The numbers that differ by just 1 can be explained by the fact that Avisynth calls the first frame 0 and other programs call the first frame 1. I don't understand the larger discrepancies. I load the VOBs into DGIndex and make a d2v straight off, so I seldom use PGCDemux.
Thanks Chewtobacca, I tried using the VOBs direct rather then creating the m2v file first and DGIndex reported the same frame count for both Star Wars and Empire so I am still scratching my head on this one.
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
Using the VOBs directly would give you PGC_2 first (15 frames) followed by PGC_1 (main feature.) So I would have thought the total frame count would be higher by fifteen. Yet you say it's the same when you use the VOB method? That's very weird indeed.
LeeThorogood said:
Is the small variation in frame count normal when working with NTSC material or have I missed something?
This statement befuddles me because I've never seen frame counts listed whenever I run PGCDemux (except in the Logfile, but you had that turned off.)LeeThorogood said:
PgcDemux read the NTSC GOUT DVD of Star Wars as 174262 frames
However, in practice you must take into account the “fuckwit factor”. Just talk to Darth Mallwalker…
-Moth3r
I make the frame count the same as Darth Mallwalker. I was just about to post too lol.
The NTSC GOUT is 99.98% film, so there is a tiny amount of hard telecined material. The total is 174276. After forced film the total is 174270.
EDIT: Sorry, Darth Mallwalker I thought you said DGIndex befuddled you, but you said PGCDemux. My mistake!
Thanks for the assistance Darth Malwalker! :D First of all my apologies I forgot to mention before that the frame counts were identical after I took the 15 frames of PGC_2 into account.Darth Mallwalker said:
Using the VOBs directly would give you PGC_2 first (15 frames) followed by PGC_1 (main feature.) So I would have thought the total frame count would be higher by fifteen. Yet you say it's the same when you use the VOB method? That's very weird indeed.
I will get these files for you and post back when I have them.Darth Mallwalker said:
Not sure you've missed anything, but emphatically I say No it's not normal, even for NTSC.
I'm glad you're asking these questions, and not simply accepting it, because it shouldn't be that way.
Hope I can help to figure out the anomally. So far I'm stumped...
Can you try turning on PGCDemux Logfile, and also "Include end time" in CellTimes.txt, then post both files here or somewhere.
PgcDemux didn't display the frame count directly, I had to convert the time code it reported for PGC_1 into frames.Darth Mallwalker said:
This statement befuddles me because I've never seen frame counts listed whenever I run PGCDemux (except in the Logfile, but you had that turned off.)
Where in PGCDemux are you seeing the frame count?
Can you make a snapshot of PGCDemux, where "174262" is displayed on-screen? That might help to give me a sense of context.
Finally if I can be so bold to list what I believe are the correct answers for STAR WARS:
PGC_1 contains 174261 frames.
Both AviSynth and VirtualDub call the first frame 0, so the last frame will be called 174260.
(Other programs might start with 1, as Chewtobacca already mentioned.)
Skipping PGCDemux, and DGIndexing the VOB files directly ought to yield 174276 frames. That's 15 from PGC_2 + PGC_1's 174261.
Again VirtualDub and AviSynth both count from zero, so
last frame of PGC_2 is called 14
first frame of PGC_1 is called 15
last frame of PGC_1 is called 174275
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
Ahhh thanks, up until now I've neglected the 'force film' factor.
If you choose 'force film' then PGC_2's 15 frames will be reduced to 12. That's the way 'force film' works--dropping frames as necessary to achieve strict adherence to 23.976 rate.
So I would have expected 12+174261=174273
And you say it's 174270 ?
