Tools for HC: GetCQ 0.17.0.1 + HIG 0.17.0.1 - now gpl

[Amnon82]

GetCQ for HCEnc by Amnon82

http://img137.imageshack.us/img137/3981/gcq017010mm.png http://img91.imageshack.us/img91/160/hig017017vg.png

This program uses HCEnc by Hank.
Use this program to predict your
movies using a HC.ini-file.

0.17.0.1 <11.04.06>
* updated for HC 0.17
* added new features of HC 0.17

0.16.1.2 <12.12.05>
* Fixed prediction engine
* Same size/%-value is fixed now

(C) 2005 - 2007 by Amnon82

Download Build 0.17.0.1 + HIG 0.17.0.1 (http://home.arcor.de/autoq2_amsoft/files/releases/GetCQ_HIG_01701.7z)

Well, I decided to release the source code of this two apps under gpl. When somebody wants, you can update it to the newest HCenc version. Since I'm on linux now I don't have time for my win32 apps. Sorry.

Source code of HIG 0.17.0.1 (http://paldo.org/~amnon/windows_source_code/delphi/video_encoding/hig/hig-src-0.17.0.1.tar.bz2) (md5sum: d7e5ff4b0b0f22042d0c47e44dfeac9e)

Source code of getCQ 0.17.0.1 (http://paldo.org/~amnon/windows_source_code/delphi/video_encoding/getcq/getcq-src-0.17.0.1.tar.bz2) (md5sum: de63a842fb2152f0a5f120f8d4203c34)

[Fluffbutt]

Nice idea, but the archive contains multiplex, the mpeg (de)multiplexer.

Any chance of the GetCQ, to try? Thanks.

Probably I'm a little early.. But why have multiplex in the archive? Oh well, I can wait.. LOL

[Prodater64]

If the program calculates with CQ and not CQ_maxbitrate parameter, final result could be not DVD compliant.
If you use CQ parameter to calculate and later CQ calculated with CQ_maxbitrate parameter, result could be messed up.
Could you explain this a little bit, how the program works?

[Amnon82]

@Prodater64: It calculates with both. Check 'Force Max.Bit.' for CQ_maxbitrate.
@Fluffbutt: Try 0.16.0.1 ;)

20:59:42 >> GetCQ for HC 0.16.0.1 started ...
21:00:36 >> AVS loaded from: C:\TEST.AVS
21:00:36 >> Desired target: SVCD 352x288
21:00:36 >> Entered wanted size in MB: 450
21:00:36 >> Entered audio size(s) in MB: 66
21:00:36 >> Authoring and ISO overhead in MB: 12
21:00:36 >> % of movie will be used: 2
21:00:36 >> Your desired final encoded sample in MB: 7
21:00:36 >> Your desired final encoded file in MB: 350
21:00:36 >> Encode after prediction enabled
21:00:36 >> _________________________________________
21:00:36 >>
21:00:36 >> _________________________________________
21:00:36 >> Start prediction run #1
21:00:36 >> First CQ-value: 4.8
21:01:39 >> Results of prediction run #1
21:01:39 >> Encoded sample in MB: 6
21:01:39 >> Encoded sample in %: 85,71
21:01:39 >> Wanted samplesize in MB: 7
21:01:39 >> Precalculated finalsize in MB: 300
21:01:39 >> Wanted finalsize in MB: 350
21:01:39 >> _________________________________________
21:01:39 >> Start prediction run #2
21:01:39 >> CQ-value of this pass: 4.0
21:02:45 >> Results of prediction run #2
21:02:45 >> Encoded sample in MB: 7
21:02:45 >> Encoded sample in %: 100,00
21:02:45 >> Wanted samplesize in MB: 7
21:02:45 >> Precalculated finalsize in MB: 350
21:02:45 >> Wanted finalsize in MB: 350
21:02:45 >> _________________________________________
21:02:45 >> Encoding final sample [E = W]
21:03:50 >> Encoded final sample in MB: 7
21:03:50 >> Wanted samplesize in MB: 7
21:03:50 >> Precalculated finalsize in MB: 350
21:03:50 >> Wanted finalsize in MB: 350
21:03:50 >> Calculated finalsize in MB: 350
21:03:50 >> _________________________________________
21:03:50 >> AVS loaded from: C:\TEST.AVS
21:03:50 >> Desired target: SVCD 352x288
21:03:50 >> Entered wanted size in MB: 450
21:03:50 >> Entered audio size(s) in MB: 66
21:03:50 >> Authoring and ISO overhead in MB: 12
21:03:50 >> % of movie will be used: 2
21:03:50 >> Your desired final encoded sample in MB: 7
21:03:50 >> Your desired final encoded file in MB: 350
21:03:50 >> _________________________________________
21:03:50 >> Start fullencode
21:03:50 >> Final Q-value: 4.0
21:04:01 >> Results of prediction run #2
21:04:01 >> Encoded final encode in MB: 3
21:04:01 >> Encoded final encode in %: 42,86
21:04:01 >> Wanted samplesize in MB: 7
21:04:01 >> Precalculated finalsize in MB: 150
21:04:01 >> Wanted finalsize in MB: 350

Most GetCQ needs only two prediction runs ;)

[Amnon82]

What features do you need?

Planed features:
*Save last settings
*Save batchlist
*Rename final encodes in batchmode

00:53:25 >> GetCQ for HC 0.16.0.2 started ...
00:53:48 >> AVS loaded from: AVI_SAMPLE.AVI.AVS
00:53:48 >> Desired target: SVCD 352x288
00:53:48 >> Entered wanted size in MB: 400
00:53:48 >> Entered audio size(s) in MB: 60
00:53:48 >> Authoring and ISO overhead in MB: 12
00:53:48 >> % of movie will be used: 2
00:53:48 >> Your desired final encoded sample in MB: 7
00:53:48 >> Your desired final encoded file in MB: 350
00:53:48 >> Encode after prediction disabled
00:53:48 >> _______________________________________
00:53:48 >>
00:53:48 >> _______________________________________
00:53:48 >> Start prediction run #1
00:53:48 >> First CQ-value: 4.2
00:54:56 >> Results of prediction run #1
00:54:56 >> Encoded sample in MB: 6
00:54:56 >> Encoded sample in %: 85,71
00:54:56 >> Wanted samplesize in MB: 7
00:54:56 >> Precalculated finalsize in MB: 300
00:54:56 >> Wanted finalsize in MB: 350
00:54:56 >> _______________________________________
00:54:56 >> Start prediction run #2
00:54:56 >> CQ-value of this pass: 3.8
00:56:03 >> Results of prediction run #2
00:56:03 >> Encoded sample in MB: 7
00:56:03 >> Encoded sample in %: 100,00
00:56:03 >> Wanted samplesize in MB: 7
00:56:03 >> Precalculated finalsize in MB: 350
00:56:03 >> Wanted finalsize in MB: 350
00:56:03 >> _______________________________________
00:56:03 >> Encoding final sample
00:57:09 >> Encoded final sample in MB: 7
00:57:09 >> Wanted samplesize in MB: 7
00:57:09 >> Precalculated finalsize in MB: 350
00:57:09 >> Wanted finalsize in MB: 350
00:57:09 >> Calculated finalsize in MB: 350
00:57:09 >> Sample renamed to AVI._005709.sample
00:57:09 >> _______________________________________
00:57:09 >>

[Fluffbutt]

Thanks man!

is this anything like your "AutoCQ 052 For HC Encoder.rar" (my rename) from a while ago?

