ProTools Speed correction - Have you the same problem?

[Pedro_vde]

When converting the soundtrack from a feature from 24fps to the 25fps PAL version for Broadcast I've encountered a strange phenomenon in ProTools.

When using the 'old digirack plug-in' Time Compression/Expansion the speed-up works fine, sounds fair and the most important, it's sync.

When using the recent 'TimeShift' the sound is crap and the file is not sync, although I'm using the same values.

Today I had for example a file with the exact lenght of 00:21:35:23
Speeded it up to 25fps with TCE gave me a lenght of 00:20:44:02
Speeded it up to 25fps with TS gave me a lenght of 00:20:46:01

besides the inaccuracy of the timing the TS sounds like crap with lot's of transient artefact's. Although I used the same settings for the Transient Fields as Digi does in its Post Production presets for Time Shift

I've also experimented with Speed from SoundToys. This one gave me the same length as TS (00:20:46:01) but the quality of the audio signal is terrible... it sounds like a verry bad compressed MP3.

When I use a TC-calculator the conversion of a 24fps to 25fps of this length gives me 00:20:44:03 as a result

I'm getting a bit confused...

Are there any alternatives ? What do you think fellow-slutz? Is it a mistake or am I not the only one who want to take out the pro in the naming of this thing that drives me nuts? How are you guys doing it?

Kind regards from Belgium

Pedro

[minister]

the original Digi TCE is actually pretty decent. Timeshift can work...just make sure your settings are right.

Speed is worthless.

X-Form is really the way to go as far as Digi products

http://www.digidesign.com/index.cfm?...=1&itemid=4954

Waves Sound Shifter is also pretty darn good.

Me? i use X-form, but usually the Sample Rate Convert speed trick. not artifacty at all. then pitch if i need to.

[Garret]

I would do as Tom says and use the sample rate conversion trick and then pitch if necessary. Pitch n time is really good too with few artifactes
Garret

[Kuba_Pietrzak]

Speed change 24<->25 fps is a complex problem and is not easy to solve so far (especially with factory plug-ins).
There are several methods to do it.
If there is no need to compensate pitch change (sometimes there is really no need), you can do the sample rate convertion trick and you are done.
If you need to compensate the pitch you may:
- lay the whole mix onto the dubber which has a fixed vari-speed playback capability, than record the whole mix back into ProTools with + or - 4% speed change THROUGH TC M6000 turned into harmoniser mode (and probably program the amount of pitch change in this device) - this way you can receive your mix with the different speed but the same pitch; be carefull about low end
another method:
- process the whole mix with Serato Pitch'n'Time (which always has accurate timing) and use it as your guide; now you have to slow down or speed up every line of dialog with e.g. TCEX (if you have dry dialog), reedit all BGs and Foleys and perhaps reedit and find the best solution to change the speed of music (perhaps sample rate conversion if the composer or any other stuff member does not have an absolute pitch) and MIX THE WHOLE THING AGAIN - this method is the most complicated but this way you receive best results;
I would not treat Dialogs or Foleys with Serato; I would not treat music or BGs with TCEX. I am also trying to figure out what is TimeShift best for, but I am affraid, it was not design for Post work
Hope this helps
All best
Kuba

[Ruudman]

I really miss Prosoniq TimeFactory (now only in Classic for Mac) which used their splendid MPEX algo....

Agreed, Speed is no good at larger tasks



ruudman

[minister]

kuba made a good point that i forgot to make...

sometimes it is best to pitch only the things that really need pitching and leave the rest alone -- assuming you made the speed change. also, one tool may be better at dialog and another better at music. so, if you can do it on stems, great! if not, well...you do what you have to!

[Pedro_vde]

Indeed, Kuba made me think a bit further... and indeed the TCE works OK for Dialogue, but not for music... To bad. We don't have the Pitch'n'Time but we do have an old DD1500 setup which is now transferring in +4% through a Pitch correction of 4% in the System 6000.
Because the master has to be on Digibeta, I don't bother on the 24 to 16 bit truncating.

I'm wondering what the ideal setup would be to do a transfer of the discrete 5.1 from the feature to a continue version in 25-fps for DVD... in 24-bits.

(a DD1500 can't handle the 24-bit, I know a DD8 can)

Any suggestions?

Kind regards from Belgium

Pedro

P.S. Thanks for the replies.

[kptkarl]

http://www.hairersoft.com/AmadeusPro/AmadeusPro.html

With Amadeus Pro one can change the characteristic(qualitly) of a .wav file using these magic numbers
NTSC to PAL: 50050.02481 kHz
PAL to NTSC: 46033.94321 kHz
24fps to PAL: 50000.00000 kHz
30fps to NTSC: 47952.04187 kHz
NTSC to 30fps: 48048.00609 kHz

first process the file using the above number(s) with "don't resample" checked. then process again back to 48khz with "don't resample" unchecked. import into pro tools and you're done.

amadeus pro also has a batch process feature that comes in handy for this.