Bill Russell
Forum Replies Created
-
Well, thanks. Oh how confusing. There is no danger of “doubling up” — adding 601 to 601? Funny — when importing Avid’s own bars tif using 601 they scope internally (Avids software scope) too low, just the opposite of what you get externally. Do you get that too? I don’t have SDI out or any external scopes to compare.
Sigh — I’m accustomed to FCP world where levels (on video, but graphics another story) are just plain correct in and out or import to export. Somehow not having the RGB / 601 /709 video options is not a disadvantage in FCP.
– B
-
Score. That’s working well. I created null sliders for all three variable, for end, start, and time. Several small steps forward in comprehension. Thanks.
“THE LOST SKELETON OF CADAVRA” –
-
Is there such a thing as “global” variables? Like, if I want to change the “endScale” value for all of the layers at once, instead of editing the expressions for each layer, is there place I can define and call and “endScale” variable globally? Thanks!
-
Hi Mike — multiplying the result by value[0]/100 seems to be the answer for getting _proportional_ zoom changes based on the intitial property value, whereas adding value to it seemed to result in different zoom speeds depending on the size of the property value. I think this is the reason: For instance, 23% + 50 is a lot more extreme than 125% + 50. That said, I HUGELY appreciate both the response and especially how to do the ease/ramps you describe!!!, that is going to really come in handly.
Enormously grateful – B
“THE LOST SKELETON OF CADAVRA” –
-
Dan, thank you, that’s perfect, it’s doing exactly what I need when I set startScale at 100. “Value”, now I understand what that means, a little edification. What is “value[0]” — or otherwise, is there a glossary or something that explains syntax and variables in a way a calcified brain like mine could understand?
Again, thank you.
“THE LOST SKELETON OF CADAVRA” –
-
Thank you, Yikes Mike! That’s a big help. One big issue though, is there a way to make it relative to the initial scale size I already have set for the layer?
(For example, before applying the expression I’ve got a particular layer set up to 23% scale, so that the 50-100% zoom dictated by the expression will be percentages of the initial property scale setting of 23%. The next layer is at 125%, so the 50-100% zoom will be a percentage of the 125% scale, etc. etc…)
-
You mean it takes 18 seconds to go from bottom to top of frame? If I’m understanding you correctly, that is a very slow crawl indeed. Anyway, seems the pixel for pixel advice is a wonderful thing!
Learning something new hourly here. I do my crawls with camera moves (since I often have graphics and animation within the roll… and don’t get buffer errors…) so I’ll have to pick brains here to see if there’s any way to achieve pixel perfection in 3D space! Isn’t CC wonderful?
“THE LOST SKELETON OF CADAVRA” –
-
There’s a little motion blur check box goes with each layer, and a little motion blur check box at the top of the composition window. I could write paragraphs trying to describe where they are. Each is just a little symbol – try hovering over the column header symbols, after a few seconds a small text pop-up will tell you what each switch is. Best thing to do is look up “motion blur” in AE help — it will show you a picture of what the motion blur switches look like.
What you want to do is enable motion blur for the composition, which means that every layer that has its motion blur enabled will render a blur between frames when there is movement. This may help your jitters.
It is also possible the text is too sharp or moving too fast. You might want to try putting an imperceptible Gaussian blur on the text layers just to see if that helps the problem. Also, don’t do text brights at 100% — do ’em at 90-95%.
If your text is a bright color, this will cause a lot of jittering too, especially if you are working in DV. Color in DV video is “blobbed” on top of grayscale and is only one-forth the resolution (which you will experience only AFTER you render out to DV), so a bright red letter will be MUCH softer and blockier than a white or grey or black letter, and maybe unreadable. In most other video formats, color is compressed this way to different degrees depending on format.
Also, in DV, in bright colors you may notice striations. So if strong color is your problem, try reigning in the saturation of your colored text quite a bit so that there is enough grayscale “showing through” to reveal sharp detail.
Thoughts.
“THE LOST SKELETON OF CADAVRA” –
-
Thank you, that is a great answer! Hugely appreciated. I’ll get to work on that.
I’ll just say for starters (before I got your response here) that simply turning off multi-processor then jacking back up the ram cache to 60% cut the render time in half, bringing it to the same performance as AE6.5. Rosetta is off for CS3 (I just checked, thank you) so seems it should do better than 6.5 in this mode, but oh well. Anyway, off to do your benchmark testing!
“THE LOST SKELETON OF CADAVRA” –
-
Well, if you want to be totally conservative, simply go with, say, Arial, Arial Black or Arial Narrow. It actually looks quite fine in my experience, and you can still work up plenty of exciting designs around simple fonts.
If your work is going to be final output and not color corrected by anybody, then you should observe a few rules. Set your working color space to HDTV (709, found in “Project Settings”, which is also a good profile for projected film). Don’t let your white letters be totally white!! They can do weird things (like vibrate, and on TVs can even cause an audible buzz). Use an almost white, around 90% (or 95% if you want really white) white. Same with any color, don’t let it’s brightness reach a full 100% — bring it down to between 90-95%.
If white is to be seen over a large area, like a white background instead of a black background, or a large dominent shape is white, then bring it all the way down to around 75%. Yes, indeed, the big “white” bar on NTSC color bars is actually only 75%! 75% looks slightly gray on a computer screen, but quite pleasing (not blaring) on a video or film screen, when it is the dominant color.
Back to styles. If you want fancier font styles — especially fonts that have custom “textures” or “roughness”, I find these fancy fonts can fool you. They look best in print or on a computer screen, but really jaggy and messy when projected or on a video screen. So go for the font style you want then “pull back” by choosing the more conservative of options — ie, fewer curlies, smoother details, etc. A little goes a long way with fonts. You can get away with more in HD. But if you are in standard def, then best to go very conservative.
Interlacing as well as the just low resolution of SD demands sufficient smoothness. For that matter, in both HD and SD, people think sharpness is the key, but really its smoothness. Even softening things very slightly can do wonders for a good steady clear result, and for getting rid of artifacts such as crawl or moire.
So there ya go. Hope there are some useful ideas in there.
– B
(P.S. Don’t forget title safe area! AE has a safe area guide box — all video applications do — the outer box is picture safe, which is the total area that will be visible on most televisions, SD or HD; the inner box is title safe, you don’t want to go beyond this. In fact, broadcast quality checks will reject video masters if text goes outside of title safe. Your layout may look great on a computer monitor, but then you’ll go to screen for an audience and shed tears when all your text is cut off at the sides and bottom. Some DLP projectors can be set to show every pixel, but is rare and not to be counted on. It’s a tricky world now — you gotta make the whole picture “safe”, but still be “within safe”. Rather Zen, huh.)
“THE LOST SKELETON OF CADAVRA” –