[Tim Parsons]

[Alex Udell] “use Premiere’s Project Manager to make a copy of that project to a new location.”

Ah, see that’s interesting! Thanks for the tip. I just might use that!

[Alex Udell] “This doesn’t handle the file naming though.”

That’s a shame. :/ So there’s no way to truly rename an MXF? None whatsoever?

[Tim Parsons]

I really appreciate your replies!

@Vince: I understand that P2 has a complex structure of necessity and that FCP transcodes into a brand new file, but as my original question states, I want the workflow without the transcoding. Read further to see more of what I mean by this.

@Alex: At the moment, I cant see the video you linked due to workplace network restrictions. And sorry for the confusion; I am on a PC now. Our Macs were pre-Intel-ancient, so we decided the most cost-effective upgrade was to go PC.

That being said, I think I understand from your step by step outline that Raylight creates a sort of proxy for each complete MXF I want to use, and I would then edit those proxies (or I suppose “shortcut” would be more accurate) in Premiere or whatever editing application and it would simply reference the original MXFs right?

Well, to me that seems like it would be ok, but I think there should be a better way.

DISCLAIMER: the following is what I think today’s technology should be able to do, based on… well, not much. So I could be wrong. I just think this should be possible.

If you were to buy any cheap consumer point-and-shoot camera, it comes with its own software that allows you to view the contents of the camera and copy the contents to your hard drive. Some of these cheap softwares even let you organize and rename the pictures.

The cheap camera software does nothing but selectively rename and selectively copy. No conversion or anything.

Similarly, I want to be able to have an application or Premiere feature that will allow me to open my P2 CONTENTS folder, view the clips on the card (not just thumbnails, but previews I can scrub through), rename the clips (which would then rename the corresponding video, audio, and metadata files to match), choose which clips I want to keep, and select a destination folder on my hard drive. So far, FCP 7 can do all this.

Here’s where my idea kicks in: instead of transcoding… simply copy. Just copy. No transcoding, only copying from one folder to another. My theoretical application/feature would copy the CONTENTS folder to your destination, bringing with it only those files which you specified to keep, and renaming the clip to your desired name on all levels of metadata.

Boom. Done. And now we have the proper P2 folder structure, full quality copies of the original MXFs, and the only things different are the names, and content of the folders (imagine that, we only have shots we want to keep!).

Call me crazy, but that sounds remarkably simple. But before you go telling me that I should develop it, let me tell you that I don’t know programming whatsoever. But I have seen similar technologies at work. I don’t think it would be too hard to combine them into one simple, streamlining tool.

Am I wrong?

[Tim Parsons]

For real?! Are you telling me there’s no way to rename P2 MXFs in Premiere without going in and renaming all the individual audio/video source files??? That’s ridiculous! I think the whole Media Browser metadata thing is cool and all, but why the heck is there no way to bake the metadata in?

For example, in FCP 7, I can rename a clip in the browser and choose to rename the original source file. Heck, I can even delete the clip form my project and choose to delete it from the hard drive!

Also in FCP 7, There’s a log and transfer window that allows me to selectively rename and selectively capture only the shots I want. Granted, FCP transcodes, but the workflow is impeccable!!

In premiere, it ought not be too much of a stretch to ask for a utility that will allow the user to preview their P2 card, selectively name, and selectively copy to the hard drive. I’m not saying transcode, I’m not saying destroy the folder structure, but let me select what I want to copy and what to name the files.

Is there any way to do anything remotely close to that? Even if it’s a 3rd party software/plugin?

[Tim Parsons]

My question is not one of “macro” workflow, but “micro workflow. Not projection systems, but of AE technique.

I know how a dome projection affects an image. I know what my image needs to look like. See my OP. That’s how it affects it and that’s what it needs to look like, only moving.

I just don’t know how to get my image to look like it!

So my question remains: How can I set up a camera or group of cameras+stitching to make a 3D environment flatten to a 2D video, so that when projected on the dome it will look 3D again (not stereoscopic, but rather “immersive”)??????

I’m looking for an answer like, “In After Effects, click this, then change this, then rotate this, then run this process, etc.”

[Tim Parsons]

Quite possibly… But I have two things working against me: 1) budget. We have AE, but can’t afford a bunch else. 2) Workflow. I know AE. I can make it look good in AE (see pics above). I don’t know 3D software, and I don’t know if I could get it to look as good in Blender, which is what I would end up using, since it’s free.

