HCEnc^n released ! [Archive]

Darksoul71

18th September 2007, 19:15

From the readme:

HCEnc^n v0.2.1 - the multicore frontend for HCEnc
====================================
by Dark$oul71

Purpose:
======
Similar to QuEnc^n HCEnc^n has be developed to
enable multi-instance encoding with HCEnc. The
idea itself is not new and was implemented prior
to HCEnc^n and QuEnc^n by Rumbah with Dual
DVDRB.

Usage / Commandline Interface:
=======================
HCEnc^n has only two valid commandline switches:
-ini "C:\MyHC.ini"
-cores 2

OK, -ini is the same as for HCEnc (the GUIless
version). HCEnc^n support a standard HC
compatible ini file. So you can generate an ini
file with the GUI version of HCEnc and use a
batchfile to launch HCEnc^n.

This can look like this:
"C:\HCEnc^n\HCENC^n.exe" -cores 2 -ini "D:\DIKO Working\eraser.ini"

If your path or the ini name has blanks in it
you have to incorporate the full path into "
Otherwise HCEnc^n will fail.

The -cores parameter allows you to define
how many instances of HCEnc shall be used.
I could have also called this threads.

Grab it here:
HCEnc^n v0.1.2:
http://rapidshare.com/files/56500995/HCEnc_n_v0.1.zip.html

Regards,
D$

nowhere2007

20th September 2007, 00:24

Cool, now I can finally take full advantage of my Core2 when using HCEnc!

Revgen

29th September 2007, 07:34

Thanks DarkSoul.

Program is fast and very easy to use. Quality looks great.

Can't do any benchmarks right now though.

Adub

29th September 2007, 10:03

So is it version 0.2.1 or 0.1.2?

Darksoul71

29th September 2007, 11:30

version should say 0.1.2 :)

Darksoul71

29th September 2007, 23:24

MR Nice

30th September 2007, 01:57

I am new to encoding so any help is welcome. I download the files but what do you do them?

Revgen

30th September 2007, 03:01

I am new to encoding so any help is welcome. I download the files but what do you do them?

Do you have HCEnc and DGMPGDec on your comp?

Do you you know how to use regular HCEnc?

If you do, then setup your encode in the HCEnc GUI and save it as an individual .ini file. Open HCEnc^n.exe and choose the .ini file you saved. I'm assuming your using 2 cores.

If you're using 4 cores than go to the CMD Prompt and type in:

HCEnc^n -cores 4 -ini "yourinifile.ini"

Revgen

30th September 2007, 05:11

MR Nice

30th September 2007, 20:17

Hi, Can use DGMPGDEC with Favc authoring program?

Revgen

30th September 2007, 20:21

Hi, Can use DGMPGDEC with Favc authoring program?

I suppose you could try it. If it doesn't work than grab the latest version from www.neuron2.net

Darksoul71

30th September 2007, 22:17

@MR Nice:
From the included HCEnc^n.txt :)

Initial setup:
=========
Setting up HCEnc^n is no rocket sience.
Simply grab HCEnc and DGIndex. Then
unzip both packages into the HCEnc^n.
At first launch HCEnc^n will find them
and store the path into the ini file.
Configure the generated HCEnc^n.ini
according to your system. For dualcore
machines I would set Cores to 2. Only
beefy multicore machines (e.g. overclocked
CoreDuo 6600 with 4 GHz) might be able
to handle more instances of HCEnc^n than
cores available. YMMV....

HCEnc homepage:
http://www.bitburners.com/HC_Encoder/

DGIndex homepage:
http://neuron2.net/dgmpgdec/dgmpgdec.html

Honestly I don´t know what version of DGIndex FAVC includes but why don´t you grab the latest DGIndex and try it ? :cool:

@Revgen:
I´ll be pleased for any other results. Esp. from a quad core such as the Intel Core 2 Quad Q6600 (anyone ?)

MR Nice

30th September 2007, 23:06

Sorry I think I am getting mixed up. I am trying to convert some avis that are nstc 29.97 to pal 25 so I can put them onto one dvd. The problem I have is when I convert the 29.97frames to pal 25 frames the video frame rate is slow and jerky.

Revgen

30th September 2007, 23:09

@Revgen:
I´ll be pleased for any other results. Esp. from a quad core such as the Intel Core 2 Quad Q6600 (anyone ?)

I'll be getting a Quad QX6850 very soon. :devil:

Darksoul71

1st October 2007, 20:27

@Revgen:
Can´t wait for your benchmarks :D

msdossys

3rd October 2007, 09:02

Excellent program, but one question... after this has completed, it appears HCEnc^n does not clean up after itself. I'm encoding a small MKV clip "ep3sample.mkv" containing H264 & AAC. After HCEnc^n is done, my directory contains ep3sample_000.avs .ini and the completed .m2v. Same three files with ep3sample_001, not to mention two folders HC_000 and HC_001 containing HCEnc.

After its all done, i have two playable .m2v's, and HCEnc^n sitting in my taskbar blinking between the Skull icon and a X. clicking on it shows a menu with Script Paused (which is checked) and exit....

Errrr... OK may have figured it out... thought I tried it before, but apparently not... So I click on Script Paused and after a while it finished up on its own. You may want to put that little tidbit in your .txt file for the next release :) Or am I just being crazy and it'll clean up on its own after it sees both HCEnc's have exited OK?

Darksoul71

3rd October 2007, 12:59

@msdossys:
Please post the resulting HCEnc^n.log file.
Sounds to me as if the HCEnc^n crashed in the middle of merging the m2v segments.

TIA,
D$

sunbeam

6th October 2007, 11:39

Hi folks,
I`m testing this nice programm extension with hcenc 0.21 for some days.
Same things as msdossys mentioned above happened with me with all the test files.
Now I went for a 3:30 h encoding from a sat-record and all went good but after 10 h the whole process finished
but with NOT making the final .m2v file. This merged file is not on my hdd`s. Umpfh... . But ok, it was a test.
Look at post #23 (http://forum.doom9.org/showthread.php?p=1052746#post1052746) for a log.
Setting the priority to idle or similar would be nice. So its hard to do things on the pc... .
The quality of the output files are in a very nice quality. :)

Greets,
sunbeam

Darksoul71

6th October 2007, 12:28

I´ll comment as soon as the attachments are approved.
Currently I have no idea though...
HCEnc^n worked perfectly for nearly 50 encodings now for me..:confused:

Regards,
D$

P.S.: BTW: Any HCEnc instance is set to idle priority already by HCEnc^n.

sunbeam

6th October 2007, 14:29

Hi Darksoul71,

nice tool you worked out here! Thanks for that. :)
Newly tested a sat-record of 1,5 gb. Pass 1+2 no problem at all.
When I touch the skull in the taskbar inbetween it starts to blink with this red X and script paused is marked.
But the encoding moves on `til the end of pass 2 but then it will not merge the 2 .m2v files 000+001 !.
Dismark it and the merging process starts then, but its not shown nowhere except hdd-led is working hard.
After 2-4 minutes the 2 files are merged and the HCEnc^n.log looks exactly same as the above one.
Could it be that the input or output file of a 3h30min record (~8gb for a DL-DVD) is too big to be handled properly?P.S.: BTW: Any HCEnc instance is set to idle priority already by HCEnc^n.It looks like this is ignored by windows. The 2 instances of hcenc_021.exe are using 50% of cpu each. So its 100% in taskmanager.
With tools like process tamer etc. there is also no way to "tame" this process. Not for 1%.
100% process aktivity shown in the taskmanager means to wait a 15sec to open notepad.exe ! Thats hard to live with... . :eek:

Darksoul71

6th October 2007, 20:15

@sunbeam:
Priority setting works for me. When I look at the HCEnc priority in the task manager it shows "idle" as priority. Gaming beside two HCEnc instances is no big problem. Surfing the web, recording Huffyuv AVI in 704x576 is no problem.
I use a Opteron 170 with 2GB RAM under WinXP SP2.

BTW: Have you tried to disable process tamer ?

What is your system ?

Regards,
D$

Darksoul71

6th October 2007, 20:53

@Sunbeam:
>>When I touch the skull in the taskbar inbetween
>>it starts to blink with this red X and script paused is marked.
Never do that ! This will pause the autoit script (which HCEnc^n is) from running.
I have no clue if this could impact the part of HCEnc^n which waits for the process
id to close.

>But the encoding moves on `til the end of pass 2
>but then it will not merge the 2 .m2v files 000+001 !.
No big wonder here....
HCEnc^n does not control HCEnc. It encodes by itself until
the encoding process finishs.

>Dismark it and the merging process starts then,
>but its not shown nowhere except hdd-led is working hard.
After unpausing the script HCEnc^n seems to notice that
all spawned HCenc instance have finished and
starts DGIndex for merging the segments. The hidden run
of DGIndex is intended since I dislike anoying "pop-up"
applications. They disturb my workflow.

>After 2-4 minutes the 2 files are merged
>and the HCEnc^n.log looks exactly same
> as the above one.
Well, I would love to see the log but still
some admin need to approve then :(
You could also post the logfile here in the thread.

>Could it be that the input or output file of a
>3h30min record (~8gb for a DL-DVD) is
>too big to be handled properly?
Unless you use FAT32 size is no issue.
I´m not aware that DGIndex or HCEnc
have any issue with larger files.

