of Staxrip and see if you get a different result. I would just use default settings for either encoder.

[poisondeathray]

Post a sample of the source and corresponding section of the encode. Don't use that program to split if it changes the quality (it's probably re-encoding) - use mkvtoolnix split & cut tab

[The.King]

Post a sample of the source and corresponding section of the encode. Don't use that program to split if it changes the quality (it's probably re-encoding) - use mkvtoolnix split & cut tab

Thanks. I've never actually used MKVtoolnix before, just Mkvmerge - so apologies for the delay while I figured out how to use it. I kept getting errors when trying to set up a time code for the clip.

Attached are the 'encoded' file & 'original' file from same timestamps. It took ages to find a decent spot in the old movie where the camera was panning, as this is when the judder is most noticeable - when camera pans.

Here is the info I get when I check the source file...

Video
ID : 1
Format : MPEG Video
Format version : Version 2
Format profile : Main@Main
Format settings, BVOP : No
Format settings, Matrix : Default
Codec ID : V_MPEG2
Codec ID/Info : MPEG 1 or 2 Video
Duration : 1h 39mn
Bit rate mode : Variable
Bit rate : 4 854 Kbps
Nominal bit rate : 7 500 Kbps
Width : 720 pixels
Height : 576 pixels
Display aspect ratio : 16:9
Original display aspect ratio : 2.35:1
Frame rate : 25.000 fps
Standard : PAL
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Interlaced
Scan order : Top Field First
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.468
Stream size : 3.37 GiB (94%)
Language : English

Audio
ID : 2
Format : AC-3
Format/Info : Audio Coding 3
Format profile : Dolby Digital
Mode extension : CM (complete main)
Codec ID : A_AC3
Duration : 1h 39mn
Bit rate mode : Constant
Bit rate : 192 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 136 MiB (4%)
Title : Stereo
Language : English

And here is what I get from the encoded file that was run through Staxrip

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L3.0
Format settings, CABAC : Yes
Format settings, ReFrames : 5 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 1h 39mn
Bit rate : 1 967 Kbps
Width : 652 pixels
Height : 572 pixels
Display aspect ratio : 1.657
Original display aspect ratio : 1.658
Frame rate : 25.000 fps
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.211
Stream size : 1.33 GiB (89%)
Writing library : x264 core 93 r1542 5b86182
Encoding settings : cabac=1 / ref=5 / deblock=1:-3:-2 / analyse=0x3:0x113 / me=esa / subme=10 / psy=1 / psy_rd=1.00:0.25 / mixed_ref=1 / me_range=32 / chroma_me=1 / trellis=2 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-4 / threads=12 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / constrained_intra=0 / bframes=5 / b_pyramid=2 / b_adapt=2 / b_bias=0 / direct=1 / wpredb=1 / wpredp=2 / keyint=250 / keyint_min=25 / scenecut=40 / intra_refresh=0 / rc_lookahead=60 / rc=2pass / mbtree=1 / bitrate=1967 / ratetol=1.0 / qcomp=0.60 / qpmin=10 / qpmax=51 / qpstep=4 / cplxblur=20.0 / qblur=0.5 / ip_ratio=1.40 / aq=1:1.00

Audio
ID : 2
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Codec ID : A_AC3
Duration : 1h 39mn
Bit rate mode : Constant
Bit rate : 192 Kbps
Channel(s) : 2 channels
Channel positions : Front: L R
Sampling rate : 48.0 KHz
Bit depth : 16 bits
Compression mode : Lossy
Stream size : 136 MiB (9%)
Language : English

I asked in another forum about Format profile : High@L3.0 in Staxrip as I can't seem to find out how or where to increase this to the regular Level 4.1 - like you can easily do in the likes of MeGUI? (I'm not sure if it would make much of a difference especially with the source?) I also tried to integrate QTGMC into Staxrip for better quality de-interlacing, but I'm hoping to get a bit of help on that as I haven't been successful myself.

Hopefully the above is enough to go on...

Thanks for the help.

