TMPGEnc: TMPGEnc Plus 2.524 (2.524.63.181) Released

glent2k3 · #1 12-17-2004, 06:48 AM

Heres the log

Code:

Specification Changes 	
[1] The formats for the project file and the template file have been changed. Therefore, project created with the new version cannot be used with older versions.
[2] [Project wizard (Step 4/5)]: in the media pull-down selection, DVD 8.5 GB (Double Layer) has been added, for the bitrate setting and size graph.
[3] In the MPEG system multiplexer, the max. bitrate is written instead of the average bitrate for the multiplexing rate. It shortens the outputting process.
[4] [MPEG Setting - GOP structure]: Even if Closed GOP is not selected, the GOP structure is now compliant with the MPEG standard.
[5] Created MPEG-2: the sequence display extension value in the header is now DVD compliant.
Image Quality 	
[1] Image quality for MPEG-1 with long-GOP, has been improved.
[2] Encoding of scenes using still-image, has been improved.
Decoding 	
[1] Decoding process of MPEG-video is faster.
[2] CRI Sofdec MPEG Decoder can read WAVE files (48 / 96 khz, 16 / 20 bit, 1 / 2 ch).
Fixed 	
[1] Errors related with encoding using multi-thread.
[2] The encoding of MPEGs with GOP structure using only I and B pictures, has been improved.
[3] [MPEG Setting - Video]: The Aspect ratio 2.21:1 for MPEG-2, was set to 2.11:1 .
[4] [Output to file - AVI file / WAVE file]: when clicking the Cancel button in the [Sound Selection] dialog, the dialog was shown again.
[5] [MPEG Setting - Quantize matrix]: if saving an existing template under a new name, it could not be deleted.
[6] [MPEG Setting - Quantize matrix]: if saving a template under no name, memory access violations occurred.
[7] When reading empty files (zero data), it could take a very long time to show the error message.
[8] MPEG source files with invalid data could sometime, make the encoding process freeze.
[9] AVI files created with TMPGEnc, when played with Windows Media Player could stop in the middle, with the message "Invalid file".

Dialhot · #2 12-17-2004, 07:18 AM

Quote:

Originally Posted by glent2k3

[1] The formats for the project file and the template file have been changed. Therefore, project created with the new version cannot be used with older versions.

ARGHHHHHH....
I hope than old templates can be used with new tmpgenc, else I can see all the newbies rushing here for help

Quote:

[1] Image quality for MPEG-1 with long-GOP, has been improved.

GREAT !
Tmpgenc was already the best encoder for MPEG1 KVCD. And they improved that ?

kwag · #3 12-17-2004, 09:41 AM

Quote:

Originally Posted by Dialhot

ARGHHHHHH....
I hope than old templates can be used with new tmpgenc, else I can see all the newbies rushing here for help

The templates work fine.
What doesn't work is CQMatic

I'll look into it later this afternoon.

-kwag

kwag · #4 12-17-2004, 11:31 AM

Fixed

Grab it here: http://up1.fastuploads.info/CQMatic-1.3.10.exe.zip

Couple of changes:

1) Forced X1 (standard. X3 disabled) prediction, until further optimized with new algorithms I'm studying.
2) Bumped version to 1.3.10, reflecting changes that it's now compatible with TMPGEnc Plus 2.524 (2.524.63.181) and should also be compatible with previous 2.5x versions.

Let me know of any problems.

-kwag

WOWIEGURL · #5 01-02-2005, 11:24 PM

i used the latest CQmatic and Calcumatic to find the cq... but when it activates Tmpgenc the latest free version.. tmpgenc only shows a sample of 240 frames.

Can someone help?

kwag · #6 01-02-2005, 11:29 PM

Quote:

Originally Posted by WOWIEGURL

Can someone help?

Exactly how

Did you let it run the complete prediction cycle

Also, the latest version of CQMatic is 1.4.03. Is that what you tried

-kwag

WOWIEGURL · #7 01-02-2005, 11:50 PM

its activating the tmpgenc screen where it shows scenes of the movie.. and I am using the version you mentioned in the post.. rite now there is 864 frames of movie being sampled.. but the frames it is showing seem to be out of order... if I let if keep going will the resulting mpg be out of order as well?

kwag · #8 01-02-2005, 11:54 PM

Post your log

Double click on the log window, and paste it here.