Whatever happened to that, it was pretty good. I seem to recall it didn't like new HC versions - does this new GetCQ allow for newer HC's to come?

[Amnon82]

AutoCQ was with a GUI. It created the HC.ini-s. I added the HCBatch.exe into the main-program. I'll think about release a new version of AutoCQ. GetCQ has the new AutoQ2 prediction mode I've enhanced to work better with HC. HC gaves me a 0.0 CQ value format. CCE has only 0. So I can create better CQ values using HC. This tool reaches always 100% ;) It has a much better prediction tech than AutoCQ. Maybe I'll join AutoCQ with GetCQ using a special form to create a HC.ini. Also I added the batchmode to it. Do you have the exit error after using an AVISource?

[kumi]

Wow, very cool, I'm going to test this as soon as possible.

Unfortunately it won't run: HC complains about "input not complete", then GetCQ throws up an error MessageBox: '4,8' ist kein gültiger Gleitkommawert. Problem with the decimal symbol? (I'm using Windows XP SP1, USA settings, period decimal symbol)

[Amnon82]

hmm. In germany it is the way around. 4.8 ist kein gültiger Gleitkommawert.

4.8 is the cqvalue. I change it to 4,8 for calculation. I think I should turn it around for USA.

[Amnon82]

http://img389.imageshack.us/img389/1272/trennzeichen1zj.png

dezimaltrennzeichen - decimal separator was on my computer a comma and not a dot.
I changed it now to a dot on my machine and changed the default cq value to 5.

[Amnon82]

00:04:36 >> GetCQ for HC 0.16.0.3b started ...
00:05:02 >> AVS loaded from: SAMPLE.D2V.AVS
00:05:02 >> Desired target: SVCD 480x576
00:05:02 >> Entered wanted size in MB: 30
00:05:02 >> Entered audio size(s) in MB: 3
00:05:02 >> Authoring and ISO overhead in MB: 0
00:05:02 >> % of movie will be used: 2
00:05:02 >> Your desired final encoded sample in MB: 1
00:05:02 >> Your desired final encoded file in MB: 50
00:05:02 >> Encode after prediction enabled
00:05:02 >> _______________________________________
00:05:02 >>
00:05:02 >> _______________________________________
00:05:02 >> Start prediction run #1
00:05:02 >> First CQ-value: 5
00:05:10 >> Results of prediction run #1
00:05:10 >> Encoded sample in MB: 0
00:05:10 >> Error 21: 0 MB! Can't go on.
00:05:10 >> GetCQ brakes the prediction mode.
00:05:10 >> AVS loaded from: SAMPLE.D2V.AVS
00:05:10 >> Desired target: SVCD 480x576
00:05:10 >> Entered wanted size in MB: 30
00:05:10 >> Entered audio size(s) in MB: 3
00:05:10 >> Authoring and ISO overhead in MB: 0
00:05:10 >> % of movie will be used: 16
00:05:10 >> Your desired final encoded sample in MB: 4
00:05:10 >> Your desired final encoded file in MB: 25
00:05:10 >> Encode after prediction enabled
00:05:10 >> _______________________________________
00:05:10 >>
00:05:10 >> _______________________________________
00:05:10 >> Start prediction run #1
00:05:10 >> First CQ-value: 5
00:05:30 >> Results of prediction run #1
00:05:30 >> Encoded sample in MB: 3
00:05:30 >> Encoded sample in %: 75.00
00:05:30 >> Wanted samplesize in MB: 4
00:05:30 >> Precalculated finalsize in MB: 19
00:05:30 >> Wanted finalsize in MB: 25
00:05:30 >> _______________________________________
00:05:30 >> Start prediction run #2
00:05:30 >> CQ-value of this pass: 4.2
00:05:53 >> Results of prediction run #2
00:05:53 >> Encoded sample in MB: 3
00:05:53 >> Encoded sample in %: 75.00
00:05:53 >> Wanted samplesize in MB: 4
00:05:53 >> Precalculated finalsize in MB: 19
00:05:53 >> Wanted finalsize in MB: 25
00:05:53 >> _______________________________________
00:05:53 >> Start prediction run #3
00:05:53 >> CQ-value of this pass: 3.5
00:06:17 >> Results of prediction run #3
00:06:17 >> Encoded sample in MB: 3
00:06:17 >> Encoded sample in %: 75.00
00:06:17 >> Wanted samplesize in MB: 4
00:06:17 >> Precalculated finalsize in MB: 19
00:06:17 >> Wanted finalsize in MB: 25
00:06:17 >> _______________________________________
00:06:17 >> Start prediction run #4
00:06:17 >> CQ-value of this pass: 2.9
00:06:43 >> Results of prediction run #4
00:06:43 >> Encoded sample in MB: 4
00:06:43 >> Encoded sample in %: 100.00
00:06:43 >> Wanted samplesize in MB: 4
00:06:43 >> Precalculated finalsize in MB: 25
00:06:43 >> Wanted finalsize in MB: 25
00:06:43 >> Sample renamed to SAMPLE.000643.sample
00:06:43 >> _______________________________________
00:06:43 >>
00:06:43 >> _______________________________________
00:06:43 >> AVS loaded from: SAMPLE.D2V.AVS
00:06:43 >> Desired target: SVCD 480x576
00:06:43 >> Entered wanted size in MB: 30
00:06:43 >> Entered audio size(s) in MB: 3
00:06:43 >> Authoring and ISO overhead in MB: 0
00:06:43 >> % of movie will be used: 16
00:06:43 >> Your desired final encoded file in MB: 25
00:06:43 >> _______________________________________
00:06:43 >> Start fullencode
00:06:43 >> Final Q-value: 2.9
00:07:53 >> Results of final encode:
00:07:53 >> Encoded final encode in MB: 19
00:07:53 >> Wanted final encode in MB: 25
00:07:53 >> Final encode renamed to SAMPLE.000753.m2v
00:07:53 >> _______________________________________
00:07:53 >>
00:07:53 >> AVS loaded from: SAMPLE.D2V.AVS
00:07:53 >> Desired target: SVCD 480x576
00:07:53 >> Entered wanted size in MB: 30
00:07:53 >> Entered audio size(s) in MB: 3
00:07:53 >> Authoring and ISO overhead in MB: 0
00:07:53 >> % of movie will be used: 16
00:07:53 >> Your desired final encoded sample in MB: 4
00:07:53 >> Your desired final encoded file in MB: 25
00:07:53 >> Encode after prediction enabled
00:07:53 >> _______________________________________
00:07:53 >>
00:07:53 >> _______________________________________
00:07:53 >> Start prediction run #1
00:07:53 >> First CQ-value: 5
00:08:13 >> Results of prediction run #1
00:08:13 >> Encoded sample in MB: 3
00:08:13 >> Encoded sample in %: 75.00
00:08:13 >> Wanted samplesize in MB: 4
00:08:13 >> Precalculated finalsize in MB: 19
00:08:13 >> Wanted finalsize in MB: 25
00:08:13 >> _______________________________________
00:08:13 >> Start prediction run #2
00:08:13 >> CQ-value of this pass: 4.2
00:08:36 >> Results of prediction run #2
00:08:36 >> Encoded sample in MB: 3
00:08:36 >> Encoded sample in %: 75.00
00:08:36 >> Wanted samplesize in MB: 4
00:08:36 >> Precalculated finalsize in MB: 19
00:08:36 >> Wanted finalsize in MB: 25
00:08:36 >> _______________________________________
00:08:36 >> Start prediction run #3
00:08:36 >> CQ-value of this pass: 3.5
00:09:00 >> Results of prediction run #3
00:09:00 >> Encoded sample in MB: 3
00:09:00 >> Encoded sample in %: 75.00
00:09:00 >> Wanted samplesize in MB: 4
00:09:00 >> Precalculated finalsize in MB: 19
00:09:00 >> Wanted finalsize in MB: 25
00:09:00 >> _______________________________________
00:09:00 >> Start prediction run #4
00:09:00 >> CQ-value of this pass: 2.9
00:09:26 >> Results of prediction run #4
00:09:26 >> Encoded sample in MB: 4
00:09:26 >> Encoded sample in %: 100.00
00:09:26 >> Wanted samplesize in MB: 4
00:09:26 >> Precalculated finalsize in MB: 25
00:09:26 >> Wanted finalsize in MB: 25
00:09:26 >> Sample renamed to SAMPLE.000926.sample
00:09:26 >> _______________________________________
00:09:26 >>
00:09:26 >> _______________________________________
00:09:26 >> AVS loaded from: SAMPLE.D2V.AVS
00:09:26 >> Desired target: SVCD 480x576
00:09:26 >> Entered wanted size in MB: 30
00:09:26 >> Entered audio size(s) in MB: 3
00:09:26 >> Authoring and ISO overhead in MB: 0
00:09:26 >> % of movie will be used: 16
00:09:26 >> Your desired final encoded file in MB: 25
00:09:26 >> _______________________________________
00:09:26 >> Start fullencode
00:09:26 >> Final Q-value: 2.9
00:10:35 >> Results of final encode:
00:10:35 >> Encoded final encode in MB: 19
00:10:35 >> Wanted final encode in MB: 25
00:10:35 >> Final encode renamed to SAMPLE.001035.m2v
00:10:35 >> _______________________________________
00:10:35 >>

