Jump to content
3DCoat Forums

V4.1 BETA (experimental 4.1.17D)


Recommended Posts

  • Member

In the latest beta (10), been getting a lot of hangs (not a hard crash) when using the e-panel closed spline tool (and others) in the Plane and 2D Paint modes (Voxel room). Happens very frequently and I'm not exactly sure what usage causes it.

And right clicking a spline point doesn't change it's mode when the point is hovering outside of the surface.

Link to comment
Share on other sites

  • Advanced Member

Can anyone tell me how to change the texture resolution in per pixel painting room? It used to be under the 'Edit->Mesh and Texture resolution' menu, but now it's gone I only have options related to mesh and pTex :(

Beta 10 OpenGL

Link to comment
Share on other sites

  • Advanced Member

Can anyone tell me how to change the texture resolution in per pixel painting room? It used to be under the 'Edit->Mesh and Texture resolution' menu, but now it's gone I only have options related to mesh and pTex :(

Beta 10 OpenGL

Ok, got it, my scene has 3 UV tiles and 3d coat thinks it's in pTex mode...

Any idea how to bring back the change resolution menu, or to make the app understant this is UV painting and no pTex? I need to up-res my textures and don't want to export the whole lot to an external app just to do that. After all, resampling a texture is the least we can ask from a painting app isn't it.

Link to comment
Share on other sites

  • Advanced Member

That's the main point. I would ask Andrew if the menus, pop-ups are all hardcoded - in the main source code - or he uses some XML or some script to compile the menu windows with all pop-ups and tabs into the source code. Then I would ask Andrew to give me a version of 3D-Coat with the hopefully external User Interface [uI]-configuration files that I can compile into 3DC and freely modify all menu texts windows tabs, their position and order of appearance. I would upload the UI-test version of 3D-Coat here that would only serve to everyone agree on one GOOD interface version with proper naming not the functionality of the tools. Then once ALL agree the new UI is a much better arrangement and naming, send it back to Andrew to integrate the new UI into the newest version of 3DC.

nice idea but in practice everyone agreeing doesn't work in my experience. I could write a ton of information on interface design. I actually work at a company where part of it service is interface design, OOOii granted much of it is fake design for movies but much of it is also real, and even the movies they want practical future UI. It needs design theory that goes from large to small, then everything is implemented based on the design theory. Once people understand the design they will always know where to look for updated features, and new users will be able to learn the software faster. But moving things around could be disastrous too. People hate change, there are many users who have no issue with the UI and aren't represented on the forums, so the design would have to also be based strongly on whats there now. Anyway I'd say the best thing we users can do is build a design document, have a forum and work from there but at the end of the day we implement nothing. Andrew would look at and approve of whatever we convinced him works.

Unfortunately, this is not exactly what I meant. I was saying that a dedicated 3DCoat employee should be the interaction designer - and if there isn't one they should hire one. Someone with significant software interaction design experience, someone like a big corporate would hire, and it shouldn't be a customer doing it. The person should be unbiased and use industry standard ideas as well as customers' ideas or needs as important input for any UI changes, but who doesn't turn on a dime unless it is the right UX thing to do. If real, big interaction design is not your day job, then you shouldn't be doing this. A good interaction designer does not have to be (and really shouldn't be a biased customer of the product) because they will listen and incorporate the best customer input with an overall picture of the user experience and development roadmap in mind. And they will have a process for future input to be recorded, prioritized and incorporated if need be. And that interaction would be in consult with the devs to determine a roadmap of the UI improvements to implement over time what is possible, big wins and low cost. Good interaction design is not democracy. It should be a meritocracy. ;)

Suggesting a user throw together a document just won't work as well, I think you'd need someone in the same location as the majority of 3DCoat development. It seems to me that 3DCoat is big enough and a stable enough mature product to warrant such critical evaluation and UI development to take it to the next level.

This can be done as a one-off few month contract or on-going as need be. But relegating interaction design to the bottom of the pile, actually dimishes the power of the amazing programming Andrew is doing. Good interaction design will show off his amazing tools much better and bring more and more users on board creating very cool stuff.

But then again, I'm just doing what everyone else does in forums and speculate. I've just seen large companies (and small startups alike) do amazing things with UX by just spending a small amount of money on usability and interaction design. I'm a big fan of 3D Coat, and was just excited enough to make a suggestion about it in case it gets some traction.

