Adolfo Rozenfeld
Forum Replies Created
-
Some time ago I had the opportunity to use the “Deband” effect, included in one of the Tinderbox plug-in sets. Pretty amazing. Especially designed to remove banding without adding so much noise. I don’t remember which specific Tinderbox set (there are a couple) it was in, since I don’t own it myself.
Of course, a project with 16/32 bit processing, and rendered to a 10 or 16 bit codec should exhibit no banding at all, with no need for a compromise solution.
Hope this helps.Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
Greg: During the last few years, it became quite clear that Divx has little technological relevance outside of its’ specific user base (movie bootlegs). It’s a propietary derivative of MPEG-4 technology, which isn’t supported by any of the big multimedia architectures (Windows Media, Quicktime and Flash).
The H-264 codec (also known as Advance Video Codec, also known as MPEG-4 Advanced Profile) is so much better and it can be enjoyed by anyone who has Quicktime 7.0 or later installed.
1024×768 or 1024×576 are completely outisde of the PAL standard sizes (or any other, for that matter). For the web, you’ll be fine (some sort of non compliant, semi HD size). But you also wanted it for broadcast TV, right?
Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
Hello, Greg (and Adam, of course).
Here’s the thing: The best way to produce a delivery (not a production) version for web distribution, etc is to first make a production quality master and then derive the different distribution versions from that. This is because delivery codecs are interframe/temporal technologies that produce much better results when they are able to “study” (2 pass compression) an already finished source (ideally with no or little compression). Good applications to do this are Apple Compressor, Sorenson Squeeze, etc.From Cinema 4D you could export a 720×480 file with 16:9 pixel aspect, or a 864×480 file with square pixels (and make FCP force it to fit). If your main concern is being practical, you should use the Quicktime codec that matches your FCP sequence settings. This means, if your FCP sequence is DV, exporting a Quicktime DV file will play in real time in FCP.
If your major concern is quality, then using the Apple uncompressed codecs are the best option, given you have the proper video hardware to use or take advantage of them. Of course if that’s the case, those files will also play in real time inside FCP.TGA/TIFF sequences are a good intermediate step. Because they allow you to stop a long render and resume it later from the last rendered frame. After that, you could use Quicktime player to import the still image sequence and export a suitable Quicktime file.
Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
Steve is correct, of course.
However, the problem you exerienced is due to something else.
True, DV is far from being an ideal codec for intermediate use in graphic animaton. But it’s not THAT bad.Here’s the thing: About six or seven years ago, when the Quicktime DV codec appeared, showing those files on a computer screen in real time was not very easy for the average computer. Because of this, Quicktime files with DV codec have by default turned on an option that shows the file at half resolution. The reason for this is no longer there, but they didn’t change it anyway. That’s why your text looks so ugly 🙂 Quicktime files captured or exported using the DV codec are shown in a low resolution preview mode.
In versions prior to Quicktime 7, you had to go to movie properties and turn on the High quality flag for the video track. In QT 7, it’s so much easier, go to preferences and enable “use high quality when available” (or similar wording).
All the best,
AdolfoAdolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
[tastypiesinc] “This is the way I understand it… please correct me if I am wrong. The animation will be broadcast on TV, so eventually it needs to be 0.9 pixel aspect ratio, right? But since I am using Illustrator files, I need to work in square pixels. I should work in 720 x 540 square pixels then fit that comp into a 720 x 486 0.9 pixel comp for rendering. Is that right?”
That is completely correct. But there’s an Illustrator specific issue you need to be aware of. Unlike most other applications, Illustrator takes the work area mostly more as a reference than as a fixed target. Because of this, if you place elements outside the work area, that will force the resulting AE composition to be larger than your original size (to acomodate the elements outside the original AI area). The opposite is true if your elements don’t fill the work area: the comp size will be reduced to the used area.
You can solve this by creating a 720×540 tranparent rectangle and making sure it’s perfectly aligned with the work area. Then select the rectangle and create Crop marks (you know, those used for cropping printed work). AE will interpret the crop area as the Composition boundaries. It’s an AI > AE convention.
Hope this helps.Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
FWIW, the scripting environment in Cinema 4D (Xpresso) is completely node based. Such visual approach to scripting it’s probably the easiest and most intuitive implementation of expression based animation I’ve seen. Yet, there is much I haven’t seen 🙂
Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
FWIW, the scripting environment in Cinema 4D (Xpresso) is completely node based. Such visual approach to scripting it’s probably the easiest and most intuitive implementation of expression based animation I’ve seen. Yet, there is much I haven’t seen 🙂
Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
One thing that could be really good (from AE’s idosincracy, which is not Shake’s ) is if it could display layers in a nested composition as sub layers for the pre-comp layer in the timeline. It could really make things much quicker. Strange it doesn’t do that yet. You’re often opening whole timelines for pre-compositions that only have one or two layers inside.
Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
One thing that could be really good (from AE’s idosincracy, which is not Shake’s ) is if it could display layers in a nested composition as sub layers for the pre-comp layer in the timeline. It could really make things much quicker. Strange it doesn’t do that yet. You’re often opening whole timelines for pre-compositions that only have one or two layers inside.
Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com -
[Sandman] “Thanks so much for your response”
No problem.
[Sandman] “So basically a node based application is just being able to have control over the render order?”
It’s my understanding that this control of rendering order has much more to it. You could, for example, route effects independentlly of the footage elements they are linked to. So you could in theory crop a gaussian blur, instead of cropping a layer which had a gaussian blur applied before. Same thing for many other scenarios.
[Sandman] “What exactly do you mean when you refer to AE having a layer based interface? Are you referring to the layers in the timeline…or layers in terms of things being rendered in a specific order? If you’re referring to layers in the timeline, how do these node based applications interact? Wouldn’t they need some sort of “timeline” with multiple layers and things of that nature…or do they do things completely differently?”
Yes, I meant the layer based timeline. If you think about it, you already know applications that are based on timelines and still are not layer based. Editing applications, for instance, are track based, not layer based. Tracks exist independently of the clips you use in them. This is not related to node based apps, but shows that not all timelines are layer based.
My experience with that kind of interface tells me that timelines are not so so much about the visual stacking order of elements. They are more like AE 7’s graph editor, a place where you see keyframes displayed, etc. They usuallly have a separate Object window, where you see how objects are stacked. These applications are usually very context sensitive. You select certain object, and the timeline switches to animating properties for that object (again, like AE 7 graph editor). And you get spatial properties for that object in an Inspector window, etc. The one thing I would love to have in AE is what they call “Heads Up Display” (like war plane cockpits!), the ability to have parameters superimposed on the composition window for you to manipulate
Many of the things I am mentioning are more frequent in 3D applications, for example.
[Sandman] “And lastly, why do people refer to masks in these other programs as being node based and being able to feather specific points? What does feathering specific points have to do with nodes?”
You know what? Do those applications allow that control over masks because they are node based? Or because they are specialized compositing applications and then it makes sense for them to be node based 🙂
I don’t know if AE needs a node tree scheme to allow that. Probably it just needs an updated masking technology. I have no idea how or where they could display the feather values for each node, but hey, thank god I am just a user.
Adolfo Rozenfeld
Buenos Aires – Argentina
https://www.adolforozenfeld.com
adolfo@adolforozenfeld.com