robcat2075

OK, I didn't see the first post was in May. I'm still on the fence. I'm think I'm going to image my Windows 7 first and then try 10 so i can revert back easy if too much stuff doesn't work. I don't want to lose After Effects.

I meant, why did it work today and not yesterday?

So i wonder what changed.

I will probably just have people message the images to me and I will set up the exhibition page.

My suggestion would be to render with SSAO, then turn out all the lights and render the glow as a separate pass and composite the two with either "lighten" or "add". I dimly recall a similar problem with glow in my "painting with light" series and I found that compositing the glow an easy fix.

For any of you Android phone users. from NPR... Major Flaw In Android Phones Would Let Hackers In With Just A Text

They did a lousy job at that but you can't curve the "2001" image onto a Cinerama screen because the camera did not shoot a wide angle of view comparable to what a Cinerama screen presents to a viewer. The woman in pink and the red couch on the edges of the 2001 frame are probably only 20-30° from the center axis of the camera. Everything in that frame is more in front of the camera than to the sides. Nothing in that frame can be properly placed on the side area of a curved screen that would wrap around you On the other hand, the left and right edges of a Cinerama image are 70° off the center axis. To actually sit in a Cinerama movie is not seeing something that truly looks like the smilebox. What you see is a wall that curves around you. The middle of the screen does not appear shorter than the sides because the middle is just as close to you as the sides.

What I'm doing is un-distorting a wide shot. Would you agree that my A:M camera is at least taking a truly wide field of view similar to Cinerama? CinemaScope could never take in that wide an angle of view. They didn't have the lenses for it and... a truly wide angle image needs to be shown on a screen curved to the same angle to not look completely weird. CinemaScope had only a slightly curved screen. And would you agree that my Posed version at least eliminates the size distortion of the single lens image? It's not complete yet, but those are the first two steps to get to a Cinerama-like image. Here is the problem with three-camera Cinerama displayed in a flat format. Notice how the shoreline and the wavelines bend at every seam. That distortion is a problem for any straight line that is not exactly in the center like the horizon line. The wrap around Cinerama screen is an attempt to re-wrap the image around the viewer to recreate how it was wrapped around the camera. It didn't work completely, the distortion was still obvious on a Cinerama screen but somewhat less so because each third of the image was turned to face the centerpoint where you sat. Also, Cinerama movies were planned to avoid horizontal lines that would cross the seam. There is very little linear horizontal element to that Debbie Reynold sample so distortion is not obvious. Smilebox hides the bends a bit better because they are curving the interior of each third.

Here is a bit more toying with the notion of using a pose. the top is the original A:M camera render, mapped onto a rectangular grid. I think this distortion is called pincushion distortion. The bottom is the same grid smooshed in a pose to negate the pincushion effect to make objects appear about the same onscreen size no matter where they are in the field of view.

Here's a quick test of a very wide angle camera in A:M. I used a lathed circle to place the vases all equidistant from a center point where I put the camera. The distortion of the vases near the edge is obvious but also note that the radiating lines of the lathed circle appear parallel in the camera view (top). It should be possible to remap that camera render onto something that could be stretched and squeezed and curved for a smile-box view.

To do it in A:M... I would note that the three-camera approach was compromise based on off-the-shelf lenses they had at the time. OmniMax is more modern attempt at the same effect. Today an OmniMax camera (not Imax) can cover the same field of view in one image and the OmniMax projector on the OmniMax screen (basically a dome) can approximately undo the distortion that result from putting that on one flat piece of film. That distortion is what we see when we set an A:M camera to a very wide angle. Idea 1) It should be possible to take that flat A:M render and re-map it onto a suitable rectangle then, with a pose, stretch and squeeze it in the right places to undo the distortion and create the curvature. Idea 2) you can do three cameras just like cinerama and remap them onto a curved screen. Because A:M cameras can occupy the same point simultaneously there should be no mismatch between adjoining edges although there will be linear distortion at each seam. Idea 3) Instead of three cameras.... many cameras, each set to take a narrow vertical slice of the panorama. Those could be mapped back onto a panorama screen. By using many narrower cameras you would reduce the curvilinear mismatch between neighboring slices.

holly, if you have further questions, let us know. I'll note that Hypercam 2 is a free screen movie software.

I have actually seen Cinerama on a true Cinerama screen. "This is Cinerama" was re-released in the early 70s and my parents and I went to see it. It is a remarkable effect IF you are sitting in the sweet spot approximately where the curvature of the screen would be centered, which we made sure to do. If you were far back in the theater you probably would see something more like the appearance of the "smile box". The seam between the three filmstrips was always pretty obvious, especially when some object with straight lines intersected it, but for a lot of material it didn't matter.

A feature I quite like about A:M, so much so that I feel inconvenienced when something HAS to be done with a menu and can't be had with a click or a double click.

I'll investigate. The short answer is 'yes' but the demonstrable answer is 'hmmm... now how did I do that again..." Edit: Actually... it's easy to demo. If you accidentally select the roll handle of the Null then it will rotate. This doesn't appear to happen upon first click but with a second click. I'll see if I can prevent it. In a model... In the properties of a null I can set Manipulator Options>Manipulation to "Translate Only" and the roll handle will not appear no matter how many times I click How are you creating your translate=-only Null?I assume the Null is part of a Model because, unless I'm missing something, Nulls cannot be limited in such a way outside of a Model. True, a Null by itself in a Chor doesn't have the same Manipulator Limits properties. For a Null by itself in a Chor I can put a Spherical Limits constraint on it with Mins and Maxs all set to 0° and the Null will not be rotatable. (Rotation Keyframes will be created if you try to rotate it but no rotation motion will happen)

Texas Lottery tickets, Caricatures by Largent, and distinguished award medals... Don't miss out on your chance to win! You still have three weekends to get your entry in tip, top, super-duper insect image shape!

In a quick test, i can't make a translate-only null go into rotation just by grabbing it. Is this something you can repeat?

The current tactic of constraining bone controllers to limit their movement should be enough for that. What specifically do you need to do?

Yeah, normally you can't have a single CP.

A work-around might be to -hide all the CPs you want to hide -select the CPs that remain visible -unhide all CPs -invert the selection (period key) -name the inverted selection Group -save the model then when you later reload the model and need the CPs hidden -select the Group -period key -H

Try increasing the shadow "darkness" of the light coming from the right.

You can have one kleig pointing down at the ground and another pointing up just to illuminate the light shade. That way you can adjust each one's brightness separately for desired effect. Sorry, Paul! I meant to reply to your post but edited it instead... If you can remember what you wrote and want to repost it that would be great.

I'll add that in 99.9% of animation circumstances you want the "A" Button to be ON (red). When it is OFF, new keyframes won't be made and any changes you make to the scene will change previous keyframes.

BTW, you may find my "Making Poses" video useful for its coverage of "drag and drop" poses which may suit your hand posing needs better than slider poses.

It's a good idea, and it has been suggested before. I've discussed it with Steffen before and basically he said that the way A:M uses the operating system to create those pose windows makes it difficult to add images for them. For now, descriptive titles like "loose pointing finger" will have to do.