-kwag

WOWIEGURL · #9 01-03-2005, 12:31 AM

CQMatic Version 1.4.03
http://www.kvcd.net
Copyright KVCD.Net, 2004.
All rights reserved.
Time: 15:20:39 Date: 01/03/2005
Ready!
Project: F:\project1.tpr

Creating: CQMatic.tpr

F:\project1.mpg
Project resolution: 352x240
Project: F:\project1.tpr

Creating: CQMatic.tpr

F:\project1.mpg
Project resolution: 352x240
Execute.
Movie Time: 90:33
Average Bitrate: 1078.3
Full Encode mode
Executing Prediction Phase...
Process started at 15:21:42
On 01/03/2005
Preparing for calibration phase...
Calibration phase started.
Step 1 of 10.
Step 2 of 10.
Step 3 of 10.
Step 4 of 10.
Step 5 of 10.
Step 6 of 10.
Step 7 of 10.
Step 8 of 10.
Step 9 of 10.
Step 10 of 10.
Calibration phase complete.
CQ set for prediction
Setting up sampling.
Phase 1 of 7, cycle 1
Encoder started...
Process time: 1.78 minutes.
Encoder end.
File size difference = 2.246575
Low fence: 60.000000
High fence: 90.000000
Last CQ = 60.00
Current CQ = 75.00
CQ difference = 15.000000
Phase 1 of 7, cycle 2
Encoder started...
Process time: 1.80 minutes.
Encoder end.
File size difference = 1.463656
Low fence: 75.000000
High fence: 90.000000
Last CQ = 75.00
Current CQ = 82.50
CQ difference = 7.500000
Phase 1 of 7, cycle 3
Encoder started...
Process time: 1.78 minutes.
Encoder end.
File size difference = 1.052403
Low fence: 82.500000
High fence: 90.000000
Last CQ = 82.50
Current CQ = 86.82
CQ difference = 4.323242
Phase 1 of 7, cycle 4
Encoder started...
Process time: 1.80 minutes.
Encoder end.
File size difference = 0.980166
Low fence: 82.500000
High fence: 86.823242
Last CQ = 86.82
Current CQ = 84.66
CQ difference = 2.161621
Phase 1 of 7, cycle 5
Encoder started...
Process time: 1.80 minutes.
Encoder end.
File size difference = 0.984389
Low fence: 82.500000
High fence: 84.661621
Last CQ = 84.66
Current CQ = 83.34
CQ difference = 1.321655
Phase 1 of 7, cycle 6
Encoder started...
Process time: 1.80 minutes.
Encoder end.
File size difference = 1.015452
Setting up sampling.
Phase 2 of 7, cycle 1
Encoder started...
Process time: 1.80 minutes.
Encoder end.
File size difference = 1.004293
Setting up sampling.
Phase 3 of 7, cycle 1
Encoder started...
Process time: 1.82 minutes.
Encoder end.
File size difference = 1.031497
Low fence: 83.339966
High fence: 90.000000
Last CQ = 83.34
Current CQ = 85.96
CQ difference = 2.624992
Phase 3 of 7, cycle 2
Encoder started...
Process time: 1.78 minutes.
Encoder end.
File size difference = 0.998587
Low fence: 83.339966
High fence: 85.964958
Last CQ = 85.96
Current CQ = 84.65
CQ difference = 1.312492
Phase 3 of 7, cycle 3
Encoder started...
Process time: 1.80 minutes.
Encoder end.
File size difference = 1.001651
Setting up sampling.
Phase 4 of 7, cycle 1
Encoder started...
Process time: 1.68 minutes.
Encoder end.
File size difference = 1.037116
Low fence: 84.652466
High fence: 90.000000
Last CQ = 84.65
Current CQ = 87.79
CQ difference = 3.141968
Phase 4 of 7, cycle 2
Encoder started...
Process time: 1.65 minutes.
Encoder end.
File size difference = 1.031497
Low fence: 87.794434
High fence: 90.000000
Last CQ = 87.79
Current CQ = 88.90
CQ difference = 1.102783
Phase 4 of 7, cycle 3
Encoder started...
Process time: 1.70 minutes.
Encoder end.
File size difference = 1.029174
Low fence: 88.897217
High fence: 90.000000
Last CQ = 88.90
Current CQ = 89.45
CQ difference = 0.551392
Setting up sampling.
Phase 5 of 7, cycle 1
Encoder started...
Process time: 1.70 minutes.
Encoder end.
File size difference = 0.949218
Low fence: 2.000000
High fence: 89.448608
Last CQ = 89.45
Current CQ = 84.91
CQ difference = 4.542412
Phase 5 of 7, cycle 2
Encoder started...
Process time: 1.70 minutes.
Encoder end.
File size difference = 0.956770
Low fence: 2.000000
High fence: 84.906197
Last CQ = 84.91
Current CQ = 81.24
CQ difference = 3.670494
Phase 5 of 7, cycle 3
Encoder started...
Process time: 1.68 minutes.
Encoder end.
File size difference = 1.084885
Low fence: 81.235703
High fence: 84.906197
Last CQ = 81.24
Current CQ = 83.07
CQ difference = 1.835251
Phase 5 of 7, cycle 4
Encoder started...
Process time: 1.72 minutes.
Encoder end.
File size difference = 0.993809
Low fence: 81.235703
High fence: 83.070953
Last CQ = 83.07
Current CQ = 82.15
CQ difference = 0.917625
Phase 5 of 7, cycle 5
Encoder started...
Process time: 1.72 minutes.
Encoder end.
File size difference = 1.032896
Low fence: 82.153328
High fence: 83.070953
Last CQ = 82.15
Current CQ = 82.61
CQ difference = 0.458809
Setting up sampling.
Phase 6 of 7, cycle 1
Encoder started...
Process time: 1.78 minutes.
Encoder end.
File size difference = 1.095811
Low fence: 82.612137
High fence: 90.000000
Last CQ = 82.61
Current CQ = 86.31
CQ difference = 3.693932
Phase 6 of 7, cycle 2
Encoder started...
Process time: 1.77 minutes.
Encoder end.
File size difference = 1.026861
Low fence: 86.306068
High fence: 90.000000
Last CQ = 86.31
Current CQ = 88.62
CQ difference = 2.318306
Phase 6 of 7, cycle 3
Encoder started...
Process time: 2.13 minutes.
Encoder end.
File size difference = 1.022267
Low fence: 88.624374
High fence: 90.000000
Last CQ = 88.62
Current CQ = 89.31
CQ difference = 0.687813
Setting up sampling.
Phase 7 of 7, cycle 1
Encoder started...
Process time: 2.43 minutes.
Encoder end.
File size difference = 0.979746
Low fence: 2.000000
High fence: 89.312187
Last CQ = 89.31
Current CQ = 87.50
CQ difference = 1.808960
Phase 7 of 7, cycle 2
Encoder started...
Process time: 2.40 minutes.
Encoder end.
File size difference = 0.982695
Low fence: 2.000000
High fence: 87.503227
Last CQ = 87.50
Current CQ = 85.99
CQ difference = 1.514206
Phase 7 of 7, cycle 3
Encoder started...
Process time: 2.38 minutes.
Encoder end.
File size difference = 0.985238
Low fence: 2.000000
High fence: 85.989021
Last CQ = 85.99
Current CQ = 84.72
CQ difference = 1.269371
Phase 7 of 7, cycle 4
Encoder started...
Process time: 2.48 minutes.
Encoder end.
File size difference = 0.986940
Low fence: 2.000000
High fence: 84.719650
Last CQ = 84.72
Current CQ = 83.61
CQ difference = 1.106415
Phase 7 of 7, cycle 5
Encoder started...
Process time: 2.42 minutes.
Encoder end.
File size difference = 1.004735
Final CQ value is 84.95
Encoding set to Full encode.
Full encode start...
CQMatic complete!
Total minutes of process: 58.12
Process ended at 16:19:49
On 01/03/2005