Sorry for all bugs in 0.16.0.3a. It is fixed now in 0.3b. Happy testing.

[Amnon82]

In the next release I'll add an enhanced GUI. The style will almost be the same like in AutoQ 1.0.8.2. If you have some sugestions, tell me ...

[Amnon82]

Released 0.5

HIG will come soon.

[neo squidward]

always error in hc about input file. wont work at all

[Amnon82]

What kind of error? Post me your HC.ini and your AVS-Script.
You can try my sample HC.ini in the INI-Folder. I tested it with an avi and it worked for me. Download the last build 0.16.0.5. This one shows you an english error message. Under the version 0.16.0.5 they are all german...

[neo squidward]

Used your ini file and it worked. Thanks! (I thought I did last night but guess not)

[Amnon82]

np. maybe you did some wrong. In the next release I'll add a INI-Gen so you won't have this problem again ;)

[klicker4546]

This program never worked for me yet! Where do I have to edit and finally place the HC-ini file? Pretty annoying so far! Is there somewhere a documentation?

[Amnon82]

GetCQ searches for the HC.ini which is placed in the INI folder. When You start GetCQ a new file HC.bak is created which is a backup of the HC.ini. You have to edit the HC.ini before starting GetCQ or search for an other INI using the '...'-button in the INI-Line. So GetCQ uses the INI path which is written in the edit.

Here is the full sample HC.ini updated for HC 016:

-------------------------------
HC.ini template
These commands are required
Use full pathnames and just use the bitrate you want
-------------------------------
*infile G:\MOVIE\MOVIE.avs
*outfile G:\MOVIE\MOVIE.m2v
*bitrate 550
*maxbitrate 850
-------------------------------
Next commands are optional,
putting a space for it disables a command
-------------------------------
Aspect ratio has the options 1:1,4:3,16:9,2.21:1
-------------------------------
*aspect 1:1
*aspect 4:3
*aspect 16:9
*aspect 2.21:1
-------------------------------
This command specifies bottom field first,
only used for interlaced encoding. Use TFF
to set top field first
-------------------------------
*bff
*tff
-------------------------------
This command tweaks the compression curve,
0 means VBR, 100 tends to CBR
-------------------------------
*bias 30
-------------------------------
This command closes all gops
-------------------------------
*closegops
-------------------------------
This command can be used to force the cpu
to use specific extensions
-------------------------------
*CPU AUTO
*CPU MMX
*CPU SSE2
*CPU SSE3
-------------------------------
This commands orders the encoder to do a
1-pass with a constant quantization factor.
Use cq_maxbitrate to be DVD-compliant.
If the actual bitrate overshoots the maximum
bitrate set by the MAXBITRATE command the
quantizer is temporally raised.
-------------------------------
*cq 4.0
*cq_maxbitrate 4.2
-------------------------------
This command defines custom intra and intermatrices
to be used and supersedes the default matrix and
the MATRIX command. The matrices must be supplied
as given in the example, 8 values per line.
-------------------------------
*custommatrix
8 32 32 32 32 32 32 32
32 32 32 32 32 32 32 32
32 32 32 32 32 32 32 32
32 32 32 32 32 32 32 32
32 32 32 32 32 32 32 32
32 32 32 32 32 32 32 32
32 32 32 32 32 32 32 32
32 32 32 32 32 32 32 32

