[Andrew Somers]

CHoose SPRITE or TEXTURED POLYGON in that pull up menu, then below that select the layer that the custom particle is on in the comp.

[Andrew Somers]

Just to chime in here – I can’t speak of earlier version of C4D – but we have adopted R13, and use the built in renderer for photo real rendering. I was surprised at how good it looked in the final comp.

I’ve never seen my client’s eyes pop out of their head like they did when they saw the final comp.

BUT – using the DEFAULT materials settings, it DID look like plastic initially – I did spend time tweaking materials and render settings, and then doing a multi-render pass, and then building and adjusting in AE to get the image colorgraded and adjusted to fit the comp.

The tools are definitely there at least for inorganic objects. And VRay is also available, and that was a key factor for us. And there are IK tools built in, plus third party ones like Cactus Dan’s.

The other thing of course is that you have the most integrated round trip integration with After Effects of any 3D program.

And the learning curve for C4D seems to be among the easiest – the interface is elegant and easy to use IMO.

We are 100% Mac, so I have not used 3Ds, but have used Lightwave and Maya. I’ve never liked Lightwave as it’s clumsy to say the least, and I have never seen a “photo real” render – decent renders to be sure, but Lightwave is king of having that “digital” look. YUK!

And Maya? Maya is fifteen programmers racing cars toward an intersection, and they all collided with each other and exploded in a ball of flame and debris – the result of the smoldering wreckage is Maya. Like most Autodesk products, the interface is a kludgey klunky kollection of krap. I’m pretty sure that Autodesk hires programmers that were fired by Microsoft.

/end of rant.

SO: Back to the OP’s original comment: For real 3D work, I suggest Cinema 4D. Best integration with AE and shortest learning curve, with professional features.

I have used Invigorator, and for a down and dirty inside AE method, it works okay – but the interface, even the new one, is still a bit clumsy. It does not ray trace, though you can still fake the look of caustics with a couple techniques. When integrated into an AE scene using an animated 3D camera though, the interface becomes cumbersome.

I have not used Element yet – but it looks pretty cool. Even so, it has quite a few limitations.

[Andrew Somers]

ALSO, if you click on the render setting (i.e. “best settings”) you can specify the specific frames you want to render.

Set the TIME SPAN to CUSTOM

In the above example,the first frame is set to frame 50, and the last one to frame 100, irrespective of the work area setting.

[Andrew Somers]

AE color management should be renamed “Color MANGLEment”. LOL

It can be enough to make your head asplode. And at the risk of causing multiple head asplosions, I’d like to add a bit that I didn’t see mentioned above.

The problem Colin is describing relates specifically to either GAMMA or BLACK/WHITE levels.

Much video is not fully “color managed” in most player applications, and despite the fact that Apple does allow for embedding ICC profiles in Quicktimes, After Effects (at least as of CS5.5) does not allow embedding the ICC profile, meaning the video clip you render out will not be color managed when played. (Not certain of CS6 will allow for embedding in QTs.)

I rarely use AVI, so I can’t speak to it’s specific quirks,

SCENE REFERRED vs OUTPUT REFERRED.

Charles Poynton relates the situation in detail here:

https://www.poynton.com/notes/PU-PR-IS/Poynton-PU-PR-IS.pdf

But here is a quick “nutshell” of the situation, using HDTV as the example.

HDTV tech specs define a SCENE REFERRED profile for program material. THis profile defines the primaries, and also a gamma curve that is somewhat close to 1.9. This profile is called REC709.

The Rec709 scene referred video (gamma 1.9) is pumped into a display (gamma 2.4) and you get a *system gamma boost* of roughly 1.2

The system gamma boost is there intentionally, so that a scene referred HDTV Camera, shooting a scene in daylight, will be “perceptually the same” once it is displayed on a screen in a darkened living room.

sRGB is an OUTPUT or DISPLAY REFERRED profile. It uses the same primaries as Rec709, but has a slightly different gamma curve, that is approx. 2.2.