[poisondeathray]

There is no difference in the frames of the encode that will account for "judder". The frames align up 1:1 when you match them up (ie. nothing duplicated, or dropped or weird business going on). So that suggests it's a playback issue on your end. Did you try other players ? At first you said only when streaming, so that suggests streaming/network issue

The profile and level don't actually do anything (it' s not limiting in x264, even when you have VBV constraints set). You have it set to "auto" and it selects the profile and level that matches your encoding settings. You can force a profile and level with the extra commandline settings - There are dropdown menus in megui, not sure about staxrip. In short, don't worry about L4.1 vs L3.0 - auto will work fine

This is a 25fps source, so if you 're in North America, you might see a slight judder on a 60Hz display, but it's the same judder that would be in the "source". You shouldn't see any "judder" in "PAL" 50Hz regions

[The.King]

@poisondeathray:

Thanks for the reply. In one sense I'm happy that the encode is ok as I thought I might have inadvertently done something unusual with my usual encode settings, so at least that's ok. I'm still baffled though because I'm not streaming anything per se and I know that my network is ok. I've even disconnected my WDTV and plugged it directly into my TV to play the content, but I'm still getting judder when the camera pans?

I've copied the file over to a laptop and plugged that directly into my TV via the HDMI out on the laptop, played it with vlc, MPC-HC x64, & WMP, and it's still the same? I just can't wrap my head around what's causing it? It's even the same with the 1080p file I encoded?

I'm in a PAL zone, so shouldn't have any issues with 60Hz displays.

[poisondeathray]

What happens when you play the source file (on the same setup) ?

[LMotlow]

You're deinterlacing and dropping alternate fields.

[The.King]

What happens when you play the source file (on the same setup) ?

The source file plays without issue or judder - whether it's the original .vob file or the .mkv made from the vob file. It's just a bit grainy like most .vob files are. In case it was something to do with makemkv, I even encoded directly from the .vob files and still got the same result - judder when the camera pans?

[The.King]

You're deinterlacing and dropping alternate fields.

I'm not even sure what that means? I'm deinterlacing because the source is interlaced and the program even auto suggests that it should be deinterlaced. As for "dropping alternate fields", I don't have a clue about that. All I did was a very minor crop to remove the tiny black bars left & right which are normally present and flicker on a .vob source. Unless you elaborate further, your information is lost on me.

[LMotlow]

You're deinterlacing and dropping alternate fields.

I'm not even sure what that means? I'm deinterlacing because the source is interlaced and the program even auto suggests that it should be deinterlaced. As for "dropping alternate fields", I don't have a clue about that. All I did was a very minor crop to remove the tiny black bars left & right which are normally present and flicker on a .vob source. Unless you elaborate further, your information is lost on me.

I fugured. And you're encoding mod-4 frames, when most encoders/decoders have problems with anything that isn't mod-8 or mod-16. I realize this doesn't make any sense to you either, if you don't know what fields and frames are. I'll opt out. Too busy to get into basics. Maybe the others will help. My guess is they'll ignore it. So should you.

Side borders make vob's flicker? Hm. Back to my lunch.

[poisondeathray]

The source you provided has progressive content, but encoded interlaced. The content is what is important. You should treat it as progressive (ie. don't deinterlace). But that has nothing to do with "judder" or your original problem in this case. (if you deinterlace, you get 25p, if you treat is as progressive you still get 25p. The cadence and frames are the same, it's just that the deinterlaced version will look worse in terms of quality). Those are important considerations, but not for your original question. The problem is any "auto" tool can make mistakes - you need to learn how to do it yourself

[jagabo]

Both files play fine on my WDTV Live + NTSC TV.

LMotlow obviously didn't download the source video. It's 25i but contains progressive frames. So you didn't lose any temporal resolution.

[The.King]

The problem is any "auto" tool can make mistakes - you need to learn how to do it yourself

Hi. Thanks for that. I thought it was a bit weird. I initially ran the source through Staxrip (x86) and the 'Field | Deinterlace' box was not automatically checked - meaning it may be progressive. When I ran the same source through Staxrip x64, the 'Deinterlace | TDeint' was auto checked. I tried it in MeGUI and after hitting the "analyse" button, I first got 'interlaced', on trying it just now - the result came back 'progressive'?? Mediainfo (as per my earlier post) also said it was interlaced and TFF.

Is there a foolproof way to confirm a source and be 100% certain? I know you said 'learn how to do it yourself', but where do I even start? Is there a quick way to do this (and by 'quick' I mean just not spending an hour trying to figure it out.)

Thanks.

[The.King]

Both files play fine on my WDTV Live + NTSC TV.

LMotlow obviously didn't download the source video. It's 25i but contains progressive frames. So you didn't lose any temporal resolution.

Hey. Thanks for taking the time to check this out on your WDTV Live. Good to know it's ok. I'm out of my depth when I say this, but maybe it being 25i and still containing progressive frames is the reason why I was getting different results from different GUI's?

If I knew a way to confirm a source's content 100%, that would be the only way I'd go. I just don't know how to do that and I defo don't know enough details about encoding to write command lines - hence why I'm still using a GUI.

