version 2.56.39.143.
Also tried 2.512.52.161.

-break

Ok, I just ran TMPGEnc 2.56, and the project file looks fine. I was able to load it in CQMatic without a problem.
Can you post your .tpr file :?:

-kwag

Hi break,
I've had the same problem for the past week. If I would run CQMatic in prediction mode, TMPGEnc would pop up only for a second and close twice. I was getting the same text message everytime ending with a CQ of 90. If I performed a full encode, TMPGEnc would give me an illegal video steam error when CQMatic tried to open it.

It seemed like CQMatic started working again when I deleted the currentbatch.tbe and currentcfg.tpr file in the Tmpgenc folder and started TMPGEnc.exe up fresh. I'm not sure if it was just one of those files or both, but when I deleted them, everything worked. Maybe those files got corrupted somehow.

flood

I have posted it you in PM, since file is very long... :wink:

@flood
Thanks flood, I'll try your suggestion :)


-break

Hi flood, thanks very much! :D :D :D
After I have deleted the currentbatch.tbe and currentcfg.tpr file in the Tmpgenc folder, I run another prediction test:
1 pass, CQ 60
2 pass, CQ 69,11!
Now CQMatic is searching the right CQ... 8)

But now, I have to delete that file all the time before using CQMatic? :?:

-break

break,
You shouldn't have to delete it every time. That problem happened to me twice, and I think it happened when I cancelled a test prediction. For some reason those files or one of those files created a problem.

flood

OK flood, thanks again! :wink:

-break

well, testing with old RC1 version now, cause the rc2 is creating too big files, doesn´t matter which values i take for MIN and MAX :(

comparing now cq values of rc1 and rc2 for same movie and settings, lets see.,..

Okay... I still have problems with CQMatic...(when I'm trying to encode avi-files!)... kwag, perhaps you find a way to make it running!

http://www.kvcd.net
CQMatic Version 1.0 RC-2
Copyright Softronex Corporation, 2003.
All rights reserved.

Creating: CQMatic.tpr

E:\Piglet#39s.Big.Movie.DiAMOND 25 Frames.m1v
Execute.
Movie Time: 72
Average Bitrate: 1271
Prediction Only mode
Executing Prediction Phase...
Using CQ of 60.00
Encoder started...
Encoder end.
File size difference = -14885753.000000
Last CQ = 60.000000
Current CQ = 90.000000
CQ difference = 30.000000
Using CQ of 90.00
Encoder started...
Encoder end.
File size difference = -14885753.000000
CQ ABOVE watermark. CQ set to 90.0
Using CQ of 90.00
CQMatic complete!

What problems 8O :!:
Your film is only 72 minutes long, and that's why CQMatic chooses 90 as CQ :!:
Just encode the movie :!:

-kwag

The same problem happened to me again. This time I selected "New Project" under File in TMPGEnc and it worked this way. This may be better than deleting those two files. From now on I will start doing this before creating a project file.

-flood

I'm sorry kwag... the CQ is about 80... I did it with ToK...

CQMatic seems to have problems with avi files...

Hi fn86,

I just noticed en error in your log, that I didn't pay attention before :!:
CQMatic is not predictiong AT ALL your source :!:
The line File size difference = -14885753.000000 is an impossible condition, because the software is doing a division of wanted file size by your sample file size. So it can never be negative :!:
So the problem seems to be the way your long path is set up in TMPEG, and I can't read it properly in CQMatic.

What is your complete path to your avi :?:

-kwag

RC-3 is up :!:

-kwag

hi!

i´d like to know, those who got fine results with cqmatic, what do you take for MIN and MAX bitrates ?

thx ;)

bye

MIN = (0.57 * Average Bitrate)
MAX = 2,000Kbs

-kwag

tested rc2 with 107 min movie,
set max 2500, min = 0.57*avg bitrate
cq predicted = 72,02
movie after muxxing = 808 mb
well a little bit too big, but just a little

test 2 :
83 min movie, max and min like above,
cq predicted = 77,31
video stream length = 727 mb, should be 715
again a little bit too high

well , think that the MAX bitrate is resulting in too high filesizes here.
testing now 83 min movie with MAX = 2000 and min = 0.57*avg bitrate,
with rc3

cq predicted here is 79,98.

will see results in 2,5 h :)

hi:

bad results :(

min=0,57*avg bitrate, max=2000, rc3
83 min movie
length of video stream = 780 MB !!! JUST the video! :(

perhaps a bug?
don´t know, please help :)

I have been getting the SAME problem consistently with both RC-1 and RC-2! I was able to get some great predictions with some of the earlier betas so I'm trying again with beta7. Haven't tried RC-3 yet though... I'm also using a max=2000 so maybe that has been a problem with the more recent RC's? However I've been using this all along, even when I got good prediction with the betas (only problem with the betas for me was the damn cq looping bug!)

I'll see if beta7 or rc-3 makes a difference, and if not I'll try max=2500. BTW this has all been in avi sources of normal movie length (90-120min), the good and the bad predictions.