(Sorry for cluttering this New Release thread - maybe the forum moderator could create a New Release Announcements thread, a New Release Bugs-ONLY thread, and a New Release discussion thread - this one is way too long to follow)....

  • Like 3
Link to comment
Share on other sites

Unfortunately, this is not exactly what I meant. I was saying that a dedicated 3DCoat employee should be the interaction designer - and if there isn't one they should hire one. Someone with significant software interaction design experience, someone like a big corporate would hire, and it shouldn't be a customer doing it. The person should be unbiased and use industry standard ideas as well as customers' ideas or needs as important input for any UI changes, but who doesn't turn on a dime unless it is the right UX thing to do. If real, big interaction design is not your day job, then you shouldn't be doing this....

Snipped for brevity.

Nailed it. Could not agree more with you.

  • Like 1
Link to comment
Share on other sites

my POV in middle of this brainstrom

- priority now is to find/kill bugs until the v4 release -roadmap say in 2 months-

- Correct some misspelling and standard naming.

- Tweak some tools location to make 3C workflow more ergonomically comfortable.

- Optimize 3DC code to speed up the work.

- after V4, to begin the interaction designer era ready for a v5 release -long term development-

  • Like 1
Link to comment
Share on other sites

  • Contributor

my POV in middle of this brainstrom

- priority now is to find/kill bugs until the v4 release -roadmap say in 2 months-

- Correct some misspelling and standard naming.

- Tweak some tools location to make 3C workflow more ergonomically comfortable.

- Optimize 3DC code to speed up the work.

- after V4, to begin the interaction designer era ready for a v5 release -long term development-

Amen bro'!

And speaking about bugs... Big fish, the biggest I've managed to catch...

This is a continuation of this: http://3d-coat.com/f...indpost&p=86182 post about "The Air Cushion Bug" (that's how I've called it since I first encountered it).

Please download this exemplary scene: https://docs.google....WlxcUZnUzQ/edit

There are two objects in it. A sphere and a thin, but high density layer that is a copy of the sphere's surface which I called the COATING. It might be a cloth with sculpted wrinkles, character's hair or whatever. This one though only has a short handwritten text just to add some details.

-> The following actions are independent of auto snap setting. You may have it set to outer surface, nearest or closest. It won't change a thing.

-> The following actions will place new or current vertices UNDER the coating instead of OVER it.

Things to try in order to reproduce the bug (I suggest undoing after each step):

1. Smooth area with Brush tool (you know, with SHIFT. I'm not sure about Brushing itself).

2. Use Split Rings.

3. Subdivide the retopo layer.

4. If you hit a certain spot with a Points/Faces tool, the point will be placed UNDER the coating. This will manifest in point not being visible after left mouse button has been clicked.

5. SOMETIMES Quad tool will place new vertices UNDER the coating. Now, I'm not sure about this one, but I remember it caused me a lot of trouble with my zombie model. It doesn't seem to be misbehaving in the attached scene though.

After you try all of the above things, you might try baking the normals. You might do this on the included retopo layer, or create your own, denser one. In fact, the denser the original mesh, the worse the result (!).

Choose Merge with NM (PP). The settings don't matter really. Just make sure you choose auto UV generation, because the scene doesn't have any.

After calculations are complete, you'll end up with normal map "chopped off". Looking similar to this:

buggednm.jpg

Tools that I found to be working fine with the layered surface structure are: Add Points, Quads (most of the time), Points/Faces (if you don't hit that bitter spot a.k.a The Air Cushion), Brush (but NOT Brush smooth), Strokes. I don't guarantee that those tools will work 100% of the time. I have not tested them that much simply because I don't have time for it.

Okay, but why do I call this bug The Air Cushion bug?

It's because the way I see it. Take a look at this schematics:

schemayf.jpg

Two objects of different density (or resolution, if you prefer this term), one covering the other. There have to be areas of pure space between them. They're not completely merged with each other and some of the tools think that it's okay to put a vertex in this place.

It's either that or the tools keep thinking that the innermost, or the farthest from the camera (within the reason of course), surface should have the new vertex snapped to it.

PS. Please forgive me any grammar or other lingual mistakes... Not an native English speaker I am.

