Aparent size - retina display

Yes, network rendering would help, but I don't really believe that we will get it for Cheetah and I'm not sure if it would be worth it for Martin (it's a lot of work and I don't really know if it would sell him enough copies).

From what I've seen, nvidia optix realy would be great, only on a mac we'll never get it.

AMD denoiser on the other hand never got me convinced. I can't see much difference between it's result and a simple gaussian blur in photoshop (where I have more control, btw., and it's faster, too). The denoiser kills the crisp detail, too ... Maybe it will get better with time.

Intel denoiser: I think it would be better, but the example pics show again a big loss of detail through the denoiser. And I'm not so sure how long Apple will use Intel processors instead of their own. So ...
 
Yes, network rendering would help, but I don't really believe that we will get it for Cheetah and I'm not sure if it would be worth it for Martin (it's a lot of work and I don't really know if it would sell him enough copies).

That's too bad. I have no idea the amount of work it would take, but what EIAS does is have the renderer as a standalone app. The main app exports a package with all the scene data that can be used by any number of render apps in different machines. I think you could even open a copy of the renderer per processor core. It was beautiful. The master renderer assigned the frames as the slaves sent back one finished frame.
 
I think you could even open a copy of the renderer per processor core. It was beautiful. The master renderer assigned the frames as the slaves sent back one finished frame.
Yes the network rendering of EIAS is a marvelous thing. You can render high rez stills in strips per node and the master will splice the final image together. Network rendering is my most hoped for feature from Cheetah 3D.
 
Network rendering is great and with my hardware I'm dependent on it (no Cheetah obviously). For me it works like a breeze and I can put some older pcs to use, but my experience shows me what a can of works it could be to open for Martin if he would implement network rendering.

First off, who could profit from it? I myself have a mixed network consisting of macs and windows pcs (mostly I do use only a mac as server and one pc as client). The 2nd mac is most of the time in use, so it wouldn't help me at all if Cheetah got network capability. But this wouldn't change if I had some older macs around, because of Apple. The machines have to be old enough to have them replaced, new enough to support metal. Martin doesn't have much choice there. So, with Metal in mind, how many of us could and would use network rendering?

Second, a lot of people complain about it not working (whatever software and os they use). With routers, switches, different computers, software in general (for example on 1 pc we had a printer driver that took 100 % of the available cpus), special software like firewalls and cables there are many components that can be faulty. And sometimes everything seems ok, but network rendering will not work. Of course a lot of users think it's the renderer's fault. (In my experience it's very often some cable (mostly the crimping). Just had after a rewire of my network on another place 3 defect cables most people wouldn't have found because everything did seem just fine and peachy (2 cat. 5, 1 cat. 6). So, I believe, there would be a lot of supporting to do.

So, no, in my opinion it wouldn't be worth it for Martin.

About the new mac pro: Like many, many others I'm in the same boat with you (it's quite full, this boat. Do not rock, please). Even if may main income did come from 3d I couldn't justify such a forbidding price (I would of course want more RAM). With an amd ryzen you could get the same or more raw power for something between a third or half the price of a mac pro.
 
I may get some weird looks for saying this haha.. but the way I see it the problem is that Cheetah has only one programmer. It's a one man show. That's the main issue and what's holding back a lot of features that users have craved for a long time, not just network rendering..

Please don't ban me from the forums :p
 
Once you move away from the big boys (Autodesk, Foundry/Modo) and freeloaders (Blender), I'm guessing that most of the 3D companies have very small teams. Lightwave, EIAS, Cheetah etc. I think that's just the reality. I don't believe that Martin makes a living purely from sales of Cheetah licences (the numbers don't add up IMO), so for him to employ others and make a bigger team would probably mean the end of Cheetah. For those of us who have been around for a while, we know many of those bigger companies have user requests going back 10 years or more that have not been addressed.

Also, I think that Martin has to pick and choose which new features to implement, since accommodating everybody's needs would be impossible. With only a limited amount of time, I'm guessing he must be very judicious with how he improves Cheetah. And don't forget, some of the slowness and delays is due to factors completely beyond his control (e.g Apple changing course every 30 seconds).
 