Note that sRGB, an output referred profile, specifies a gamma of 2.2 instead of Rec709’s lower 1.9, because sRGB was developed and intended for computer monitors, which are used in a brighter work environment, as opposed to a dark living room viewing environment.

BOTH sRGB and Rec709 are assumed to be displayed on a monitor with a power transfer factor of around 2.4. sRGB’s transfer function in intended for brighter viewing environments.

16/235 vs 0/255

Another critical issue is the black/white levels. Does your codec expect them to be 16/235 or 0/255? Some video players may not interpret one or the other correctly. And some encoders may force the video into 16/235 even if you are sending it 0/255.

Broadcast Rec709 has the black set at 16 and the white set at 235 for both RGB and YCbCr. If you try and send 0/255 into this system, you will have your blacks clipped and too dark, and the whites clipped or too bright.

WHICH SHOULD YOU USE ?

Working Profile

As Darby indicated, your working profile is not the critical problem here. Choosing the best working profile is another topic entirely, but in a nut shell, choose a profile that contains all of your colors without clipping. Larger profiles such as ProPhoto *require* the use of high bit depths (i.e. 32 bit). If you are headed for the web or for HDTV, then sRGB is often a good choice – but only for your WORKING profile.

Output Profile

If you were going to play back your video in a COLOR MANAGED player and you had embedded your profile in the video media, then it would not matter what output profile you chose, as the player would manage it and display it appropriately.

HOWEVER, as we said at the start, video is typically not color managed for the purposes of distribution. (You may notice that “Embed Profile” is greyed out when using most output modules in AE). And besides, if you cannot control how people are viewing, you’d want your video to play correctly in non-colormanaged environments.

So, that means that we must choose an output profile in After Effects’ output module that creates a video file with the kind of color/gamma that the PLAYER application EXPECTS to see.

And what would a player application usually expect? TYPICALLY A SCENE-REFERRED VIDEO STREAM.

Since sRGB is an OUTPUT referred profile, it is “generally incorrect” to use to output your video if you are sending your video someplace that is expecting a scene referred profile such as Rec709.

However, Adobe FLASH expects sRGB, so the implication there is to use sRGB as the output profile, and that is what the flash environment *expects*.

But most player apps are expecting either Rec601 OR Rec709. As a result, Rec709 is usually the choice for a video output.

And Then the 16/235 thing:

Now, as it happens, different codecs and different player apps may handle the 16/235 and 0/255 and gamma issues differently. Some do not correctly interpret the correct black/white levels.

Colin: Since I don’t use the AVI container much, and don’t which which variant of “uncompressed” you are using, I can’t say definitively – BUT from what you describe it sounds like you are outputting to a 0/255 black/white point colorspace (sRGB), and the codec & player you are using wants to see a 16/235.

So do this: TRY USING the colorspace: Rec709 16-235 as your OUTPUT colorspace in the output module. Note that while Rec709 has a lower gamma (which increases contrast for darker environments) than sRGB, using the 16-235 variant should fix your black levels if this is what the codec you are using is expecting.

Here are some links with some discussion on scene vs output referred colorspaces, gamma, and color:

https://provideocoalition.com/index.php/cmg_keyframes/story/scene_vs_display_referred_profiles/

https://www.poynton.com/PDFs/GammaFAQ.pdf

This little tidbit from page 6 of the gamma faq:
Ambient lighting is rarely taken into account in the exchange of computer images. If an image is created in a dark environment and transmitted to a viewer in a bright environment, the recipient will find it to have excessive contrast.

https://www.poynton.com/PDFs/ColorFAQ.pdf

[Andrew Somers]

The timewarp effect will NOT change the length of the layer that you APPLY it to. THis is obviously a problem if you slow footage down using Timewarp, your footage will run out before the layer end. There are TWO ways to fix this:

1) Precomp the layer, and then apply timewarp to the Precomp in the original composition – set the precomp length to be long enough to encompass all the video that you are slowing down.

OR

2) In “Interpret footage” LOOP the footage several times, so that it will make a longer layer to begin with, then when you apply the slomo TimeWarp, you will have enough room to see the full footage.