A "healthy" HCEnc^n log will look like this:
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>HCEnc^n v0.1.2 - the multicore frontend for HCEnc
27/09/2007 02:17:31 >>Using AutoIt Version 3.2.8.0
27/09/2007 02:17:31 >>HCEnc^n is stored in path C:\Eigene Dateien\AutoIt Projekte\HCEnc^n v0.1.2 (ini-browsing + HCEnc search)
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>HcEnc^n was called with the following commandline -cores 2 -ini "D:\Punisher\MPEG2\Punisher_movie_2cores.INI"
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>Adding file D:\Punisher\MPEG2\Punisher_movie_2cores.avs to logfile
27/09/2007 02:17:31 >>LoadPlugin("C:\PROGRA~1\AutoGK\DGMPGDec\DGDecode.dll")
27/09/2007 02:17:31 >>mpeg2source("D:\Punisher\agk_tmp\Punisher.d2v")
27/09/2007 02:17:31 >>
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>Adding file D:\Punisher\MPEG2\Punisher_movie_2cores.INI to logfile
27/09/2007 02:17:31 >>*MAXBITRATE 8000
27/09/2007 02:17:31 >>*PROFILE best
27/09/2007 02:17:31 >>*ASPECT 16:9
27/09/2007 02:17:31 >>*GOP 15 2
27/09/2007 02:17:31 >>*CLOSEDGOPS
27/09/2007 02:17:31 >>*WAIT 0
27/09/2007 02:17:31 >>*INFILE D:\Punisher\MPEG2\Punisher_movie_2cores.avs
27/09/2007 02:17:31 >>*OUTFILE D:\Punisher\MPEG2\Punisher_movie_2cores.M2V
27/09/2007 02:17:31 >>*LOGFILE D:\Punisher\MPEG2\Punisher_movie_2cores.LOG
27/09/2007 02:17:31 >>*BITRATE 2000
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>Commandline #0: "D:\Punisher\MPEG2\HC_000\HCenc_021.exe" -ini "D:\Punisher\MPEG2\Punisher_movie_2cores_000.INI"
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:31 >>Adding file D:\Punisher\MPEG2\Punisher_movie_2cores_000.INI to logfile
27/09/2007 02:17:31 >>*MAXBITRATE 8000
27/09/2007 02:17:31 >>*PROFILE best
27/09/2007 02:17:31 >>*ASPECT 16:9
27/09/2007 02:17:31 >>*GOP 15 2
27/09/2007 02:17:31 >>*CLOSEDGOPS
27/09/2007 02:17:31 >>*WAIT 0
27/09/2007 02:17:31 >>*INFILE D:\Punisher\MPEG2\Punisher_movie_2cores_000.AVS
27/09/2007 02:17:31 >>*OUTFILE D:\Punisher\MPEG2\Punisher_movie_2cores_000.M2V
27/09/2007 02:17:31 >>*LOGFILE D:\Punisher\MPEG2\Punisher_movie_2cores.LOG
27/09/2007 02:17:31 >>*BITRATE 2000
27/09/2007 02:17:31 >>----------------------------------------------------------------------------
27/09/2007 02:17:32 >>Instance #0 of HCEnc has been started !
27/09/2007 02:17:32 >>Commandline for HCEnc Instance #0: "D:\Punisher\MPEG2\HC_000\HCenc_021.exe" -ini "D:\Punisher\MPEG2\Punisher_movie_2cores_000.INI"
27/09/2007 02:17:32 >>Commandline #1: "D:\Punisher\MPEG2\HC_001\HCenc_021.exe" -ini "D:\Punisher\MPEG2\Punisher_movie_2cores_001.INI"
27/09/2007 02:17:32 >>----------------------------------------------------------------------------
27/09/2007 02:17:32 >>Adding file D:\Punisher\MPEG2\Punisher_movie_2cores_001.INI to logfile
27/09/2007 02:17:32 >>*MAXBITRATE 8000
27/09/2007 02:17:32 >>*PROFILE best
27/09/2007 02:17:32 >>*ASPECT 16:9
27/09/2007 02:17:32 >>*GOP 15 2
27/09/2007 02:17:32 >>*CLOSEDGOPS
27/09/2007 02:17:32 >>*WAIT 0
27/09/2007 02:17:32 >>*INFILE D:\Punisher\MPEG2\Punisher_movie_2cores_001.AVS
27/09/2007 02:17:32 >>*OUTFILE D:\Punisher\MPEG2\Punisher_movie_2cores_001.M2V
27/09/2007 02:17:32 >>*LOGFILE D:\Punisher\MPEG2\Punisher_movie_2cores.LOG
27/09/2007 02:17:32 >>*BITRATE 2000
27/09/2007 02:17:32 >>----------------------------------------------------------------------------
27/09/2007 02:17:32 >>Instance #1 of HCEnc has been started !
27/09/2007 02:17:32 >>Commandline for HCEnc Instance #1: "D:\Punisher\MPEG2\HC_001\HCenc_021.exe" -ini "D:\Punisher\MPEG2\Punisher_movie_2cores_001.INI"
27/09/2007 03:57:03 >>All instances of HCEnc have been closed !
27/09/2007 03:57:03 >>Filesize of file D:\Punisher\MPEG2\Punisher_movie_2cores_000.M2V in Byte: 888873108
27/09/2007 03:57:03 >>Filesize of file D:\Punisher\MPEG2\Punisher_movie_2cores_001.M2V in Byte: 888847100
27/09/2007 03:57:03 >>Commandline for DGIndex: -IA=1 -FO=0 -YR=2 -TN=1 -OM=0 -BF=[D:\Punisher\MPEG2\Punisher_movie_2cores.LST] -DSD=0 -DRC=0 -DSA=0 -OFD=[D:\Punisher\MPEG2\Punisher_movie_2cores] -EXIT -HIDE
27/09/2007 04:01:26 >>Filesize of file & D:\Punisher\MPEG2\Punisher_movie_2cores.M2V in Byte: 1777720212
27/09/2007 04:01:26 >>----------------------------------------------------------------------------
27/09/2007 04:01:26 >>Adding file D:\Punisher\MPEG2\Punisher_movie_2cores.LST to logfile
27/09/2007 04:01:26 >>D:\Punisher\MPEG2\Punisher_movie_2cores_000.M2V
27/09/2007 04:01:26 >>D:\Punisher\MPEG2\Punisher_movie_2cores_001.M2V
27/09/2007 04:01:26 >>----------------------------------------------------------------------------
27/09/2007 04:01:26 >>Merging of M2V files finished !