Dialhot · #10 01-03-2005, 03:47 AM

What do you mean by "out of order" ?

Note: I'm warning you, if this is again a stupid question you just ask to spoil people time, I'll have strong discussion in PM about you with Karl.

rds_correia · #11 01-03-2005, 04:18 AM

Watch out Wowie,
You're watching Tmpgenc's preview window of a sliced video: sometimes it can seem "out of order"...
Cheers

Dialhot · #12 01-03-2005, 04:21 AM

Quote:

Originally Posted by rds_correia

Watch out Wowie,
You're watching Tmpgenc's preview window of a sliced video: sometimes it can seem "out of order"...
Cheers

Perhaps you can tell me the meanings of "out of order" then because I never saw the preview "out of order" by any meanings I know.
Preview is always the image of what is encoded. And is always correct if you had tested your script before to start wasting your and others time in using it in CQMatic

rds_correia · #13 01-03-2005, 05:27 AM

Calm down Phil.
I meant to say that it looks funny when you are looking at the preview of a sliced movie while it is encoding.
Just that.
And I told Wowie to watch out and not doing the same mistake as in the wmp thread.
By no means I meant to say that just because a movie is sliced it looks out of order.
"Out of order" could mean a "universe" of problems.
Like pink spots all over the movie: it would be out of order too, right?
No reason to sweat.
I was just backing up your theory.
Cheers

Dialhot · #14 01-03-2005, 06:22 AM

Quote:

Originally Posted by rds_correia

I meant to say that it looks funny when you are looking at the preview of a sliced movie while it is encoding.