I fully agree with Paul (except with Modo belonging to the big boys. They have a relatively small team. For a long time it was The Foundry's stepchild that didn't get much love. It's changing, though, but their money cow still is nuke). For example that quite soon we wouldn't be able to use older macs for rendering isn't Martin's fault. Even 50 programmers wouldn't change that fact. It's Apple who deprecated OpenGL and just with that created a lot of work for developers (who, on the other hand, almost all should have seen it coming sooner or later. They just shied away from the cost for a relatively small user base).

In my opinion Cheetah isn't a professional software (you can use it for some stuff professionally, though). Professionals aren't the target group, but people who love working with 3d without the wish to spend too much money on it but don't want to use blender. And it's Mac only. So the user base is very small (even smaller than for all the other 3d packages) and can't grow much, but very beautiful things are done with it.

So if Martin could and would get some additional programmers, do a windows version, creating thereby a bigger user base, it wouldn't be Cheetah anymore (even if it would survive. I think, like Paul, that it would be the end of Cheetah). For starters it would get more expensive, and gone were the days where the sole programmer would be able to react to any reasonable bug report. Almost immediately. And more often than not, maybe after some questions, the thread about the bug will end with his words that the next version will not have this bug anymore. This you usually don't get with bigger developers. It just would be one of many.

The other thing: Cheetah is very, very stable. In my opinion part of the reason is that it's running on one OS exclusively.

So yes, it's missing some important features (that's why I use something else as my main app), but there is a lot to like about it. And for that price it's more than a fair trade, very good for learning 3d or coming back to it after some time. For a lot of uses it's more than enough.

And what Paul writes about waiting for 10 years or more for sometimes quite simple requests is all too true (sometimes even for reported bugs). Just look at any forum of any 3d software (or, if you use photoshop. How many years did it take adobe to implement control z as we're used to in any other software? At least they showed some humor when they presented it big time as a new feature). I do agree, that SSS etc. should be standard for any 3d suite.

As I needed some features I had to change to another software as my main 3d app. I do keep Cheetah around and try to stay 'fluent' in it, because I'll use it more when I'll finally buy a new computer. While that will be occupied with rendering I'll use Cheetah on this one. And if I want (or have to) show someone what 3d is or even teach a few basics, something like my main app is plain too much. People get afraid when they see a lot of tabs and possible commands. So maybe you'll sooner or later outgrow Cheetah, especially if you'll be able to get some customers for your work. We can't have affordable software that get's better and better for free (well, zbrush is an exception, but that will change, and of course blender where it's not sure if it will work on a future metal only mac), but expect the full spectrum in its function and the best quality. There is a reason why everything else (except blender) cost many times more.
 
I agree with everything said, and I know Martin likes to keep a tight grip on his creation. There are options like outsorcing development of certain things, like a network renderer, but I guess it won't happen so I'm not holding my breath. I like the app as we all do here, it's inevitable to wish for more.

Cheers!
 
Very good points Hasdrubal. The other thing to bear in mind is that very few if any 3D pros (or semi pros) will use only 1 package. Many will have, at an absolute minimum, 2 3D apps plus Photoshop. When you visit the various 3D forums, you will see many folks have 2,3,4 3D apps, which they use together or will pick the most appropriate one for the particular project. In such a setup / workflow, Cheetah could fill a very neat gap in someone's 3D arsenal.
 
Last edited:
Paul, I do agree, but I see a tendency to use one main app plus a ton of plugins and specialized tools like vray, octane, zbrush, marvelous designer, substance, maxwell and so on (and of course unity or unreal). Because all those apps got very complicated in the meantime and they do cost a heck. Of course there are still a lot who do use several suites, especially with modo (it's in my opinion the best modeler around, it's great for uv-ing, but there is almost no vfx (so houdini is used for that) and for animation maya still is better suited. A lot of Pros have blender around, be it just as a free alternative to realflow.

More and more do use affinity instead of photoshop by the way.

No disrespect meant, but I do ask you, what gap do you have in mind that Cheetah could fill? I know there are people who do use it for converting models from one format to another. Apart from that (and my (partly intendended) use of it) I honestly don't see what it can do faster or better than one of the bigger suites. How would you sell it to somebody who already has cinema4d or modo?