Link to comment
Share on other sites

  • Advanced Member

If hiring someone is out of the question, another option could be to set up a voting system for each suggested change to the UI. It would be important to send an e-mail to everyone who has purchased 3D Coat though, to make sure there are enough people voting so that a clear picture emerges of what the majority wants. As incentive to take some time out to vote, those who participate in all the necessary voting sessions could be offered a discount on the v4 upgrade and/or new licenses as well. It all depends on what Andrew wants to do since he naturally has the final say in everything. Democracy may not be the perfect solution, but it at least solves the problem of disagreements over how things should be changed. As to what should be changed... well I'm not really sure how this should be handled.

Link to comment
Share on other sites

sorry I disagree with you

but this is an commercial app, not an open source one.

A dedicated 3DCoat employee should be the interaction designer - and if there isn't one they should hire one. Someone with significant software interaction design experience, someone like a big corporate would hire, and it shouldn't be a customer doing it.

A good interaction designer does not have to be (and really shouldn't be a biased customer of the product) because they will listen and incorporate the best customer input with an overall picture of the user experience and development roadmap in mind.

And that interaction would be in consult with the devs to determine a roadmap of the UI improvements to implement over time what is possible, big wins and low cost.

Good interaction design is not democracy. It should be a meritocracy. ;)

Suggesting a user throw together a document just won't work as well, I think you'd need someone in the same location as the majority of 3DCoat development.

  • Like 1
Link to comment
Share on other sites

  • Advanced Member

I suggested hiring an User-Experience-Designer years ago already.

So I can only agree to what photonvfx says.

While the program got a lot nicer to look at (cosmetic level) the underlying interaction-architecture

has remained more or less the same over the years.

  • Like 1
Link to comment
Share on other sites

  • Member

If hiring someone is out of the question, another option could be to set up a voting system for each suggested change to the UI. [..]

Good idea! Set up a voting system here on the forum. Every major game developer - especially Kickstarter projects - use voting system nowadays. There are the WebDesigner - Andrew Zakolukin, and programming - Vladimir Popelnukh who can accomplish this. Then after you vote, display the result so you can see what EVERYONE wants, how many people like the new names of commands, menus, windows, 3DC-operations, etc..

Link to comment
Share on other sites

  • Advanced Member

If hiring someone is out of the question, another option could be to set up a voting system for each suggested change to the UI. It would be important to send an e-mail to everyone who has purchased 3D Coat though, to make sure there are enough people voting so that a clear picture emerges of what the majority wants. As incentive to take some time out to vote, those who participate in all the necessary voting sessions could be offered a discount on the v4 upgrade and/or new licenses as well. It all depends on what Andrew wants to do since he naturally has the final say in everything. Democracy may not be the perfect solution, but it at least solves the problem of disagreements over how things should be changed. As to what should be changed... well I'm not really sure how this should be handled.

While I understand your point and desire to have your needs met by the software, I have to disagree with the approach.

I still maintain that we as customers should concentrate on our art and what we use the application for, not on its software architecture. We should not be designing by committee or email . I personally would trust any decent and competent interaction-designer who does due diligence and balances customer issues with best practices and dev/Pilgway business needs.

We shouldn't be worrying about how so much -- that is the interaction designer's job to figure out. We just need to communicate our issues, offer suggestions and let him/her deal with the priority and severity of the issue in balance with the overall application. A customer suggestion that seems great out of context (and for that person's particular need) may not work in the big picture. This is why interaction designers are needed. It also prevents the "playing catchup" as every different customer has a different opinion. You can't please everyone, but if you follow best practices you will please most of them and they won't even know there was an issue in the first place. :)

  • Like 1
Link to comment
Share on other sites

  • Advanced Member

Good idea! Set up a voting system here on the forum. Every major game developer - especially Kickstarter projects - use voting system nowadays. There are the WebDesigner - Andrew Zakolukin, and programming - Vladimir Popelnukh who can accomplish this. Then after you vote, display the result so you can see what EVERYONE wants, how many people like the new names of commands, menus, windows, 3DC-operations, etc..

I won't beat this to death, but I strongly disagree. Software design shouldn't be democracy - it should be based on the merit of the idea. Plus what if you are on a project and have a deadline and didn't get a chance to vote? There's too many ways for this to fall over and too much admin overhead for Pilgway.

