guys, just to keep you in mind than MCG isn’t the saint graal at all…
just to open your eyes
Actually you have (or in the near future) two wonderfull tool
-
Fabric Engine, a cross platform, cross DCC solution.
-> the tools you write for example Max IS compatible with Maya or Houdini or Modo UE4…
-> the core computing is CPU + GPU !
-> the tool provide in version 2.0 CANVAS programming (think about nodes/graph etc)
some reads:
http://fabricengine.com/2015/05/fabric-at-fmx-2015/
http://fabricengine.com/fabric-for-3ds-max/ -
Houdini Engine, a cross platform, cross DCC solution
-> the tools you write in Houdini are compatible with Maya, Max, Unity, UE4…
Come on Autodesk, with this really good alternatives on the road, with your so called on develloppment “bitfrost external engine”, you can’t make a tool for Max wide open for the futur ??? something great once ?
All i’m saying is MCG is a joke because
- there is not a bit of interactivity with this tool, you just have to work blindly and compile and… well we are on 2015 !
- I’m a softimage artist who already had a tool like MCG, ICE, and all my work, my tools, just goes to the trash now the software is discontinued -> we need a common language a common format, just like we do with Alembic / Photoshop files / Open VDB / GLSL etc…
what was happened?
where are new-cool-useful-helpful-engraved MCG toys?
is the clown dead?
The party has moved to the MCG Facebook group.
it’s a private party.
err well no, if you don’t have 2016 you’re not. and as I don’t need another cloning tool I’ll give it a miss thanks. (still waiting on my suggestions btw and i thought it was supposed to be quick and easy :surprised bash em’ out in 30 mins ‘n’ all that, It’s only been 2 weeks)
it’s a month left since 3ds max 2016 was released. no doubts the key of the release is the MCG.
let’s check where we are and what we have now.
i’ve browsed all known to me max resources and found all together about thirty made with MCG and shared plugins. only thirty…
all off them are written by experienced or at least very familiar with tools development people.
no one shared anything really useful or valuable.
no new it plugins development artist posted any of his works (or attempts)
why?
this is my point and it can be wrong…
the tools is NOT for a verdant developer. in many cases it needs better knowledge of the system than mxs
good solutions are not trivial and don’t expect to be shared
autodesk doesn’t really and seriously promote the tool
the tool is very limited and it’s hard to acknowledge this fact
the tool is raw and was pushed without any workflow idea
what is the difference between a tools and a toy?
the tool is designed to make what the user want.
the toy has to only entertain the owner.
MCG today is a toy.
ok I see things a little differently. I think 1 month is probably not enough for develope new and cool stuffs. The main reason is that MCG is too fresh. It has no backward compatibility, and because of this if you use max as the main pipeline software, it takes time to change from the previous version. (Bugs and its fixes, testing the new tools if they are working properly in the pipeline etc…) So be patient… Soon we will see cool stuffs (not just random clone tools) if the users learn how MCG dataflow works. It will need a year or two for the users to get familiar with is.
that’s exactly what i worry about.
the tools development can’t wait a year or two. it always needs a solution right now.
that’s why i don’t see the MCG in a tools development pipeline.
@denisT, Klunk: Denis, as developers you know better than anyone how important feedback is to improve a tool. So instead of complaining and arguing against MCG, why not give proper feedback to Autodesk on how to improve it? I’m sure MCG came to stay and it’s likely to be the main focus of development for a while, so what do you need from it to be useful to you? You guys might not like nodes but guess what? Most hardcore Houdini users code in VEX directly and usually forget about VOP nodes entirely. It allows even a mixed environment where you can connect nodes normally but there’s an “Inline” node where you can put code there. This can simplify node trees a lot. I find this approach really nice. All of this is possible with no performance loss because everything is compiled in the end. As MCG is compiled too I guess a similar thing could be done. For performance, I can’t imagine anyone better than you guys to give a decent feedback of what’s wrong and why.
@Paul Neale: I’m impressed about what you said in post #25. I think I wouldn’t believe it if it didn’t come from you. Anyway, aren’t you a bit too positive about some features in max 2016? I mean, OSD (OpenSubdiv) is slow like hell to me and in GPU mode it’s still slower than Turbosmooth and yet can’t show textures, so how can it be better than in Maya? Also, XRef for rigs seems to be getting close to be useful but it’s not quite there yet. When XRef’ing a rig with tracks in Local mode, it breaks instanced tracks, something I try to use whenever possible. World Space Modifiers still can’t be XRef’ed and get merged. And while custom attributes are now exposed when modifiers are XRef’ed, you still can’t access modifier properties. So it’s not possible to use your PEN Attribute Holder preset stuff, which is an annoying limitation. I know it’s old and you probably don’t use it anymore, but for be it works very well. If at least we could merge the modifiers of specific objects, but looks like it’s all or nothing. Ideally XReference should expose everything and allow overrides per parameter. In Softimage you could save these overrides as deltas to load on other files. Very powerful. Maybe max is designed in a way this is not possible.
that’s exactly a feedback. i don’t trust MCG and i can’t see it in a tools development pipeline in the near future.
btw i’ve answered the question “how to make it better”… rewrite sdk on a modern way.
today is 21st century… it’s not 90th where the max sdk still lives.
i don’t believe that anyone likes a ‘chief special’ made with rotten ingredients.
You’re right, but if it tastes good and doesn’t make me sick, who cares? In the end of the day artists don’t care how developers get their job done just like clients don’t care how artists get their job done, as long as expectations are met. In case of MCG, if Autodesk manages to get speeds like Genome out of it, I’m happy. If Thinkbox can do it without access to the source code, they should be able to do that too.
it sounds like you are OK to have rotten but not poisoned.
it makes you an ideal 3ds max client.