But again, we run into the HOW. The 3D vs the 2.5D isn’t the issue, it’s how to set up a virtual camera, or set of virtual cameras to do the job. Even if I knew 3D software, I would be asking the same question: How can I set up a camera or group of cameras+stitching to make dome projection motion graphics work?

I’m pretty sure AE can do what I want, but I can’t figure it out.

[Tim Parsons]

@Roland, Actually, I’m trying to figure out which way is best. I think creating the fly-through would be easier, but I just don’t know. I’m not familiar enough with AE’s capabilities in dome projection, and I don’t know a workflow that will work. That’s what I’m looking for: how-to with a workflow. Because I don’t know how. That’s why I’m asking the COW.

Fact is, I have seen plenty of their planetarium projects, I have a basic grid idea (hence the grid in my OP), have studied what dome projections look like, and even help out with the occasional After Effects troubleshoot, but most of their graphics are individual elements placed around the screen. Not immersive digital 3D worlds created solely in AE. So this one is new to all of us who work on the planetarium shows.

I tried creating a camera with a 180 degree lens so that I could just fly a single camera through a single digital world. As expected, the 180 degree camera lens didn’t work as planned.

The clouds are all 2D images of clouds, placed randomly throughout 3D space. Here is what they look like with a standard 35mm camera:

Then I tilt that same camera 90 degrees upward and change it to these settings:

The result is this:

The lens horribly distorts anything that is nearby that goes above its field of vision. If i try to move the camera upwards, I can’t see the clouds anymore.

Am I doing something wrong? Any other ideas?

[Tim Parsons]

Well, they have batch software for taking the image and slicing it up and distributing to their projectors. That’s not my issue. I need to know how to create dome-looking animations. Not 2D video that I want to wrap around a domed surface, but actually creating a domed video in virtual 3D space.

Hence my ideas about creating various angles of the same 3D fly-through and stitching them together to look like a domed video.

I already know their desired format, fps, ration, and resolution. I simply need to know how to create animations that can be properly projected onto a dome.

[Tim Parsons]

“Some” aliasing, i could live with, but this is really bad. Can you give me and the OP some recommended settings for minimal aliasing?

[Tim Parsons]

I, too am having the same problem as the OP, but my HD footage is progressive, not upper field.

We are experimenting with a new HD workflow at work. Previously, everything has been 4:3 SD. Now we are testing HD shooting, editing, and delivery. We are experimenting using Adobe CS5.

We still need to distribute some on DVD, however.

Our problem arises when we have pristine HD footage (both 720p and 1080p) and burn it to a DVD… Aliasing. Like crazy.

We have tried nesting our HD timeline inside an SD timeline. Aliasing. We applied antialiasing filter. First, there are no controls for the filter, and second, it makes the video blurry.

We tried direct Dynamic Link of the HD timeline to Encore and burning straight to DVD, letting Encore down-convert, but it aliases on the DVD.

We tried to export to an SD m2v then burning in Encore. The export looks great, but the DVD looks like junk. Aliasing.

I believe that it could be one of two things… or both.

The first theory is, it’s taking a native 16:9 signal and cramming it into an NTSC SD widescreen (which is not true widescreen, it’s anamorphic), so when playback from the DVD stretches it back out to widescreen, it loses a ton of information.

The second theory is, it’s taking a progressive signal and interlacing it, taking half the lines and “staggering” them to be out of sync with the other half of the lines, making it looked aliased.

We have tried exporting with progressive, upper field dominance, and lower field dominance. Nothing works in getting a clean video down-converting to DVD.

Please help! We need to have a working proposal for the new HD workflow by the 16th. If we don’t it means another year of 4:3 SD!!

[Tim Parsons]

Actually, the PNG compression setting is not the format setting. It ends up a fully self-contained “standard” MOV. It’s not a “PNG sequence.”

The purpose is to be able to test the viability of a new HD workflow at my workplace. We have one computer capable of handling HD (hardware and software) and we are rendering some test videos to see if we want to make the full jump.

The full uncompressed setting is designed to show just what the computer can do. We may likely end up exporting at Photo-JPEG once we get a workflow down, but for now, we want the quality maxxed out to show the execs just how efficient our proposal will be.

Look at the specs. It’s a beast. But for some odd reason, we are having problems. And a beast that lags like that isn’t convincing anyone, at the moment. We’d like it to run full-capacity for when we pitch the idea.