sunbeam

6th October 2007, 22:05

@Darksoul71:
Priority setting works for me. When I look at the HCEnc priority in the task manager it shows "idle" as priority. Gaming beside two HCEnc instances is no big problem. Surfing the web, recording Huffyuv AVI in 704x576 is no problem.
I cant believe this. My CPU is 100%, HCenc_021 is 2x ~390.000kb with low priority, thats true.
There is not much air for other processes :eek:
BTW: Have you tried to disable process tamer ?
Sure, most the time he`s not running.
What is your system ?
P4 2,8 GH, 1GB RAM Dual Channel, HD SATA 500GB, W2K-SP4+ in NTFS.
@Sunbeam:
>>When I touch the skull in the taskbar inbetween
>>it starts to blink with this red X and script paused is marked.
Never do that ! This will pause the autoit script (which HCEnc^n is) from running.
I have no clue if this could impact the part of HCEnc^n which waits for the process
id to close.OK, understand.
The hidden run of DGIndex is intended since I dislike anoying "pop-up"
applications. They disturb my workflow.
It is easy for me to agree in this now.
I´m not aware that DGIndex or HCEnc have any issue with larger files.
Same with me. At least I made about 11 encodings these days without any disturbances.
This is a very nice Tool. The thing with the 3h30min encoding I will repeat this night.
You could also post the logfile here in the thread.
Here it is:
06/10/2007 00:20:10 >>----------------------------------------------------------------------------
06/10/2007 00:20:10 >>HCEnc^n v0.1.2 - the multicore frontend for HCEnc
06/10/2007 00:20:10 >>Using AutoIt Version 3.2.8.0
06/10/2007 00:20:10 >>HCEnc^n is stored in path H:\Sandbox\HCEnc
06/10/2007 00:20:10 >>----------------------------------------------------------------------------
06/10/2007 00:20:10 >>HcEnc^n was called with the following commandline -cores 2 -ini "H:\DVD\BATCH\BATCH_2\2_HCENC.ini"
06/10/2007 00:20:10 >>----------------------------------------------------------------------------
06/10/2007 00:20:10 >>----------------------------------------------------------------------------
06/10/2007 00:20:10 >>Adding file h:\dvd\batch\batch_2\2_b_template.avs to logfile
06/10/2007 00:20:10 >>Loadplugin("H:\Sandbox\DGMPGDEC\DGDecode.dll")
06/10/2007 00:20:10 >>LoadPlugin("H:\Sandbox\PLUGINS\Decomb.dll")
06/10/2007 00:20:10 >>mpeg2source("H:\DVD\BATCH\BATCH_2\FILM_2.d2v",cpu=6,idct=6,iPP=false)
06/10/2007 00:20:10 >>FieldDeinterlace(blend=false)
06/10/2007 00:20:10 >>Crop(16,8,-16,-8)
06/10/2007 00:20:10 >>LanczosResize(672,544)
06/10/2007 00:20:10 >>ConvertToRGB32()
06/10/2007 00:20:10 >>Import("H:\DVD\BATCH\BATCH_2\2_B_VD_DELOGO.avs")
06/10/2007 00:20:10 >>Addborders(24,16,24,16)
06/10/2007 00:20:10 >>#ConvertBackToYUY2()
06/10/2007 00:20:10 >>ConvertToYV12()
06/10/2007 00:20:10 >>----------------------------------------------------------------------------
06/10/2007 00:20:10 >>----------------------------------------------------------------------------
06/10/2007 00:20:10 >>Adding file H:\DVD\BATCH\BATCH_2\2_HCENC.ini to logfile
06/10/2007 00:20:10 >>*INFILE h:\dvd\batch\batch_2\2_b_template.avs
06/10/2007 00:20:10 >>*OUTFILE H:\DVD\BATCH\BATCH_2\2_b_template.m2v
06/10/2007 00:20:10 >>*LOGFILE H:\DVD\BATCH\BATCH_2\2_b_template.log
06/10/2007 00:20:10 >>*BITRATE 4846
06/10/2007 00:20:10 >>*MAXBITRATE 9100
06/10/2007 00:20:10 >>*FRAMES 0 306741
06/10/2007 00:20:10 >>*PROFILE best
06/10/2007 00:20:10 >>*ASPECT 4:3
06/10/2007 00:20:10 >>*AUTOGOP 15
06/10/2007 00:20:10 >>*DC_PREC 10
06/10/2007 00:20:10 >>*INTERLACED
06/10/2007 00:20:10 >>*TFF
06/10/2007 00:20:10 >>*CLOSEDGOPS
06/10/2007 00:20:10 >>*CUSTOMMATRIX
06/10/2007 00:20:10 >> 8 16 19 22 26 27 29 34
06/10/2007 00:20:10 >> 16 16 22 24 27 29 34 37
06/10/2007 00:20:10 >> 19 22 26 27 29 34 34 38
06/10/2007 00:20:10 >> 22 22 26 27 29 34 37 40
06/10/2007 00:20:10 >> 22 26 27 29 32 35 40 48
06/10/2007 00:20:10 >> 26 27 29 32 35 40 48 58
06/10/2007 00:20:10 >> 26 27 29 34 38 46 56 69
06/10/2007 00:20:10 >> 27 29 35 38 46 56 69 83
06/10/2007 00:20:10 >>
06/10/2007 00:20:10 >> 16 17 18 19 20 21 22 23
06/10/2007 00:20:10 >> 17 18 19 20 21 22 23 24
06/10/2007 00:20:10 >> 18 19 20 21 22 23 24 25
06/10/2007 00:20:10 >> 19 20 21 22 23 24 26 27
06/10/2007 00:20:10 >> 20 21 22 23 25 26 27 28
06/10/2007 00:20:10 >> 21 22 23 24 26 27 28 30
06/10/2007 00:20:10 >> 22 23 24 26 27 28 30 31
06/10/2007 00:20:10 >> 23 24 25 27 28 30 31 33
06/10/2007 00:20:10 >>----------------------------------------------------------------------------
06/10/2007 00:20:11 >>Commandline #0: "h:\dvd\batch\batch_2\HC_000\HCenc_021.exe" -ini "h:\dvd\batch\batch_2\2_b_template_000.INI"
06/10/2007 00:20:11 >>----------------------------------------------------------------------------
06/10/2007 00:20:11 >>Adding file h:\dvd\batch\batch_2\2_b_template_000.INI to logfile
06/10/2007 00:20:11 >>*INFILE h:\dvd\batch\batch_2\2_b_template_000.AVS
06/10/2007 00:20:11 >>*OUTFILE h:\dvd\batch\batch_2\2_b_template_000.M2V
06/10/2007 00:20:11 >>*LOGFILE H:\DVD\BATCH\BATCH_2\2_b_template.log
06/10/2007 00:20:11 >>*BITRATE 4846
06/10/2007 00:20:11 >>*MAXBITRATE 9100
06/10/2007 00:20:11 >>
06/10/2007 00:20:11 >>*PROFILE best
06/10/2007 00:20:11 >>*ASPECT 4:3
06/10/2007 00:20:11 >>*AUTOGOP 15
06/10/2007 00:20:11 >>*DC_PREC 10
06/10/2007 00:20:11 >>*INTERLACED
06/10/2007 00:20:11 >>*TFF
06/10/2007 00:20:11 >>*CLOSEDGOPS
06/10/2007 00:20:11 >>*CUSTOMMATRIX
06/10/2007 00:20:11 >> 8 16 19 22 26 27 29 34
06/10/2007 00:20:11 >> 16 16 22 24 27 29 34 37
06/10/2007 00:20:11 >> 19 22 26 27 29 34 34 38
06/10/2007 00:20:11 >> 22 22 26 27 29 34 37 40
06/10/2007 00:20:11 >> 22 26 27 29 32 35 40 48
06/10/2007 00:20:11 >> 26 27 29 32 35 40 48 58
06/10/2007 00:20:11 >> 26 27 29 34 38 46 56 69
06/10/2007 00:20:11 >> 27 29 35 38 46 56 69 83
06/10/2007 00:20:11 >>
06/10/2007 00:20:11 >> 16 17 18 19 20 21 22 23
06/10/2007 00:20:11 >> 17 18 19 20 21 22 23 24
06/10/2007 00:20:11 >> 18 19 20 21 22 23 24 25
06/10/2007 00:20:11 >> 19 20 21 22 23 24 26 27
06/10/2007 00:20:11 >> 20 21 22 23 25 26 27 28
06/10/2007 00:20:11 >> 21 22 23 24 26 27 28 30
06/10/2007 00:20:11 >> 22 23 24 26 27 28 30 31
06/10/2007 00:20:11 >> 23 24 25 27 28 30 31 33
06/10/2007 00:20:11 >>----------------------------------------------------------------------------
06/10/2007 00:20:11 >>Instance #0 of HCEnc has been started !
06/10/2007 00:20:11 >>Commandline for HCEnc Instance #0: "h:\dvd\batch\batch_2\HC_000\HCenc_021.exe" -ini "h:\dvd\batch\batch_2\2_b_template_000.INI"
06/10/2007 00:20:11 >>Commandline #1: "h:\dvd\batch\batch_2\HC_001\HCenc_021.exe" -ini "h:\dvd\batch\batch_2\2_b_template_001.INI"
06/10/2007 00:20:11 >>----------------------------------------------------------------------------
06/10/2007 00:20:11 >>Adding file h:\dvd\batch\batch_2\2_b_template_001.INI to logfile
06/10/2007 00:20:11 >>*INFILE h:\dvd\batch\batch_2\2_b_template_001.AVS
06/10/2007 00:20:11 >>*OUTFILE h:\dvd\batch\batch_2\2_b_template_001.M2V
06/10/2007 00:20:11 >>*LOGFILE H:\DVD\BATCH\BATCH_2\2_b_template.log
06/10/2007 00:20:11 >>*BITRATE 4846
06/10/2007 00:20:11 >>*MAXBITRATE 9100
06/10/2007 00:20:11 >>
06/10/2007 00:20:11 >>*PROFILE best
06/10/2007 00:20:11 >>*ASPECT 4:3
06/10/2007 00:20:11 >>*AUTOGOP 15
06/10/2007 00:20:11 >>*DC_PREC 10
06/10/2007 00:20:11 >>*INTERLACED
06/10/2007 00:20:11 >>*TFF
06/10/2007 00:20:11 >>*CLOSEDGOPS
06/10/2007 00:20:11 >>*CUSTOMMATRIX
06/10/2007 00:20:11 >> 8 16 19 22 26 27 29 34
06/10/2007 00:20:11 >> 16 16 22 24 27 29 34 37
06/10/2007 00:20:11 >> 19 22 26 27 29 34 34 38
06/10/2007 00:20:11 >> 22 22 26 27 29 34 37 40
06/10/2007 00:20:11 >> 22 26 27 29 32 35 40 48
06/10/2007 00:20:11 >> 26 27 29 32 35 40 48 58
06/10/2007 00:20:11 >> 26 27 29 34 38 46 56 69
06/10/2007 00:20:11 >> 27 29 35 38 46 56 69 83
06/10/2007 00:20:11 >>
06/10/2007 00:20:11 >> 16 17 18 19 20 21 22 23
06/10/2007 00:20:11 >> 17 18 19 20 21 22 23 24
06/10/2007 00:20:11 >> 18 19 20 21 22 23 24 25
06/10/2007 00:20:11 >> 19 20 21 22 23 24 26 27
06/10/2007 00:20:11 >> 20 21 22 23 25 26 27 28
06/10/2007 00:20:11 >> 21 22 23 24 26 27 28 30
06/10/2007 00:20:11 >> 22 23 24 26 27 28 30 31
06/10/2007 00:20:11 >> 23 24 25 27 28 30 31 33
06/10/2007 00:20:11 >>----------------------------------------------------------------------------
06/10/2007 00:20:11 >>Instance #1 of HCEnc has been started !
06/10/2007 00:20:11 >>Commandline for HCEnc Instance #1: "h:\dvd\batch\batch_2\HC_001\HCenc_021.exe" -ini "h:\dvd\batch\batch_2\2_b_template_001.INI"
06/10/2007 10:46:05 >>All instances of HCEnc have been closed !
06/10/2007 10:46:06 >>Filesize of file h:\dvd\batch\batch_2\2_b_template_000.M2V in Byte: 3717045356
06/10/2007 10:46:06 >>Filesize of file h:\dvd\batch\batch_2\2_b_template_001.M2V in Byte: 3715606687
06/10/2007 10:46:06 >>Commandline for DGIndex: -IA=1 -FO=0 -YR=2 -TN=1 -OM=0 -BF=[h:\dvd\batch\batch_2\2_b_template.LST] -DSD=0 -DRC=0 -DSA=0 -OFD=[H:\DVD\BATCH\BATCH_2\2_b_template] -EXIT -HIDE
06/10/2007 10:46:06 >>Filesize of file & H:\DVD\BATCH\BATCH_2\2_b_template.m2v in Byte: 0
06/10/2007 10:46:06 >>----------------------------------------------------------------------------
06/10/2007 10:46:06 >>Adding file h:\dvd\batch\batch_2\2_b_template.LST to logfile
06/10/2007 10:46:06 >>h:\dvd\batch\batch_2\2_b_template_000.M2V
06/10/2007 10:46:06 >>h:\dvd\batch\batch_2\2_b_template_001.M2V
06/10/2007 10:46:06 >>----------------------------------------------------------------------------
06/10/2007 10:46:06 >>Merging of M2V files finished !

Until the end everything worked fine except that there was no merged .m2v file.
We will see about this tomorrow... :)

Greets,
sunbeam

Darksoul71

6th October 2007, 22:14

@sunbeam:
Oh oh, running two instances of HCEnc on a hyperthreading CPU might not be a good idea.
The concept of parallel multi-instance encoding (QuEnc^n and HCEnc^n) only work well
on at least a dual-core CPU or a dual CPU system.

The log you´ve posted seems ok beside the fact that the resulting M2V is 0 Byte big :)

Hm, you can try two things:
1) Open the first segment manually via DGIndex and then add the second one. Then try saving a D2V project.
This should show wether the generated segments are fine or not.
2) You could enable the Debug value in the HCEnc^n.ini by setting it to 1.
This way DGIndex is not run hidden and you can see any errors showing up.

