And another from the DVD anaglypherized with the ALS targeted depth plus. Things that seem to help for landscapes and in general. Avoid protruding objects close to the frame. Make the lighting a bit more even than you would with a normal image. Try to avoid strong blue and red colours being used in the scene. There are other considerations to do with image separation and the placement of the focus point - hence the semi-automation of the entire process with filters, lenses and mirrors. I trust that you can see it works.
Nesting TLA’s? (Three Letter Acronyms) Is that conventional or should it not be ALSAA? Either way, it is good news, since the ALS, although less user friendly than the targeted version, offers less image distortion and so works with wider FOV’s or AOV’s. Although neither works well with DOF, both are fine with TA! Probably, given the nature of the effect, DOF wouldn’t be a bad idea anyway, and liable to cause instant headaches.
Depth perception experiment with four kites (ever since Graham Dresch turned up with his home made kites I’ve been thinking about kites). So anyway, on my monitor at the viewing distance I sit at, the foreground kite measures the following in each scene respectively. ALST+ Now all I need to do is figure out how to animate this…
image 1 - 2 1/4 inches
image 2 - 2 3/8 inches
image 3 - 2 1/2 inches
image 4 - 2 7/8 inches
iamge 5 - 4 1/2 inches (final calculation performed by bringing a kite spar forwards to measure directly on the monitor surface)
Interesting series of experiments. Best strategy when viewing is to find the object which is easiest to get 3D, then also the more difficult align easily. For the last example, the kite farthest away is a bit difficult to bring in line. However, the nearest one is easy. Once you get it right, the rest of the picture follows without pain.
I also been experimenting along these lines using the ALS. The first one uses 2.1° “toe-in”, is easy on the eyes but the ivy ia a bit flat in depth. the second one has 10°, look much more natural as far as depth is concerned, but is more difficult to behold - particularly towards the left and right edge. There are ghosts from the red. Also, the specular mapped HDRI in the background adds some distutbance.
I see what you mean. The first ivy example is easier to look at, but perhaps not realistic as far as the perceived depth is concerned. Like the final example of the kites, finally the terrain is beginning to look large and the sky distant - unlike the first, where the landscape and the sky barely reach beyond the screen. There are many choices for the artist to make with 3D viewing - which means… it is a very good thing.
Haven’t been commenting but I’ve been watching.
Very nice, and I like that Peasant Brinnen stated “So anyway, on my monitor at the viewing distance I sit at ...”.
A few years back I posted a Vue render on here where I’d added some post-work that was supposed to make it pop at a certain distance from the screen. I never did continue with that test because it didn’t work as expected. The whole idea of it was sort of related to anaglyph, but it wasn’t supposed to be anaglyph. The look you get from aberration is very similar, so that made me wonder whether exaggerating aberration would have a similar (although admittedly crude) effect of 3D, and whether it would be distance dependent from the monitor.
No one said a word when I mentioned it because I think they all thought I was talking crap.
Turns out I was, because I never did get it to work.
Thanks Len, yes well, here’s the thing, thinking about it, we manage to cope with TV where we can accept we are looking at large things even when they fit on the screen, then there is no reason not to suppose there is some leeway with the same once depth is added? Well… I don’t know that for certain, but 3D cinema seems to work OK mostly… Anyhow, here’s Kite scene 3 with some proper volumetric clouds. Took a tad longer to render as you might imagine.
Nice work @lantis, the depth setting is certainly a challenge in order to convey the correct sense of scale. Here this is a tricky topic and one I’ve dedicated a lot of clock cycles to recently.
1 - TA render
2 - IBL render
3 - Standard render (on my old old computer this same render would have taken about two to three weeks). It actually took about half an hour. It is an “upgrade” of an old Bryce 5 file, which is a fun use for the Anaglypherizer - Anaglyph Lens System (targeted) if you rather.
