DVD-RB 0.62 cce_encoding_init() failed: error

[brashquido]

Hi All,

I'm wanting to set DVD-RB going in batch process mode to rebuild about a dozen or so movies while I am away on holidays. However, when I have Half D1 and half space for extras enabled on some movies I get an "cce_encoding_init() failed:" error as discussed here (http://forum.doom9.org/showthread.php?s=&threadid=74901&perpage=20&pagenumber=1), and here (http://forum.doom9.org/showthread.php?s=&postid=480704#post480704) during the encode phase. Not a huge problem as it seems to go on encoding if you click the ok button, but as I won't be here to do that it will be a bit of a show stopper.

From what I could tell, wmansir's fix of changing non_linear=1 to non_linear=0 requires you to run the prepare stage for each project first. As I will be on holidays and will be looking to use the one click mode used for batch processing, I was just wondering if there was anything I could do to prevent this from happening? Maybe a hidden setting someone could point me too?

The details of the section giving me grief are;

The Cut (PAL Region 4)
VTS 3
Segment 3
Aspect Ratio 4:3

Software Environment

DVD-RB .56
CCE SP 2.66.01.07
EclCCE 1.81
AVISynth 2.5

item.ecl

; Cinema Craft Encoder SP -- Encoder Control List
; Created by DVD ReBuilder

[item]
title=V03000300002002
aud_out=0
vaf_file=H:\TEMP\CUT\D2VAVS\V03000300002002.vaf
aud_file=H:\TEMP\CUT\D2VAVS\V03000300002002.mpa
file_focused=0
packet_size=2048
width=352
height=576
frame_rate_idx=3
cbr_brate=6000
vbr_brate_avg=100
vbr_brate_min=0
vbr_brate_max=9000
seq_endcode=0
dvd=0
half_width=0
half_height=0
lum_level=0
aspect_ratio=2
gop_m=3
gop_nm=4
gop_hdr=12
seq_hdr=1
all_closed_gop=0
fix_gop_length=0
samples_per_sec=44100
stereo=2
brate_idx=7
crc=1
progressive=0
alternate_scan=1
intra_dc_prec=2
aud_mode=0
tc_ref_frm=0
drop_frame=0
fix_vbv_delay=0
letter_box=0
pulldown_detect=0
offset_line=0
create_new_vaf=1
credits_tweak=0
credits_start=0x00000
credits_brate=1000
h_filter=0
h_filter_idx=8
dither=0
dither_max=8
qmat_idx=0
quality_prec=16
video_type=4
vid_file0=H:\TEMP\CUT\D2VAVS\V03000300002002.m2v
vid_file1=H:\TEMP\CUT\D2VAVS\V03000300002002.m2v
vid_out=1
vaf_out=1
timecode=0x0000000
opv_q_factor=20
opv_brate_min=0
opv_brate_max=7200
vbr_bias=25
vbr_pass=5
use_filter=0
filter_val=6
non_linear=1
top_first=0
mpeg1=0
mpeg1_cps=1

[file]
name=H:\TEMP\CUT\D2VAVS\V03000300002002.avs
frame_first=0
frame_last=20
encode_first=0
encode_last=20


AVS of problem section

#------------------
# AVS File Created by DVD Rebuilder
# VOBID:02, CELLID:02
#------------------
mpeg2source("H:\TEMP\CUT\D2VAVS\V03.D2V")
trim(27381,27400)
BilinearResize(352,576)
ConvertToYUY2(interlaced=true)
AudioDub(BlankClip())

[brashquido]

Anyone? :(

[wmansir]

I can't think of any way to prevent this from happening before running a batch job.

How often do you encounter this error? I've done several dozen movies, using that same version of CCE, using the half&half option often, and only received it 2 or 3 times.

[jdobbs]

So it only happens when you use the half/half option, and setting it to non-linear=0 always fixes it? What would be the downside of DVD-RB always setting non-linear=0 when half/half is selected?

[wmansir]

I've only encountered in in DVD-RB while using Half-D1, but I have recreated the error in CCE with a Full-D1 clip.

