NuEnc 0.01b Released

[Peter Cheat]

I've finally finished with the GUI (I hate GUI programming) and here it is, NuEnc 0.01b.

Changes include:
- CQ mode now vbv compliant (hopefully)
- GUI Re-arranged (and fixed)
- New encoding window displays comprehensive encoding statistics
- Now able to change more options such as motion search diameter and minimum bitrate
- Now less than half the size compared to previous versions (~300KB)
- Constant quality encoding added (experimental at the moment)
- Custom matrix support via 'matrix.ini' and external XviD style matrix files
- Field order can be changed for interlaced encoding
- Implements the latest changes I've made to avcodec
- Fixes to multi-pass encoding. More passes will never decrease quality.
- Some other features Nic added to QuEnc

Known Issues: Keyframe forcing does not work correctly (will NOT force consecutive B-frame to I-frame), but frames that do become I frames DO have a GOP sequence header.

Briefly tested and works as expected.

http://www.petercheat.host.sk/libav/

Files mirrored http://www.petersplace.000k2.com/projects/


Edit by Moderator:
If you are looking for working download link (original web-site doesn't exist anymore), please see this (http://forum.doom9.org/showpost.php?p=1611885&postcount=236) post!

[Peter Cheat]

Freelock7 - Could you please test to see if the sequence you had that was encoded poorly with 0.00d encodes properly with 0.01?

[Paced]

Hey Peter Cheat, thanks for the new release, but I seem to be having some problems with it. Whenever I click on the 'Advanced' button, I keep getting a popup saying: Please enter a number between 0.2 and 1.2 :confused: Then, after the 'warning', I press OK, and it finally allows me to access the Advanced screen. After I make some changes, I press 'Apply', and the very same popup arises again. Wierd.

//Edit

Additionally, when I go to 'Quit' the program, I get this popup: Please enter an integer between 4 and 15000.

[Peter Cheat]

I've sort of fixed this. Redownload it and try again. The first time you use it, set it to all the settings you want, then close it. Reopen it, and it will work as it should. I said i sucked at GUI programming :D

[Paced]

Originally posted by Peter Cheat
I've sort of fixed this. Redownload it and try again. The first time you use it, set it to all the settings you want, then close it. Reopen it, and it will work as it should. I said i sucked at GUI programming :D

Hehe, thanks for the quick fix. The first popup I spoke of is now gone, but the second one is still there - Please enter an integer between 4 and 15000. It pops up whenever I close/open the program and open an AVS file. Not that it should affect anything, but just thought I'd let you know. Thanks again.

//Edit

Just found out that in 2 pass VBR mode, the program finishes the first pass, but then idles and does not start the second pass (this didn't happen to me in the previous version).

[Peter Cheat]

Hehe, thanks for the quick fix. The first popup I spoke of is now gone, but the second one is still there - Please enter an integer between 4 and 15000. It pops up whenever I close/open the program and open an AVS file.


Make sure that:
minimum bitrate is between 4 and 15000kbits <- probably the problem
average bitrate is between 4 and 15000kbits
maximum bitrate is between 4 and 15000kbits

It should at least start (or possibly crash with an error), scaling is a little slower in this version - but if it doesn't start for a (very) long time there is a problem. I'll probably add extra debugging if it seems to be a common problem. It is not possible to idle indefinitely (there are precautions taken so this can't happen).

[freelock7]

Some problems:
1) I can download only the first file 'NuEnc0.01 based on QuEnc'.
The two others don't load correctly (<source> and <ffmpeg source>).
2)nothing works when I try to <encode>.
3)The 312 ko size of the new libavcodec is correct (small size)?

[Peter1234]

Peter Cheat,
Is there a P4 problem? When I click the icon to start the program nothing happens (it blinks, but no window opens).

EDIT: It creates an empty debug.log and then closes.

[Bogalvator]

I've got the same thing as Peter1234, and I'm using an Atlon-XP :(

[clickit]

I have the same problem
except that I'm running winXP SP2 on a Celeron machine
It also creates an empty debug.log

[fabrice]

Hi,

I'm using an Athlon XP, and work perfectly! (trying the CQ mode, and compare it with 2-pass VBR! :) )
Using W2K. What are your OS?

Fabrice

[Peter1234]

I am using Windows XP SP1 with SSE2.

[freelock7]

Other trouble with <gui appearence>:
It becomes suddenly blue when opening <advanced options>.
Always the same bug: the encoder refuse to start.
---------------
AthlonXP2200
XPProfessionnal

[vinetu]

Well, NuEnc starts here (P3 Celeron-Tualatin MMX,SSE, winXP SP2)
and I can set "Advanced Setings"-no problem,but when I click "Encode" a Warning popup:

------------------------------------------------------------------------
Could not open Video Codec
Please check 'debug.log' for more information.
Take note of settings and report if a serious bug exists.
------------------------------------------------------------------------

in the debug file I see this:
-----------------------------------------------------------------
[mpeg2video @ 6552A00C]MPEG1/2 doesnt support 1000000/41599 fps
-----------------------------------------------------------------

The .avs file is a simple "avisource->resize->addborders" and is reported in VDubMod as
---------------------------------------------
720x576, 24.039 fps (41599 micro seconds)
816 frames (0:33.94)
FourCC : [YV12]
---------------------------------------------

The "41599 fps" in debug file and "(41599 micro seconds)" in VDubMod file info is obvious!

If I add "AssumeFPS(25) at the end in the script - Encoding is started and work as expected!

Best Regards
Marin

[Peter Cheat]

Originally posted by vinetu
Well, NuEnc starts here (P3 Celeron-Tualatin MMX,SSE, winXP SP2)
and I can set "Advanced Setings"-no problem,but when I click "Encode" a Warning popup:

------------------------------------------------------------------------
Could not open Video Codec
Please check 'debug.log' for more information.
Take note of settings and report if a serious bug exists.
------------------------------------------------------------------------

in the debug file I see this:
-----------------------------------------------------------------
[mpeg2video @ 6552A00C]MPEG1/2 doesnt support 1000000/41599 fps
-----------------------------------------------------------------

The .avs file is a simple "avisource->resize->addborders" and is reported in VDubMod as
---------------------------------------------
720x576, 24.039 fps (41599 micro seconds)
816 frames (0:33.94)
FourCC : [YV12]
---------------------------------------------

If I add "AssumeFPS(25) at the end in the script - Encoding is started and work as expected!

Best Regards
Marin

24.039 is not a valid frame rate for MPEG-2. 23.976, 24, 25, 29.97 should all work. This is not a bug, but at least the log file helped.

@Peter1234
I don't have a P4 so I can't test on it. If it opens and then just closes, I don't know what the problem could be. Unless I removed P4 specific optimisations that are necessary (highly doubt it). Maybe its my screwed up GUI and registry settings.

@Bogalvator & clickit
I am using an Athlon XP with WinXP SP2 without problems. I'll fix up the GUI bugs, and maybe it will work. I shuld have waited longer before releasing.

@freelock7
When you start off, things are a bit screwed, I know. Change to all the settings you want (VBR 2-pass, set the advanced options to what you want, and set the colour to grey). Then close it (click on Quit)then open it, everything should work fine (hopefully).

EDIT: I've just uploaded my settings here (http://www.petercheat.host.sk/libav/files/settings.zip). Try using this and see if it helps.

[vinetu]

24.039 is not a valid frame rate for MPEG-2.

Correct! -my bad... :)

It seems if the source is very compressible (MAX bitrate is lower than choosed "Average bit Rate")
-You'll get a Warning message "Cannot open Video codec"
or something and an empty log file at the end of 1-st pass.

I have success with P4 (Nordwood core -MMX,SSE,SSE2, winXP SP2)!

Best Regards!

Edit: I forget to say - I've tested multipass 3 and 4 passes VBR only

[Peter Cheat]

@vinetu
Yes I know. The problem is, what do to in this situation. FFMPEG by default just quits if it can't reach its target. I could make it continue, but then I'll get heaps of complaints saying that encodes don't reach target size. This is why NuEnc 'crashes' when starting second pass if you are encoding a (mostly) black screen.

I think I will change it so it continues, but leaves an error in the log stating that target size could not be reached.

Minimum quantiser could be changed to 1, BUT, this can lead to buggy output (quantiser 1 is not recommended!).

[freelock7]

It doesn't.

[Peter Cheat]

Open debug.log and see if there is an error. I can't pinpoint the problem.

[freelock7]

Debug_log says:
[mpeg2video @ 6552A00C]removing common factors from framerate"

[Peter Cheat]

What is the frame rate of your source? It might not be valid.

[Bogalvator]

@clickit & Peter1234,

I just tried Nuenc 0.01 this morning, and it worked.

It might be because I disable "Extreme & High Quality seetings" in QuEnc 0.56 (guessing) - so perhaps uncheck all settings in QuEnc and then try to load NuEnc again?

[Bogalvator]

By the way,

The "modified FFMpeg source" downloads up to 50% then disconnects.

[Peter Cheat]

I don't think that would be a problem. I'll separate the registry entries though just in case for next release, which will have to be soon since there seems to be a few bugs.

EDIT:
Download problems are because of host.sk. They haven't been reliable for a while now. I've intended to move hosts, but couldn't find any decent ones. Use a download manager.

[freelock7]

PAL source.
It works fine with NuEnc0.00a.

[Peter Cheat]

Does an error message show up when you try to encode? What settings are you using. I might have bungled up some options.

[freelock7]

Any error message.
It opens a new window with <encoding> and <statistics> but it remains blank. The encoder don't start.
Peter, why the new libavcodec is so small (312ko)?

[Peter Cheat]

The reason it is so small is because I've removed MPEG-4 specific code (and code relating to other codecs). This is just to save bandwidth and to make it easier to download. If its larger than ~700KB, the download seems to dropout all the time.

Technically it would also make it faster, but saving ~100 cycles per frame is not going to make a difference on a machine that does 1,500,000,000 cycles per second :D

I can't figure out why it doesn't work. Try a different source maybe, and see if that works. I can't see how it can freeze with the encoding window without an error, and without starting.

[buzzqw]

same here
stat is blank at start but if i move window i can see "something"

winxp sp1, p4 2.8ghz

(@freelock7 avcodec.dll is upxed, but even if upxed is small

BHH

[Peter Cheat]

What do you mean by "something"? Do the statistics become partially visible (is it encoding, but the stats are just not correctly displayed?).

[freelock7]

The problem seems to come from this new compiled libavcodec.
Maybe I'm wrong but if i were you, I'll keep the NuEnc0.00d libavcodec file and add in it the new 0.01 changes.

[Peter Cheat]

That is unlikely. I did remove a lot of code, sure. But 90% was related to QPEL and GMC. The rest was audio codecs, and video codecs like mpeg4, snow, cinepak etc. All code relating to MPEG1&2 were left alone. If I wasn't sure, I also left it. Others are not having problems so I doubt its a problem with the DLL. I know the GUI is a little bit dodgy, and just realised that if you minimise the window while encoding, then maximise it sometime later, the encoding stats might disappear. This will be fixed now. I'll have to release a new version tomorrow with all these fixes :o

EDIT:
I won't change back to NuEnc 0.00d DLL, I've fixed all the bugs that were found in it. I could recompile without removing any redundant code, but the only difference will be that it's twice as big.

[clickit]

it works with seetings.zip

[buzzqw]

look at this
http://www.64k.it/andres/Immagine.png

after a while (sometimes never) it display something but fps is like 0.14 per sec...
(p4 2.8ghz)

great appz btw !

BHH


after a while
EDIT: http://www.64k.it/andres/post.png

I have just clicked on setting.reg launch a conversion
Isn't strange "peak bitrate" ?

[Peter1234]

Peter Cheat,
After loading your settings the program starts. When I load an avs file and click the Encode button with your settings, the encode window opens up but encoding does not start. Encode window has no statistics and there is an empty box on it. Debug.log is empty.

[Peter Cheat]

I've now fixed GUI problems in NuEnc 0.01a. It should start up correctly.

There are 2 versions. A small one with most codecs removed and a large one without anything removed. If you have a P4, the large one should work. I may have deleted P4 specific code when removing unused codecs :o.