I'm not even commenting on LMotlow.

[poisondeathray]

There are 2 different topics here

One is your playback, the other identifying and treating the source properly.

In your specific case, they are unrelated (you encode actually is ok cadence wise). In other cases, deinterlacing may change the cadence.

There used to be a mini guide from neuron2. His website is down. http://neuron2.net/faq.html#analysis . Check the wayback machine it might be archieved

I don't have time to type out a long explanation right now. Bascially you need to see if there is motion in each field by either using separatefields() or bob() . If there is fwd/back motion you have the wrong field order. SO use either AssumeBFF() or AssumeTFF() (the opposite of what you currently have). If there is motion every field (a different moment in time represented), you have interlaced content. If they are pairs of the same thing (but slight up/down motion from top/bottom field) you have progressive content. Field order doesn't matter with progressive content (you get the same thing) . There are dozens of other field patterns , pulldown cadences, but you need to at least distinguish between progressive and interlaced - those are the big ones.

[The.King]

@poisondeathray,

No problem. Thanks for posting back. I'll see if I can dig out that guide somehow.

[The.King]

There used to be a mini guide from neuron2. His website is down. http://neuron2.net/faq.html#analysis . Check the wayback machine it might be archieved

Ok, so I found the mini guide (or at least part of it), but it's kind of skipping the very basics of writing a script for someone like me who uses GUI's and never wrote a script before. Having just read the following, I don't know if there should be two scripts or just one? I know you guys do this in your sleep, but from an FNG's perspective, it's confusing before you even begin. i.e., The guide starts...

How do I figure out the nature of my source video?

It's important to know whether your video is progressive, interlaced, telecined, field-blended, etc., when making decisions about how to process it. Here's a reliable method for analyzing your video:

(1) The first step is to create an Avisynth script that sources your video and separates the fields. (2) If you have MPEG video, you can use DGIndex to make a project with "Honor pulldown flags". (3) Then make a script to step through the fields:
MPEG2Source("file.d2v")
AssumeTFF() # or AssumeBFF(). See the next FAQ.
SeparateFields()

Ok, so
(1) I open up word pad or edit pad, etc. to source my video and if I remember from years ago, this eventually gets saved as an .avs file and not as a .txt file. But the guide doesn't even tell me how to load the video or what kind of script is required? So in my case, what script do I need to point things to my source, in this case an .mkv files sitting in C:\MakeMKV?

(2a) Will DGIndex work with my .mkv file as the source and if not, what do I use instead? (When I'm encoding original DVD's, I use makemkv to produce an .mkv file from all the .vobs as it's apparently more stable than loading x-amount of vobs.)
(2b) What the heck is a project with "Honor pulldown flags", and did the author of the post really need to confuse the life out of an FNG before they even got started?

(3) I then load this - I assume (into the same script created in (1) above)...
MPEG2Source("C:\MakeMKV\file.d2v")
AssumeTFF()
SeparateFields()

Except my file is .mkv not whatever .d2v is?

For simplicity, lets just say I have a DVD of the movie Goodfellas. I create an .mkv file from the original .vobs with makemkv. I now have the following file in the following location - C:\MakeMKV\Goodfellas.mkv.

Can someone tell me what .avs script I need to write to just load this source file into whatever program/plugin (like DGIndex) so that I can add the other lines like AssumeTFF() & SeparateFields(), etc. to learn & determine whether my source is interlaced, progressive, telecined, field-blended, etc.? I'm trying to learn, but some of the guides don't make it easy.

Thanks.

[jagabo]

DgIndex and Mpeg2Source() often don't work well with MPEG 2 in MKV. I recommend you extract the MPEG 2 video as an elementary stream with MkvToolnix. That will give you an M2V file. Load that in to DgIndex, make sure Video -> Field Operation -> Honor Pulldown Flags is enabled. Select File -> Save Project. That will give you a D2V file (it's a text file, you can open it with Notepad and look at it if you want). You open that file with Mpeg2Source() in an AviSynth script:

Mpeg2Source("filename.d2v")
Bob() # or SeparateFields()

If the AVS file is in the same folder as the D2V file you don't have to use the full path name. If you create the AVS scrip in a different folder you'll need to provide the full path the the D2V file. You can open that AVS script with any editor or media player that supports AVS scripts. I use VirtualDub.

[The.King]

DgIndex and Mpeg2Source() often don't work well with MPEG 2 in MKV.

That seems to be the case as DGIndex is stuck on 11% and nothing is happening with the .mkv file created from makemkv.