Now it gets really weird
However, in practice you must take into account the “fuckwit factor”. Just talk to Darth Mallwalker…
-Moth3r
STAR_WARS LogFile said:
[General]
Total Number of PGCs in Titles=2
Total Number of PGCs in Menus=27
Total Number of VobIDs in Titles=6
Total Number of VobIDs in Menus=22
Total Number of Cells in Titles=52
Total Number of Cells in Menus=22
Demuxing Mode=by PGC
Demuxing Domain=Titles
Total Number of Frames=217822
Selected PGC=1
Number of Cells in Selected PGC=51
Selected VOBID=None
Number of Cells in Selected VOB=None
[Demux]
Number of Video Packs=3125913
Number of Audio Packs=259584
Number of Subs Packs=3290
Number of Nav Packs=14197
Number of Pad Packs=0
Number of Unkn Packs=0
[Audio Streams]
Audio_1=0x80
Audio_2=0x81
Audio_3=0x82
Audio_4=None
Audio_5=None
Audio_6=None
Audio_7=None
Audio_8=None
[Audio Delays]
Audio_1=0
Audio_2=0
Audio_3=0
[Subs Streams]
Subs_01=0x20
Subs_02=0x21
Subs_03=0x22
Subs_04=0x23
Subs_05=None
Subs_06=None
Subs_07=None
Subs_08=None
Subs_09=None
Subs_10=None
Subs_11=None
Subs_12=None
Subs_13=None
Subs_14=None
Subs_15=None
Subs_16=None
Subs_17=None
Subs_18=None
Subs_19=None
Subs_20=None
Subs_21=None
Subs_22=None
Subs_23=None
Subs_24=None
Subs_25=None
Subs_26=None
Subs_27=None
Subs_28=None
Subs_29=None
Subs_30=None
Subs_31=None
Subs_32=None
STAR_WARS CellTimes.txt said:
635
3435
8791
13451
16040
20430
27680
34760
41756
45320
49305
51916
57955
62115
65745
69686
74350
75835
79111
84535
88290
90565
92925
93160
98025
102010
109141
114705
118466
125970
128976
134405
137870
142065
148385
152036
157255
164206
169355
171956
174665
177881
180066
183647
192046
202117
206566
207811
210976
217807
217822
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
I get 174270 after forced film. This is what Avisynth says if I add Info() to the script, and HC Enc says the same when I load the d2v into it.
DGIndex using the VOBs directly and with Forced Film checked gives a total frame count of 174276 for Star Wars. This is in the information window of DGIndex.
If I then open the resulting d2v file in AviSynth/VirtualDub I get 174270 frames total.
Thanks for the help also Chewtobacca!
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
No problem! Now getting working on ESB. :-D
So can I just confirm that 174270 is the correct frame-count I should be seeing in AviSynth/VirtualDub when using the VOBs direct with Forced Film selected in DGIndex for Star Wars?
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
That's right yes.
While we're on the topic, TFM().TDecimate() interestingly gives a frame count of 174269, so it's obviously handling the hard telecined material differently to forced film.
Lovely thanks a lot! :D
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
By the way, all the non-film material occurs in the first hundred frames before the 20th Century Fox logo appears.
Please forgive me but I dont understand exactly what this means. :( As I am re-authoring the GOUT DVDs using VobBlanker should I edit out this footage or leave it be? As always many thanks for the help! ;)Chewtobacca said:
By the way, all the non-film material occurs in the first hundred frames before the 20th Century Fox logo appears.
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
You're following g-force's script, so don't worry about it. It's irrelevant really. I shouldn't have posted it in your thread; it was just that we talking about frames and I was interested, so I checked it. It's trivial, and I was thinking out loud, so to speak. Just ignore me. :-)
Ah right, ok no worries ;)
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
Sorry if you're already on top of this, but since I used Vobblanker myself recently to do the same thing (only with DJ's v3 video) I might be able to save you a couple of further headaches when it comes to recombining!
The original GOUT video has 53-54 cells - that's 50 'reported' chapter headers (i.e. the ones you see in the menu), one unreported (which amounts to the last couple of seconds of the end credits), and a couple of split chapters (sometimes to assist with the chapter break, sometimes for no good reason!)
Since VB replaces video by matching cells, you need to make sure your own video has the same number - this doesn't appear to be as simple as remuxing the original celltimes with your video, however, since the cells should start and end at I frames in the MPEG stream (which will be in different places in your video).
