David Battistella
Forum Replies Created
-
Thanks Bob. I had a lengthy discussion the other day with an AVID editor who insisted it had to do with the speed of the audio being different at 23.976 vs 23.98. the fraction of .004% apparently makes a difference when the pulldown is put back to 29.97.
David
-
Walter,
I agree with you. One of the fisrt tests I did with DVCPROHD was with moving boris titles around. I thought the exact same thing. The codecs perform beautifully. I assume that the codec was primarily designed as an aquisition codec so it is most likely optimized for moving (camera) images. There is a huge difference between Boris and the FCP titles. So hopefully that is all the problem is/was.
David
-
Rich,
The best way to get the most out of the codec is to put the most in. If you use a vector based titler (like boris) or photoshop files or Illustrator via AE for titles there will still be a loss whn you are going to the DVCPROHD codec. It was designed to make images look fantastic and sometime graphics suffer.
Since these shows are going to Beta SP why not consider rendering out a 10 bit uncompressed SD sequence and doing your titling on that? That way you get the benifit of clean titles. Any compressed title will suffer because its just a 72 dpi image.
The only other thing I would experiment with in the DCVPROHD codec would be to use some sharpening filters on the text or try a natress smoothing filter. EXPERIMENT a bit, see what happens.
David
-
Walter reported this but I a pretty sure the 1.1.2 driver for the Kona 2 fixed this.
David
-
Probably none.
David
-
I think that one way to deal with the TC issue is to actualy record a TC signal to one of the audio tracks of the HDV camera. This would then be the master TC. The TC laid down as the long GOP tc would not be used and the timecode from track two could be the master code. This would require a small TC generator and you would loose an available audio track for recording BUT you could then recapture the footage because the audio TC would be the master TC for the tape.
TC is an audible audio signal so recording it to a track doe smake sense, especially in a controlled studio type set up that you are gong for.
You could also split that same signal onto the tape or convert it to RS-422 to be read by deck control as a live time code source.
FCP is a bit behind shen it comes to live time code signals but this is one workaround that would allow you to shoot HDV and eliminate problem of not being ablr to recapture.
I remember seeing these types of setups in the early days of DV when SMPTE TC was not the norm .
David
-
Scott,
This is as far as I understand it. 59.94 is the refresh rate for the “fields per second” this means that at 29.97 there are 59.94 fields in that second. I do not think that the KONA2 captures individual flieds so the preset would be designed to capture a frame rate of 29.97 frames per second. 59.94 tells you that the footage was captured as an interlaced picture with 59.94 fields per second and a frame rate of 29.97 fps.
it makes sense that a 59.94 capture preset would not work as I do not think any deck out there actually splits fields.
David
-
I have heard that for 29.97 finished 24pN is the way to go, but it would be hard to do a filmout, etc in this mode. Graeme natress has developed an advanced pullown converter that helps with 24Pa footage for a 29.97 sequence.
for a 24frame DVD you should use the 24Pa mode.
Noah, thouhgts on this?
David
-
great thread guys,
I agree with Marco. They may make nice products but they are definitely on the price side. AJA has what I need. I have to say that I can not be but amazingly impressed with AJA and their level of service and how well they back their products. Every call I have made to them has been first rate information and serivce. It gives you a great comfirt level.
Who wouldn’t buy from them?
David
-
I think this might be a problem with Dynamic RT.
See if it still happens with the Dynamic RT set to High instead of Dynamic
David