I recommend you extract the MPEG 2 video as an elementary stream with MkvToolnix. That will give you an M2V file.

That will have to be another learning curve for tomorrow. I've never used MkvToolnix before until today, so I haven't got a clue how to extract anything from it after I have loaded all the original .vobs files?

I'll give it a shot tomorrow and report back. I have Virtualdub anyway, so I just got to learn how to extract elementary from vobs in mkvtoolnix. All the .vobs are loaded, I just don't know what to do next? Woohoo.

Thanks for your help.

[The.King]

Ok, burnt a candle and gave this a shot before calling it a night.

Think I figured out the mkvtoolnix thing? I just loaded all of the original .vob files, set the output folder and name to movie.m2v, added job to Q & started. It took about 10 minutes and gave me a really large m2v file. (original vobs were under 4gb, the m2v file was almost 8.5gb.)

Anyway, I created an .avs script with the following...

MPEG2Source(C:\Desktop\MAKE MKV\movie.m2v)
AssumeTFF()
SeparateFields()

I opened VirtualDub 1.10.4 -> File, open video file, and pointed to the .avs script I just created which is in the same folder as the .m2v file. I get an error to say...

Avisynth open failure:
Script error: expected a , or )
(C:\Desktop\MAKE MKV\script.avs, line1, column 14)

I'm lost and it's waaaaaay past bedtime.

[poisondeathray]

I only suggested mkvmerge to split, because I thought that was what you were dealing with (MKV source)

If you have VOB's or MPEG2 sources, just open it directly in DGIndex . DGIndex and dgdecode.dll is very reliable for MPEG2 sources

You need quotation marks for the path. Also when you index it, it will give you a .d2v file . The .d2v file is what you use with MPEG2Source, not the .m2v

MPEG2Source("C:\Desktop\MAKE MKV\movie.d2v")
AssumeTFF()
SeparateFields()

[The.King]

If you have VOB's or MPEG2 sources, just open it directly in DGIndex . DGIndex and dgdecode.dll is very reliable for MPEG2 sources

OK then, I've used DGIndex to create the .d2v file directly from all .vob files (VTS_01_1.VOB, VTS_01_2.VOB, VTS_01_3.VOB & VTS_01_4.VOB). The resulting .d2v file is a big difference size wise - only 754kb as opposed to 8.5GB for the other .m2v file. It also gave me a separate AC3 file (136MB) for the soundtrack.

You need quotation marks for the path. Also when you index it, it will give you a .d2v file . The .d2v file is what you use with MPEG2Source, not the .m2v

Thanks for the tip. OK, I've used DGIndex to create the .d2v file (as above) which I just called 'new.d2v', so I've amended my .avs script (which also sits in the same new folder) as follows...

MPEG2Source("C:\Desktop\New Folder\new.d2v")
AssumeTFF()
SeparateFields()

I then use DGIndex to load the amended .avs file. When I do, I get an error to say: "No video sequence header found!"
I then tried to load the same .avs script into VirtualDub. When I do, I get the following error...

Avisynth open failure:
Script error: There is no function named 'MPEG2Source'.)
(C:\Desktop\New Folder\script.avs, line1,)

I know that I am doing something wrong (obviously) - despite following instructions to the letter, but this is why I have never gone down the route of manually checking files and why I have never written command lines for any encoding. I've been at this now for more than 12hrs now and I can't get past loading the f*£king .avs file to check the source (which is mildly infuriating) and seem to be spending more time writing messages than actually encoding? There has to be a definitive guide on one website somewhere supporting this properly WITHOUT skipping any steps for FNG's, so why can't I find it?

Thanks.

[The.King]

And in case anyone suggests - I've read this link http://avisynth.nl/index.php/FAQ_loading_clips#How_do_I_load_MPEG-1.2FMPEG-2.2FDVD_VOB...to_AviSynth.3F a million times.

I've even tried amending my .avs script a couple of ways to include other command lines (to see if that would get it going) like...

LoadPlugin("C:\Program Files\DGMPGDec\dgdecode.dll")
MPEG2Source("C:\Desktop\New Folder\new.d2v")
AssumeTFF()
SeparateFields()

AND

# DGDecode:
LoadPlugin("C:\Program Files\DGMPGDec\dgdecode.dll")
MPEG2Source("C:\Desktop\New Folder\new.d2v")
AssumeTFF()
SeparateFields()

I still get the same errors?

[manono]

There has to be a definitive guide on one website somewhere supporting this properly WITHOUT skipping any steps for FNG's, so why can't I find it?