BTW:
>>HCenc_021 is 2x ~390.000kb
With 1 GB of RAM it´s very likely that HCEnc^n will most likely not only max out your system in terms of available CPU time but also from a RAM perspective. Four instances of HCEnc for example require at least 3GB of RAM. You could also try QuEnc^n since QuEnc is much less demanding in terms of RAM access.

My rules of thumb:
2 Core Systems -> 2GB RAM
4 Core Systems -> 4GB RAM

sunbeam

6th October 2007, 22:46

@Darksoul71:
Oh oh, running two instances of HCEnc on a hyperthreading CPU might not be a good idea.
Now I know. But it is working so far... . I liked to tryout your work and the HCEnc 0.21 :)
Hm, you can try two things:
1) Open the first segment manually via DGIndex and then add the second one. Then try saving a D2V project.
This should show wether the generated segments are fine or not.
They were gone short after the encoding process. The whole folder was "cleaned up" by itself.
2) You could enable the Debug value in the HCEnc^n.ini by setting it to 1.
This way DGIndex is not run hidden and you can see any errors showing up.
I try this. Maybe it helps to find out why there is a .m2v file with 0 byte.
Thank you for your patience in this... . :thanks:

Darksoul71

6th October 2007, 22:50

@Sunbeam:
They were gone short after the encoding process. The whole folder was "cleaned up" by itself.
Psst, don´t tell anyone:
Clean up of files is disabled when Debug is set to 1 ;)

@all:
Even if this sounds a bit strange:
HCEnc^n (and his cousin QuEnc^n) are only quite dumb implemented "middlewares" between the command interpreter (e.g. batch file or frontend) and DGIndex & HCEnc (or QuEnc). Unless there is a stupid bug in my code (which of course happens from time to time :)) it´s really not very likely that a crash is caused by HCEnc^n itself. It might be a problem in your avisynth script, it might be "wrong" characters in your path and so on.

I´ll try my best to make both as robust as possible but the encoding process itself is out off my hands. The same is true for merging.

I still have the feeling that a implementation similar to HCEnc^n should be part of the encoder itself and not a separate middleware / third party tool. This would make it much more robust.

sunbeam

7th October 2007, 12:24

@Darksoul71:
This time no problem with the 3h30min file. Maybe it`s better not to touch the skull in the taskbar while encoding. I think that was the fault.
Ah, yes, last night I dreamt of a nice core 2 quad cpu board... :)

Lorax2161

10th October 2007, 19:06

@all:
I´ll try my best to make both as robust as possible but the encoding process itself is out off my hands. The same is true for merging.

Point well taken, and thank you for your hard work with HCEnc^n. If you have more than one core, it definitely fills a void. Getting free time back from the encoding process is almost as good as free money. :)

I am having a problem, however. In over a week of testing (several times a day), my success rate stands at zero percent.

Here is my experience: No crashes for HCEnc or HCEnc^n. Both m2v files are created without an apparent problem. File appears to be merged successfully. Size of final file is consistent with specified average bitrate for two-pass process.

Now, when I open the m2v file with any media player (Windows Media Player, Zoom Player, Nero Showtime) the file duration is reported at half the true duration. The file will play successfully up until the halfway mark, and then it will rapidly accelerate and play the second half of the file in hyperspeed. This behavior has been consistent for over a week despite various tweakings to rectify the situation. In fact, this is the only result I have ever had. I have had no problems with the single instance of HCEnc, however.

Disclaimer: I am a noob. I've tinkered with video encoding on and off for a few years so I am familiar with the basics, but that does not rule out that I've done something quite dumb somewhere along the way.

That said, my suspicion is that both pieces of the video file are true and correct, but that bad information is being written into the header of the merged m2v file. Every player I try these files with behave exactly the same way--reports half the duration, plays the first half correctly and then goes hypersonic to finish the second half.

My system: AMD Athlon 64 X2 Dual Core Processor 4200+, 2 GB RAM, Windows XP Home Edition SP2. Although D$ mentioned this front end is not intended for HT computers, I had nothing to lose at this point and ran it on my P4 3.00GHz HT, 2 GB RAM, Windows XP Home Edition SP2. Same exact results. Very strange for two different computers with different processors since no one else has commented on having a problem like mine.

I am using the latest DGIndex 1.49 and HCgui 0.21 to make my ini file, then saving and exiting. I have tried the batch file method and manually starting HCEnc^n, with no difference in the results.

Because this error has been consistent, and reproducible, I'm confident it can be eliminated. Unfortunately, my reach for a solution exceeds my grasp of expertise, and I am stuck.

I will include my log files, both the one for HCEnc^n and HCEnc, because they disagree in the number of frames to encode so I suspect that may be relevant.

==================================================

10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>HCEnc^n v0.1.2 - the multicore frontend for HCEnc
10/10/2007 12:08:07 >>Using AutoIt Version 3.2.8.0
10/10/2007 12:08:07 >>HCEnc^n is stored in path C:\Program Files\HCEnc_n
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>HcEnc^n was called with the following commandline -cores 2 -ini "C:\Program Files\HCEnc_n\test_HC.ini"
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>Adding file l:\104\test.avs to logfile
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>AviSource("L:\104\ufo.05.avi")
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >> crop(8,42,-8,-62,align=true)
10/10/2007 12:08:07 >> addborders(8,44,8,60)
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>#--------------------------------------------------------------------
10/10/2007 12:08:07 >># Loaded Plugins
10/10/2007 12:08:07 >>#--------------------------------------------------------------------
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>loadplugin("C:\Program Files\aviSynth 2.5\plugins\RemoveGrainSSE3.dll")
10/10/2007 12:08:07 >>loadplugin("C:\Program Files\aviSynth 2.5\plugins\UnFilter.dll")
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>#--------------------------------------------------------------------
10/10/2007 12:08:07 >># Trimmed Video
10/10/2007 12:08:07 >>#--------------------------------------------------------------------
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>trim (13366,15166)
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>#--------------------------------------------------------------------
10/10/2007 12:08:07 >># Applied Filters
10/10/2007 12:08:07 >>#--------------------------------------------------------------------
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >># Levels (clip, int input_low, float gamma, int input_high, int output_low, int output_high, bool "coring")
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>Levels(5, 1, 255, 0, 245)
10/10/2007 12:08:07 >>RemoveGrain(mode=17).unfilter(20,20).RemoveGrain(mode=1)
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>Adding file C:\Program Files\HCEnc_n\test_HC.ini to logfile
10/10/2007 12:08:07 >>*INFILE l:\104\test.avs
10/10/2007 12:08:07 >>*OUTFILE K:\my_temp\test2.m2v
10/10/2007 12:08:07 >>*LOGFILE K:\my_temp\test2.log
10/10/2007 12:08:07 >>*BITRATE 3688
10/10/2007 12:08:07 >>*MAXBITRATE 8800
10/10/2007 12:08:07 >>*FRAMES 0 1800
10/10/2007 12:08:07 >>*PROFILE best
10/10/2007 12:08:07 >>*ASPECT 4:3
10/10/2007 12:08:07 >>*AUTOGOP 15
10/10/2007 12:08:07 >>*INTERLACED
10/10/2007 12:08:07 >>*BFF
10/10/2007 12:08:07 >>*CLOSEDGOPS
10/10/2007 12:08:07 >>*BIAS 20
10/10/2007 12:08:07 >>*LASTIFRAME
10/10/2007 12:08:07 >>*MATRIX mpeg
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>Commandline #0: "l:\104\HC_000\HCenc_021.exe" -ini "l:\104\test_000.INI"
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>Adding file l:\104\test_000.INI to logfile
10/10/2007 12:08:07 >>*INFILE l:\104\test_000.AVS
10/10/2007 12:08:07 >>*OUTFILE l:\104\test_000.M2V
10/10/2007 12:08:07 >>*LOGFILE K:\my_temp\test2.log
10/10/2007 12:08:07 >>*BITRATE 3688
10/10/2007 12:08:07 >>*MAXBITRATE 8800
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>*PROFILE best
10/10/2007 12:08:07 >>*ASPECT 4:3
10/10/2007 12:08:07 >>*AUTOGOP 15
10/10/2007 12:08:07 >>*INTERLACED
10/10/2007 12:08:07 >>*BFF
10/10/2007 12:08:07 >>*CLOSEDGOPS
10/10/2007 12:08:07 >>*BIAS 20
10/10/2007 12:08:07 >>*LASTIFRAME
10/10/2007 12:08:07 >>*MATRIX mpeg
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>Instance #0 of HCEnc has been started !
10/10/2007 12:08:07 >>Commandline for HCEnc Instance #0: "l:\104\HC_000\HCenc_021.exe" -ini "l:\104\test_000.INI"
10/10/2007 12:08:07 >>Commandline #1: "l:\104\HC_001\HCenc_021.exe" -ini "l:\104\test_001.INI"
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>Adding file l:\104\test_001.INI to logfile
10/10/2007 12:08:07 >>*INFILE l:\104\test_001.AVS
10/10/2007 12:08:07 >>*OUTFILE l:\104\test_001.M2V
10/10/2007 12:08:07 >>*LOGFILE K:\my_temp\test2.log
10/10/2007 12:08:07 >>*BITRATE 3688
10/10/2007 12:08:07 >>*MAXBITRATE 8800
10/10/2007 12:08:07 >>
10/10/2007 12:08:07 >>*PROFILE best
10/10/2007 12:08:07 >>*ASPECT 4:3
10/10/2007 12:08:07 >>*AUTOGOP 15
10/10/2007 12:08:07 >>*INTERLACED
10/10/2007 12:08:07 >>*BFF
10/10/2007 12:08:07 >>*CLOSEDGOPS
10/10/2007 12:08:07 >>*BIAS 20
10/10/2007 12:08:07 >>*LASTIFRAME
10/10/2007 12:08:07 >>*MATRIX mpeg
10/10/2007 12:08:07 >>----------------------------------------------------------------------------
10/10/2007 12:08:07 >>Instance #1 of HCEnc has been started !
10/10/2007 12:08:07 >>Commandline for HCEnc Instance #1: "l:\104\HC_001\HCenc_021.exe" -ini "l:\104\test_001.INI"
10/10/2007 12:09:35 >>All instances of HCEnc have been closed !
10/10/2007 12:09:35 >>Filesize of file l:\104\test_000.M2V in Byte: 13821964
10/10/2007 12:09:35 >>Filesize of file l:\104\test_001.M2V in Byte: 13812452
10/10/2007 12:09:35 >>Commandline for DGIndex: -IA=1 -FO=0 -YR=2 -TN=1 -OM=0 -BF=[l:\104\test.LST] -DSD=0 -DRC=0 -DSA=0 -OFD=[K:\my_temp\test2] -EXIT -HIDE
10/10/2007 12:09:39 >>Filesize of file & K:\my_temp\test2.m2v in Byte: 27634420
10/10/2007 12:09:39 >>----------------------------------------------------------------------------
10/10/2007 12:09:39 >>Adding file l:\104\test.LST to logfile
10/10/2007 12:09:39 >>l:\104\test_000.M2V
10/10/2007 12:09:39 >>l:\104\test_001.M2V
10/10/2007 12:09:39 >>----------------------------------------------------------------------------
10/10/2007 12:09:39 >>Merging of M2V files finished !

