[Brian Jones]

ah, a volumetric effect. SSS or something maybe

[Brian Jones]

sounds like a job for BodyPaint or an I missing something?

[Brian Jones]

morphs work best when the two objects are very close to the same number of points. Have you tried taking the building to the letter rather than letter to building? It may work better that way.
Have you looked at the ShrinkWrap deformer – wrap the building around the letter then transform/make things visible at the right time etc –
Maybe the geometry is just not going to work out and you’ll have to combine partial transforms in post

[Brian Jones]

it’s hard to tell without the scene file but that look like what you get if you have two surfaces in exactly the same place

[Brian Jones]

you’re using GI, looks like the Physical renderer. All GI produces noise which cleans up with time. With Physical that means pure time if you are using the Progressive sampler (you just wait until it’s clean enough) and if using Adaptive you get less noise as you increase the sampling subdivision setting that takes more time as the samples go up and it has less noise but you just ‘get what you get’ when it’s done so if it’s still too noisy you have to try again. I use Corona usually and it’s just the same – more time equals cleaner render but it has controls to stop when a chosen noise level is reached and it has a good denoise function as well to shorten renders by cleaning the noise after the render. (and if you preplan and set things up right you can render then change lighting after the render is done without having to re-render etc.)
Scale is usually best if it’s real scale, it depends on the renderer but usually it looks more real if it is real scale – that assumes the lights are real scale too. If you are using GSG light kit lights they would give better results if the model is scaled down because they are meant to be lights in a studio, to look better the model needs to be that size too.
It depends on the look you are going for though if you want it like your example render then scaling it down so it looks like it’s lit in a studio could be right. For that scene to be lit normally (ie. outside) you’d need a sky (imagery or physical sky) which would mean sharper shadows (sunny day)…

[Brian Jones]

not enough info, how did you make the water bottle? Is it polygons or is it still a lathe (or something)? If it’s polygons then the Split command is easiest. You need to be in Polygon mode, select the polys you want to separate then run the Split command (Mesh/Commands/Split – at least in R19 it’s there – or just right-click and it’s in the menu there). That will give you a new object with a copy of the split polys *but* the original object still has the polys you selected (in case you wanted to do it more than once I guess) so you would have to delete those polys still in the original object to get your two halves.

[Brian Jones]

sure

[Brian Jones]

it depends on how much time you have to produce the content. Sure a threadripper or dual xeons (or quad/oct xeons etc) will allow you to do more (assuming your particles are multithreaded – xparticles is, TP is not) but regardless of how much power you have you can always bring your machine to it’s knees if you try something complex enough. Don’t be afraid to cheat. Are you really going to need every piece of grain that would be in the machine in your scene or just enough for the camera view in each shot?
Does it need to be a perfect render of grain or will much simpler particles do for most of the shots?
Your timeline will tell you what you can get away with – if the most efficient way to do a scene still takes too long to produce then no, the 1700X isn’t fast enough but that’s why render farms exist (and also the threadrippers/xeons/etc/etc)

[Brian Jones]

I am unaware of anything that would do that (which may not mean much) but if you have a ‘new.c4d’ file or ‘template.c4d’ file in the Maxon folder (at least in a Mac, wherever that would be in a PC) delete them so C4D starts with the actual defaults and see what happens.

[Brian Jones]

sure xparticles could work but xp dynamics, while good, are very particle based rather than the object. (at least in xp3.5, I don’t know what it will be like in 4). I’d stick with dynamics either with a TP emitter or an XP emitter. You could do a lot to quiet the excess particle bounce by making the ‘grain’ the size the actual size you want to see (or very close) and not have such a large object scaled down by the particle system.