Yes... funny... not "out of order".

Quote:

Like pink spots all over the movie: it would be out of order too, right?

Right. And did you ever see Tmpgenc preview having pink spots ?

rds_correia · #15 01-03-2005, 06:56 AM

Quote:

Originally Posted by Dialhot

Right. And did you ever see Tmpgenc preview having pink spots ?

Nope

First thing that popped into my mind

.
Cheers

kwag · #16 01-03-2005, 09:13 AM

I'll remove the left over test sample on the next update.
It's really unnecesary to leave it, because it was the last test strip encoded on the last phase and really not encoded at the final CQ that the movie will be encoded.
So I'll delete it to avoid any future confusions.

-kwag

Dialhot · #17 01-03-2005, 09:15 AM

Quote:

Originally Posted by kwag

I'll remove the left over test sample on the next update.
It's really unnecesary to leave it, because it was the last test strip encoded on the last phase and really not encoded at the final CQ that the movie will be encoded.
So I'll delete it to avoid any future confusions.

Do you mean that there won't be any .sample file at the end of the process

This file is absolutly needed if you do "predicton" only. Else how can we check the quality and decide to keep the result or not ?

kwag · #18 01-03-2005, 09:27 AM

Quote:

Originally Posted by Dialhot

Do you mean that there won't be any .sample file at the end of the process

Yes.

Quote:

This file is absolutly needed if you do "predicton" only. Else how can we check the quality and decide to keep the result or not ?

Good point

I'll look into making a small sample after the prediction phases, just prior to the full encode

Then the sample left will be done at the CQ that the movie will be encoded.
Expect CQMatic 1.4.04 that later tonight

-kwag

rds_correia · #19 01-03-2005, 02:31 PM

Quote:

Originally Posted by kwag

Good point

I'll look into making a small sample after the prediction phases, just prior to the full encode

Then the sample left will be done at the CQ that the movie will be encoded.
Expect CQMatic 1.4.04 that later tonight

-kwag

Just a suggestion.
The sample file is really helpful when you change to prediction mode only, right?
I mean, if you leave cqmatic set for full encode, I don't see any point in having a sample file.
That is, unless you left it set for full encode by mistake and you want to stop it when it starts encoding, right?
Then I think that it would be cool if you could leave a sample file with the predicted CQ figure, only when cqmatic is set for prediction mode.
Otherwise you could do something magnificent

.
You could rewrite cqmatic for something different.
1st we would choose the tmpgenc.exe path, and a countdown clock box with 05:00 minutes already set but with the ability to change such value.
2nd we would write the length of the movie and point cqmatic to tmpg's project file.
3rd we'dd have a start button on it, as simple as that.
4th cdqmatic would do the prediction modules and would stop with a big noise coming from the built-in PC speaker.
Also there would be an on screen message box stating the predicted CQ figure and it would leave a sample file for us to inspect the quality.
5th cqmatic would still be running with a countdown clock of 5 minutes (or the chosen time from step 1) and with a "start encode" and "cancel encode" buttons that would override the countdown clock.
If the clock reaches 0:00 then the full encode would start. At that point the sample is erased.
If you don't care for checking the quality and you trust your "feeling" that the achieved CQ is ok, you could override the clock by pressing "start full encode". At that point the sample is erased.
If you check the sample quality on your favorite media player and you think it's not good enough you can press the "cancel encode". Maybe you leave the sample for more investigation...
How about that?
I know, I know: I'm nuts

.
But maybe something that I just wrote can give someone else a good idea

.
BTW, can someone split this thread and put the cqmatic related posts on the cqmatic part of the forum?
Cheers guys.

Dialhot · #20 01-03-2005, 02:44 PM

Quote:

Originally Posted by rds_correia

Then I think that it would be cool if you could leave a sample file with the predicted CQ figure, only when cqmatic is set for prediction mode.

Not agreed. Let the file anytime as it does not cost anything to let it.
I often launch a prediction on my PC before to go to sleep (prediction take one hour on my PC). In the morning, the encoding is at 30% (I need 7h to encode a movie). If I see that the sample is too bad, then I switch off the PC while I go to work, else I let the encoding go till its end.