CQ Mode is now also linear (CQ 50 produces files about half the size of CQ 100).

[freelock7]

No more success!
Both of them don't work.
Always the same bug report:
<[mpeg2video @ 6551400C]removing common factors from framerate>

[Peter Cheat]

I have a feeling that the encoding window is doing something weird. Try to encode, then press CTRL-ALT-DEL, click on the processes tab and see how much CPU NuEnc.exe is using. If its ~99% then its stuck in an endlesss loop somewhere...

[freelock7]

Cpu=99%
NuEnc seems to block the system (stand by).

[Peter Cheat]

Ok, were getting somewhere then.
Try adding AssumeFPS(25) to the last line of the AVS script. Does the error still appear in the log?

[freelock7]

Already tested.
The debug log is empty now.

[Sectie.B]

Peter

I'm having the same problem: encoding window appears but just sits there and cpu usage is way up 98/99 until I hit Cancel (XP1700 + XP SP2)

One difference between 0.01 and 0.01a: with 0.01 I do get a small output (m2v)-file but it is only the first (3 I think) frames. With the a-version I don't get any output at all.

The same scripts (tried several) all work with NuEnc 0.00d and Quenc 0.56.

[Peter1234]

NuEnc 0.01a and 0.01a for P4 both open first window. When avs file is entered and encode button is clicked a new window titled Encoding opens with no text. If this window is draged on screen the text appears. All values on screen seem to be zero. Seems to be in loop. CPU usage 50%. Terminates when Quit button is clicked. No Debug.log generated. I think it is stuck trying to find avs file. XP SP1 on P4 with SSE2.

[vmesquita]

Hi Peter Cheat,

Thanks for the effort of improving avcodec, looks awesome so far! :D
Could you please check the ffmpeg modified sources download link? I am trying to download so I can recompile FreeEnc with your improvements, but it looks like the file is corrupted... :(

[Peter Cheat]

@vmesquita
Try using a download manager. host.sk seems to be a bit overloaded. I having trouble uploading to it, and often have to resume uploads. I'm pretty sure its all there.

@P4 owners
It seems that somehow the GUI screws up with P4 only. Might also be related to a newer Visual C++ dll. I'll redo the encoding window, it might be that it doesn't work because of my dodgy programming. Strange that it works on my AMD flawlessly, but others have problems.

Maybe Nic could tell me why it doesn't work. I don't really know C++.

[Peter Cheat]

@freelock7, Peter1234 & Sectie.B
Try this (http://petercheat.host.sk/libav/files/nuenc-test.zip). I removed the pause function and compiled in VC++ without any optimisations (debug mode). I don't know if this will work, but have to try something.

EDIT: I've been able to download all the files without any problems using FlashGet. If anyone knows of a good free web host / file host where I can store these files, let me know.

[Peter1234]

Peter Cheat,
New test version produces same results as 0.01a and 0.01a for P4 for me. XP SP1 on P4 with SSE2. An empty debug log is not much help. Can you add some more entries at the begining of program that might narrow down where the problem is. Maybe some entries like: program started, first window was generated, avs file name valid, avs file found, avs file loaded, encoding button clicked, encoding window opened, first pass started.

[dragongodz]

i gave it a try aswell. yep 0 byte debug file and frozen encoding window. and that on windows xp with athlon xp.

tried downloading the ffmpeg source and it quits at 50% every time. tested with flashget and getright. also does not resume.

for free web hosts look here
http://www.free-webhosts.com/

check out the user reviews where you can find out what limits(file types, file sizes etc) there are aswell.

EDIT - infact all the rar files seem to cut in half while the zips download ok. so maybe just using zips for all files is worth trying.

[Peter Cheat]

I've now added mirrors for larger files.

@dragongodz
Is it ok to program in VC++ like this:

m_EncodeDlg.Create(IDD_ENCODE_DIALOG); // Create the encoding window
m_EncodeDlg.ShowWindow(0); // Hide it for now

...when encode button pressed...
m_EncodeDlg.ShowWindow(1); // Show the encoding window

Maybe this is the problem. It works perfectly for me, but other people have trouble. Maybe it is related to the Visual C++ runtime library I am using. I am using 6.02.3104.0 (msvcp60.dll). What version do you have dragongodz?

EDIT: I also have version 7.10.2240.8 (msvcp70.dll). Not sure which one is actually being used.

EDIT2: Debugging shows that msvcp60.dll is being used. Dragongodz, could you try to compile it, maybe then it would work???

[dragongodz]

did you Invalidate(); after showing the encoding window ? that will cause a repaint/refresh.

msvcp60.dll version here is 6.0.8972.0

i cant compile it because i cant get the nuenc source code. its in a rar which as i said cuts out before completing downloading.

[Sectie.B]

@freelock7, Peter1234 & Sectie.B
Try this. I removed the pause function and compiled in VC++ without any optimisations (debug mode). Same result I'm afraid :( (Amd XP1700 & XP+SP2): empty encoding window; high cpu-usage and no output.

[freelock7]

Only the first frame is encoded at low priority.
Too much cpu use in <high priority>.
Now, it shows that in debug_log:
[mpeg2video @ 6551400C]Warning min_rate > 0 but min_rate != max_rate isnt recommanded!
Bitrateviewer report: 27660 kbs at the beginning!

[vmesquita]

@Peter Cheat
Thanks! I managed to download the source using the lycos link. There's a referer problem, i.e., you most visit http://members.lycos.co.uk/petercheat/ and then go to http://members.lycos.co.uk/petercheat/modified-ffmpeg-src-2004-11-14.rar but other than that it's working fine. ;)

[Audionut]

The last 2 versions that you have posted, both encode only 3 frames on my p4.

I B P


Also place "http://members.lycos.co.uk/petercheat/" in front of file name, and flashget downloads "rar" file fine.

[dragongodz]

no good for the nuenc source though since it is not mirrored.

[Peter Cheat]

The source is now zipped instead and also mirrored. I think you have a point there about rar files dragongodz. I've added Invalidate() but I can't tell if it has actaullt done anything. I removed key frame forcing just in case this loop was the problem. Since 3 frames are encoded, this could be the problem. Key frame forcing didn't work 100% anyway.

Try this file (now updated): NuEnc-Test (http://www.petercheat.host.sk/libav/files/nuenc-test.zip)

Let's all hope this one works for everyone.

@freelock7

[mpeg2video @ 6551400C]Warning min_rate > 0 but min_rate != max_rate isnt recommanded!


This is not an error. VBR encoding with a min/max bitrate wasn't recommended before my changes. Should be fine now though ;), provided you can get NuEnc to work at all. I removed this warning from the test version.

Debug.log contains error output from the avcodec library. I redirected the errors there so problems could be solved easier. Hasn't helped much yet though...

EDIT:
Another version with a large amount of debugging info is here (http://www.petercheat.host.sk/libav/files/nuenc-debug.zip)

[fabrice]

Hi,

Got your source, and I think the problem could be with m_nForceKey array: you don't have a size, and don't allocate it dynamically.
So you can do 2 things: declare it as a array of integer, with an index, or as a pointer.

Fabrice

[Peter Cheat]

I was unure about this when writing it, but according to Microsoft documentation, it is legal.


A Microsoft extension allows the last member of a C or C++ structure or class to be a variable-sized array. These are called unsized arrays. The unsized array at the end of the structure allows you to append a variable-sized string or other array, thus avoiding the run-time execution cost of a pointer dereference.

struct PERSON
{
unsigned number;
char name[]; // Unsized array
};


I don't know much C++, I spend most of the time reading info in the MSDN Library. This array began being used in NuEnc 0.00b, and it wasn't a problem then. If it was a memory allocation problem, surely it would freeze before encoding, when the array is declared, no?

Using a pointer would be nicer though, good point.

[freelock7]

It doesn't change anything.
Can you look at the new options you created in the gui.
Sure there is a bad command line or something like this.

[Peter Cheat]

@Freelock7
Grab the debug version from the link above, try to run it, then post the contents of debug.log. It should have quite a bit of information in it.

[freelock7]

[NuEnc]Reading Registry Keys
[NuEnc]Drawing Display
[NuEnc]Initialising Codecs
[NuEnc]Loading Matrices
[NuEnc]Creating Encoding Dialog
[NuEnc]Loaded AVS File C:\Documents and Settings\Freelock7\Mes documents\DVD.avs
[NuEnc]Encoding Starting...
[NuEnc]Displaying Encoding Window
[NuEnc]Successfully Loaded AVS File
[NuEnc]CQ Mode Set
[NuEnc]Beginning Encode
[NuEnc]Encoding Frame 0
[NuEnc]Encoding Frame 1
[NuEnc]Encoding Frame 2
[NuEnc]Encoding Frame 3
[NuEnc]Encoding Frame 4
[NuEnc]Encoding Frame 5
[NuEnc]Encode Complete
[NuEnc]Quitting...

[Peter Cheat]

wtf, apparently it only encodes the first 5 frames, then decides it has finished, and exits normally...Must be a break statement.

Thankyou freelock7.

Edit: It is crashing just before updating the encoding window.

[Peter Cheat]

Freelock7 - update the debug exe with this one NuEnc.exe (http://www.petercheat.host.sk/libav/files/NuEnc.exe) and can you try again?

[freelock7]

[NuEnc]Reading Registry Keys
[NuEnc]Drawing Display
[NuEnc]Initialising Codecs
[NuEnc]Loading Matrices
[NuEnc]Creating Encoding Dialog
[NuEnc]Selected AVS File C:\Documents and Settings\Freelock7\Mes documents\DVD.avs
[NuEnc]Encoding Starting...
[NuEnc]Displaying Encoding Window
[NuEnc]Successfully Loaded AVS File
[NuEnc]CQ Mode Set
[NuEnc]Beginning Encode
[NuEnc]Encoding Frame 0
[NuEnc]Encoding Frame 1
[NuEnc]Encoding Frame 2
[NuEnc]Encoding Frame 3
[NuEnc]Encoding Frame 4
[NuEnc]Updating Statistical Information
[NuEnc]Encoding Frame 5
[NuEnc]Encode Complete
[NuEnc]Quitting...

[Peter Cheat]

The problem is therefore somewhere here:


...
fprintf(stderr, "[NuEnc]Updating Statistical Information\n");
DoEvents();
int nPercent = (int)((((double)i)/((double)AVSReader.GetNumFrames()))*100.0);
bEncodeDlg->m_bEncodeProgress.SetPos(nPercent);

if ( bFlip && bEncodeDlg->IsWindowVisible() )
{
fprintf(stderr, "[NuEnc]Updating Encoding Window\n");
...


"[NuEnc]Updating Encoding Window" is not displayed, its either crashing when trying to update the progress bar, or at the if statement. SoI'll comment out the "bEncodeDlg->IsWindowVisible()" part and try again. Thanks again freelock7.

Just updated here:
http://www.petercheat.host.sk/libav/files/NuEnc.exe (http://)

[freelock7]

The link don't match (invalid url)

[Peter Cheat]

I think its fixed.Try this (http://www.petercheat.host.sk/libav/files/nuenc-fixed.zip). The encoding window can't be closed, you have to move it and press stop on the main window if you want to stop. Hopefully, it will work though.

[freelock7]

17 frames encoded.
Bloks more & more (no response from control panel). No report log.

[Peter Cheat]

Ok, one more try. See if this this (http://www.petercheat.host.sk/libav/files/nuenc-fixed.zip) works. I am out of ideas.

[freelock7]

Freezing...

[Peter Cheat]

Ok, back to the drawing board. Thanks anyway :). Perhaps someone else will find that it works for them. I don't know :confused:.

[freelock7]

The new GUI seems to be unable to work with the new options.
If you delete some of them, maybe you'll find out.

[Sectie.B]

I'm sorry to have to confirm each of freelock7's findings. I hope it has nothing to do with both of us living in Belgium? Now wouldn't that be a rather special, eh bug :p :D

Is it perhaps an option/solution to remove the seperate encoding/statistics-window and just use a simple 1-window-gui as it was in 0.00d?