16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
16 16 16 16 16 16 16 16
-------------------------------
This command defines the intra and intermatrix
to use.
-------------------------------
*matrix HC
*matrix NOTCH
*matrix QLB
*matrix BATCH1
*matrix JAWOR1CD
*matrix HVSGOOD
*matrix HVSBETTER
*matrix HVSBEST
*matrix AVAMAT6
*matrix AVAMAT7
-------------------------------
This command defines the DC-precision to be used,
range 8 - 10
-------------------------------
*dc_prec 10
-------------------------------
This command specifies the frames to be encoded,
if this command is not pressent all frames will
be encoded. In case of avisynth imput don't use this
command but use the trim option in the avisynth file.
-------------------------------
*frames 1001 2500
-------------------------------
This command defines the GOP structure. Max GOP lenght is 36,
max B-frames is 2. If this command is omitted, HC will run in
AUTOGOP mode.
-------------------------------
*gop 12 2
-------------------------------
Use this command if your source is interlaced, it might improve
quality. If this command is used also the alternate scanmethod
is set.
-------------------------------
*interlaced
-------------------------------
This command disables output of the sequence endcode (00 00 01 B7)
*noseq_endcode
-------------------------------
This command disables the scene change detection. Scene change detection
is enabled by default, at each scene change an I-frame is inserted
and the GOP is closed.
-------------------------------
*noscd
-------------------------------
This command disables the VBV checking.
-------------------------------
*novbv
-------------------------------
This command defines the encoding quality.
-------------------------------
*profile fast
*profile normal
*profile best
-------------------------------
This command defines the scanning method to be used.
If interlaced is flagged it defaults to ALT but can
be set to ZIGZAG with this command.
-------------------------------
*scanmethod ZIGZAG
*scanmethod ALT
-------------------------------
This command will shutdown the system in 300 seconds
after finishing encoding
-------------------------------
*shutdown
-------------------------------
This command sets the timecode
-------------------------------
*timecode 0 0 0 0
-------------------------------
this command sets the waiting time before HC exits
after an encoding session, range 0 - 99
-------------------------------
*wait 4
-------------------------------
This command defines the log filename, if omitted
no logfile will be written. If the logfile already
exits the logs will be appended.
-------------------------------
*logfile G:\MOVIE\MOVIE.log
-------------------------------
HC.ini by Amnon82
-------------------------------

[Amnon82]

http://img523.imageshack.us/img523/5155/hig4hj.png

HIG will come on the WE ;)

[AMED]

I think i've found a bug in the 0.16.1.0 release.

I've used HIG to create a .ini file for HC.
______________________________________________
HIG 0.16.1.0 INI.template
______________________________________________
*infile F:\Family Guy\Episode 101 - Death has a Shadow.avi - avisynth.avs
*outfile F:\Family Guy\Episode 101 - Death has a Shadow.m2v
*logfile F:\Family Guy\test.log
*bitrate 4000
*maxbitrate 8000
*profile best
______________________________________________
*matrix AVAMAT6
*frames 0 0
*autogop
______________________________________________
*dc_prec 10
*cpu AUTO
*scanmethod ZIGZAG
*aspect 4:3
*bias 30
*timecode 0 0 0 0
*TFF
______________________________________________
I have tested the "Episode 101 - Death has a Shadow.avi - avisynth.avs" script and it plays fine in MPC.

I use the above INI file in GetCQ and when i click start HC says that the AVS script couldn't be loaded. I had a look at the GCQPREDICTION.avs file that is used for the prediction and found a few problems.
#AVS Created by GetCQ0.16.1.0
#_________________________________
function GetCQPrediction(clip input,int percent,int Goplenght,int Gopmulti,string "ping")
{
PercCount=(Framecount(input)/100)*percent
period=int(Framecount(input)/PercCount)*(Goplenght*Gopmulti)
input=(ping=="pong")?input.trim((period/2),Framecount(input)):input
selectrangeevery(input,period,(Goplenght*Gopmulti))
}
ping = Import("F:\FAMILYGUY\EPISODE101-DEATHHASASHADOW.AVI-AVISYNTH.AVS").GetCQPrediction(2,12,1,"ping")
pong = Import("F:\FAMILYGUY\EPISODE101-DEATHHASASHADOW.AVI-AVISYNTH.AVS").GetCQPrediction(2,12,1,"pong")
ping++pong
Subtitle("GetCQ Ver.0.16.1.0 Prediction - (C) 2005 by Amnon82", 35, 35, 25, 125, font="georgia", size=12, text_color=$ffFFFF)
the two lines to note are ping and pong, it looks like when GetCQ makes the prediction script it removes all the spaces of the location to and the filename of the original AVS script.

Also, what is going on with setting the size out the output MV2 ?
http://img225.imageshack.us/img225/7441/019gb.th.png (http://img225.imageshack.us/my.php?image=019gb.png)
Is there any way of making it optional for calculating the authoring and iso overhead.

does the output file size have to be a multiple of 25? Is there a chance of fine tuning down to say 1MB ?

e.g.
If i set a target filesize of 500MB
The desired final encode size is 375MB

If i set a target filesize of 510MB
The desired final encode size is 400MB

[Amnon82]

The program is in the early state. I thougth it would be easier for beginners that GetCQ do the calculations for the overhead. It is based on AutoQ2 new prediction mode. I removed the edits where the overhead can be edit so the gui will be easier for beginners.

SVCD overhead is dynamic. I took 12 MB for 795 MB. If you enter 385 for example it would be around 6 MB. Also the audio size edit gets subtracted. You have to write zero in it.

Maybe I'll add an advanced GUI so you can tweak it.

For the spacesbug. I added a new code for the detection called uppercase. It don't like space as you found out. I used never spaces in my scripts so I didn't found out.

I'll think about it to edit it in the next release.
Let's see what changes hank will do for the next release of his encoder. His new features I'll add also.

[klicker4546]

Hey Amed,

I've also played around with Amnons tool. Where's the problem to remove the spaces? Maybe consider it not as a bug, you'd better of to think of it as a feature. ;-)

I like that peace of software although there's work left to do. I'm looking forward for new releases.

[Amnon82]

@AMED: Version 0.16.1.1 fixes the problem you mentioned.

[AMED]

Thank you for fixing that

[Firesurfer]

Hi,

I tried v0.16.1.0 and after hours of running it stopped without encoding. It seems that a little sanity checking on CQ-Values should be done :).
The story in short:

run #1: CQ=5 -> undersized
run #2: CQ=4,138 -> undersized
run #3: CQ=3,996 -> way oversized
run #4: CQ=5,627, but even CQ=4,138 was already undersized....

The script I'm running:

SetMemoryMax(192)

LoadPlugin("E:\cut_streams\filter\colormatrix.dll")
LoadPlugin("E:\cut_streams\filter\autocrop.dll")
LoadPlugin("E:\cut_streams\filter\undot.dll")

mpeg2source("tatort.d2v")
ColorMatrix()
AutoCrop(mode=0,threshold=50,samples=10)
h=(720-width(last))/2
v=(576-height(last))/2
Undot()
AddBorders(h,v,h,v)
ColorYUV(levels="pc->tv")
FadeIO(4)


The log-file (very long, but needed to make my point):