[Andrew Somers]

Let me echo what Walter is saying here – it is *generally* the case that you don’t want to “remove the rig” – you want to cut out the subject and then comp them back into a clean element.

I’m not sure what you are going for here, or why this was done stop motion, or what your intent is. I don’t understand why you would do stop motion with a live actress against a live plate – this makes no sense to me at all, unless you are trying to get that “king kong” effect to match with other stop motion animation??

TO MAKE A CLEAN PLATE: Do you care if the plate is static, or do you want the trees to wiggle and shimmy as they do there in “stop motion” ?

If static, then simply take several frames from various points in your clean element, and stitch them into a single tall still element, and then keyframe that to match the tilt down.

If you want the jiggly/jittery trees, the create a synthetic plate using your A element, and the clean element, time warping the clean element as needed to match the A element tilt down, and then use garbage masking to combine the two, using the natural geometry of the scene to hide the transition.

If you want natural waving leaves, then shoot multiple elements at normal speed for several positions through the tilt. Make sure each one is long enough for the full shot. Overlap each one about 50%, then stitch together and keyframe the motion as per the static plate above.

ROTO:

Use MOCHA to roto the entire girl, paying particular attention to the top edge of her, and wherever the black guide poles enter frame.

It is best to use multiple separate rotos for various body parts. You can do a basic garbage roto around the lower white skirt thingy.

You will need a separate roto around the top of the dress where the rope extends, and this needs to re replaced with a dress element that is not “pulled up”.

(for the pulled up dress, try the MESH WARP efect to flatten it, or paint a flatter element to lay over in association with the roto).

[Andrew Somers]

What you are looking at there is fairly basic motion graphics – I recommend the excellent book:

https://www.crishdesign.com/books/Creating-Motion-Graphics/index.html

And look at any number of Andrew Kramer tutorials:

https://www.videocopilot.net/tutorials/all/

[Andrew Somers]

Matt: Just ran this test and when I look at the activity monitor the 4 BG instances are are shooing .5% CPU usage, while the main AE ap is using 120%. This nearly quadruples the expected render time.

This sounds like you are choked for RAM.

Is your disk cache turned on? It should be set to 20 GB or so.

[Andrew Somers]

Matt said: ” It’s still using all of your processors, but it’s using them to render each frame individually (not multiple at the same time). “

No Matt, this is not a correct assumption. Even multi-threaded code will not neessarrly take advantage of all processors – a fact that becomes clear as you add processors. Even with a multi threaded app, when you go from 2 ores to 4, and 4 cores to 8, you will see an overall reduction in core utilization.

One way around this is to launch multiple instantiations in the background, so that each instantiation can take advantage of the unused processor time that other instantiations are not using, and also make more efficient use when dealing with single-threaded operations.

This is the reason that “render multiple frames simultaneously” was added as a feature.

EVEN SO – there are effects that will not multi thread, and I believe even some that will not work with muilt-processing at all.

In the jpeg you posted, it appears that you have something that is operating as a single thread. Not seeing your project, it’s hard to know what/why – but don’t assume that everything – plugin/effect what have you – is multi threaded.

ALSO, when posting images of activity monitor, the “System Memory” tab usually has some of the best clues.

Try This:

STEP ONE: Quit all other apps and log out (or reboot), then log back in and *only* start After Effects.

STEP ONE: Turn Multiprocessing ON. Since you only have 16 GB of RAM, limit the processors to 4, and set each to 2GB of RAM. Set “reserved for other apps” to 4 or 5 GB.

This should launch 4 background instances and you should then be rendering with 4 cores. You *may* be able to increase the number of cores, but generally (in practice) 16 GB is really only enough for efficient use of 4 to 5 cores when using MP. If not, then you may be using one of the effects that drags AE down to a single threaded situation. Without knowing your project or effects chain, it’s hard to say.

[Andrew Somers]

If render multiple frames simultaneously is off then you will not be using all processors. It must be ON, and you must have enough RAM for each process, etc etc.

What machine and OS are you running, how many physical cores, and how much RAM do you have?