Please, don't get it wrong. I don't say, there is no use for Cheetah in such setups. It's just that I don't see it.
 
I agree @Hasdrubal. Cheetah is a great MAC app for 3D beginners. It's very friendly. It has a good toolset. Still very basic in many aspects but it's good enough to produce very good results. It's also a good app for the casual 3D user, the hobbyist/entusiast. The price is right. Blender is free and powerful, but overly complicated, not good for beginners at all even with their new interface. Even I don't feel in the mood to dive in it, it's just too much lol. So Cheetah is great for mac users that want to learn 3D for casual use, that's the way I see it.
 
Paul, I do agree, but I see a tendency to use one main app plus a ton of plugins and specialized tools like vray, octane, zbrush, marvelous designer, substance, maxwell and so on (and of course unity or unreal). Because all those apps got very complicated in the meantime and they do cost a heck. Of course there are still a lot who do use several suites, especially with modo (it's in my opinion the best modeler around, it's great for uv-ing, but there is almost no vfx (so houdini is used for that) and for animation maya still is better suited. A lot of Pros have blender around, be it just as a free alternative to realflow.

More and more do use affinity instead of photoshop by the way.

No disrespect meant, but I do ask you, what gap do you have in mind that Cheetah could fill? I know there are people who do use it for converting models from one format to another. Apart from that (and my (partly intendended) use of it) I honestly don't see what it can do faster or better than one of the bigger suites. How would you sell it to somebody who already has cinema4d or modo?

Please, don't get it wrong. I don't say, there is no use for Cheetah in such setups. It's just that I don't see it.

It fills a gap in your heart that you never even knew existed, between on the one hand, the joyless cross that one must bear using 3D software dictated by your school, place of work, status etc., and on the other hand, nothing. :)

3D users come in many shapes and sizes, and they will all find their own way to the software setup that suits them, their clients, skill levels etc. Yes, the bells and whistles that are present in other packages are not present in Cheetah, but how many users would use even half the features in C4d, Modo, Maya etc. Cheetah won't be used in the next Hollywood blockbuster, but none of us work in those studios, so it's probably a moot point.

I have been reading your posts for a while, and there is no question that you have a very discerning and critical eye, and an ability to see flaws, qualities etc. in 3D work that is well beyond most 3D users, and certainly most clients. There was a fellow on here years ago whom we all loved, Luke Bocchinelli, who was making money using Cheetah and Photoshop for archviz, and we've had other guys that have come and gone do the same.

So the question really is : what are you trying to achieve, and who is the client, and how much time / money are they willing to spend to get what they're after ? For smaller clients, you can get very good renders with Cheetah and a little Photoshop, and they will never know the difference between that and a render with a more expensive package. You may notice the difference, but then, who is the audience ? You ? Other 3D experts ? The one who pays ?

I think also one must accept that if a package appeals to hobbyists and general 3D enthusiasts, then the work that is generated will be at the hobbyist level. So is that a limitation of the software, or the user ? I'm convinced that if we could get Cheetah into the hands of a few pros, and get them to knuckle down and try to do some great work, then it could be done. My personal feeling is that the potential of what Cheetah can do has not been shown as yet. And the points I make here are not limited to Cheetah. If someone can get a decent archviz render out of Cinema 4D, without using vray, maxwell etc., then they deserve a medal.

So for simple projects (still or animated), assisting with import and 3d scene building, nice renders with minimal overhead and minimal outlay, then Cheetah would I think work well.
 
Paul, what you write is more or less what I myself would give as reasons to use Cheetah instead of another 3d package. Depending on what you do, it can be enough even for professional use. I never said something contrary. There I do agree with you, even with the fact that there is more potential in it than we have already seen. There are quite a few pros in here who don't need anything else for their work.

But my question was, what could someone do with Cheetah professionally who already has another suite in use (and of course knows how to use it). Why should someone who does work with say Lightwave or Maya get Cheetah (except the file conversion already mentioned or for teaching reasons)?