09:16:09 >> GetCQ for HC 0.16.1.0 started ...
09:16:13 >> AVS loaded from: !TATORT.AVS
09:16:13 >> Desired target: DVD 720x576
09:16:13 >> Entered wanted size in MB: 2200
09:16:13 >> Entered audio size(s) in MB: 122
09:16:13 >> Authoring and ISO overhead in MB: 117
09:16:13 >> % of movie will be used: 4
09:16:13 >> Your desired final encoded sample in MB: 78
09:16:13 >> Your desired final encoded file in MB: 1950
09:16:13 >> Encode after prediction enabled
09:16:13 >> _______________________________________
09:16:13 >>
09:16:13 >> _______________________________________
09:16:13 >> Start prediction run #1
09:16:17 >> First CQ-value: 5
09:23:58 >> Results of prediction run #1
09:23:59 >> Encoded sample in MB: 58
09:23:59 >> Encoded sample in %: 74,36
09:23:59 >> Wanted samplesize in MB: 78
09:23:59 >> Precalculated finalsize in MB: 1450
09:23:59 >> Wanted finalsize in MB: 1950
09:23:59 >> _______________________________________
09:23:59 >> Start prediction run #2
09:24:05 >> CQ-value of this pass: 4,138
09:31:41 >> Results of prediction run #2
09:31:41 >> Encoded sample in MB: 73
09:31:41 >> Encoded sample in %: 93,59
09:31:41 >> Wanted samplesize in MB: 78
09:31:41 >> Precalculated finalsize in MB: 1825
09:31:41 >> Wanted finalsize in MB: 1950
09:31:41 >> _______________________________________
09:31:41 >> Start prediction run #3
09:31:48 >> CQ-value of this pass: 3,996
09:39:25 >> Results of prediction run #3
09:39:25 >> Encoded sample in MB: 98
09:39:25 >> Encoded sample in %: 125,64
09:39:25 >> Wanted samplesize in MB: 78
09:39:25 >> Precalculated finalsize in MB: 2450
09:39:25 >> Wanted finalsize in MB: 1950
09:39:25 >> _______________________________________
09:39:25 >> Start prediction run #4
09:39:31 >> CQ-value of this pass: 5,627
09:47:06 >> Results of prediction run #4
09:47:07 >> Encoded sample in MB: 58
09:47:07 >> Encoded sample in %: 74,36
09:47:07 >> Wanted samplesize in MB: 78
09:47:07 >> Precalculated finalsize in MB: 1450
09:47:07 >> Wanted finalsize in MB: 1950
09:47:07 >> _______________________________________
09:47:07 >> Start prediction run #5
09:47:13 >> CQ-value of this pass: 4,657
09:54:44 >> Results of prediction run #5
09:54:45 >> Encoded sample in MB: 73
09:54:45 >> Encoded sample in %: 93,59
09:54:45 >> Wanted samplesize in MB: 78
09:54:45 >> Precalculated finalsize in MB: 1825
09:54:45 >> Wanted finalsize in MB: 1950
09:54:45 >> _______________________________________
09:54:45 >> Start prediction run #6
09:54:51 >> CQ-value of this pass: 4,498
10:02:25 >> Results of prediction run #6
10:02:25 >> Encoded sample in MB: 73
10:02:25 >> Encoded sample in %: 93,59
10:02:25 >> Wanted samplesize in MB: 78
10:02:25 >> Precalculated finalsize in MB: 1825
10:02:25 >> Wanted finalsize in MB: 1950
10:02:25 >> _______________________________________
10:02:25 >> Start prediction run #7
10:02:31 >> CQ-value of this pass: 4,344
10:10:03 >> Results of prediction run #7
10:10:03 >> Encoded sample in MB: 73
10:10:03 >> Encoded sample in %: 93,59
10:10:03 >> Wanted samplesize in MB: 78
10:10:03 >> Precalculated finalsize in MB: 1825
10:10:03 >> Wanted finalsize in MB: 1950
10:10:03 >> _______________________________________
10:10:03 >> Start prediction run #8
10:10:17 >> CQ-value of this pass: 4,195
10:17:57 >> Results of prediction run #8
10:17:57 >> Encoded sample in MB: 73
10:17:57 >> Encoded sample in %: 93,59
10:17:57 >> Wanted samplesize in MB: 78
10:17:57 >> Precalculated finalsize in MB: 1825
10:17:57 >> Wanted finalsize in MB: 1950
10:17:57 >> _______________________________________
10:17:57 >> Start prediction run #9
10:18:03 >> CQ-value of this pass: 4,051
10:25:38 >> Results of prediction run #9
10:25:38 >> Encoded sample in MB: 73
10:25:38 >> Encoded sample in %: 93,59
10:25:38 >> Wanted samplesize in MB: 78
10:25:38 >> Precalculated finalsize in MB: 1825
10:25:38 >> Wanted finalsize in MB: 1950
10:25:38 >> _______________________________________
10:25:38 >> Start prediction run #10
10:25:45 >> CQ-value of this pass: 3,912
10:33:25 >> Results of prediction run #10
10:33:26 >> Encoded sample in MB: 98
10:33:26 >> Encoded sample in %: 125,64
10:33:26 >> Wanted samplesize in MB: 78
10:33:26 >> Precalculated finalsize in MB: 2450
10:33:26 >> Wanted finalsize in MB: 1950
10:33:26 >> _______________________________________
10:33:26 >> Start prediction run #11
10:33:32 >> CQ-value of this pass: 5,509
10:41:02 >> Results of prediction run #11
10:41:02 >> Encoded sample in MB: 58
10:41:02 >> Encoded sample in %: 74,36
10:41:02 >> Wanted samplesize in MB: 78
10:41:02 >> Precalculated finalsize in MB: 1450
10:41:02 >> Wanted finalsize in MB: 1950
10:41:02 >> _______________________________________
10:41:02 >> Start prediction run #12
10:41:07 >> CQ-value of this pass: 4,559
10:48:42 >> Results of prediction run #12
10:48:42 >> Encoded sample in MB: 73
10:48:42 >> Encoded sample in %: 93,59
10:48:42 >> Wanted samplesize in MB: 78
10:48:42 >> Precalculated finalsize in MB: 1825
10:48:42 >> Wanted finalsize in MB: 1950
10:48:42 >> _______________________________________
10:48:42 >> Start prediction run #13
10:48:49 >> CQ-value of this pass: 4,403
10:56:24 >> Results of prediction run #13
10:56:24 >> Encoded sample in MB: 73
10:56:24 >> Encoded sample in %: 93,59
10:56:24 >> Wanted samplesize in MB: 78
10:56:24 >> Precalculated finalsize in MB: 1825
10:56:24 >> Wanted finalsize in MB: 1950
10:56:24 >> _______________________________________
10:56:24 >> Start prediction run #14
10:56:31 >> CQ-value of this pass: 4,252
11:04:08 >> Results of prediction run #14
11:04:08 >> Encoded sample in MB: 73
11:04:08 >> Encoded sample in %: 93,59
11:04:08 >> Wanted samplesize in MB: 78
11:04:08 >> Precalculated finalsize in MB: 1825
11:04:08 >> Wanted finalsize in MB: 1950
11:04:08 >> _______________________________________
11:04:08 >> Start prediction run #15
11:04:17 >> CQ-value of this pass: 4,106
11:12:00 >> Results of prediction run #15
11:12:00 >> Encoded sample in MB: 73
11:12:00 >> Encoded sample in %: 93,59
11:12:00 >> Wanted samplesize in MB: 78
11:12:00 >> Precalculated finalsize in MB: 1825
11:12:01 >> Wanted finalsize in MB: 1950
11:12:01 >> _______________________________________
11:12:01 >> Start prediction run #16
11:12:08 >> CQ-value of this pass: 3,965
11:19:55 >> Results of prediction run #16
11:19:55 >> Encoded sample in MB: 98
11:19:55 >> Encoded sample in %: 125,64
11:19:55 >> Wanted samplesize in MB: 78
11:19:55 >> Precalculated finalsize in MB: 2450
11:19:55 >> Wanted finalsize in MB: 1950
11:19:55 >> _______________________________________
11:19:55 >> Start prediction run #17
11:20:03 >> CQ-value of this pass: 5,583
11:27:39 >> Results of prediction run #17
11:27:39 >> Encoded sample in MB: 58
11:27:39 >> Encoded sample in %: 74,36
11:27:39 >> Wanted samplesize in MB: 78
11:27:39 >> Precalculated finalsize in MB: 1450
11:27:39 >> Wanted finalsize in MB: 1950
11:27:39 >> _______________________________________
11:27:39 >> Start prediction run #18
11:27:47 >> CQ-value of this pass: 4,620
11:35:28 >> Results of prediction run #18
11:35:28 >> Encoded sample in MB: 73
11:35:28 >> Encoded sample in %: 93,59
11:35:28 >> Wanted samplesize in MB: 78
11:35:28 >> Precalculated finalsize in MB: 1825
11:35:28 >> Wanted finalsize in MB: 1950
11:35:28 >> _______________________________________
11:35:28 >> Start prediction run #19
11:35:36 >> CQ-value of this pass: 4,462
11:40:53 >> Results of prediction run #19
11:40:53 >> Encoded sample in MB: 0
11:40:53 >> Error 21: 0 MB! Can't go on.
11:40:53 >> GetCQ brakes the prediction mode.
11:40:53 >> _______________________________________
11:40:53 >>
11:40:53 >> _______________________________________

