Chris Smith
Forum Replies Created
-
Okay, I ended up dumping in all the material regular HDV (not AIC). If I set the output to BM HD 1080 output I get black. If I set it to BM NTSC 8 bit, I get a picture on my monitor. However, although it plays real time on the FCP Canvas, I only get bad looking frames on my output when the playhead is stopped. It looks similar to the single field detting but that setting is off.
I did what BM suggested and re-installed FCP 5 then the BM FCP 5 drivers last. No difference. Furthermore the SD I have always had in the machine (elements for my reel) won’t play out through the Monitor from the QT player. They did before. I went to check the ‘use QT legacy’ and it was already checked and greyed out. However, it will not play out the output.
Any advice?
G5 DP 2.0 / OSX 10.4.1 / QT 7.0.1 / BM Drivers 5.0 / FCP 5
Chris Smith
https://www.sugarfilmproduction.com -
Just drop your HD comp into an SD comp and scale the comp size using the normal scale parameter.
Chris Smith
https://www.sugarfilmproduction.com -
Wow, a wealth of great info. Yes, I’d love it just converted to BM Uncompressed SD. It was only shot HDV to get a better picture and DOF than DV, but from here on out I’d love it to be SD letterboxed.
I dump in tomorrow. I may have some more questions.
Thanks Again.
Chris Smith
https://www.sugarfilmproduction.com -
-
If that is the treatment: You are looking at a Paper stuck to the CRT tube then you transition to the image itself, then it’s not really a move. It’s kind of like one of those shots in The Matrix where you are aware of the TV’s scan lines and frame, then the shot transitions to the clean image.
In that case, I’d take a good digital still of the edges of a real TV to get the frame. Then mask out the inside. Then Shoot a digital still of say an open window from inside a living room. Now lay this over your video using screen transfer mode and drop the opacity so it feels like a reflection off of glass. Then look up in the Cow archives for suggestions for getting the TV scan lines look. One of my favorites is using the Venitian Blind transition in AE set to scan line thickness with feather on. Then put the paper/tape layer on top of all this. Then I guess just dissolve this whole mess into the clean image of the video footage.
If you still want to physically move the camera PAST the paper and towards your person, then you are looking at a 3D matchmove (if you shoot a real dolly move) but then logically having the paper taped to the screen doesn’t work. Or the thing where you push past 2 layers in AE. Too your resolution question, if you see the footage small then the camera moves close enough to make it fill the frame then you don’t lose resolution. If you want the footage to always fill the frame, then you will need to shoot a lock off of the person but smaller in frame than you would normally frame it. Then push into the shot a bit. You will lose resolution. However if your move isn’t too severe, the loss in resolution may not be noticeable. If you want to hold your resolution true than you will need to use a higher res format than your output format (like transfer or shoot HD to output SD).
Chris Smith
https://www.sugarfilmproduction.com -
If you shot the stuff as a lock off. Bring that layer into AE 3D space. Put the paper layer in 3D space as well. Move the paper forward in Z space.
Now you can’t zoom the camera, you need to move the camera. Huge difference. Actually animate the camera’s position towards the paper, past it then closer to the video layer.
What I am still unclear about: If the paper is attached to the TV, How can the camera go past it? Does it just fade away, or un attach itself?
Chris Smith
https://www.sugarfilmproduction.com -
Well I did a Straight Alpha pass and just rendered out normal (not as multipass) and turned everything off but the objects I wanted. In AE I had it ignore it’s own alpha channel. Then I used the object ID as a luma matte. This worked. However, I guess now my question is: Is this the normal/most efficient way to do this?
Whenever an Object ID matte is created, is the normal workflow to always render out the rgb images as straight alpha, then override this alpha in the compositor? Or are ppl living with the double anti-aliasing creating fringes?
Chris Smith
https://www.sugarfilmproduction.com -
Okay, I have some questions:
1. Was the footage already shot? Is it definitely doing a dolly push in?
2. Confused on the “Taped to the TV screen thing” if it’s taped to the screen how can we pass by it? I guess I’m confused as to what it’s supposed to be taped to. In your test it is “floating in space” which is a very dramatic effect I think, but not sure if this is what your intention is.If your real camera actually dollied in to the subject, then I think you would need to spend some time trying to match move it to look like it’s locked to the image. If the shot was sone as a lock off, I think it’s a cool effect to bring in the footage into AE 3D space along with your paper/tape offset in Z position and do a AE camera move past the paper to the video image. Then use the DOF blur in AE to rack between the layers when it is appropriate.
Chris Smith
https://www.sugarfilmproduction.com -
Well I was waiting for a response myself. But have you tried just turning off all objects, lights, and post effects? I would think with nothing to render you could blaze through it pretty quick.
Chris Smith
https://www.sugarfilmproduction.com -
Post a QT for folks to see so they can help you from here or offer alt solutions. Also for me, because I’m as slow minded as a house cactus, to understand completely what you speak of.
Chris Smith
https://www.sugarfilmproduction.com