I'm not sure what impact changing the non-linear setting would have on quality, I haven't experimented with it myself, or if I did it was so long ago I don't recall the differences. Changing it has always fixed the problem in my experience, but I think someone in the thread brashquido linked to above reported it not working.

Another solution might be to set all clips less than 60 frames to CBR. That avoids this error too. I don't think VBR has much impact on such a short clip, and the majority of <60 frame clips are just blank frames anyway.

[jdobbs]

That would be easy to do. I already set single frames to CBR.

[brashquido]

Pardon my ignorance, but why would it still be happening then? Of the 4 projects I'm currently doing;

28 Days Later = No problems
Charlies Angels 2 = Corrupt ISO
Whale Rider = 2 x cce_encoding_init() failed: error
Veronica Guerin = 1 x cce_encoding_init() failed: error

I only ever really use the Half D1 option for movies where the original is over 7GB to try and give as much space as possible to the main feature, so I don't really know how many movies would be effected by it. Is there anything you would like me to do. I've tried using .57 as well with the same results. Perhaps it would be helpful to turn the logging feature on and post the results here?

[pg55555]

jdobbs

First of all Thanks for DVD-Rebuilder. It have helped me to produce great quality backups.

I have been experimenting this "cce_encoding_init() failed: error".
It allways seems to ocurr for small cells (15 to 45 frames in my case) within a VOBID. Usually this cell are just blank (black) at the beginning or ending of the VOBID
It still happened with RB 0.57a and CCE SP 2.66 Trial

I have resolved the situation or by changing the encoding of the cell to CBR or, as wmansir indicated, by changing non_linear=1 to non_linear=0 in the REBUILDER.ECL file.

So, it is not a critical issue, but when you find in the morning the "cce_encoding_init() failed: error" instead of the finished job ...

That would be easy to do. I already set single frames to CBR.
If you can try this solution, maybe as one of the options for CCE...

Thanks in advance.

Pablo

[pg55555]

I got again the "cce_encoding_init() failed: error" when trying to back up The Lion King. (the extras have lots of small cells)

I tried changing non_linear=1 to non_linear=0 for the less than 60 frames segments in the REBUILDER.ECL file, but this time it does not work for the 15 frames cells.

Changing the encoding from VBR to CBR in the REBUILDER.ECL file ( changing video_type=4 to video_type=1 for the problematic cells, also reduced bitrate to 1000 for the segment by changing to CBR=1000) DOES work.

Using RB 0.62, CCE SP 2.66 trial. The Lion King was pre-processed with IFOEdit 0.96 to strip multiangles.

[brashquido]

Thanks for the info pg55555. I just tried changing non_linear=1 to non_linear=0 for section I was having this problem with in Whale Rider, however I am still getting the error. I'll try changing video_type=4 to video_type=1 for the problem cells, and also bring the bitrate down to 1000 for the segment by changing to CBR=1000. I let you know how I get on :)

[pg55555]

Really changing the CBR rate to 1000 is not necessary. I do it to avoid a too oversized output.
In my experience those short cells where the error appears are just blank (black) frames, so reducing the bitrate does not produce any harm.

[brashquido]

Well, it seems to have worked for me :) . Mind you I got the same error on the next VTS as well :rolleyes: . At least I know how to fix it now :) .

@jdobbs

I seem to remember reading in a thread somewhere where you said that DVD-RB already detected cells less than 60 frames and set them to CBR? If further testing of what pg55555 has suggested bare fruit, would it be also worth adding these additional changes? Sorry if this is old news, it's just this is a regular problem I come across when setting the half D1 option for the extras.

[wfn1]

i agree that adding something like that would prove beneficial. setting questionable segments to CBR also helped me with the problem I outlined earlier in this thread http://forum.doom9.org/showthread.php?s=&threadid=83006

I'm not sure it has to be hardcoded in dvd-rb, maybe an option that would be available through a toggle in CCE advanced settings as well as something that would appear if you pressed the stop/pause button. in that case with errors like mine where cce idles you'd have an option to stop it, kill cce and then set the current segment to cbr and resume encoding. i maybe asking for too much here so feel free to cut me off :)

[brashquido]

I think is seems to be pretty universal that this problem only happnes on short cells with a only a few seconds of video (approx 60 frams or less). It also seems that these cells are always blank/black and look to be used as spacers between the real footage.