[Amnon82]

It is still in the beginnig. The prediction with HCEnc is still for testing. SVCD you can't encode cos of some overridings of the max bitrate.

HCEnc is developed for DVD encoding. Hank added cq_maxbitrate in his last release. We've to test it. Maybe I'll work closer with hank and we implent my prediction directly into HCEnc.

I read your log. It seems it jumps around and don't stop prediction. You got in 5 passes 'Encoded sample in MB: 73'. Normaly it should stop in the second pass of this loop. I've to check the sources to find the bug.

If you have space on your hdd you can keep your project-files. I hope I can release on wednesday a fixed version. Maybe tonight, but I'm realy not shure.

Edit: I found some time and released the fixed version now ;)

[Firesurfer]

Thanks for the quick fix! :)

I had the files still on my hd, so I tried again with the new version. It started the full encode, but the final file came out about 180 MB too small. Unfortunately I had to delete the source today for space reasons, so no more tests for this one. My guess is, that the prediction is not going well with my noisy tv captures. WinTV pvr350 is a real pita when it has to deal with grainy stuff. I will try on my "nightmare before christmas" cap tomorrow :)

have a nice one



12:33:49 >> GetCQ for HC 0.16.1.2 started ...
12:34:12 >> AVS loaded from: !TATORT.AVS
12:34:12 >> Desired target: DVD 720x576
12:34:12 >> Entered wanted size in MB: 2200
12:34:12 >> Entered audio size(s) in MB: 122
12:34:12 >> Authoring and ISO overhead in MB: 117
12:34:12 >> % of movie will be used: 4
12:34:12 >> Your desired final encoded sample in MB: 78
12:34:12 >> Your desired final encoded file in MB: 1950
12:34:12 >> Encode after prediction enabled
12:34:12 >> _______________________________________
12:34:12 >>
12:34:12 >> _______________________________________
12:34:12 >> Start prediction run #1
12:34:14 >> First CQ-value: 5
12:42:30 >> Results of prediction run #1
12:42:31 >> Encoded sample in MB: 58
12:42:31 >> Encoded sample in %: 74,36
12:42:31 >> Wanted samplesize in MB: 78
12:42:31 >> Precalculated finalsize in MB: 1450
12:42:31 >> Wanted finalsize in MB: 1950
12:42:31 >> _______________________________________
12:42:31 >> Start prediction run #2
12:42:45 >> CQ-value of this pass: 4,138
12:53:45 >> Results of prediction run #2
12:53:45 >> Encoded sample in MB: 73
12:53:45 >> Encoded sample in %: 93,59
12:53:45 >> Wanted samplesize in MB: 78
12:53:45 >> Precalculated finalsize in MB: 1825
12:53:45 >> Wanted finalsize in MB: 1950
12:53:45 >> _______________________________________
12:53:45 >> Start prediction run #3
12:53:56 >> CQ-value of this pass: 3,996
13:01:57 >> Results of prediction run #3
13:01:57 >> Encoded sample in MB: 98
13:01:57 >> Encoded sample in %: 125,64
13:01:57 >> Wanted samplesize in MB: 78
13:01:58 >> Precalculated finalsize in MB: 2450
13:01:58 >> Wanted finalsize in MB: 1950
13:01:58 >> _______________________________________
13:01:58 >> Start prediction run #4
13:02:03 >> CQ-value of this pass: 5,627
13:09:34 >> Results of prediction run #4
13:09:35 >> Encoded sample in MB: 58
13:09:35 >> Encoded sample in %: 74,36
13:09:35 >> Wanted samplesize in MB: 78
13:09:35 >> Precalculated finalsize in MB: 1450
13:09:35 >> Wanted finalsize in MB: 1950
13:09:35 >> _______________________________________
13:09:35 >> Start prediction run #5
13:09:45 >> CQ-value of this pass: 4,657
13:17:12 >> Results of prediction run #5
13:17:12 >> Encoded sample in MB: 73
13:17:12 >> Encoded sample in %: 93,59
13:17:12 >> Wanted samplesize in MB: 78
13:17:12 >> Precalculated finalsize in MB: 1825
13:17:12 >> Wanted finalsize in MB: 1950
13:17:12 >> _______________________________________
13:17:12 >> Start prediction run #6
13:17:19 >> CQ-value of this pass: 4,498
13:24:49 >> Results of prediction run #6
13:24:49 >> Encoded sample in MB: 73
13:24:49 >> Encoded sample in %: 93,59
13:24:49 >> Wanted samplesize in MB: 78
13:24:49 >> Precalculated finalsize in MB: 1825
13:24:49 >> Wanted finalsize in MB: 1950
13:24:49 >> Sample renamed to !TATORT.132449.sample
13:24:49 >> Same % value
13:24:49 >> _______________________________________
13:24:49 >>
13:24:49 >> _______________________________________
13:24:49 >> AVS loaded from: !TATORT.AVS
13:24:49 >> Desired target: DVD 720x576
13:24:49 >> Entered wanted size in MB: 2200
13:24:49 >> Entered audio size(s) in MB: 122
13:24:49 >> Authoring and ISO overhead in MB: 117
13:24:49 >> Your desired final encoded file in MB: 1950
13:24:49 >> _______________________________________
13:24:49 >> Start fullencode
13:24:49 >> Final Q-value: 4,344
16:16:59 >> Results of final encode:
16:16:59 >> Encoded final encode in MB: 1777
16:16:59 >> Wanted final encode in MB: 1950

[Prodater64]

ATM, There is not a reliable prediction method that can work fine with HC Encoder. HC Encoder watch for max btr for keep itself inside of DVD compliant. When max btr is higher than compliant, HC cuts it. This bits are partially bringed back to the encoder, but the method still doesn't work in an exact way.
For that reason, error in filesize could be most of times between +/- 4%.
Some times more than this.