==================================================

Note the number of frames in the HCEnc log below. I am unsure whether this is a log of the last (2nd) instance of HCEnc (overwriting itself?) or if this is merged data from both instances, but the interesting thing is that this test clip is 1801 frames, but this log reports only 900, and the midpoint is where the trouble seems to start. (I have tried this with both open and closed GOP, and it didn't make a difference.)

==================================================

-----------------------------------------
| HCenc - MPEG2 encoder - rel. 0.21
-----------------------------------------
| HCenc - MPEG2 encoder - rel. 0.21.0.0 |
-----------------------------------------

MPEG profile@level: MP@ML
input: l:\104\test_000.avs
output: l:\104\test_000.m2v

--------------------
| encoder settings |framerate: 29.970
aspect ratio: 4:3
bitrate Kb/s: 3688
max. bitrate Kb/sframerate: 29.970
aspect ratio: 4:3
bitrate Kb/s: 3688
max. bitrate Kb/s: 8800
pulldown: no
closed gops: yes
VBV check: yes
scene change det.: yes
interlaced: yes, BFF
goplen,B-pic: AUTO
dc_precision: 9
scan method: ALTERNATE
bias: 20
chapter frames: 1
time code: 0 0 0 0
CPU: SSE3
matrix: MPEG
luminance gain: no

------------------
| source stats |
------------------

nr. of frames in source: 901
width*height: 720x480
fps:
---------------------
| encoding - pass 1 |
---------------------

pass1 encoding time: 0:00:33 (33 s)
fps: 27.1

--------------------------------
| encoding - intermediate pass |
--------------------------------

bitrate set to: 3688 kb/s
est. outfile length: 13519 kB
intermediate encoding time: 0.0 s

---------------------
| encoding - pass 2 |
---------------------

pass 2 encoding time: 0:00:31 (31 s)
fps: 29.3

------------------
| encoding stats |
------------------

total CPU time: 0:01:05 (65 s)

intra matrix used
8 16 19 22 26 27 29 34
16 16 22 24 27 29 34 37
19 22 26 27 29 34 34 38
22 22 26 27 29 34 37 40
22 26 27 29 32 35 40 48
26 27 29 32 35 40 48 58
26 27 29 34 38 46 56 69
27 29 35 38 46 56 69 83

non-intra matrix used
16 17 18 19 20 21 22 23
17 18 19 20 21 22 23 24
18 19 20 21 22 23 24 25
19 20 21 22 23 24 26 27
20 21 22 23 25 26 27 28
21 22 23 24 26 27 28 30
22 23 24 26 27 28 30 31
23 24 25 27 28 30 31 33

nr. of gops: 65
nr. of frames: 900
nr. of I-frames: 65
nr. of P-frames: 302
nr. of B-frames: 533
average bitrate: 3678
minimum bitrate: 1256
maximum bitrate: 8060

bytes in bitstream: 13812452
bits in bitstream: 110499616

average Quantizer: 4.666

total elapsed time: 0:01:10 (69.80 s)
110575712

average Quantizer: 5.855

total elapsed time: 0:01:10 (69.73 s)

==================================================

Apology for the long post. I am hopeful a solution to this problem exists. I appreciate all the work the contributors of the video community have made, especially Hank315 and Darksoul71 for both of the tools I've spoken about in this post. Now if I can just get them to play nice with each other. :)

Thank you.

Darksoul71

10th October 2007, 22:22

Hi Lorax2161,

a near perfect example how a problem description should look like :thanks:

Believe it or not but your description drew my attention to something I didn´t take into account.

I had two issues with HCEnc^n encoded files I never had before:
1) MPLEX (which I use for my HCEnc^n batch tool) failed muxing the files completely. To be more specificially:
The resulting muxed file was exactly as big as one segment plus audio.

2) TMPEG DVDAuthor was unable to determine the correct length of the resulting correctly. The movie was only detected as half as long as the true playtime.

I "solved" the problem 2) by muxing audio and video via TMPEGEnc. Hmmmm....two times we see indications for the header code being screwed up exactly in the middle of the two segmented encoding.
Unfortunately I´ll have no big time

My suggestion:
Grab an older version of DGMPEGDec from Neurons site:
http://neuron2.net/dgmpgdec/dgmpgdec.html

Update 1: Try 1.4.6...I think this was the version which I´ve used with HCEnc^n and QuEnc^n prior my last big update

Update 2: I´ve posted over at the DGIndex forum and asked neuron about the mpeg2 header handling of DGIndex.....

I´m pretty confident that we soon know what is happening....

Unfortunately I do not know when Neuron added the -hide parameter which I use to actually prevent the DGIndex windows from showing up when the mpeg segments are merged.
You can do one thing to avoid that DGIndex is called with -hide parameter which is to set Debug in the HCEnc^n.in to 1 like this:

[Common]
Debug=1
Cores=2
DGIndex=C:\Eigene Dateien\AutoIt Projekte\HCEnc^n v0.1.2 (ini-browsing + HCEnc search)\DGIndex.exe
HCEnc=C:\Eigene Dateien\AutoIt Projekte\HCEnc^n v0.1.2 (ini-browsing + HCEnc search)\HCenc_021.exe

Oh well, I have to find some time at the weekend to do further testing but I have the strong feeling that DGIndex is causing this problem. May be the segments are only merged and the header informations are not corrected.
I will run a test encode over night and then merge the M2V segments also via TMPEGEnc stream copy. If my two problems vanish (what I expect them to do because those issues were basically non-existent with older versions of QuEnc^n) DGIndex it "guilty" and I have to discuss with neuron.

Stay tuned,
D$

Lorax2161

10th October 2007, 22:47

Hi Lorax2161,

a near perfect example how a problem description should look like :thanks:

Thanks, Darksoul71. I can be verbose at times, but "this doesn't work" isn't very helpful to a developer, is it? :)

Believe it or not but your description drew my attention to something I didn´t take into account.

Hmmmm....two times we see indications for the header code being screwed up exactly in the middle of the two segmented encoding.

Similar problems to be sure. Even if not identical, it could be the same root cause. Resolving this could fix these problems, and perhaps some that haven't come up yet.

May be the segments are only merged and the header informations are not corrected.

That is my best semi-educated guess at this point. ;)

I will try your suggestions, and I will stay tuned. Thank you for looking into this. Do not rush yourself, I can encode single-core style without problems. Still, if you crack this problem it will be a good day for many of us.

Regards,

Lorax2161

Darksoul71

10th October 2007, 23:01

Indeed...."doesn´t work" without additional informations will never help me and esp. the one posting to solve his problem. :rolleyes:

I just would be happy if more people reporting (potential) problems would do it like you. Saves my time and their time....

Things on the DGIndex do not look so good though. Neuron has replied to my question but the merging code was not touched. DGIndex does not correct the timecode but if this would be causing the problem I would have had the problems I discribed before with QuEnc^n....hmm....

I will do two test encodings with HCEnc^n using DGIndex 1.4.6 because I have them lying around anyway. When problems vanish with v1.4.6, we are at least able to nail the that the new version of DGIndex might cause the problem. After this I´ll get back to neuron.

Stay tuned,
D$

Lorax2161

10th October 2007, 23:16

Things on the DGIndex do not look so good though. Neuron has replied to my question but the merging code was not touched. DGIndex does not correct the timecode but if this would be causing the problem I would have had the problems I discribed before with QuEnc^n....hmm....

I will do two test encodings with HCEnc^n using DGIndex 1.4.6 because I have them lying around anyway. When problems vanish with v1.4.6, we are at least able to nail the that the new version of DGIndex might cause the problem. After this I´ll get back to neuron.

Stay tuned,
D$

I will do the same re: testing & v1.4.6.

Another thought... if DGIndex is not causing this problem could it be receiving info incorrectly and making decisions based on that? (ie: "I was just following orders." ha ha)

For example, somewhere there could be the correct timecodes being divided by the number of cores or subtracted unnecessarily? It would explain why the media players think the video is over at 50%, even though they are joined to a length of 100%. So the duration in the scroll bar is incorrect, but (in my problem) when the needle reaches the end the video keeps playing, although at a crazy speed increase until it finishes.

Thanks for your continued research,

Lorax2161

Darksoul71

10th October 2007, 23:19

no problem...I was just too lazy to think about what might be causing my problems. I´ll run QuEnc^n too with the two movies. TMPEGEnc seems to correct the timecode when muxing the video.

Lorax2161

11th October 2007, 00:04

Sorry, got kicked off temporarily.

no problem...I was just too lazy to think about what might be causing my problems. I´ll run QuEnc^n too with the two movies. TMPEGEnc seems to correct the timecode when muxing the video.