[dragongodz]

if nothing comes up between now and then i will have a look on thursday. :)

[fabrice]

Hi,

Running in debug mode, with VC7 gives me a lot of failed assertion (with windows, and showWindow?!).
Could be because of different versions of VC++?
I'll change what I see, and send you the changes...

Fabrice

[dragongodz]

Peter - quick dirty fix. :)

in EncodeDlg.cpp change
m_bBufferStatus.Create( WS_VISIBLE | WS_CHILD | PBS_SMOOTH, rect, this, IDC_BUFOC );
to
// m_bBufferStatus.Create( WS_VISIBLE | WS_CHILD | PBS_SMOOTH, rect, this, IDC_BUFOC );

will stuff up the progress bar but it encodes and the other information is updated properlly.

EDIT - ok heres a quick temporary build done with the above mentioned change. so the people who have had problems with it working can try this out until Peter fixes it properly etc. :)
http://www.pcpages.com/dragongodz2/

[Peter Cheat]

So its the buffer status that is the problem. Thankyou very much dragongodz for spending time to work out the problem, and to upload a solution. I'll fix this bit up, and upload a new version asap.

[freelock7]

@Dragongotz
The nuenc0.01-d build works!
Good job.
Go for testing now.

[Peter1234]

Hooray for dragongodz. His version seems to work. XP SP1 SSE2 on P4. I haven't done any testing other than to check that it works.

[dragongodz]

ye sorry i didnt do a proper fix but things are a bit on the busy side.

i also noticed in avsenc.cpp that there is both a showwindow() and enablewindow() together. the showwindow() should(i say should) enable it anyway so the enable is a bit redundant. just something you may want to try Peter. :)

[Peter Cheat]

Yeah, that was a paranoid check. I removed "bEncodeDlg->EnableWindow()" now. And your fix wasn't "quick and dirty". It sets the control back to default. Nothing dirty about that :D. I tried to make it "smooth", but that really doesn't matter...

I've updated here (http://www.petercheat.host.sk/libav/files/nuenc-0.01b.zip). If someone can confirm that this one now works when 0.01a didn't that would be good. CQ mode is now vbv compliant (report buffer underflows if they occur).

[freelock7]

First (quick) results (from Dragongodz link):
-the average don't work in 1Pass.
-In 2Pass, it's ok but quality could be better(low bitrate).
-very slow

(from Peter link):
Same results

[Peter Cheat]

1 pass VBR is broken. I intend to replace it with CQ, targetting a bitrate doesn't work well.

What settings are you using (thats probably what is making it slow).

At least it works, so I can make the release official.

[freelock7]

VBR:2600(2Pass)_KVCD Notch
Turbo:on
interlaced
scene detection
treillis:on
LeastBits
SATD
DIAMOND3(1)
first pass:7.5fps-SecondPass:3.5fps
(1)Not tested other <motion search> options.
Follow up Tomorrow.

[Peter Cheat]

They are slow settings try:

VBR:2600(2Pass)_KVCD Notch
Turbo on
interlaced
scene detection
trellis off
Comp. Fn.
SAD
Adaptive6

I use this and get ~25fps for SVCD and ~15fps for DVD. Depends on your AviSynth script though. I find that trellis is no good for low bitrate, and that Adaptive search is better (because it is so much faster).

[Peter1234]

Peter Cheat,
Version 0.01a did not work for me, but your version 0.01b and dragongodz version do work. With only Mpeg2Source as script I am getting about 30 fps indicated for pass 1 and 2 in 2 pass VBR for 2000 kbps. I have not done any testing other than to check that it works. I am using XP SP1 on 3 GHz P4 with SSE2.

[Fl0ppy]

I'm trying to download the last file but host.sk send me: error 404 file not found.
And other thing, if i try to add a matrix in external file, the quality it's worst.
But if i edit the ini file and i add it, the quality it's superb.
:confused:

[Peter Cheat]

@Fl0ppy
Make sure the file is in XviD format. A simple text file doesn't work, which is why I also added the ini file for easy usage (for n00bs like me who can't be bothered converting to some format, copy&paste is much easier :D). Btw, the link does work, host.sk sometimes gets a bit bogged down.

@Peter1234
Thanks for confirming. Hopefully there will be less problems/complaints now :).

[lanius]

The download from this file doesn't work :(
NuEnc 0.01b based on QuEnc

all other files works :)

[Peter1234]

Peter Cheat,
I have been experimenting with version 0.01b and everything seemed to be working until I tried an xvid matrix. I was able to load the matrix and the program seemed to use it and generated a good output file. I closed the program while testing the output file and now the program will not open. I have reloaded the registry settings you posted but it still will not open. I would like to clean the registry entries but I do not know where to find them. Can you tell me where they are located (and if possible, what changes I should make). Thanks.

EDIT 1: I found the registry entry and deleted the data for m_csCustom. Program will now open but I do not think I will try an xvid matrix again.

EDIT 2: Program is running, but I noticed that selecting an ini matrix from the pull down box does not enter a value in REG_SZ. It still says (value not set) from when I erased the entry. Shouldn't it enter a value for the selected matrix?

[Peter Cheat]

Here is a mirror (http://members.lycos.co.uk/petercheat/nuenc0.01b.zip). You'll have to go to http://members.lycos.co.uk/petercheat/ first, then add nuenc0.01b.zip to the url. I'm still looking for free reliable webspace.

@Peter1234
This is odd. I've use XviD matrices without any problems. I followed this procedure.
- Opened NuEnc
- Clicked on advanced button and selected matrix from file, then found an xvid matrix
- Encoded a trailer 2-pass VBR
- Closed NuEnc
- Reopened NuEnc (no problem opening it)

Are you doing anthing different?

The Xvid matrix code was taken from Nic. I hope its not another problem with the controls I used :rolleyes:.

[Peter1234]

We were typing at same time. Please see my above edit.

[Peter Cheat]

m_nMatNum <- stores matrix selected from ini.
m_csCustomMatrixFile <- stores a custom matrix from file

m_csCustomMatrixFile is chosen if m_nMatNum points to the last entry (From File...)

[Peter1234]

Peter Cheat,
I am using a matrix that worked in QuEnc 0.56. I tried again. I selected the xvid matrix and ran the program and I got an output file. I am using CQ mode. I closed program and it will not re-open. But, there may be a regristry problem now because the address for the xvid matrix did not get entered in m_csCustom this time and program will still not open.

[Peter1234]

Peter Cheat,
m_nMatNum is 0x0000000(0) and m_csCustom is empty. Should I remove and reinstall NuEnc 0.01b?

[Peter Cheat]

Initially there are no keys when NuEnc isrun for the first time. You can't really uninstall it, you could try deleting ALL the keys relating to NuEnc. But I doubt this will solve the problem.

m_nMatNum is 0 and m_csCustomMatrixFile is empty, NuEnc should load without problems, selecting the default matrix (I've tried it and no problems opening it).

If m_csCustomMatrixFile and m_nMatNum are not being written to registry, then NuEnc might not be closing properly (crashing). Is the output file ok (not blocky)?

[Peter1234]

Peter Cheat,
Output video looked good. I will reinstall backup and wait to see if anyone else has a problem. Before I tried the xvid matrix I did select different ini matrix files and that seemed to worked OK.

[lanius]

Thanks !!!! :thanks:

[freelock7]

Final results of my first test on PC (not read on DVD player yet)
-----------------------------------------------------------------
In VBR-2Pass, at low bitrate, I did not see any real improvement in picture quality compared to the preceding versions even if the choice of the advanced options is much vaster and more pointed.
QuEnc0.54 from Amnon's EasyCD/DVD 0.61 with custom matrix produced the best picture.
But, in CQ (30) NuEnc0.01b produces a great quality with a fast response in high bitrate. Very good surprise and it is very fast!
Disadvantage: one does not know the final size of the video.
Undoubtedly NuEnc 0.1b has a great potential with this new CQ setting.
Great job Peter! You must continue to still make it more powerful!
------------
PS: VBR-1Pass not tested with NuEnc0.01b (cause known).
NuEnc0.00a is the best >4000kbs.

[evade]

@Peter Cheat

mencoder is my encoding application of choice on my linux system. would you be so kind as to provide the source for your changed version?

[dragongodz]

In VBR-2Pass, at low bitrate, I did not see any real improvement in picture quality compared to the preceding versions even if the choice of the advanced options is much vaster and more pointed.
remember that the new rate control is trying to do exactly that, control the bitrate. the old rate control does not do a real good job. so the old RC can produce rather big spikes which will look great for those scenes but may not play happy on hardware. while the new RC should produce a more consistant bitrate which should produce a more consistant quality. sort of like xvids new stricter scaling option. thats the hope atleast. :)
as for all the extra options Peter has added to the gui....no comment. ;)

[Peter1234]

Peter Cheat,
Is there a VBV buffer size problem? When I set VBV to 224 in NuEnc, BitRateViewer shows output VBV as 112.

[freelock7]

@Dragongodz
Yes.This is why the new version of QuEnc0.56 functions less better in the top bitrate.
My test are based on PC results not on TV. I know the underflow and VBV problems but I think that the goal of the new libavcodec-on the strict level of quality- is to succeed at the same result of the old version.It would be absurd to evolve and obtain worse performances, no?
But I was not satisfied with your conclusion.
So I tested once again in the morning and I found this strange result(I'm lost here):
QuEnc0.55Alpha -with the new modified libavcodec????-produces the same great quality as its little brother QuEnc0.54.
But-last but not least-QuEnc0.56 is worse!!!
What kind of libavcodec 0.55Alpha is using?

[Peter Cheat]

@Peter1234
I thought this too, but BV reports 1835008 / 1024 / 16 (not in KB) whereas NuEnc is asking for a size in KB (1835008 / 1024 / 8). Just there to confuse us ;).

@evade
I intend to upload all my sources, but I don't see a point in uploading everything while I don't have a reliable host. But I have found an awesome free host, just need to get things set up.

@freelock7
The bit distribution (and quality) depend on the rc equation used. Did you try increasing the value of rc variability to 1 (hint hint). QuEnc uses 0.5 by default, but because of the bitrate tolerance, its value is somewhere between 0.5-1 and varies throughout an encoding. Setting RC Variability to 1 is the same as CQ Mode (but not for 1-pass VBR). The bonus being that you know what file size to expect. The downside is you need to do at least 2 passes. RC variability changes the bitrate response - 0 means no change (CBR), 1 means change to keep constant quality (average quantiser).

@dragongodz
I added more options to clutter the GUI :D. Actually, I added the extra options because I was annoyed that I had to recompile every time I changed a parameter. The fact that it is more complicated to use doesn't hurt me :D.

I've done some testing for macro block based rate control (for potentially higher quality). The problem is that first pass stats file becomes MUCH larger than the actual MPEG. Might be better to just forget about that one. There is no real point for me to port the XviD rate control. I stole the good bits :D.

EDIT:
My aim was to fix the underflow issues. As a consequence, quality may improve as the bitrate follows a strict rule (the rc equation).

[Peter Cheat]

@freelock7
QuEnc0.56 may appear to be worse because Nic is using constant quant 2 for first pass, but I don't think he reduced the scene change threshold to compensate. You then get fewer scene changes detected (avcodec uses a stupid scene change algorithm which detects more scene changes with higher quantisers. Scene changes do not vary with quantiser to my eyes ;))

[dragongodz]

It would be absurd to evolve and obtain worse performances, no?
no. the aim is to stick to that standards and be hardware compatible. who cares how great it looks if when you go to play it on hardware it doesnt play or stutters like crazy. so yes everyone wants the quality to stay as good as possible but there will be times when it has to be reduced.

QuEnc0.55Alpha -with the new modified libavcodec????-produces the same great quality as its little brother QuEnc0.54.
that would be because 0.55 and 0.56 do not use the new rate control. Nic said he was planning 0.60 to be released in 2 versions, 1 with old RC and 1 with new RC.

I've done some testing for macro block based rate control (for potentially higher quality). The problem is that first pass stats file becomes MUCH larger than the actual MPEG. Might be better to just forget about that one
for 2 pass MB rate control is probably overkill. for 1 pass VBR however it would be very good.