Soooo... you really need to use a simple authoring program to mark your own 51 chapter points in your version (matching those of the GOUT to the closest available frame), export that to a VIDEO_TS folder, then open it in VB to mark the additional cells using the split cells function (using the GOUT cells list as reference - VB gives the clearest view of this in my experience so perhaps open that up for a peek first).
Again, sorry if this is teaching you to suck eggs but I was scratching my head over this for a good while so I thought it might help!
174276 = 15 + 174246 + 15
That's the number of raw frames encoded on the disc (what you'd get using 'Ignore' option in DGIndex.)
Using 'force film' you get
174270 = 12 + 174246 + 12
So I'll respectfully disagree with Chew, there's non-film bits at both ends.
The first 15/12 comprise PGC_2 which is located at the head of the first VOB.
The last 15/12 are the last cell of PGC_1 (IfoEdit calls it Cell51) at the tail of VTS_03_7.VOB
For those interested in frame counting, be sure to try DGIndex's feature
"Tools --> Parse D2V" and then read the parse.txt file it generates.
The non-film bits at beginning & end can be spotted easily, even by a rookie like myself, because there's no asterisks.
However, in practice you must take into account the “fuckwit factor”. Just talk to Darth Mallwalker…
-Moth3r
You're right. I overlooked the possibility that there might be non-film material at the end when comparing the videos. I was looking at the videos; I should have checked the parsed D2V. That would explain why there is three frames of difference between ignore pulldown flags and the inverse telecine all the way through. Thanks!
*Pretends to fully understand what has just been said* So is 174270 still the correct frame count for Star Wars in AviSynth/VirtualDub?
Jonno thanks for the info ;) when you say a simple authoring program would muxman do the trick? *sigh* this project just seems to get more and more complicated all the time!
Original Trilogy in Replica Technicolor Project
Star Wars PAL LaserDisc Project
174270 is correct because you're using g-force's script, and he used forced film.
LeeThorogood said:
Jonno thanks for the info ;) when you say a simple authoring program would muxman do the trick? *sigh* this project just seems to get more and more complicated all the time!
Yes and no ;)
Yes because Muxman would certainly allow you to add your own list of celltimes, but no because you need to place those cell breaks in the context of your own edit/encode, requiring a visual timeline interface.
Ideally you'd use something like Adobe Encore to put this part together - it's got a very flexible timeline and (if you don't mind using its inbuilt MPEG encoder) can even encode around the chapter points you've placed on your AVI, so those can be positioned on precise frames (and the encoder will just force I-frames on those).
Simpler software will probably just let you mark chapters on existing I-frames in the MPEG stream - obviously you won't know what these are until you try navigating around the timeline (you'll only be able to 'park' the cursor on an I-frame) - and then build the video and audio streams into DVD-legal specs.
Perhaps have a browse through these programs for one that will fit the bill/budget (unless someone else can weight in and supplement my own limited knowledge!)
http://www.videohelp.com/tools/sections/authoring-dvd
Of course I could simply offer to do this part for you, if there's an easy way for us to exchange the materials. Sounds like you might need to sort out the issues with your video streams first, but let me know :)
Chewtobacca said:
174270 is correct because you're using g-force's script, and he used forced film.
Sounds correct from a certain point of view ;)LeeThorogood said:
VirtualDub via AviSynth read the d2v file DGIndex created as 174258 frames
However, in practice you must take into account the “fuckwit factor”. Just talk to Darth Mallwalker…
-Moth3r
Darth Mallwalker said:
Yes 174270 is what you get using Forced Film on the VOB files directly, without using PgcDemux. However if I'm reading g's script correctly it isn't meant to include PGC_2 first, only PGC_1.
So would you use PGCDemux? I assumed from this line that g-force was loading the VOBs into DGIndex:
Mpeg2Source("C:\Documents and Settings\Owner\Desktop\AviSynth Scripts\VTS_03_1.d2v") # <-----Set path
VTS_03_1.d2v is the default name that DGindex gives when you load in the VOBs and make a .d2v. When I do it this way, I get 174270 with forced film. I'm confused now lol.