Interesting. For what it's worth, I just replayed the clip in "Nero Showtime." I was correct when I said it displays the wrong video duration of the merged file (50% of true duration) but I was too hasty in another way--the clip plays properly even with the wrong duration info. So at least for that player, it displayed the incorrect video duration in the timeline, but overrode the instructions in regards to playing the merged file.

Sounds like TMPEGEnc overrides what it finds when it's "lied to", as well. Other players do exactly as they are told and "ruin" the clip during the last segment of playback.

Lorax2161

11th October 2007, 01:56

I have a workaround until something better comes along:

Just used ReStream v090 and was able to fix my file. Loaded the m2v, and only made two changes: "horizontal size" had to change from 540 to 480 for me in NTSC country, and a check in the box for "reset timestamps." If you're in PAL-land, you would only need to check the "reset timestamps" box.

After writing a new file, I pulled them both up one at a time in VirtualDubMod. I noticed the file with the new timecode loaded much faster.

Oh, besides that, it played correctly! This also corrects the duration timeline in your media player. I'm off to do a longer test encode than my quick test encode (1 minute) and will post the results.

The two video files do appear to have been written correctly, but there is some incorrect information being inserted/corrupted into the merged m2v header.

Hope this helps...

Lorax2161

Lorax2161

15th October 2007, 23:17

Did some testing over the weekend. ReStream helped somewhat but does not resolve the problem entirely.

The Good News:

ReStream set to "reset timestamps" did just that, and the clips I made played fine in most media players and with the proper duration listed. They also burned without problems in Imgburn when authored by DVDlabPro, and played correctly on my set top DVD player, a Pioneer DV-525.

Additionally, when I removed "Sequence End Codes" via ReStream, it enabled DVDlabPro to use the auto-chapter feature--without removing this end code it would stop in the middle where the two segments were joined. I could not manually override this by choosing "find next I-frame"... it just wouldn't budge after the mid-point.

Unfortunately, I did not find the following helpful piece of info until after I had bumped into it the hard way, but this may be useful for some of you:

http://forum.doom9.org/archive/index.php/t-94862.html

The Bad News:

If I open the merged m2v file with VirtualDubMod and select "FILE", "FILE INFORMATION", it crashes instantly whether I reset the timecodes and removed the sequence end codes or not. (I experience no crashes with my captured NTSC DV source material, nor with an m2v file created with a single instance of HCEncoder.)

Looming over my head is the notion that even though my player can handle these merged clips, I believe that they are not DVD compliant. (Sharp as a tack, aren't I?) As most of my source material is from normal tv captures and family vacations, these DVDs will be expected to play on various set top players. Lack of compliance is a factor for me. The backload I have of material to encode is what had me looking for a tool like HCEnc^n in the first place.

Which leads me to my current speculation of what is possible:

1) The concept of joining two VBR m2v files from the same source, but encoded separately, is not possible while staying DVD compliant. Yes, my set top player can handle it, but I had to reset the time codes and take out the sequence end codes from both segments. A better solution would be to remove the sequence end code from Segment One and leave it in for Segment Two... or maybe write a new one. I don't know. :confused:

2) If it is possible to join two VBR m2v files, then maybe the problem lies in the last GOP being cut off at the end of Segment One. If the source file is split directly in half, this would happen more often than not. Maybe if the slicing were to say, "Find the middle of the file, then back down X number of frames until the number of frames in Segment One is evenly divisible by the number of GOPs and slice the file there" it could merge more successfully than it currently does with Segment Two.

Again, I wish to thank Darksoul71 for his work on this. I agree with his previous comments that you can probably only do so much with a middle-ware solution to multi-core/multi-processor support for a program as complex as HCEncoder. (Thanx Hank.)

I will remain available for testing if that is useful. For now, I have a mountain of encoding to do and because of my DVD compliance concerns with merged m2v's, I will revert back to a single instance of HCEncoder until either there is a breakthrough with the joining problem, or I have thought up another work-around to another work-around.

Regards,

Lorax2161

Revgen

18th October 2007, 01:57

Lorax2161

18th October 2007, 02:26

Okay I tried it on my Quad-Core QX6850 using 8000kbps. I always use high-bitrate.

I got 40fps on the first pass for all 4 cores.

And 20fps on the 2nd pass for all 4 cores.

So basically it's 160fps on the first pass and 80fps on the 2nd pass which combines for 60 total fps over 2 passes. So it's 2.5 times quicker than realtime 24fps despite having to use 2 passes to encode. So a 1 hour 50 minute movie took me about 40 minutes to encode in 2 passes. This definitely is the tool to use for now until Hank makes a multi-threaded HCEnc that does multi-processing internally without splitting the jobs.

@Revgen: That's great news! I cannot hope to match your speed with only two cores, but I would love to see the same kind of success on my system.

I won't repeat the bad luck I've had so far with this, as it's detailed in my previous posts in this thread.

Would you be willing to post your ini & log files please? I am still having a joining problem at merge time, and would appreciate the opportunity to review detailed settings which are proven effective.

Thanks,

Lorax2161

Revgen

18th October 2007, 03:25

Sorry to bust your bubble, but I use TMPGEnc for muxing, so like Darksoul said, it probably is corrected there.

Here's my HCEnc^N log.

17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>HCEnc^n v0.1.2 - the multicore frontend for HCEnc
17/10/2007 13:26:30 >>Using AutoIt Version 3.2.8.0
17/10/2007 13:26:30 >>HCEnc^n is stored in path E:\HC021
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>HcEnc^n was called with the following commandline
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Adding file e:\tcm encode.avs to logfile
17/10/2007 13:26:30 >>LoadPlugin("E:\ColorMatrix.dll")
17/10/2007 13:26:30 >>LoadPlugin("E:\mvbob2\NNEDI.dll")
17/10/2007 13:26:30 >>LoadPlugin("E:\TIVTC.dll")
17/10/2007 13:26:30 >>avisource("E:\Don Juan\ds000.avi")
17/10/2007 13:26:30 >>assumetff()
17/10/2007 13:26:30 >>greyscale()
17/10/2007 13:26:30 >>converttoyv12(interlaced=true)
17/10/2007 13:26:30 >>Interp=nnedi(field=1,threads=1)
17/10/2007 13:26:30 >>tfm(order=1, mode=0, pp=6, field=1, slow=2, mchroma=false,clip2=interp)
17/10/2007 13:26:30 >>tdecimate(chroma=false)
17/10/2007 13:26:30 >>assumefps(24000,1001,sync_audio=true).ResampleAudio(48000)
17/10/2007 13:26:30 >>trim(3910,165278)
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Adding file E:\HC021\juan.ini to logfile
17/10/2007 13:26:30 >>*INFILE e:\tcm encode.avs
17/10/2007 13:26:30 >>*OUTFILE E:\Don Juan\juan.m2v
17/10/2007 13:26:30 >>*LOGFILE E:\juan.log
17/10/2007 13:26:30 >>*BITRATE 8000
17/10/2007 13:26:30 >>*MAXBITRATE 9500
17/10/2007 13:26:30 >>*FRAMES 0 161368
17/10/2007 13:26:30 >>*PROFILE best
17/10/2007 13:26:30 >>*ASPECT 4:3
17/10/2007 13:26:30 >>*AUTOGOP 14
17/10/2007 13:26:30 >>*DC_PREC 10
17/10/2007 13:26:30 >>*PULLDOWN
17/10/2007 13:26:30 >>*MATRIX mpeg
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Commandline #0: "e:\HC_000\HCenc_021.exe" -ini "e:\tcm encode_000.INI"
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Adding file e:\tcm encode_000.INI to logfile
17/10/2007 13:26:30 >>*INFILE e:\tcm encode_000.AVS
17/10/2007 13:26:30 >>*OUTFILE e:\tcm encode_000.M2V
17/10/2007 13:26:30 >>*LOGFILE E:\juan.log
17/10/2007 13:26:30 >>*BITRATE 8000
17/10/2007 13:26:30 >>*MAXBITRATE 9500
17/10/2007 13:26:30 >>
17/10/2007 13:26:30 >>*PROFILE best
17/10/2007 13:26:30 >>*ASPECT 4:3
17/10/2007 13:26:30 >>*AUTOGOP 14
17/10/2007 13:26:30 >>*DC_PREC 10
17/10/2007 13:26:30 >>*PULLDOWN
17/10/2007 13:26:30 >>*MATRIX mpeg
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Instance #0 of HCEnc has been started !
17/10/2007 13:26:30 >>Commandline for HCEnc Instance #0: "e:\HC_000\HCenc_021.exe" -ini "e:\tcm encode_000.INI"
17/10/2007 13:26:30 >>Commandline #1: "e:\HC_001\HCenc_021.exe" -ini "e:\tcm encode_001.INI"
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Adding file e:\tcm encode_001.INI to logfile
17/10/2007 13:26:30 >>*INFILE e:\tcm encode_001.AVS
17/10/2007 13:26:30 >>*OUTFILE e:\tcm encode_001.M2V
17/10/2007 13:26:30 >>*LOGFILE E:\juan.log
17/10/2007 13:26:30 >>*BITRATE 8000
17/10/2007 13:26:30 >>*MAXBITRATE 9500
17/10/2007 13:26:30 >>
17/10/2007 13:26:30 >>*PROFILE best
17/10/2007 13:26:30 >>*ASPECT 4:3
17/10/2007 13:26:30 >>*AUTOGOP 14
17/10/2007 13:26:30 >>*DC_PREC 10
17/10/2007 13:26:30 >>*PULLDOWN
17/10/2007 13:26:30 >>*MATRIX mpeg
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Instance #1 of HCEnc has been started !
17/10/2007 13:26:30 >>Commandline for HCEnc Instance #1: "e:\HC_001\HCenc_021.exe" -ini "e:\tcm encode_001.INI"
17/10/2007 13:26:30 >>Commandline #2: "e:\HC_002\HCenc_021.exe" -ini "e:\tcm encode_002.INI"
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:30 >>Adding file e:\tcm encode_002.INI to logfile
17/10/2007 13:26:30 >>*INFILE e:\tcm encode_002.AVS
17/10/2007 13:26:30 >>*OUTFILE e:\tcm encode_002.M2V
17/10/2007 13:26:30 >>*LOGFILE E:\juan.log
17/10/2007 13:26:30 >>*BITRATE 8000
17/10/2007 13:26:30 >>*MAXBITRATE 9500
17/10/2007 13:26:30 >>
17/10/2007 13:26:30 >>*PROFILE best
17/10/2007 13:26:30 >>*ASPECT 4:3
17/10/2007 13:26:30 >>*AUTOGOP 14
17/10/2007 13:26:30 >>*DC_PREC 10
17/10/2007 13:26:30 >>*PULLDOWN
17/10/2007 13:26:30 >>*MATRIX mpeg
17/10/2007 13:26:30 >>----------------------------------------------------------------------------
17/10/2007 13:26:31 >>Instance #2 of HCEnc has been started !
17/10/2007 13:26:31 >>Commandline for HCEnc Instance #2: "e:\HC_002\HCenc_021.exe" -ini "e:\tcm encode_002.INI"
17/10/2007 13:26:31 >>Commandline #3: "e:\HC_003\HCenc_021.exe" -ini "e:\tcm encode_003.INI"
17/10/2007 13:26:31 >>----------------------------------------------------------------------------
17/10/2007 13:26:31 >>Adding file e:\tcm encode_003.INI to logfile
17/10/2007 13:26:31 >>*INFILE e:\tcm encode_003.AVS
17/10/2007 13:26:31 >>*OUTFILE e:\tcm encode_003.M2V
17/10/2007 13:26:31 >>*LOGFILE E:\juan.log
17/10/2007 13:26:31 >>*BITRATE 8000
17/10/2007 13:26:31 >>*MAXBITRATE 9500
17/10/2007 13:26:31 >>
17/10/2007 13:26:31 >>*PROFILE best
17/10/2007 13:26:31 >>*ASPECT 4:3
17/10/2007 13:26:31 >>*AUTOGOP 14
17/10/2007 13:26:31 >>*DC_PREC 10
17/10/2007 13:26:31 >>*PULLDOWN
17/10/2007 13:26:31 >>*MATRIX mpeg
17/10/2007 13:26:31 >>----------------------------------------------------------------------------
17/10/2007 13:26:31 >>Instance #3 of HCEnc has been started !
17/10/2007 13:26:31 >>Commandline for HCEnc Instance #3: "e:\HC_003\HCenc_021.exe" -ini "e:\tcm encode_003.INI"
17/10/2007 14:46:19 >>All instances of HCEnc have been closed !
17/10/2007 14:46:19 >>Filesize of file e:\tcm encode_000.M2V in Byte: 1675198679
17/10/2007 14:46:19 >>Filesize of file e:\tcm encode_001.M2V in Byte: 1675370479
17/10/2007 14:46:19 >>Filesize of file e:\tcm encode_002.M2V in Byte: 1674318943
17/10/2007 14:46:19 >>Filesize of file e:\tcm encode_003.M2V in Byte: 1671393441
17/10/2007 14:46:19 >>Commandline for DGIndex: -IA=1 -FO=0 -YR=2 -TN=1 -OM=0 -BF=[e:\tcm encode.LST] -DSD=0 -DRC=0 -DSA=0 -OFD=[E:\Don Juan\juan] -EXIT -HIDE
17/10/2007 15:06:19 >>Error: DGIndex failure ! DGIndex was closed by HCEnc^n after 20 minutes timeout !
17/10/2007 15:06:19 >>Filesize of file & E:\Don Juan\juan.m2v in Byte: 0
17/10/2007 15:06:19 >>----------------------------------------------------------------------------
17/10/2007 15:06:19 >>Adding file e:\tcm encode.LST to logfile
17/10/2007 15:06:19 >>e:\tcm encode_000.M2V
17/10/2007 15:06:19 >>e:\tcm encode_001.M2V
17/10/2007 15:06:19 >>e:\tcm encode_002.M2V
17/10/2007 15:06:19 >>e:\tcm encode_003.M2V
17/10/2007 15:06:19 >>----------------------------------------------------------------------------
17/10/2007 15:06:19 >>Merging of M2V files finished !