My aim was to fix the underflow issues. As a consequence, quality may improve as the bitrate follows a strict rule (the rc equation).
or decrease in scenes that the old RC threw copious amounts of bits at. :)

[freelock7]

@Peter
VBR-2Pass comparison (2500kbs)
I confirm with Bitrateviewer: QuEnc0.55Alpha is better than 0.56 with a versatile response in high bitrate.
Peak in high motion picture at 8200 (5500 for QuEnc0.56).
[Settings:Scene detection/GOP15/DC10/QLB Matrix/interlaced]
@Dragongodz
no. the aim is to stick to that standards and be hardware compatible. who cares how great it looks if when you go to play it on hardware it doesnt play or stutters like crazy. so yes everyone wants the quality to stay as good as possible but there will be times when it has to be reduced.
I do not agree
I encoded with FreeEnc and it is read by my DVD player (some minnor underflow can appear sometimes).
I'm looking for the best picture quality on my TV screen. If not, I choose another encoder which works better.
So why CCE is so expensive?
Because it works fine with a good quality.

[Peter Cheat]

@freelock7
Set the RC variability to 1. The quality will be much better. Try it.

The idea is to maximise quality, and stick within the standards. NuEnc does both, but you have to chose RC variability to your own personal tastes. 0.5 introduces heavy curve compression, 1 is just linear even compression. FreeEnc uses 1 by default (rc_eq=tex).


So why CCE is so expensive?
Because it works fine with a good quality.


No, because they are rip-off merchants :D.

[Easy123]

@Peter Cheat:

If you´ve got a free minute could you look after the Link for NuEnc 0.01b on your Homepage. I´d like to give it a go, but the only thing I get is "Error 404" :(

[Peter Cheat]

Another mirror:
NuEnc0.01b (http://www.petersplace.000k2.com/projects/nuenc0.01b.zip)

[johnnyquid]

Peter,

Since there are a lot more options in the GUI that might confuse people it might be a good idea to have a button that resets all the settings back to default settings.

[Mug Funky]

I do not agree
I encoded with FreeEnc and it is read by my DVD player (some minnor underflow can appear sometimes).
I'm looking for the best picture quality on my TV screen. If not, I choose another encoder which works better and everybody can tell you that.

not so. ask anyone who encodes for a living.

people don't worry about how good the quality is if the thing glitches out on their equipment. they'll only notice the glitches. that's a guarantee.

between a slightly blocky DVD and a slightly glitchy one, i can guarantee that the glitchy one will be returned more often (people feel like snobs if they complain about the quality... don't know why).

personally, i think standards compliance is THE ONLY thing stopping lavc based codecs from being used in the mainstream. once that hurdle is overcome, THEN we can make the quality even better. not the other way around, because you must bear in mind that the good quality we spend time trying to achieve might not be possible in a standard compliant model. we should work with the standards and put in as much quality as they'll allow, being careful not to break any possible compliance in the process.

otherwise we might as well author DVDs with VP6, or RV10, or any number of codecs that a DVD player will not understand.

it seems to me that the point of using MPEG-2 rather than a superior codec is precisely one of standard compliance - who wants good looking mpeg-2 if it doesn't play? we might as well be using a better codec.

[edit - minor grammatical adjustments]

[Peter1234]

Peter Cheat,
After you get all of the important things fixed, could you do something to more clearly indicate when encoding is complete. The progress bar color is not very different when starting and finished, and I usually end up looking at the Remaining Time number to confirm that it is finished. Maybe make progress bar change from gray to red. Ringing a bell when finished would also be a nice touch. I know you have already put a lot of work into this, hopefully you are willing to put in a little more. Thanks for all the work you have done already.

[freelock7]

@Mug Funky
It is explained much better but I never said anything else.
Quality & compliance are working together.
@Peter
No change with RC variability <1>.

[dragongodz]

It is explained much better but I never said anything else.
maybe english isnt your first language so you misunderstood but Mug Funky just said basically what i had already said. to which you replied "i do not agree". so actually you DID say something else.

[freelock7]

Sure! Here I'm talking french! So, excuse me for my bad explanations.

What you have to know, that's my research of the quality.
Everybody is trying to obtain such result by developing new product!
I believe in the great libavcodec potential.
That's all.

[Peter Cheat]

@johnnyquid
Good point. Probably why some are complaining about are quality issues. I'll throw that in next release.

@Peter1234
Another good point. I'll make it more obvious that encoding has finished too.

@freelock7
If there is no change when using RC variability = 1, then you are probably doing something wrong. Btw, did you try more than 2 passes? 3 passes gets to over 99% optimal. Also what settings are you using in FreeEnc that provides better results than NuEnc?

EDIT:
Perhaps I should add support for FreeEnc's templates and reduce the number of options in the GUI? Is this a better idea?

[dragongodz]

reduce the number of options in the GUI? Is this a better idea?
the eternal dilema, how much is too little and how much is too much. :)
its better IMHO to have people saying they wish an option was there than to have them complaining the output is bad because they frelled with things they dont understand(but think they do). :D

add support for FreeEnc's templates
what would be the point though ? to give people back access to things they shouldnt be playing with unless they really know what they are doing ?

[Peter Cheat]

Adding some sort of template support would be beneficial I guess because known settings could be put into files and users would select a template, bitrate, mpeg version etc and the output would be more predictable. Editting those templates would not be a good idea for newbies, but on the other hand, the only way to learn is to try, make a mistake, then figure out what you did wrong. For example, settings used for high quality in avcodec doesn't work well at low bitrates, especially trellis and RD MB decision. But I didn't know that until I tried.

Here is a mirror (http://www.petersplace.000k2.com/projects/) to all the files relating to avcodec. I haven't finished uploading all the sources, but I do intend to. I've rewritten the scene detection algorithm in avcodec, and now intend to implement closed gop with scene detection, the same way as done in CCE (only closing gops when there is scene change). Just need to figure out what the current code actually does in that area.

[dragongodz]

Adding some sort of template support would be beneficial I guess because known settings could be put into files and users would select a template, bitrate, mpeg version etc and the output would be more predictable.
yes i can see the benefit of templates but i just dont agree with the freeenc model of having every option possible, buts thats me. :)

now intend to implement closed gop with scene detection, the same way as done in CCE (only closing gops when there is scene change).
cool. another minor problem is open GOP streams do not start with a closed GOP even when B frames are used, so starts IP instead of IBBP(for example). so the first GOP is always shorter than specified. other encoders such as CCE and Tmpgenc close the first GOP in such cases from memory.

[Fl0ppy]

Hi!! i'm just trying to make a xvcd with this great tool, but i have some questions about it:
Where i can found some explanations about the options like
Mb decision (compare fn, least bits, rd), Compare fn (SAd and satd)and wich it's the best search size?

i tried to make a simple comparasion with an xvid file to vcd, but if i want a loot of quality i need to put Min quant in 1 (but when it plays i see some artifacts).

i tried this conf:
gopsize 12
Min Quant:2 Max quant: 31
Md decision: Copmnf
Compfn: SaTD
Search size: Diamond 0
Aspect ratio 4:3
Noise reduction: 0
Rate control: 0.5
Min bitrate 300
Max bitrate 2500
CQ: 100

Any mini-howto-explanation?

[dragongodz]

SATD is slightly better quality than SAD.
least bits good for low bitrate while rate distortion(rd) may be better for high bitrates.
diamond size should be atleast 1 really if you expect reasonable quality. higher should increase quality(slightly) each step.


how technical do you want the information ?

[freelock7]

Encountered problem with NuEnc01.01b mirror:
If <nuEnc.exe> don't run after installation, just go in the registry at
1)<HKey_LOCAL_MACHINE_/SOFTWARE/NuEnc>
2)<HKey_LOCAL_MACHINE_/SOFTWARE/QuEnc>
3)Delete both of them.
It resets all default parameters.

[Peter1234]

dragongodz,
Thanks for the helpful hints. I tried the following and got much better results at low bit rates (and much longer encode times).

Mode = VBR
Passes = 2
Scene Detection = checked
MB Decision = Least bits
Compare Fn = SATD
Search Size = Diamond 6
RC Variability = 0.5
Matrix = KVCD Notch
Noise Reduction = 0
Use Trellis Quantisation = unchecked
Turbo 1st Pass = checked

[Peter Cheat]

@Peter1234
A search size of 6 isn't worth the extra encode time. Try 3 or 4 (or less). The difference between 3 and 6 is small.

@freelock7
The mirror is the exactly the same file. Shouldn't be any problems.

@dragongodz
Ok, first GOP will always start IP then. It isn't a good idea to have only 1 reference for b-frames ;).

Some of the options in FreeEnc don't even refer to MPEG1 or 2. That is what I think is silly.

[Fl0ppy]

i'm looking the best (if it's possible) configuration for convert mpeg4 to mpeg 1 and 2.
Now tmpgenc offer better quality (i now it's relative).:D
And if it's possible a doc with explanations about the options in Advanced.

When a version with sse2-p4-ht?:rolleyes:

Cheers.

[dragongodz]

Ok, first GOP will always start IP then.
sorry Peter i didnt explain that well enough. what i meant is that when closed GOP is not selected then MainConcept, CCE and Tmpgenc all have the first GOP closed and then following GOPs are open. with avcodec the first GOP is open aswell. as i said just a minor thing really which may not make much difference in the scheme of things. :)
i also checked their first GOP patterns and they all did IP.

When a version with sse2-p4-ht?
avcodec has mmx,sse,sse2(bit buggy) already. as for HT well thats also meant to be rather buggy so should probably wait until everything else is ok. 1 problem at a time. :D

[Peter Cheat]

Is there a reason for closing the first GOP? I don't see any benefits of closing the first gop (unless it is a new scene).

@Fl0ppy
Try the 'HVS Good' matrix.

[dragongodz]

Is there a reason for closing the first GOP? I don't see any benefits of closing the first gop (unless it is a new scene).
no idea. have had a search for a possible reason but havnt found any yet. its just strange that they all do it. as i said it may not make much difference in the scheme of things. if i do eventually find a reason i will let you know. :)

[dragongodz]

hmm ok found this snippet

http://www.geocities.com/medinotes/vcd/chapters.html

if you go down to the Tmpgenc part it says
Another excellent MPEG encoder all round. This one does insert regular MPEG sequence headers, but there is a setting should probably be changed from default in the template:

1. If encoding a VCD, make sure that either the PAL VCD or NTSC VCD template is loaded. Then click on "Configure"
2. Click on GOP structure
3. Now, click and tick the option "Create bitstream for editing"

This will ensure that there are regular closed GOPs and thus regular MPEG sequence headers (as you can see in figure 3, TMPGEnc automatically puts in 1 sequence header per GOP).

If this setting isn't ticked, TMPGEnc can sometimes encode the MPEG in such a way that the GOP isn't closed for an extended interval (> 2 seconds). During that interval, no MPEG sequence header will be present and thus there may be an issue with entrypoint (and thus chapter) placement.
so maybe thats why Tmpgenc does the first GOP closed(big maybe i say).

still for dvd all GOPs should have sequence headers anyway.
http://www.pioneeraus.com.au/computer/dvdInfo/dvdInfo_9.html

so long as sequence headers are neing written every GOP then i dont think it really matters. still as i said before it is strange that the other encoders do it aswell.

[Mug Funky]

FWIW, premiere pro has problems with i-frames. editing is fine, but on export it places the b-frame after the i-frame on the other side of it, causing a kind of "double-take" at scenechanges. this only happens when the user has placed an edit on that scenechange.

i learnt this one the hard way. :(

whether this has anything at all to do with sequence headers and p-frames immediately proceeding i-frames on closed GOPs, i don't know.

[hank315]

Is there a reason for closing the first GOP?
No, there is no real reason.
If you create the first GOP as IPBBPBB... than the closed gop flag may be set to 0 or 1.
But if the first GOP looks like IBBPBB... and the closed gop flag is not set, a lot of decoders and editors will complain even if the first B-frames are backward predicted so it could be decoded fine.