As an aside: It's true that I see flaws others do miss (actually more so than I ever cared to show in here; I think, at least sometimes, I also see some strengths that get missed by others), and I'm most critical with my own work. Sometimes I have to show something to others (who I trust to be honest) to see if I really can get away with something or if I have to do more work (and maybe I myself am still not satisfied with the revisited work). For a customer I always have to keep in mind what I do get paid, how much time it costs me, etc. And the audience is first the customer, second his target group. The third group I always have in mind are my future customers (that is, if a customer wants something I can't really subscribe to, I'll tell him that. In the those cases I couldn't convince him or her I made sure that the result couldn't be traced back to me (or not as easy)). In a way, to me, the future customers are the most important audience and the reason why I sometimes did (and still do) a bit more than the customer paid ;). That said, I still don't earn enough money and don't spend enough time with doing 3d work to qualify as a pro at this moment and never was more than a semi-pro anyways.

But, to stress the other point of my question, I do have some reasons to keep Cheetah around. And sometimes I like to open a model in a 2nd app. But to be honest, what I do I can do faster and in a better quality in my main app than I could do it with Cheetah, and of course I can do some things I see as necessary but that aren's possible in Cheetah (that's why I shell out quite a bit of money). And there is absolutely nothing I would have to fall back on Cheetah to get done. (as an aside, I don't have a problem a problem that my main app were to difficult for me or overcomplicated. And I do like it very much and feel at home with it; that said, some things I could do as easily and as good with Cheetah). So I still don't see this gap you mentioned in your other post, where Cheetah can be helpful as additional tool in a professional setup (some years ago that was exactly what I recommended: That Cheetah should get something, whatever it is, that in other packages can't be done as easily or only with an expensive plugin).
 
I wasn't going to say anything since this is a two way conversation but..

I think there is too much talk, not by you but by every one of us, and too little to show in the gallery forum and in the website gallery. Pros and many people in general won't be interested in trying a small 3D app if users don't show what it is capable of. What I'm saying is that I wish there was more posts in the gallery of finished works, not just experiments and test renders.

I'm sure that the pro work that many people do here can't be shown because of legal issues, but still.. where are the finished personal projects? I wish users would show more of what Cheetah enable them to do.
 
Joel, this is a public forum, and you or anyone else is absolutely entitled and welcome to join in and have your say (although we have drifted somewhat from the original topic ). It just so happens that Mr Hasdrubal and I are blabbermouths. :p

The issue of a great gallery for Cheetah has been around for as long as I can remember. It's a shame, but I'm sure Martin is aware of it
 
I actually like the behavior of Cheetah 3D. a 1920x1080 image is 1920x1080 actual pixels. If I want a larger image with the same rendering time, I can just render 3840x2160 and cut the number of samples to 1/4.
 
I'd definitely like to see the ability to quickly zoom in and out on an image (and/or to be able to set a default zoom) in the Render Manager window. At the moment I use control-drag to quickly zoom the entire screen, but it's a pain to have to do that every time I check on a render. I get that there are workarounds and I'm sure it would be some degree of bother to implement this, but surely it would be a lot less tricky than the really complicated things that C3D is able to do in modelling, texturing and rendering an image?

I would hazard a guess that folks new to C3D and 3D in general are quite likely going to have retina-display MacBooks or Macs so it makes sense to be able to have Render Manager display a 2 x zoomed image (perhaps by default) for those users.
 
There could be some zoom buttons in the menu bar, but the default should always be pixel-perfect.

I work on user interfaces and it's important to me to be able to pixel-perfectly map textures and make sure that they're accurately presented on the screen.

While w're at it, what I would also like to see is the ability to re-render a part of the image by drag-selecting the target area (Maya has a feature like that).

That would be great for tweaking materials where you want to see a closeup with many samples.
 
While w're at it, what I would also like to see is the ability to re-render a part of the image by drag-selecting the target area (Maya has a feature like that).

This is a great feature. At least in another app the re-rendered region gets comped automatically into the last render, which is great if you find something wrong with your image that had rendered for hours, like a forgotten texture you have overseen in test-renders. So you have the corrected render maybe in a few minutes instead of doing everything anew. But I mostly use it for tweaking (for example materials, highlights, caustics etc.)
 
Back
Top