Lorax2161

18th October 2007, 04:17

Sorry to bust your bubble, but I use TMPGEnc for muxing, so like Darksoul said, it probably is corrected there.

Thank you very much--I'll look it over. :thanks:

Don't worry, my bubble has taken quite a beating over this!

Centurione

22nd October 2007, 17:19

I never noticed the problem, but looking at the log file of muxman I noticed the "seq end code found" in the middle of the m2v file.
I looked into the GUIparameter (that I use to generate the HC.ini) and I found the setting "write seq end code" checked.
I dechecked it and everything seems to work fine.

Hope this helps.

A little request, at least to know if it is possible to find a way to mantain the chapters on the second half of the m2v file.
Now it happen that the chapter are copied equally on both the HC.ini generated, so the chapter (the insertion of an I frame) is correct for the first half, but is incorrect on the second half (it should be needed to delete the chapter relative to the first half and offset the chapter of the second half of the frame of the first half). I hope that my english is clear enough to understand.

Regards, Centurione

Darksoul71

22nd October 2007, 20:08

@Centurione:
Yes, this helps perfectly ! The corresponding parameter is *NOSEQ_ENDCODE

Thanks a lot for pointing out !

I´ll try to release new updated version which adds this parameter to every generated ini-file if it does not exist. Unfortunately my time is currently quite limited since my job is really demanding.

In regard to your second request:
I´m afraid that I will not be able to provide this since the I frames generated for each segment are quite different to those you get when doing a normal HCEnc encode.

As I pointed already out:
HCEnc^n is a "dumb" middleware. Nothing more, nothing less...

Best regards,
D$

yair

24th October 2007, 18:17

thanks for a great app.
have a candy ;) (http://www.pleix.net/movies/Birds.mov)

music video involving 400fps dogs thrown in the air, some laZ0rs action (.mov 15mb)

Darksoul71

24th October 2007, 22:07

@all:
Give HCEnc^n v0.1.3 a try:
http://www.megaupload.com/?d=S4A1Q1M4

*NOSEQ_ENDCODE will be added automatically to the end of your segment ini file if missing in the source ini.

Didn´t have the time to do much testing so. Right now two movies run over night and I´ll check wether TMPEG DVD Author detects the movie length correct then.

:thanks:
go out again to Centurione for pointing out.... :D

yair

24th October 2007, 22:13

hi, megaupload doesn't like my country, hardly ever slots are open.
can has a mirror?

Darksoul71

26th October 2007, 20:37

@yair:
Try rapidshare instead:
http://rapidshare.com/files/65377329/HCENC_n.zip.html

@all: I´m quite busy due to my job until end of November. So don´t expect any releases prior to this date.