According to the specs:
closed_gop
This is a one-bit flag which indicates the nature of the predictions used in the first consecutive B-pictures (if any) immediately following the first coded I-frame following the group of picture header .
closed_gop is set to ‘1’ to indicate that these B-pictures have been encoded using only backward prediction or intra coding.
This bit is provided for use during any editing which occurs after encoding. If the previous pictures have been removed by editing, broken_link may be set to ‘1’ so that a decoder may avoid displaying these B-Pictures following the first I-Picture following the group of picture header. However if the closed_gop bit is set to ‘1’, then the editor may choose not to set the broken_link bit as these B-Pictures can be correctly decoded.
broken_link
This is a one-bit flag which shall be set to ‘0’ during encoding. It is set to ‘1’ to indicate that the first consecutive B-Pictures (if any) immediately following the first coded I-frame following the group of picture header may not be correctly decoded because the reference frame which is used for prediction is not available (because of the action of editing).
A decoder may use this flag to avoid displaying frames that cannot be correctly decoded.

To be on the safe side and make the stream fully compatible with all decoders just create the first GOP in the same manner as after a scene change and make it IPBBPBB... and set the closed gop flag.

[Peter Cheat]

AFAIK, closed GOP is about how a GOP ends, not how it starts.

Not Closed
I B B P B B P B B I B B P B B

Closed GOP
I B B P B B P B P I B B P B B

(In decoded or display order)

The B-frames in bold are referencing a frame which is unrelated to the current frame - which is wasteful. A closed GOP means that a P frame precedes the next I frame.

CCE and TMPGEnc both start:
I B B P B B P in decoded order (or display order)
1 2 3 4 5 6 7

I P B B P B B in encoded order
1 4 2 3 7 5 6

Avcodec already does this, so I don't have to change anything there. Avcodec also inserts a sequence header with every I frame, regardless of whether the GOP is closed or not, so no problems there.

[hank315]

AFAIK, closed GOP is about how a GOP ends, not how it starts.
It's about the start of a GOP. If a GOP is closed you can make a cut before that GOP and both parts will decode OK.
Also in the case of a scene change you start with a closed GOP so you can make a cut before that GOP.
The order I was referring to was the encoding order so:
IBBPBB --> open GOP --> temporal reference 2 0 1 5 3 4
IPBBPBB --> closed GOP --> temporal reference 0 3 1 2 6 4 5

If you write in display order I B B P B B P B B I B B P B B than the bold B-pics actually belong to the next GOP.

If AVcodec starts with IPBBPBB... (encoding order) it's OK but I still think you should set the closed gop flag.

[dragongodz]

wow what a can of worms i opened with 1 little observation. :D

actually something else i was thinking about Peter, would it be possible for you to make your RC a seperate and selecatable RC ? that way have both old RC and new usable from the 1 program. no need for Nic to release 2 versions of QuEnc for people to test and compare the 2 RC's then. :)

i know Michael(from FFMpeg) said thats what he planned for a second ,better, RC but as far as i know he has still only half done it(the second RC that is).

[Peter Cheat]

Originally posted by hank315
If you write in display order I B B P B B P B B I B B P B B than the bold B-pics actually belong to the next GOP.

True, but this would depend on the definition of GOP (true if considering decoded order).
EDIT: Since the B-frames are referring to 2 different GOPs, how can you say which GOP it belongs to. Is a bisexual person gay or straight?

@dragongodz
Can be easily done. I've finished the code for closed gop and it seems to work as it should. I'll separate RC next.

[Fl0ppy]

Hi again :D
After severals tests (converting xvid and dvd to mpeg1) i tried nuenc 0.01b and tmpgenc.
If in nuenc i put CQ 100, the quality is great, but if i put CQ 80 or 90 the quality in less than tmpgenc at same quality (80 or 90)
And the file created by nuenc/ffmpeg is half of tmpgenc :confused:
And the speed is 32fps (at 352x288 mpeg1) (i can wait);)

[dragongodz]

if i put CQ 80 or 90 the quality in less than tmpgenc at same quality (80 or 90)
And the file created by nuenc/ffmpeg is half of tmpgenc
if the nuenc encodes are half the size then i am not surprised they are of less quality. you can only fairly judge the quality with encodes of the same size.

two things to also remember.
1. mpeg1 encoding with avcodec is not meant to be the absolute best.
2. the CQ mode is brand new so will probably take a little time and tweaking to get towards optimal.

[Fl0ppy]

1. mpeg1 encoding with avcodec is not meant to be the absolute best. :rolleyes:
okay, it's better codify with tmpgencoder actually?,'coz i'm trying to migrate to nuenc.
Thanks.:D

[Mug Funky]

also, TMPG uses a different scale. so 80 in TMPGenc ≠ 80 in QuEnc.

i'd be interested in seeing what equal sized encodes look like between TMPG and QuEnc.

[spase]

like Fl0ppy, i too am encoding xvid to mpeg-2 for dvd purposes... looks like at about 2300kbps abr (maybe a little higher or lower).

i have been testing with quenc to see what bitrates produce acceptable quality (acceptable to me that is) and i decided to give NuEnc a try, just for the fact that it has more than 2 passes (encode time isn't really an issue for me right now) and it has all these matrices built in via the ini file.

in any case, i am test encoding right now @ 2600kbps 3 pass vbr with...

tellis quant. enabled
DC precision: 10
search size: diamond 1
compare fn: satd
MB decision: RD
dynamic b-frames enabled
scene detection enabled
4:3 aspect
noise reduction: 0
rc variability: 0.5
quantiser: 2-31
matrix HVS Good (based on peter cheat's comment to Fl0ppy earlier in this thread)

i plan on comparing this to the encode of similar settings in QuEnc 0.56... i posted in another thread regarding this.

any suggestions for tweaking these options @ about 2300kbps?

anyhow, my REAL questions is: how do i choose a custom matrix from the ini from the command line?... or is this not possible?

the reason i ask, is i am leaving for a few days and would like to write a batch file to let my computer encode while i am gone, whether it be using QuEnc or NuEnc.

[Peter Cheat]

@spase
Custom matrices can not be selected via the commandline yet. The last matrix chosen will be used. If you want to use the same matrix for all the files you won't have a problem.

MB decision: RD is very slow. Least bits may be better.
setting rc variability to 1 gives better quality
trellis is a bit buggy, but should work ok for the settings you've chosen.

The settings you've chosen will make for extremely slow encoding.

@Fl0ppy
CQ Mode in NuEnc is not same as TMPGEnc for a fair comparison, file size must be the same (or at least very close). In NuEnc, CQ Mode is linear while in TMPGEnc its not. This is for easy prediction. Infact, the value represents the proportion of the filesize to expect, where CQ100 is maximum (100%), and CQ80 is very roughly 80% of the size of CQ100. The accuracy depends on what you are encoding. Initially, I had made it very much the same as TMPGEnc, but changed it as requested on the KVCD forum.

@dragongodz
MPEG-1 encoding with avcodec is on par with TMPGEnc's MPEG-1 encoding. I don't know how or why it has a bad name. Try a visual comparison of CQ mode in NuEnc with TMPGEnc (same matrix/filesize).

[dragongodz]

MPEG-1 encoding with avcodec is on par with TMPGEnc's MPEG-1 encoding. I don't know how or why it has a bad name.
possibly because of the RC ?

[Fl0ppy]

I tried a few samp`les (5520frames from xvid file) and a CBR 700, with the same matrix (standard). Results:
The file size it's the same, but the quality it's much much better in tmpgenc. :confused:

[Peter Cheat]

CBR encoding does suck with NuEnc and every avcodec derivative. For a comparison of optimal quality attainable by both encoders, you have to use CQ mode, and try to get the same filesizes.

CBR @700kbit/s is never going to look good anyway.

[spase]

@Peter Cheat: thanks for the quick reply!

now off to encode video forever and author a DVD (using all freeware!)

[Peter1234]

Peter Cheat,
The more I use NuEnc 0.01b, the more impressed I get with it. Several times it has been recommended to use RC Variability = 1 for better results. My tests seem to indicate that this is true. So, when should I not use RC Variability = 1?

[Paced]

Originally posted by Peter1234
Peter Cheat,
The more I use NuEnc 0.01b, the more impressed I get with it. Several times it has been recommended to use RC Variability = 1 for better results. My tests seem to indicate that this is true. So, when should I not use RC Variability = 1?

You should tend to go towards a value of 0 (not necessarily 0, just something lower than 1) when working with lower bitrates.

[Peter1234]

Paced,
OK. That seems reasonable. Thanks.

EDIT: So a value of zero would allow more variablility than a value of 1. Is that correct? I was assuming it was the other way around.

from READ ME file: RC Variability - 1 = Constant Quality, 0 = Constant Bitrate. 0.5 is default (Use 1 for high bitrates).

For low bit rates wouldn't you want to have the bit rate vary a lot so that only those parts of video needing high bit rates would use more bits?

I guess I am still confused.

EDIT 2: After thinking about this for a while I have come to the following conclusion (which may be completely wrong). At low bit rates a larger variability is better since only those portions of video needing more bits should have bits allocated to them. Setting RC Variability to less than 1 should only be done for CBR, and CBR does not give as good a quality of output as VBR. Therefore, for best quality set RC Variability to 1 and only set to less than 1 if CBR is desired. If I am guilty of fuzzy thinking please correct me.

[Peter1234]

Since I was not clear about what RC value was best at low bit rates, I did a test at 773 kbps video with 96 kbps audio using 352x480 mpeg2. The results show that always using RC=1.0 seems to be best. Also shows that CQ mode gives better quality than VBR mode for same file size.

mode_________peak_kbps____ave_kbps____peak_Q______ave_Q____size_____passes


VBR_RC=0.2______2402_________899_______15.0________7.01___16.3_MB______2

VBR_RC=0.5______2121_________897_______12.7________6.85___16.2_MB______2

VBR_RC=1.0______1898_________895_______11.0________6.94___16.3_MB______2

CQ__RC=1.0______1753_________869_______5.94________5.83___15.6_MB______1


Results from BitRateViewer. All paramters except mode and rate control were the same for all tests. kbps values are for video plus 96 kbps audio.

[Peter Cheat]

You are simply looking at Q's and not inspecting the encodes visually. RC alters the curve compression. < 1 means that higher motion scenes will get higher Q's and lower motion scenes will get lower Q's. This is often better visually. But not always.


VBR_RC=1.0______1898_________895_______11.0________6.94___16.3_MB______2

This shows that you need to do more passes. When RC=1.0 you should expect an almost constant quantiser.

Generally for low bitrate encodes, the idea is to take bits away from high motion parts which take up too many bits, and give them back to the low motion parts which we can actually focus on. This is of course not general, and you may find RC=1.0 best for all situations. Hope this better explains what its for.

[Peter1234]

Peter Cheat,
Thanks for the explanation. Yes, I was assuming RC=1 results were better because they had lower peak Q values. But, I understand now what you are saying. Since the average Q values were about the same for all tests, the visual differences were not very noticable. That is why I was using the data values. I ran another test using RC=1 and 4 passes and the results were as you predicted.

mode_________peak_kbps____ave_kbps____peak_Q______ave_Q____size_____passes
VBR_RC=1.0______1873_________907_______7.54________6.61___16.4_MB______4

Thanks again for your help and the work you have put into this program.

[Peter Cheat]

Not a problem :D.

Hopefully I'll be able to find what is causing the min quant=1 + trellis + notch matrix bug, fix it, then release a new update with half a dozen changes and new features.

[buzzqw]

could be possible to add a fuction to log encoding stat ?
Thanks

BHH

[spase]

i have noticed the following behavior with mpeg-2 files i have encoded with NuEnc:

encoded files with average bitrate of "3000" (i put it in quotes because it doesn't necessarily get to that point). In any case, i author a dvd with these files, using dvdauthorgui.

the dvd plays fine on my computer, but is stuttery on my standalone. i thought it was a problem with a peaking bitrate over the maximum dvd allowed, so i encoded with a lower maxbitrate, and i got the same problem.

i opened the files in bitrate viewer and it turns out the bitrate doesnt go over 8900 kbps or so (peak).

so i opened the files with restream and it tells me the vbv is 112, although i set it to 224 in nuenc.

so basically, i think nuenc is not setting the vbv buffer size properly, as it ends up at 112 (according to restream) even though i set it at 224.


if there is any other reason my dvds would be stuttering, please let me know.