They're in the three docs included in the DGMPGDec package. And all the required pieces you're using are from the same package, right? You can't mix and match DGIndex, MPEG2Source, and DGDecode.dll. And don't move anything around. The D2V refers to VOBs in a specific location. If you set up a new folder, like on the 'C' or other named drive instead of on the desktop (I would if I were you), then make a fresh D2V file.

I don't know what an 'FNG' is.

[The.King]

They're in the three docs included in the DGMPGDec package.

Thanks, but that's the info on DGMPGDec. I was referring to an overall definitive guide for writing commands within an .avs file - i.e. if my source was .mp4, was .mkv, was M2TS, was AVI, etc. For all the advice I've received thus far, everyone failed to mention adding the command line LoadPlugin("C:\Program Files\DGMPGDec\dgdecode.dll") first, before I continued with adding MPEG2Source("C:\Desktop\New Folder\new.d2v") etc. These are the kinds of tiny oversights that make a new & difficult process seems almost impossible to an FNG.

And all the required pieces you're using are from the same package, right? You can't mix and match DGIndex, MPEG2Source, and DGDecode.dll. And don't move anything around.

Yes, everything is from the same package and no, I defo did not mix & match anything.

The D2V refers to VOBs in a specific location. If you set up a new folder, like on the 'C' or other named drive instead of on the desktop (I would if I were you), then make a fresh D2V file.

This is exactly what I did. I created a new folder, copied all original .vob files to the new folder, used DGIndex to create a new .d2v file from the .vobs and saved it in this folder, opened Edit Pad, created a new .avs script and saved it as an .avs file in the same location. Still the same result - got the errors below whether I tried to open it in DGIndex or VirtualDub??

I don't know what an 'FNG' is.

It's an old term within military circles meaning "Fu@king New Guy". It's used in a lot of movies based around 'Nam. Full Metal Jacket, Platoon, etc.

Thanks.

[LMotlow]

Copy dgdecode.dll into the Avisynth plugins folder. It's in your Program Files under Avisynth (or Avisynth 2.5, or Avisynth 2.6, or however Avisynth was installed), in the "plugins" subfolder.

You can load a vob into VirtualDub and do the same thing with the deinterlace and bob filters. However, if you're getting errors with VOBs (sequence headers) then I suspect you have a bum DVD to begin with. Nobody shoots movies at 25fps except in PAL land. It's obviously been speeded up (Donald Sutherland with a squeaky voice? No, please.). The original movie is 23.976 fps progressive.

Both files play fine on my WDTV Live + NTSC TV.

LMotlow obviously didn't download the source video. It's 25i but contains progressive frames. So you didn't lose any temporal resolution.

Unwarranted assumption, not to mention incorrect. The mkv didn't play "fine" on my LED or my plasma, and the re-encode was stuttery. On a hard drive or a USB stick, none of my players cared much for either of them.

I'm sorry to have alluded to quality issues in the earlier post. I realize the Q word freaks out people in this forum, so I'll try to contain myself. I usually stay away from these "I want my perfectly good DVD to fit on a 1.4MB floppy drive" threads, but it seems to be almost the only thing going on in videohelp these days, and searching the archives for tips on the way things used to be done here can get tiresome. I recall running into info about mod-8,mod-16 and encoders and decoders, etc., from one of the clique members here (jagabo, poisondeathray, or whoever). But you guys don't even follow your own rules, so why should I rummage for those posts?

[ADDED]..and noticed that in the re-encode, the first GOP is 250 frames. The second GOP is 288 frames. I didn't even know 288-frame GOPs was allowed. Depending on the playback decoder and the way they unravel the frames (they aren't all stored in sequence), that wouldn't make for smooth motion on some systems. For the web, maybe, but I never saw anything bigger than 250-frame GOPs.

What bout mpg2cut2 for a sample of the unpocessed original?

[blud7]

1. I've had progressive DVDs ripped with makemkv "appear" as interlaced before. Don't know if the problem is makemkv, packing a vob as an mkv or mediainfo having a tough time with mpeg in the mkv container. Try encoding your video without deinterlacing using ffmpeg. Something like: ffmpeg -y -i "Original.mkv" -crf 20 -tune grain -b:a 112k "Original.mp4" just to see what you get.

2. Probably try ripping your DVD to disk with another tool.

3. Sometimes graphics cards prevent you from seeing errors in video encoding. So what appears on your TV may be different from what you see on the PC. e.g. AMD has a feature called smooth video play back and ATI cards can also detect and remove interlace on playback.