Something is still not working as expected:
24/10/2007 21:41:52 >>----------------------------------------------------------------------------
24/10/2007 21:41:52 >>HCEnc^n v0.1.3 - the multicore frontend for HCEnc
24/10/2007 21:41:52 >>Using AutoIt Version 3.2.8.0
24/10/2007 21:41:52 >>HCEnc^n is stored in path C:\Eigene Dateien\AutoIt Projekte\HCEnc^n v0.1.3
24/10/2007 21:41:52 >>----------------------------------------------------------------------------
24/10/2007 21:41:52 >>HcEnc^n was called with the following commandline -cores 2 -ini "F:\Werner DVD\Werner - gekotzt wird spter_3400.INI"
24/10/2007 21:41:52 >>----------------------------------------------------------------------------
24/10/2007 21:41:52 >>----------------------------------------------------------------------------
24/10/2007 21:41:52 >>Adding file F:\Werner DVD\Werner - gekotzt wird später_3400.AVS to logfile
24/10/2007 21:41:52 >>LoadPlugin("C:\Winapp\FitDVD\Plugins\DGDecode.dll")
24/10/2007 21:41:52 >>LoadPlugin("C:\Winapp\FitDVD\Plugins\mpasource.dll")
24/10/2007 21:41:52 >>V=mpeg2source("F:\Werner DVD\Werner - gekotzt wird später.D2V")
24/10/2007 21:41:52 >>A = MPASource("F:\Werner DVD\Werner - gekotzt wird später MPA T01 DELAY 0ms.mpa")
24/10/2007 21:41:52 >>AudioDub(V,A)
24/10/2007 21:41:52 >>DelayAudio(0.000)
24/10/2007 21:41:52 >>Reverse.FadeOut(50).Reverse.FadeOut2(50)
24/10/2007 21:41:52 >>LoadPlugin("C:\Winapp\VCF2AVS\Plugins\SmoothDeinterlacer.dll")
24/10/2007 21:41:52 >>SmoothDeinterlace(lacethresh=15, edgethresh=15)
24/10/2007 21:41:52 >>UseSharpen=1
24/10/2007 21:41:52 >>Denoiselevel=1
24/10/2007 21:41:52 >>ConvertToYV12()
24/10/2007 21:41:52 >>Import("C:\Winapp\FitDVD\Denoiser\DivX (Deblock only).AVS")
24/10/2007 21:41:52 >>Import("C:\Winapp\FitDVD\Denoiser\TV Capture.AVS")
24/10/2007 21:41:52 >>----------------------------------------------------------------------------
24/10/2007 21:41:52 >>----------------------------------------------------------------------------
24/10/2007 21:41:52 >>Adding file F:\Werner DVD\Werner - gekotzt wird spter_3400.INI to logfile
24/10/2007 21:41:52 >>*MAXBITRATE 8000
24/10/2007 21:41:52 >>*PROFILE best
24/10/2007 21:41:52 >>*ASPECT 4:3
24/10/2007 21:41:52 >>*GOP 15 2
24/10/2007 21:41:52 >>*CLOSEDGOPS
24/10/2007 21:41:52 >>*WAIT 0
24/10/2007 21:41:52 >>*INFILE F:\Werner DVD\Werner - gekotzt wird später_3400.AVS
24/10/2007 21:41:52 >>*OUTFILE F:\Werner DVD\Werner - gekotzt wird später_3400.M2V
24/10/2007 21:41:52 >>*LOGFILE F:\Werner DVD\Werner - gekotzt wird später_3400.LOG
24/10/2007 21:41:52 >>*BITRATE 3400
24/10/2007 21:41:52 >>----------------------------------------------------------------------------
24/10/2007 21:41:53 >>Commandline #0: "F:\Werner DVD\HC_000\HCenc_021.exe" -ini "F:\Werner DVD\Werner - gekotzt wird später_3400_000.INI"
24/10/2007 21:41:53 >>----------------------------------------------------------------------------
24/10/2007 21:41:53 >>Adding file F:\Werner DVD\Werner - gekotzt wird später_3400_000.INI to logfile
24/10/2007 21:41:53 >>*MAXBITRATE 8000
24/10/2007 21:41:53 >>*PROFILE best
24/10/2007 21:41:53 >>*ASPECT 4:3
24/10/2007 21:41:53 >>*GOP 15 2
24/10/2007 21:41:53 >>*CLOSEDGOPS
24/10/2007 21:41:53 >>*WAIT 0
24/10/2007 21:41:53 >>*INFILE F:\Werner DVD\Werner - gekotzt wird später_3400_000.AVS
24/10/2007 21:41:53 >>*OUTFILE F:\Werner DVD\Werner - gekotzt wird später_3400_000.M2V
24/10/2007 21:41:53 >>*LOGFILE F:\Werner DVD\Werner - gekotzt wird später_3400.LOG
24/10/2007 21:41:53 >>*BITRATE 3400
24/10/2007 21:41:53 >>
24/10/2007 21:41:53 >>*NOSEQ_ENDCODE
24/10/2007 21:41:53 >>----------------------------------------------------------------------------
24/10/2007 21:41:53 >>Instance #0 of HCEnc has been started !
24/10/2007 21:41:53 >>Commandline for HCEnc Instance #0: "F:\Werner DVD\HC_000\HCenc_021.exe" -ini "F:\Werner DVD\Werner - gekotzt wird später_3400_000.INI"
24/10/2007 21:41:53 >>Commandline #1: "F:\Werner DVD\HC_001\HCenc_021.exe" -ini "F:\Werner DVD\Werner - gekotzt wird später_3400_001.INI"
24/10/2007 21:41:53 >>----------------------------------------------------------------------------
24/10/2007 21:41:53 >>Adding file F:\Werner DVD\Werner - gekotzt wird später_3400_001.INI to logfile
24/10/2007 21:41:53 >>*MAXBITRATE 8000
24/10/2007 21:41:53 >>*PROFILE best
24/10/2007 21:41:53 >>*ASPECT 4:3
24/10/2007 21:41:53 >>*GOP 15 2
24/10/2007 21:41:53 >>*CLOSEDGOPS
24/10/2007 21:41:53 >>*WAIT 0
24/10/2007 21:41:53 >>*INFILE F:\Werner DVD\Werner - gekotzt wird später_3400_001.AVS
24/10/2007 21:41:53 >>*OUTFILE F:\Werner DVD\Werner - gekotzt wird später_3400_001.M2V
24/10/2007 21:41:53 >>*LOGFILE F:\Werner DVD\Werner - gekotzt wird später_3400.LOG
24/10/2007 21:41:53 >>*BITRATE 3400
24/10/2007 21:41:53 >>
24/10/2007 21:41:53 >>*NOSEQ_ENDCODE
24/10/2007 21:41:53 >>----------------------------------------------------------------------------
24/10/2007 21:41:53 >>Instance #1 of HCEnc has been started !
24/10/2007 21:41:53 >>Commandline for HCEnc Instance #1: "F:\Werner DVD\HC_001\HCenc_021.exe" -ini "F:\Werner DVD\Werner - gekotzt wird später_3400_001.INI"
25/10/2007 00:32:08 >>Instance #0 of HCEnc has closed !
25/10/2007 01:49:28 >>Instance #1 of HCEnc has closed !
25/10/2007 01:49:39 >>All instances of HCEnc have been closed !
25/10/2007 01:49:39 >>Filesize of file F:\Werner DVD\Werner - gekotzt wird später_3400_000.M2V in Byte: 993979038
25/10/2007 01:49:39 >>Filesize of file F:\Werner DVD\Werner - gekotzt wird später_3400_001.M2V in Byte: 994042130
25/10/2007 01:49:39 >>Commandline for DGIndex: -IA=1 -FO=0 -YR=2 -TN=1 -OM=0 -BF=[F:\Werner DVD\Werner - gekotzt wird später_3400.LST] -DSD=0 -DRC=0 -DSA=0 -OFD=[F:\Werner DVD\Werner - gekotzt wird später_3400] -EXIT
25/10/2007 01:53:03 >>Filesize of file & F:\Werner DVD\Werner - gekotzt wird später_3400.M2V in Byte: 1988021172
25/10/2007 01:53:03 >>----------------------------------------------------------------------------
25/10/2007 01:53:03 >>Adding file F:\Werner DVD\Werner - gekotzt wird später_3400.LST to logfile
25/10/2007 01:53:03 >>F:\Werner DVD\Werner - gekotzt wird später_3400_000.M2V
25/10/2007 01:53:03 >>F:\Werner DVD\Werner - gekotzt wird später_3400_001.M2V
25/10/2007 01:53:03 >>----------------------------------------------------------------------------
25/10/2007 01:53:03 >>Merging of M2V files finished !

According to this log the *NOSEQ_ENDCODE was added to both ini´s used for parallel encoding. The resulting merged file has still the same behaviour (e.g. TMPEG DVD Author still does not detect the movie with full length).

Without further intense testing I have no clue what is happening.

Oh, one idea that struck my mind:
Could it be that QuEnc does not add a end sequence to the last MPEG2 header ? This would explain why I´ve used QuEnc^n for so long without any muxing problems with MPLEX and movie lengths were always correctly detected by TMPEG DVD Author without problems ?
In this case it would not matter which version of DGIndex was used.....

Darksoul71

5th November 2007, 22:23

Hi all,

I´ve found a nice "workaround". There is a tool called Restream by ssh which is able to fix both issues which HCEnc^n produces.
It corrects the timestamps for the merged video and removed the segment end code. Caveat though: It does not have a native commandline interface and no source code is avaible.

I had to hack together a cludgy AutoIt interface by accessing the controls directly. To make things worse: A few controls within Restream aren´t even accessible via AutoIt as control. So I have to do some "funky" mouseclick magic.

IMPORTANT NOTE: (no, I´m not screaming here ;))
Be patient if the Restream window doesn´t close imediately (may take 4-5 seconds). This is caused by the problem discribed above. I have added a loop to check the filesize of the target M2V when Restream runs. If the filesize doesn´t change anymore then Restream will be closed. Not perfect but worked for me in a few tests.

The final merged M2V is detected with correct playtime in TMPEGEnc DVD Author and MPLEX doesn´t "hickup" anymore.

Keep your fingers crossed but things look good *ggg*

Grab HCEnc^n v0.1.4 here:
http://rapidshare.com/files/67694008/HCEnc_n_v014.zip.html

I´ve added the same stuff to QuEnc^n but have to do further testing prior releasing the new version.

Have fun....

-D$

Lorax2161

6th November 2007, 14:35

Hi all,

I´ve found a nice "workaround". There is a tool called Restream by ssh which is able to fix both issues which HCEnc^n produces.
It corrects the timestamps for the merged video and removed the segment end code.
-D$

That is a nice workaround, Darksoul71. See posts #35 & #36 in this thread when you have a chance. ;)

Bottom line: it solves several problems but not all of them. I don't believe the "fixed" file should immediately crash VirtualDubMod when loaded and you choose "File" and "File Information." This indicates to me that there is still some dimension of the problem that Restream does not address. But it is the best workaround we've found so far.

Darksoul71

6th November 2007, 16:50

Hi Lorax,

I´m stupid as hell ! You pointed out that you´ve used Restream.

Grmpf, if I´ve payed more attention to your postings I would have focused on Restream much earlier. Anyway, I´m unable to reproduce any sort of "crash" with VirtualDubMod here. All my "fixed" M2V files generated with HCEnc^n v0.1.4 are opened fine in VDubMod, show the correct playlength and file info works also correctly. This was with Restream v0.9.0 also.

Ok, here is a updated version of HCEnc^n v0.1.4 which has some code tweaks to keep better focus on the Restream window.
This is not the case with the version I´ve released yesterday.
http://rapidshare.com/files/67852739/HCEnc_n_v014.zip.html

Oh yes, I´ve also changed the behaviour when you click on the tray icon (the scull). Now HCEnc^n doesn´t pause anymore unlike it did before. When you click on the tray icon right now only the menu will show up.

In regard to your problem I have no clue what could be happening.

Best regards,
D$

Lorax2161

6th November 2007, 22:40

Hi Lorax,

I´m stupid as hell ! You pointed out that you´ve used Restream.

Grmpf, if I´ve payed more attention to your postings I would have focused on Restream much earlier.

Not to worry--we know you aren't stupid, just busy. :cool:

Since my earlier posts about it, I've continued to use your utility since it appears that the joining problem is solved at the authoring stage. At the very least, it does not seem to have a problem when I watch the burned DVD on my set top player.

I'm glad you had good luck with Restream. That makes me want to go back and examine things more closely on my end since the merged files do not crash VirtualDubMod for you.

Thanks as always for your contributions, and the new version.

- Lorax2161

Darksoul71

13th September 2008, 20:46

Hi all,

I´ve updated HCenc^n to support the lossless mode for segmented mode also.
I´ve also removed the dependency on DGIndex for merging the segments.
The segments are now merged via binary copy. Time code is still corrected
via DGPulldown.

Grab it here:
http://www.megaupload.com/de/?d=M1H45WN4
or here:
http://rapidshare.de/files/40465421/HCEnc017f.zip.html

Have fun,
D$