[Amnon82]

As I told. Prediction is still in the beginning. We've to work closer with hank on this.

[john3voltas]

For that reason, error in filesize could be most of times between +/- 4%.
Some times more than this.
Well but sometimes the error is also lower than 2%.
Actually most of the times.
But not with GetCQ!
Avalon, I think you still haven't got how the ping pong prediction works.
And this prooves my statement:
16:14:24 >> GetCQ for HC 0.16.1.0 started ...
16:19:32 >> AVS loaded from: MATRIX_I.AVS
16:19:32 >> Desired target: DVD 720x576
16:19:32 >> Entered wanted size in MB: 2238
16:19:32 >> Entered audio size(s) in MB: 358
16:19:32 >> Authoring and ISO overhead in MB: 117
16:19:32 >> % of movie will be used: 2
16:19:32 >> Your desired final encoded sample in MB: 35
16:19:32 >> Your desired final encoded file in MB: 1750
16:19:32 >> Encode after prediction disabled
16:19:32 >> _______________________________________
16:19:32 >>
16:19:32 >> _______________________________________
16:19:32 >> Start prediction run #1
16:19:33 >> First CQ-value: 5
16:25:22 >> Results of prediction run #1
16:25:22 >> Encoded sample in MB: 37
16:25:22 >> Encoded sample in %: 105,71
16:25:22 >> Wanted samplesize in MB: 35
16:25:22 >> Precalculated finalsize in MB: 1850
16:25:22 >> Wanted finalsize in MB: 1750
16:25:22 >> _______________________________________
16:25:22 >> Start prediction run #2
16:25:24 >> CQ-value of this pass: 5,541
16:31:07 >> Results of prediction run #2
16:31:07 >> Encoded sample in MB: 37
16:31:07 >> Encoded sample in %: 105,71
16:31:07 >> Wanted samplesize in MB: 35
16:31:07 >> Precalculated finalsize in MB: 1850
16:31:07 >> Wanted finalsize in MB: 1750
16:31:07 >> _______________________________________
16:31:07 >> Start prediction run #3
16:31:09 >> CQ-value of this pass: 6,140
16:36:37 >> Results of prediction run #3
16:36:37 >> Encoded sample in MB: 31
16:36:37 >> Encoded sample in %: 88,57
16:36:37 >> Wanted samplesize in MB: 35
16:36:37 >> Precalculated finalsize in MB: 1550
16:36:37 >> Wanted finalsize in MB: 1750
16:36:37 >> _______________________________________
16:36:37 >> Start prediction run #4
16:36:38 >> CQ-value of this pass: 5,744
16:42:22 >> Results of prediction run #4
16:42:22 >> Encoded sample in MB: 37
16:42:22 >> Encoded sample in %: 105,71
16:42:22 >> Wanted samplesize in MB: 35
16:42:22 >> Precalculated finalsize in MB: 1850
16:42:22 >> Wanted finalsize in MB: 1750
16:42:22 >> _______________________________________
16:42:22 >> Start prediction run #5
16:42:23 >> CQ-value of this pass: 6,365
16:48:27 >> Results of prediction run #5
16:48:27 >> Encoded sample in MB: 31
16:48:27 >> Encoded sample in %: 88,57
16:48:27 >> Wanted samplesize in MB: 35
16:48:27 >> Precalculated finalsize in MB: 1550
16:48:27 >> Wanted finalsize in MB: 1750
16:48:27 >> _______________________________________
16:48:27 >> Start prediction run #6
16:48:28 >> CQ-value of this pass: 5,954
16:54:50 >> Results of prediction run #6
16:54:50 >> Encoded sample in MB: 37
16:54:50 >> Encoded sample in %: 105,71
16:54:50 >> Wanted samplesize in MB: 35
16:54:50 >> Precalculated finalsize in MB: 1850
16:54:50 >> Wanted finalsize in MB: 1750
16:54:50 >> _______________________________________
16:54:50 >> Start prediction run #7
16:54:52 >> CQ-value of this pass: 6,598
17:01:01 >> Results of prediction run #7
17:01:01 >> Encoded sample in MB: 31
17:01:01 >> Encoded sample in %: 88,57
17:01:01 >> Wanted samplesize in MB: 35
17:01:01 >> Precalculated finalsize in MB: 1550
17:01:01 >> Wanted finalsize in MB: 1750
17:01:01 >> _______________________________________
17:01:01 >> Start prediction run #8
17:01:04 >> CQ-value of this pass: 6,172
17:06:54 >> Results of prediction run #8
17:06:54 >> Encoded sample in MB: 31
17:06:54 >> Encoded sample in %: 88,57
17:06:54 >> Wanted samplesize in MB: 35
17:06:54 >> Precalculated finalsize in MB: 1550
17:06:54 >> Wanted finalsize in MB: 1750
17:06:54 >> _______________________________________
17:06:54 >> Start prediction run #9
17:06:56 >> CQ-value of this pass: 5,774
17:12:34 >> Results of prediction run #9
17:12:34 >> Encoded sample in MB: 37
17:12:34 >> Encoded sample in %: 105,71
17:12:34 >> Wanted samplesize in MB: 35
17:12:34 >> Precalculated finalsize in MB: 1850
17:12:34 >> Wanted finalsize in MB: 1750
17:12:34 >> _______________________________________
17:12:34 >> Start prediction run #10
17:12:36 >> CQ-value of this pass: 6,398
17:18:27 >> Results of prediction run #10
17:18:27 >> Encoded sample in MB: 31
17:18:27 >> Encoded sample in %: 88,57
17:18:27 >> Wanted samplesize in MB: 35
17:18:27 >> Precalculated finalsize in MB: 1550
17:18:27 >> Wanted finalsize in MB: 1750
17:18:27 >> _______________________________________
17:18:27 >> Start prediction run #11
17:18:29 >> CQ-value of this pass: 5,985
17:24:05 >> Results of prediction run #11
17:24:05 >> Encoded sample in MB: 37
17:24:05 >> Encoded sample in %: 105,71
17:24:05 >> Wanted samplesize in MB: 35
17:24:05 >> Precalculated finalsize in MB: 1850
17:24:05 >> Wanted finalsize in MB: 1750
17:24:05 >> _______________________________________
17:24:05 >> Start prediction run #12
17:24:06 >> CQ-value of this pass: 6,632
17:29:43 >> Results of prediction run #12
17:29:43 >> Encoded sample in MB: 31
17:29:43 >> Encoded sample in %: 88,57
17:29:43 >> Wanted samplesize in MB: 35
17:29:43 >> Precalculated finalsize in MB: 1550
17:29:43 >> Wanted finalsize in MB: 1750
17:29:43 >> _______________________________________
17:29:43 >> Start prediction run #13
17:29:45 >> CQ-value of this pass: 6,204
17:29:49 >> Results of prediction run #0
17:29:49 >> Encoded sample in MB: 0
17:29:49 >> Error 21: 0 MB! Can't go on.
17:29:49 >> GetCQ brakes the prediction mode.
17:29:49 >> _______________________________________
17:29:49 >>
17:29:49 >> _______________________________________
17:29:49 >> AVS loaded from: MATRIX_I.AVS
17:29:49 >> Desired target: DVD 720x576
17:29:49 >> Entered wanted size in MB: 2238
17:29:49 >> Entered audio size(s) in MB: 358
17:29:49 >> Authoring and ISO overhead in MB: 117
17:29:49 >> % of movie will be used: 2
17:29:49 >> Your desired final encoded sample in MB: 35
17:29:49 >> Your desired final encoded file in MB: 1750
17:29:49 >> Encode after prediction disabled
17:29:49 >> _______________________________________
17:29:49 >>
17:29:49 >> _______________________________________
17:29:49 >> Start prediction run #1
17:29:50 >> First CQ-value: 6,204
17:29:54 >> Results of prediction run #1
17:29:54 >> Encoded sample in MB: 0
17:29:54 >> Error 21: 0 MB! Can't go on.
17:29:54 >> GetCQ brakes the prediction mode.
17:29:54 >> _______________________________________
17:29:54 >>
17:29:54 >> _______________________________________