[hank315]

so basically, i think nuenc is not setting the vbv buffer size properly, as it ends up at 112 (according to restream) even though i set it at 224. Nuenc sets the buffer size properly, the maximum is 112, some programs set it to 224 but that's wrong and not DVD-compliant.
VBV buffer = 112 means 112 * 16 *1024 = 1835008 bits (229376 bytes) which is the max. VBV buffer for DVD.
Maybe you should try author it with another program like IFOEDIT and see if it reports any VBV buffer underflows.
What I've seen so far, Peter Cheat did a real good job with Nuenc's rate control and preventing VBV underflows.

[dragongodz]

Nuenc sets the buffer size properly, the maximum is 112, some programs set it to 224 but that's wrong and not DVD-compliant.
224KB is shown as 112 in programs such as restream and bitrate viewer and is correct for dvd. i know it would be simpler if everyone stuck to the using just the 1 number but different programs can use either in their options.

[hank315]

@dragongodz
Absolutely right, the buffer is 224 kB and the "magic number" is 112, so a lot of confusion about it. It would be easier if those MPEG guys just put in an 8 instead of 16 in the formula...

[Peter1234]

Peter Cheat explained this on page 6.
"BitRateViewer reports 1835008 / 1024 / 16 (not in KB) whereas NuEnc is asking for a size in KB (1835008 / 1024 / 8)."

[spase]

ok sounds good.

any ideas why the stuttering then?

[Peter1234]

spase,
You haven't provided much information, so I am just guessing, but it sounds like video is at 24 fps and you haven't set pulldown flag. WinDVD will play 24 fps but a set top player will not.

[Peter Cheat]

Originally posted by spase
ok sounds good.

any ideas why the stuttering then?

Yes as Peter1234 said you need to give more details such as frame rate, resolution and GOP size. Some players follow the standards very strictly, and some do not.

[spase]

the fps is 25 (as far as i can tell)

i popped it back in the dvd player to see if i could check out some of the behavior of the dvd, and it seems to play fine on more complex scenes, but on simple scenes where there is maybe just some text on the screen, it gets all stuttery.

[freelock7]

It appears when the quantisation is too low .
Put the min_Quant at 2

[dragongodz]

It appears when the quantisation is too low .
it could infact be a lack of bitrate aswell. some dvd players(yes the good old sony etc brands) have problems if the bitrate drops too low. some require the minimum to be over 150 while i have read others prefer over 300. so setting a higher min bitrate(300+) may also help.

[spase]

hmm thank you i will try this min bitrate option....


come to think of it, is there anyway to pad streams?

(eg. add bitrate without re-encoding...)

i am looking it up on google now... feel free to answer tho.

[spase]

nevermind... i decided i would re encode again... mostly because i want to see what kind of quality i get with jawor's 1CD quant matrix, as well as using some other filters in avisynth (namely UnDot and FluxSmooth)

*crosses fingers*

(and waits another day or so)

[Peter Cheat]

I suggest encoding a small sample section that was affected by stuttering and testing that on a DVD-RW. The you will know if setting the minimum bitrate will help. I'm pretty sure it will though.

[IcarusX75]

Spase:

The problems you are having sound very familiar to the problems I am having. Are you using a Pioneer DVD player by any chance?

I've tried changing min/max bitrate, Min. Quant, GOP, VBV, etc... but whatever I do, I can't get an MPEG-2 encoded with any of the AVCODEC front-ends to work in my player without extreme freezing (ie: unwatchable). MPEG-2 XSVCDs created with TMPGenc work fine.

If you find a solution, let me know!

Cheers,
Icarus

[Peter Cheat]

@IcarusX75
The problem spase is having isdifferent to yours. Scenes that are almost static are only affected, indicating a minimum bitrate related problem.

Your problem occurs all the way through the video. And you did pulldown correctly, which gave me no obvious remaining possibilities.

[dragongodz]

well since nobody else know what he did, since thats his first actual post, i guess we have to stab in the dark. :)

this wouldnt have anything to do with it ?
http://forum.doom9.org/showthread.php?threadid=80636
http://forum.doom9.org/showthread.php?threadid=75923

[Peter Cheat]

From KVCD Forum:


I'm trying to use the same settings in TMPG and NuEnc. GOP size of 12, framerate of 23.976 (adding pulldown post-encode with DoPulldown GUI). Resolution 480x480 (I'm in NTSC-land).

I have never had a problem with an original DVD, never had a problem with any XVCDs, or SVCDs that I haven't made myself.


I assume Icarus3000 (from KVCD forum) and IcarusX75 are the same person :).

[spase]

alas, i have no dvd-rw... plenty of dvd+r though

and no, it is a panasonic standalone, which has been pretty good to me in the past (albeit a little picky)... it has played any sort of disc i have thrown at it, except DVD+R DL.

as for the encoding, it is going well. i figure i might as well re encode everything, that way if it DOES fix the problem, then i will have a nice shiny new dvd to play.

in any case i am in no hurry.

one other thing tho. i have found that using these custom matrices (HVS_good, QLB, or Jawor's 1CD) lowers the resulting average bitrate by quite a bit. Jawors has lowered it by about 14% or more. setting the target 14% higher hasn't changed this, so i am setting it higher still on subsequent encodes. any idea how i can more accurately predict the actualy resulting average bitrate with regards to the use of custom matrices that increase compressibility?

EDIT: one other question i just thought of.

25 fps is generally used in PAL sources right? is it a problem to use 25 fps on NTSC sources? (size, aspect ratio, etc are all ok, i checked that)

[dragongodz]

I assume Icarus3000 (from KVCD forum) and IcarusX75 are the same person
ah that would explain things. :)

Icarus try setting the "program sequence" to interlaced in dopulldown gui and see if that helps.

i have found that using these custom matrices (HVS_good, QLB, or Jawor's 1CD) lowers the resulting average bitrate by quite a bit.
yes Jawors 1cd and QLB(Quenc Lower Bitrate) are for lower bitrate encodes. just using standard for higher bitrate encodes is fine.

setting the target 14% higher hasn't changed this, so i am setting it higher still on subsequent encodes.
then it has probably hit saturation for that matrix. the only solution is to not use it.

25 fps is generally used in PAL sources right? is it a problem to use 25 fps on NTSC sources? (size, aspect ratio, etc are all ok, i checked that)
some of the modern dvd players probably wouldnt mind but some, such as sony and panasonic etc, which are more fussy may not like it. the only way to know for your player is to try. you would probably be better off converting to 23.976fps and doing pulldown though. especially if you want to make sure it works on any dvd player you may buy in the future. :)

[IcarusX75]

@Dragongodz:

My source is 23.976 progressive... I should still set program stream to interlaced in DoPulldown?

I should mention that my player CAN play straight 23.976 from TMPGenc, so I don't think that is the problem, but I'm willing to try anything.

(Yes, I am Icarus3000 on KVCD board - but my Icarus3000 account here would not let me post, so I created a new one)

Thanks,
Icarus

[dragongodz]

My source is 23.976 progressive... I should still set program stream to interlaced in DoPulldown?
yes set the program sequence to interlaced. it just changes that flag so doesnt harm the actual footage at all. read the 2 links i posted to see why thats needed if the authoring program doesnt do it.

I should mention that my player CAN play straight 23.976 from TMPGenc, so I don't think that is the problem, but I'm willing to try anything.
how is that authored though ? is the authoring program doing the pulldown maybe(but not program sequence) ?

[spase]

Originally posted by dragongodz
yes Jawors 1cd and QLB(Quenc Lower Bitrate) are for lower bitrate encodes. just using standard for higher bitrate encodes is fine.


then it has probably hit saturation for that matrix. the only solution is to not use it.


some of the modern dvd players probably wouldnt mind but some, such as sony and panasonic etc, which are more fussy may not like it. the only way to know for your player is to try. you would probably be better off converting to 23.976fps and doing pulldown though. especially if you want to make sure it works on any dvd player you may buy in the future. :)

glorious.

well we shall see what happens... i dont really know how to do the conversion and pulldown etc. i will find out for myself (im sure doom9 has a nice little guide or something) if this current round of encodes doesnt work. the more i think about it though, the more i think that it is a min bitrate problem, because upon looking at the troublesome files in bitrate viewer, the approximate spots with problems dropped down to ~250kbps. i am assuming panasonic might have a 300kbps min bitrate on this player, but i couldnt find anything in the documentation.

i guess i will switch back to a regular matrix... unless there are any suggestions of a matrix to use for bitrates of about 2500kbps (2148 to be exact).

[dragongodz]

you could probably add a little noise using an avisynth filter(dont ask which one, its 1:28AM here and my memory isnt that good at this time of the morning :) . that should help push up those low areas.

actually 2148 is pretty low for dvd size picture so using a lower bitrate matrix would normally be a good idea. you do have to take in to account the type of footage aswell though. near black scenes compress down pretty small without any help from such a matrix while bright scenes can be greatly helped by them using the lower bitrate.

now i am off to bed. :D

[IcarusX75]

@Dragongodz:

this wouldnt have anything to do with it ?
Ahh... very interesting links, thanks! Now I can't wait to get home and try to re-mux with the interlaced stream set...

I hope that solves my problem, because I really appreciate the speed and quality of NuEnc!

[spase]

thanks for all your help dragon.

i am encoding right now with standard matrix. we will see how the quality compares. i have a few scenes in this one file that i have been using for testing. i will see how they come out.

so far i like the way the jawor matrix looked. maybe i will just use that and add more footage to the dvd than originally expected. hate to leave like a decent amount of empty space on a dvd+R

[IcarusX75]

@DragonGodz:

Should I set any of the other parameters in DoPulldown, or just the program sequence? If so, how should I set:

Program Sequence: (Interlaced, as you mentioned already)
Program Frames: (Interlaced/Progressive?)
Drop Frame: (Enabled/Disabled?)
No Timecodes: ?
No RFF: ?

Thanks,
Icarus

[Fl0ppy]

I found this problem:

When i try to open a divx311 with 24.95fps nuenc says:

"Could not open Video codec
Please check debug.log for more information (...)

And the debug.log:

[mpeg1video @ 6551400C]MPEG1/2 doesnt support 1000000/40071 fps
[mpeg1video @ 6551400C]MPEG1/2 doesnt support 1000000/40071 fps
[mpeg1video @ 6551400C]MPEG1/2 doesnt support 1000000/40071 fps

The script avs it's too easy:

AviSource("c:\xxxx\potato.avi")
BicubicResize(352,288,0,0.75)

But i can encode with quenc :confused:

[Bogalvator]

Fl0ppy,

Add the line assumefps(25,sync_audio=true) to the end of the script ;)

You probably also want to add SSRC(44100,fast=false) after that to return the audio to the correct sample rate (replace 44100 with 48000 etc if required).

[Fl0ppy]

Ok thanks, i tried it and no problem, but tmpgenc gives me more quality
(in this cas, the conversion mpeg4 to mpeg1).

I proved one (mpeg2 to mpeg1) and when multiplex with tmpegnc it did not give error (buffer underflow), but when reproducing it in dvd it was stopped. I multiplexed again with bbmpeg like vcd VBR, gave many "error underflow", but it works perfectly.

:confused:

[dragongodz]

Should I set any of the other parameters in DoPulldown, or just the program sequence?
living in PAL land i have no idea. its not a problem we have. :)

going by what was said in the other threads the program sequence needs to be changed but the program frames dont. also i remember in the Rejig thread when using its authoring sometimes the drop frame had to be set for NTSC so that may have to be aswell. all i can suggest is try 1 change at a time and test it.

[Peter Cheat]

I think you only drop frames if the original frame rate was 29.97 and you convert to a 23.976 progressive stream. So you start with a 29.97 fps source, perform Telecide() to get a progressive stream (with duplicated frames), and you drop the duplicated frames (with Decimate()). I've only ever handled NTSC sources once in my life (I'm in PAL land too). I used AviSynth. I think you have a 23.976 interlaced source though? Correct me if I'm wrong.