Personally I think it would be better if there were something hard coded, as once I set a encode going I don't want to hae to keep an eye on my computer to see if it's waiting for input from me. I think this is one of THE largest advantages of DVD-RB over something like the big 3. With the Big 3 method (as good as it is) I was finding I was barely capable of completeing a single project in 24 hours simply because of the amount of times manual input was required, and it was DVD-RB's ability to do an entire project with a single click that made me sit up and take notice.

If jdobbs was able to find a way to detect these small cells and set them to CBR already, then I wouldn't have thought that i would be a great stretch to also change the other options discussed in this thread. Or am I of track here?

[brashquido]

Just wanting to say that the method pg55555 suggested above has worked for me now on two separate DVD's (Whale Rider and The Italian Job).

[jdobbs]

I meant to try and get this into this version -- but I got behind. I'll try and put it in on the next one. Honestly, though, I've never run into this problem so I can't really test/confirm whether it truly works -- I'll have to take your word for it.

[wfn1]

i only had this problem once so far but i managed to get rid of it following pg55555's method so i guess that makes 3 of us.

[brashquido]

Originally posted by jdobbs
I meant to try and get this into this version -- but I got behind. I'll try and put it in on the next one. Honestly, though, I've never run into this problem so I can't really test/confirm whether it truly works -- I'll have to take your word for it.

No worries jdobbs,

I don't know why, but it seems that my last dozen or so projects have been large in size (7GB or more), so I've been using the Half D1 option quite a bit. Of that dozen I reckon I've come across this issue maybe 5 or 6 times, and even multiple instances per movie in a few of them. I used to just try setting non_linear=1 to non_linear=0 with mixed results, but setting the additional points above I haven't had one fail yet over two movies and 4 instances of the problem. Anything you can do would be great :)

[jdobbs]

That's really odd. I don't think I've ever had a project go above 4.32GB except when using OPV (then it might go over 1 out of 10 times). In fact I don't think I ever see a size swing greater than about 0.2GB

[pg55555]

jdobbs

That's really odd. I don't think I've ever had a project go above 4.32GB except when using OPV (then it might go over 1 out of 10 times). In fact I don't think I ever see a size swing greater than about 0.2GB

Maybe you got confused about the thread?:D

The "cce_encoding_init()failed: error" is not related (to my knowledge)to size problems.

[wfn1]

one thing i forgot to mention, the project i got this error on was large over 8gb and it had episodes so i did not use half-d1 and still got the error.

[jdobbs]

Yep. I quick scanned and saw "have been large in size" and completely answered the wrong question -- I'll have to start reading deeper... unfortunately my cognitive processor doesn't seemed to have been designed for multiprocessing.

[brashquido]

Originally posted by wfn1
one thing i forgot to mention, the project i got this error on was large over 8gb and it had episodes so i did not use half-d1 and still got the error.

Really? That makes your problem a little different from me then :? . The ONLY times I have ever gotten this error is when using the half D1 option. I'm actually starting to wonder why I'm getting it so often as I imagine a lot of people would utilize the Half D1 option to save space for the main feature. With that in mind you'd think it would be reported in just about every second thread. I've just gotten the error again in Secret Window. Could versions of AVISynth, CCE or MPEG2DEC3DG.DLL effect this at all?


[i]Yep. I quick scanned and saw "have been large in size" and completely answered the wrong question -- I'll have to start reading deeper... unfortunately my cognitive processor doesn't seemed to have been designed for multiprocessing.[/B]

:D Not to worry, happens to the best of us.

[brashquido]

Just trying to keep track of the moviews I'm experiencing this problem with. I got this error twice when doing 'Secret Window' last night.

[wfn1]

@brashquido

yes, no half-d1 and here's the details: used .62, avsynth 2.5.5.0, mpeg2dec3 1.0.1.0

and in fact here's the important thing, when i use cce 2.50.01.00 i get "cce_encoding_init() failed" but when i use 2.66.01.07 it just stops encoding with the progress meter stuck and no cpu activity.

i got this error while doing futurama s3 d1 (r1 ntsc)