Approximatly 1h15m and still no go?
Worse, you use many times CQ values that have been droped in previous runs...
Too many bugs, man.
You're supposed to do something like this:
2% ping
2% pong
check the average filesize
2% ping
2% ping
...
2% ping -> bullseye
2% pong just to check all is fine.
Manually I hit an almost perfect CQ in 5-7 runs plus the initial and the last ping+pong.
You're obviously very confused on how this works...
See ya

[Amnon82]

yeah. I know how it works. I tried Prodator64s ping+pong way:

#AVS Created by AutoQ2 050106.1085;
#_________________________________
function AutoQ2Prediction(clip input,int percent,int Goplenght,int Gopmulti,string "ping")
{
PercCount=(Framecount(input)/100)*percent
period=int(Framecount(input)/PercCount)*(Goplenght*Gopmulti)
input=(ping=="pong")?input.trim((period/2),Framecount(input)):input
selectrangeevery(input,period,(Goplenght*Gopmulti))
}
ping = Import("'+openavs.filename+'").AutoQ2Prediction('+floattostr(strtoint(PVAL.text)/2)+','+Gop.text+',1,"ping")
pong = Import("'+openavs.filename+'").AutoQ2Prediction('+floattostr(strtoint(PVAL.text)/2)+','+Gop.text+',1,"pong")
ping++pong

For CCE it works. For HC not.
When I've time I'll fix it ...

[techmule]

As I told. Prediction is still in the beginning. We've to work closer with hank on this.

HC ver 0.17 is out, will GetCQ still work or things need to change???.

Were you able to work "close enough" with hank ???

[Amnon82]

I don't know. You can test it if you want. Atm I've lots to do in my real life. Also I put more power into autoq2. It should work cos it is now free for all hcbatch.exe-files. I don't compile it including the hcbatch.exe anymore.

I don't found time to look into HC017 and its new cq-features. Maybe I'll compile a new version soon.

[techmule]

IAlso I put more power into autoq2. It should work cos it is now free for all hcbatch.exe-files. I don't compile it including the hcbatch.exe anymore.

I don't found time to look into HC017 and its new cq-features. Maybe I'll compile a new version soon.

Does Autoq2 work with HCbatch.exe encoder ??, I thought Autoq2 was only for CCE.

If yes, how ???

[Amnon82]

AutoQ2 is ONLY for CCE. HIG and GetCQ is for HC. I don't plan to add HC to AutoQ2 for the moment, but who knows ;)

[techmule]

I don't found time to look into HC017 and its new cq-features. Maybe I'll compile a new version soon.

Any update ???? now that AutoQ2 has got a stable release, perhaps you can give some time to HCEnc CQ prediction(GetCQ).

[Amnon82]

ATM I'm still in deep work with AutoQ2. When I find time for HC a new version will come out.

[Amnon82]

First screenshot of HIG 0.17.0.1

http://img91.imageshack.us/img91/160/hig017017vg.png

... will come today :D

[techmule]

Thanks Amnon, keep up the good work :)

I am waiting to test the new prediction routines eagerly and I hope they would be more precise than the earlier version ;).

[Amnon82]

It depends on the HC version. 0.16 didn't had the p/b factor in it. The routine will be the same. I only add the new features to HIG and change the HC.exe in GetCQ. We must first test if I've to change the routine for 0.17.

[techmule]

It depends on the HC version. 0.16 didn't had the p/b factor in it. The routine will be the same. I only add the new features to HIG and change the HC.exe in GetCQ. We must first test if I've to change the routine for 0.17.

Oh!, I thought you will add/change some of them from the autoQ2 experience based on their prediction reliability and precision.

Since the CQ value could be used independently as well in DVD-RB free, the p/b values should not be changed during prediction, so make it more flexible in its usage.(RB doesnt have a configurable p/b parameter)

[Amnon82]

You can only set them up. During prediction only the CQVALUE changes. I only updated my gui.

[Amnon82]

Well, I decided to release the source code of this two apps under gpl. When somebody wants, you can update it to the newest HCenc version. Since I'm on linux now I don't have time for my win32 apps. Sorry.

Source code of HIG 0.17.0.1 (http://paldo.org/~amnon/windows_source_code/delphi/video_encoding/hig/hig-src-0.17.0.1.tar.bz2) (md5sum: d7e5ff4b0b0f22042d0c47e44dfeac9e)

Source code of getCQ 0.17.0.1 (http://paldo.org/~amnon/windows_source_code/delphi/video_encoding/getcq/getcq-src-0.17.0.1.tar.bz2) (md5sum: de63a842fb2152f0a5f120f8d4203c34)

[45tripp]

Thanks Amnon.

If someone would like to spend some time with it, i think a few minor adjusrments could improve the package greatly.
I have these comments:

HIG
"Opens with list index out of bounds (39)" error.
the only thing really needed to make it HC 21 compliant is to change the frames setting. By default *frames 0 0 is written to the hc ini file. This would encode all frames in 17 but with 21 the line needs to be removed. A tick box to enable writting "frames to the ini or not would be nice. Or just have the end frame default setting at an obsenely high frame number.
Add lumgain option.

GetCQ
GetCQ needs to be put in a path without spaces.
Not friendly at all. Setting paths within both hig and getcq to ini files, video files, encoders, there is no problem with having spaces in the path. But if you put getcq in a path with a space the ini file is not loaded to HC.
An option to set custom overhead size is needed. GetCq defaults to 117 MB regardless of target size.
Maybe the prediction formula avs script could be tweaked with.

I also had HC 21 sort of hang on first prediction pass for a while before beginning with encoding. Can't faithfully reproduce though.

ty

[Amnon82]

Seems I deleted the source-files from my server. I reupped them. Should work again ...