[Fl0ppy]

It's normal?

BBMPEG
=======

Video: none

Audio: none

Output MPEG information
Video: none

Audio: none

Multiplexing: video and one audio stream
program stream type: VCD
video file: c:\nuenc\test1.m1v
audio 1 file: c:\nuenc\audio.mp2
program file: temp.mpg


Multiplexing video and audio:

Scanning video stream for a sequence header ...

Scanning video stream for pictures ...
Found 177896 picture headers.
Video stream information
Stream length : 657448648
Total time (seconds) : 7115
Sequence start : 15980
Sequence end : 1
No. Pictures : 177896
No. Groups : 15980
No. I Frames : 15980 avg. size 12454 bytes
No. P Frames : 104434 avg. size 3419 bytes
No. B Frames : 57482 avg. size 1761 bytes
No. D Frames : 0 avg. size 0 bytes
Horizontal size : 352
Vertical size : 288
Aspect ratio : 0.9157
Picture rate : 25.000 frames/sec
Bit rate : 312500 bytes/sec (2500000 bits/sec)
Computed avg rate : 92500 bytes/sec (740000 bits/sec)
Computed max rate : 359250 bytes/sec (2874000 bits/sec)
Vbv buffer size : 40960 bytes
CSPF : 0

Scanning audio stream for access units information
Found 272166 audio frame headers.
MPEG audio stream information
Stream length : 113754006
Syncwords : 272166
Frames : 11108 size 417 bytes
Frames : 261058 size 418 bytes
Layer : 2
CRC checksums : no
Bit rate : 16000 bytes/sec (128 kbit/sec)
Frequency : 44.1 kHz
Mode : 0 stereo
Mode extension : 0
Copyright bit : 0 no copyright
Original/Copy : 0 copy
Emphasis : 0 none

Multiplexing information
Video stream data rate : 312500 bytes/sec (2500000 bits/sec)
Audio stream 1 data rate : 16000 bytes/sec (128000 bits/sec)
Overhead data rate : 4300 bytes/sec (34400 bits/sec)
Total data rate : 332800 bytes/sec (2662400 bits/sec)

Multiplexing file temp.mpg
video DTS (6390144.00ms) underflow at pack 290217 by 11.24ms
video DTS (6390184.00ms) underflow at pack 290222 by 6.20ms
video DTS (6390624.00ms) underflow at pack 290290 by 41.73ms
video DTS (6390664.00ms) underflow at pack 290294 by 29.70ms
video DTS (6390704.00ms) underflow at pack 290297 by 10.68ms
video DTS (6390824.00ms) underflow at pack 290315 by 16.56ms
video DTS (6390864.00ms) underflow at pack 290322 by 25.50ms
video DTS (6390904.00ms) underflow at pack 290327 by 20.47ms
video DTS (6390944.00ms) underflow at pack 290331 by 8.44ms
video DTS (6391064.00ms) underflow at pack 290349 by 14.31ms
video DTS (6391104.00ms) underflow at pack 290359 by 44.24ms
video DTS (6391144.00ms) underflow at pack 290366 by 53.20ms
video DTS (6391184.00ms) underflow at pack 290371 by 48.17ms
video DTS (6391224.00ms) underflow at pack 290374 by 29.14ms
video DTS (6391264.00ms) underflow at pack 290377 by 10.12ms
video DTS (6391344.00ms) underflow at pack 290392 by 35.01ms
video DTS (6391384.00ms) underflow at pack 290395 by 16.00ms
17 video underflows (SCR >= PTS or DTS)

NOTE: The resulting file may not play back correctly.
Try reducing the video bitrate or increasing the video and
audio startup delays.
Finished multiplexing temp.mpg

With VBV 40 (not 224)

Another question: Vbv afects to the quality? :confused:

[IcarusX75]

@Peter:

Yes, I have a 23.976 progressive film source.

I tried changing the Program Sequence Flag to interlaced, but that didn't seem to correct my problem.

I'll keep trying...

- Icarus

[dragongodz]

try setting the drop frame aswell.

what are you using to author with ?

[Peter Cheat]

@Fl0ppy
Are you using 2-pass VBR (or 3 or more passes?). This should not happen in x-pass VBR, in 1-pass it can (don't use 1 pass VBR), and it CQ mode it could possibly.

A smaller VBV will affect quality because it reduces how much the bitrate can vary.

@IcarusX75
If its already 23.976fps, dropping frames is not an option (and shouldn't be). Could you explain exactly what software you are using to author the DVD and the settings you are using. Setting the program sequence to interlaced won't help, your output is progressive. Don't use pulldown unless you want to convert to 29.97fps interlaced. Its already progressive, Pulldown is used to convert 29.97fps interlaced to 23.976 progressive or vice versa. You have a progressive sequence already. Try without pulldown.

[dragongodz]

Setting the program sequence to interlaced won't help, your output is progressive. Don't use pulldown unless you want to convert to 29.97fps interlaced. Its already progressive, Pulldown is used to convert 29.97fps interlaced to 23.976 progressive or vice versa. You have a progressive sequence already.
23.97fps is not a valid dvd framerate and can only be used with pulldown. in the links i already gave tests show that the program sequence needs to be set to interlaced even for progressive encodes unless the authoring program goes and does it.
also way back in the huge Rejig thread the drop frames had to be added to the authoring section for some problems with some NTSC outputs. i dont remember the exact cases and have no intention of going back through it to find it. i will leave that to someone else. :)

[spase]

so i just got around to authoring and burning the DVD i encoded with the minimum bitrate set (i set it to 512k just for the hell of it) and i still get the jitteriness with these files.

could the problem be with dynamic b frames?

maybe it really is a max bitrate issue. i set the max bitrate to 9550 and bitrate calc shows no spikes over that on all of the files. the audio is one track of 192kbps.

it always seems to take place at the beginning of the files, especially in moments that are less complex than most (i.e. a lot of black area)

i havent checked the ends of the clips where there is also a lot of black areas, but i will check that later.

does anyone else have ideas of whats going on?

one other thing of note: it seems all the quantisers average to about 2.00 or so. is that extremely low?

[Fl0ppy]

Originally posted by Peter Cheat
@Fl0ppy
Are you using 2-pass VBR (or 3 or more passes?). This should not happen in x-pass VBR, in 1-pass it can (don't use 1 pass VBR), and it CQ mode it could possibly.


I'm using (testing) with CQ and quantize values, but if i do:

Min. Quantiser 1-31 with CQ 60 (the quality it's great for a vcd) but more than 137 buffer overflows, if i try to multiplex with bbmpeg.....:eek:

I think it's better encoding with CQ, not with VBR 2 at 800 o 950 (x) passes. The quality looks better. (always im' talking for a vcd)

But it's my experience.

Anyway::thanks:

[mean]

Dragongodz is right,same for SVCD

Only for VCD 23.976 is valid (and not always supported correctly)

The patch to add pulldown flag to lavcodec has been available for some time and reposted recently on lavcodec ml

[Peter Cheat]

@mean
I just found the discussion on ffmpeg devel list. I'll add it. I've used 23.976 as a frame rate for SVCD, I didn't even realise it wasn't actually part of the standard.

@Fl0ppy
I've improved CQ mode compliance, I know it wasn't perfect in the release version. I'll release a fixed update in the next few days.

[Yuri]

@ Peter Cheat
Astounding results! Great work!
A pity is that NuEnc inputs only 'avs' but not 'avi'.
I was encoding through 'SegmentedAviSource' in an avs and avisynth caught an access violation error. This might not have happened if the input were an avisynth-free avi.

The script:

SegmentedAviSource("I:\video-1.avi") + SegmentedAviSource("S:\video-2.avi")
ConvertToYV12()

[Bogalvator]

Yuri,

AviSource(...) + AviSource(...) should work as expected, no need for "Segmented".

Check out the docs for proper usage of SegmentedAviSource (I had a few troubles myself getting it to work first time)

[Yuri]

@Bogalvator
...no need for "Segmented".

In my case I do need "Segmented"

I use the 'SegmentedAviSource("I:\video-1.avi") + ...' script because on my disk D: I have files: video-1.00.avi, video-1.01.avi, video-1.02.avi and so on up to video-1.07.avi - disk D: is full and the continuation is on disk S: - video-2.00.avi, video-2.01.avi, video-2.02.avi and so on up to video-2.12.avi - this is Huffyuv's lossless job of a 1.30 hour film for a 3 pass encoding.

So, if I use 'AviSource' command, I am pretty sure to get an error message like this: Avisynth open failure: AviSource autodetect: couldn't open file...

Regards to all

[Bogalvator]

Yuri,

Fair enough, you do know what you're doing!

Perhaps try having a video-all.01.avs script containing the line "Segmentedavisource("i:\video-1.avi")" and video-all.01.avs containing "Segmentedavisource("i:\video-2.avi")".

Then have a final video.avs contaning the line "segmentedavisource("video-all.avs") and see you have any joy with this.

[Yuri]

Fair enough, you do know what you're doing!

Perhaps try having a video-all.01.avs script containing the line "Segmentedavisource("i:\video-1.avi")" and video-all.01.avs containing "Segmentedavisource("i:\video-2.avi")".

Then have a final video.avs contaning the line "segmentedavisource("video-all.avs") and see you have any joy with this.

Fair enough, I do know what you're explaining! Sorry!
Anyway? I still need 'SegmentedAviSource' command.
I was following the avisynth doc in the segmentedsource.html
--------------------------
SegmentedAVISource / SegmentedDirectShowSource
SegmentedAVISource(string base_filename[,...], bool "audio", string "pixel_type")
SegmentedDirectShowSource(string base_filename[,...][,fps])

The SegmentedAVISource filter automatically loads up to 100 avi files per argument (using AVISource) and splices them together (using UnalignedSplice). If "d:\filename.ext" is passed as an argument, the files d:\filename.00.ext, d:\filename.01.ext and so on through d:\filename.99.ext will be loaded. Any files in this sequence that don't exist will be skipped.

SegmentedDirectShowSource works the same way.

From v2.04 up there is bulit-in support for ACM (Audio Compression Manager) audio (e.g. mp3-AVIs).
------------------------------------


BTW, this flame does not contribute to the issue that NuEnc inputs only 'avs' but not 'avi'.

Best of the best regards.

[Peter Cheat]

@Yuri
SegmentedDirectShowSource is not the same as SegmentedAVISource, one uses DirectShow, the other uses Video For Windows (I think). Try SegmentedDirectShowSource and see if the same error comes up (it may or may not). Adding AVI adds complications, the easiest thing to do would be to automatically create an AVS script to handle the AVI. This would obviously not solve your problem. I suck at VFW programming, and suck even more at DirectShow. Sometimes DirectShow is faster, depending on the codec being used (ffdshow has a Huffyuv DirectShow Decoder).

[Yuri]

Thank you, Peter, for your explanation and your contribution.
I'll try SegmentedDirectShowSource.

Best regards
Yuri

[dragongodz]

if you look back you will find QuEnc has had a few problems with avisource() that were fine with directshowsource() aswell. hopefully those problems are fixed now in QuEnc 0.56. did you try segmented with QuEnc 0.56 aswell by the way ?

[spase]

ok so i have tried both NuEnc and QuEnc with the encodings I was doing, and the stuttering I described was only present on the NuEnc versions.

Here are the command lines I used:
(note: QLB was used on all as the matrix)

NuEnc Command Line Options:

-i ep1.avs
-o ep1.m2v
-b 2148
-maxbitrate 9550
-dc 10
-priority 5
-pass 3
-mpeg2
-vbr
-scene
-qmin 2
-qmax 31
-mbd 2
-cmp 1
-dia 7
-trell
-nocgop
-nointerlaced
-gopsize 12
-maxbframes 2
-aspectratio 4:3
-buffer 224
-dynamic
-rc_var 1
-auto
-close

(minimum bitrate of 512 was set in the GUI)


QuEnc Command Line Options:

-i ep1.avs
-o ep1.mpg
-b 2148
-maxbitrate 8000
-dc 10
-priority 5
-2
-mpeg2
-hq
-vbr
-scene
-nocgop
-nointerlaced
-extreme
-gopsize 12
-maxbframes 2
-qlb
-aspectratio 4:3
-mpeg2mux dvd
-audiocodec ac3
-audiobitrate 160
-auto
-close

the issue doesn't appear to be a problem with over shooting the 9800kbps limit, as some of the QuEnc files jump to over 9550kbps (above the upper limit I set in NuEnc)

additionally, dropping to an extreme low bitrate doesn't seem to be the source of the problem either, as the QuEnc encodes drop below 250 kbps at some points.

for some reason i have a feeling the problem has its roots in the use of dynamic b-frames, but i have absolutely no idea why i think that... must be some sort of intuition or something. I havent had the time to play around with encoding video without dynamic b frames tho.

[dragongodz]

yes i would suggest trying with scene detection and dynamic B frames turned off and see if it still happens.

[spase]

ill give it a go with no dynamic b frames, but scene detection on in QuEnc was ok....

[Peter Cheat]

Originally posted by spase
ill give it a go with no dynamic b frames, but scene detection on in QuEnc was ok....

You could also try only 2 passes in NuEnc. There is the possibility that I've overlooked something when implementing >2 passes that doesn't make the stream 100% compliant. I just checked through the code and it seems fine, and its been tested on my standalone, but this doesn't mean I haven't bungled up something.

[dragongodz]

but scene detection on in QuEnc was ok
from memory Peter changed scene detection sensativity(thats right isnt it Peter ?) so that may also be a factor which is why i said you may want to try disabling that aswell.

[Peter Cheat]

It will only throw in more I frames, it shouldn't bugger anything else up.

@spase
What is the source frame rate?, and is the source interlaced?

[spase]

source is 25 fps and is not interlaced.

i am going to play around with this stuff when i get a little more time. finals week is here at the university of maryland, and i will be writing some papers on dogen's zen essays (shobogenzo) and even more other things, so once im done with that i am thinking of doing several test encodes with NuEnc with some opeions enabled/disabled etc.

will report back soon.

[danpos]

@PeterCheat

Hello! I just wanna know if you´ve got a forecast for NueEnc new release. I like of this encoder very much and I´m anxious to try it with new improvements and features.

Keep up on work!:)

TIA,

[Peter Cheat]

Well, I am without a PC atm, my Athlon blew up, and I have to wait for a replacement, *hopefully* in the next couple of days. That's what the salesman told me anyway, and before that he told me he'd have the parts in last week :rolleyes:.

[danpos]

@Peter Cheat

I already knew about your PC issue (you talked about that at kvcd.net ...) and I hope so that you get to assemble your new machine ASAP.:) Anyway, I´m playing with NuEnc 0.21 (test release)that you gave to me and I´m getting very, very good results at low bitrate encodings (1 CD target-XSVCD - I´m getting do it for a DVD movie action source, 104 minutes long, with a good CQ - yep, I´m doing OPV with NuEnc :p ), but that you alredy knowing (because of our talks at kvcd.net forum;) )

Keep up on work!

Kudos,

[Xero_M]

@Peter Cheat

Is ther i chans that you can send ore have a link fore the test release of nuenc 0.21. I cant whait fore the next realise of nuenc, in my opinion the best encoder att low bittrate encoding, and I have tested manny encoders.

[Peter Cheat]

I've borrowed a PC for now, I'll update asap (tomorrow).

[reboot]

What's with all the busted download links?
1b doesn't work at all.
mencode and mirror only give partial/corrupt .rar's/doesn't work.
I think if you want people to use this, you should find some hosting that's reliable, and someone to post a link that stays working.

[Bogalvator]

There's an alternative mirror here:
http://www.petersplace.000k2.com/projects/
(Peter Cheat should probably add it to his sig like he did for kvcd.net...... ;))

If you still have problems, try using a download manager like GetRight etc

[Nic]

@Peter:
Hi Peter,

Hope you get up and running at full pelt soon. Let me know when you have your latest libavcodec changes up on your website, as I want to start integrating a bit of your ratecontrol...as I think ratecontrol is now the biggest issue with QuEnc (now I've got "most" of the bugs out of it ;) )

Cheers,
-Nic

[Trahald]

Originally posted by Nic
@Peter:
Hi Peter,

Hope you get up and running at full pelt soon. Let me know when you have your latest libavcodec changes up on your website, as I want to start integrating a bit of your ratecontrol...as I think ratecontrol is now the biggest issue with QuEnc (now I've got "most" of the bugs out of it ;) )

Cheers,
-Nic

any news on this?

[dragongodz]

someone is better off asking on the kvcd forum. previous comments in this very thread suggest thats where Peter is posting all his latest news etc.

[Bogalvator]

Just checked and Peter last posted about NuEnc at KVCD.NET forums on 31 Dec, so don't worry, there's no new developments that you're missing out on.

I'm quite sure any news/updates about his work on NuEnc etc will be announced here as soon as it's known as well as on kvcd.net

[dragongodz]

Peter last posted about NuEnc at KVCD.NET forums on 31 Dec
yes and besides the post on the 9th of Jan saying his pc had blown up and the 18th of Jan saying he had borrowed a pc he had not been heard from here since the 10th of December.
also danpos said Peter had already talked about his pc blowing up and he was trying out a 0.21 version he had been sent.

make of that what you will. :sly:

[manolito]

I just found out that Danpos has published a short NuEnc tutorial in Portuguese at
http://www.vmesquita.com/forum/index.php?topic=1577.msg12176#msg12176
The nice thing is that he also hosts the latest NuEnc 0.21 test version.

I don't know if it is alright with Peter Cheat to post the link since he has not officially released version 0.21, but Danpos has only good things to tell about it.

Cheers
manolito

[danpos]

@manolito

Hi, guy! I just want to clarify that question. I was also unsure with respect to host a version not formally released for all, but thinking more over I did conclude that I can to do it for the following reasons: 1-The NuEnc version is a test one (beta), which is aimed at development of encoder (test of options, bug tracks etc); so, in this manner,if more people can to test it, more easy and rapdily the encoder development will go ahead; 2-I did a explicit communication that the download version is a non-official release (which will be provide for the owner, Peter Cheat); 3-As a admiring from Peter's work, I did want divulge it in brazillian community; so, I thought that a new version, with more features and fixes will help in disseminate it.

Obviously, I'll like to talk with NuEnc owner (Peter Cheat), but as you've noted, the same is off a very long time. My intentions are the best as possible. If all think that I did a fault, I simply deleted the file from my webspace.

I hope have got clarified that issue.

Cheers,

[freelock7]

Hope to get news about Peter Cheat soon:)
We're all still waiting for its new RC.

[manolito]

NuEnc has a slightly different command line interface from QuEnc, and if
NuEnc encounters a parameter it does not know, it simply quits with a "Bad
Parameter" message. Result: DVD Rebuilder and DVD2SVCD do not work with NuEnc.

Peter Cheat seems to have gone into hibernation, so there is little hope that
he might fix the command line interface any time soon. Since I am a little too
stupid to modify the NuEnc sources and recompile it I decided to write a small
utility that sits between the application and NuEnc and translates the command line
parameters so NuEnc can understand them.

Download at http://scifi.pages.at/manolito/NuCLI/NuCLI.zip

Tested under Win XP with NuEnc 0.01b and NuEnc 0.21. Works with the latest builds
of DVD2SVCD and DVD Rebuilder 0.74 (could not test the VIP versions).


@danpos
I like version 0.21 very much, do you have any idea what Peter Cheat changed
since 0.01b? There is no changelog...


Enjoy
manolito

[danpos]

Originally posted by manolito

@danpos
I like version 0.21 very much, do you have any idea what Peter Cheat changed
since 0.01b? There is no changelog...


Enjoy
manolito

Hi, manolito! I've no idea of modifications applied to NuEnc by Peter, because of I had access only the Nuenc.exe, matrix.ini etc ... I do agree with you with respect to that release, it is really very, very good! I've got very good encodings for MPEG-2 targets (DVD,XSVCD), although for MPEG-1 targets something must be improved. BTW, it is my favourite encoder ATM.

Cheers,

[nnigam]

I have read a lot of good things about this encoder, but where do I get the latest version. All links that I followed go to version .01. I currently am using QuEnc and would like to try this as well.

[manolito]

@nnigam:
Just go 5 posts upward, there you will find a link to a NuEnc tutorial by danpos. At the bottom of this tutorial there is a link to NuEnc 0.21.

Cheers
manolito

[nnigam]

Thanks

I had looked earlier, but obviously not well enough. This time I looked more thouroughly and found it.

[manolito]

@danpos

I have done a couple of encodes with NuEnc 0.21, mainly low bitrate VHS capture to SVCD encodes, and I must say, I am impressed. I used the settings from your tutorial with the BVCD matrix, and I cannot find any quality decrease compared to the original VHS tape.

But I do have a couple of questions, and since Peter Cheat is not available ATM, I guess you are the person to ask:

The settings you propose in your tutorial are all optimized for speed. If I want a better visual quality without too much of a slowdown, which setting would be the best to change? MB Decision from Comp Fn to Least Bits or RD, Compare Fn from SAD to SATD, or Search Size from Diamond 1 to 3 or even higher? Or is Adaptive better? I found that Adaptive 4 is about as fast as Diamond 1, which one would look better?

I understand that you did test NuEnc's CQ mode, and I'd like to know if CQ mode is VBV and bitrate compliant. In other words, is the resulting stream DVD (or SVCD) compliant by observing the Max bitrate? If so, we could have someone (Amnon82, JDobbs or Tylo) write a prediction tool for NuEnc which would give us a high quality one pass VBR encoder which also happens to be FREE.

Cheers
manolito

[danpos]

Originally posted by manolito
@danpos

I have done a couple of encodes with NuEnc 0.21, mainly low bitrate VHS capture to SVCD encodes, and I must say, I am impressed. I used the settings from your tutorial with the BVCD matrix, and I cannot find any quality decrease compared to the original VHS tape.

I'm glad that my humble tutorial has been useful for you (and mayble others :);)).

The settings you propose in your tutorial are all optimized for speed. If I want a better visual quality without too much of a slowdown, which setting would be the best to change? MB Decision from Comp Fn to Least Bits or RD, Compare Fn from SAD to SATD, or Search Size from Diamond 1 to 3 or even higher? Or is Adaptive better? I found that Adaptive 4 is about as fast as Diamond 1, which one would look better?

Well, you must to play a little bit with the settings and then choose the better "cost-benefit" ... I'm gonna reproduce down a piece of readme.txt from NuEnc 0.1:

MB Decision - Alters how the encoder selects a MC MB. Least Bits=high quality (fast), RD=very high quality (slow).
Compare Function - SAD=good(fast), SATD=best(slow)
Search size - Adaptive is faster, Diamond is a little better. Larger value corresponds to larger search size.

In other words, is the resulting stream DVD (or SVCD) compliant by observing the Max bitrate?

Yep, for sure. Since that buffer underflows not occurs, the max bitrate is guaranteed and you get a elementary stream fully compliant. In this way, you must take care with the matrix chose, in order to prevent buffer underflows to occurs.

If so, we could have someone (Amnon82, JDobbs or Tylo) write a prediction tool for NuEnc which would give us a high quality one pass VBR encoder which also happens to be FREE.

Cheers
manolito

I think that will be great a such tool. :) Lets wait that someone have interests to do it!;)

Greetings,

[Amnon82]

If I find time, I'll add NuEnc-Support to one of the next AutoCQ-Releases ...

[danpos]

Originally posted by Amnon82
If I find time, I'll add NuEnc-Support to one of the next AutoCQ-Releases ...

WOW, that's a great news! :eek: I'd love this AutoCQ release! ;)

Kudos,

[DMagic1]

Any working link for 0.01b?

[L.H.V.F.]

Hi

For anyone interested in NuEnc, and their original variants (NuEnc 0.01b - NuEnc 0.21 - NuCLI), I had solicited to manolito forum member:

Binary and Sources (http://www.mediafire.com/?64c1b0ymgvtbnjo)

Best regards,
L.H.V.F. .

[Guest]

[bump after re-open]