Have you heard the expression "too many cooks spoil the broth"? ;)

I will shut up on this topic now.

  • Like 1
Link to comment
Share on other sites

  • Advanced Member

sorry I disagree with you

but this is an commercial app, not an open source one.

Assuming you're replying to me, I wasn't suggesting that voting is what we should do. Heavens no. Hiring someone whom specializes in interfacing is by far the better path indeed. What I suggested was just an alternative should hiring not be viable for whatever reason. It stems from earlier comments I remember reading about how difficult it is to reach a consensus regarding UI alterations (unfortunately for some reason I cannot find those specific comments to quote).

Zbrush has a large userbase because it was the first, the toolset is awesome, but the ui is terrible (but the userbase is already there so everyone stick to it)

Mudbox has a large userbase (much less larger but still) because it has an UI any traditional sculptor can jump in without hours of lecture. <(actually this is exactly the opposite of 3dcoat right now: you NEED to look at tutorials to understand many of the features, not only because they're special but also because the commands are scattered all over the place)

I wouldn't call the UI in ZBrush terrible per se. It's not a layout artists are familiar with and this can be upsetting for some people, very much so for those among us unwilling to adapt. Despite that fact, it is still a very clean, streamlined UI and one that matches ZB's core purpose fairly well IMHO. The one thing it's definitely not is clunky.

Coming from 3ds Max and Photoshop as I did gave me a little bit of grief when I first used it. Mostly it was the necessity of the edit button and lack of a traditional viewport that threw me off initially. Once I got past that, the rest was a breeze. I rarely needed to consult the manual, something which I feel is the mark of a well designed interface. Mudbox is even easier to pick up for obvious reasons.

3D Coat is on the right track so far as general layout goes, it's just the details that need some love for the most part. As others have discussed, naming conventions and descriptions need to be overhauled in order to be clearer (i.e. less confusing) and this begins with looking at commonly used terms people expect to see (AUTOPO being one example already discussed).

There is a lot of redundancy as well. Exporting a mesh from the retopo room is one example, which leaves one wondering which they should be using and whether there is a difference between them. In some cases how a feature works in one room sometimes doesn't work/appear the same in another either. I think this more than anything is what leads to feelings of "clunkiness".

Doing anything is 3D Coat shouldn't be a matter of repeated trial and error just to achieve whatever it is a user is after, especially when it's something most consider fairly basic such as exporting, generating maps, rendering, etc.

Link to comment
Share on other sites

Writing scripts in 3D-Coat

Scripts in 3D-Coat are based on Angelscript that is very close to c++ syntax.

Using scripts you may do virtually everything that is possible within UI.

Scripts are intended to perform batch actions, make own macro actions, create interactive tutorials

I remember Pixo asking for scripting. Finally is here :)

This is a milestone production oriented

Link to comment
Share on other sites

But why not Python dammit? Cpp and its derivatives are too hard! I wouldn't call Cpp a scripting language. That's freakin' hardcore programming!

EDITED a dozen of times. :D

It is bacause all good changes may be easily moved to engine internally. The huge other reason is that afterSDK will be released user will have almost same language for SDK and scripts. And c++ morks much much faster. It is enough to spend 2-3 hours to understand general syntax. No more need to write scripts.

Link to comment
Share on other sites

  • Advanced Member

Thank you for adding a scripting language ,

But imho it means that all the lib that was dev for maya,nuke,houdini and mari in python would have to be rewrite (ui via pyside or pyqt and access to db for instance).

Script are in general use for simple action not to iterate over a big object ,mainly TD and pipeline guys use it.

Anyway this is still a good news.

Take care,

Pixo

Link to comment
Share on other sites

  • Contributor

It is bacause all good changes may be easily moved to engine internally. The huge other reason is that afterSDK will be released user will have almost same language for SDK and scripts. And c++ morks much much faster. It is enough to spend 2-3 hours to understand general syntax. No more need to write scripts.

Ah, I see. So basically any tool that someone writes and it meets with acclaim could be easily integrated into the code? Sounds good. And yeah, you're right. Comparing speed of Python or other interpreted languages to C++ is out of the place. :)

Thanks for adding this to 3D Coat. I guess I'll have to dust off my C++ books.

Link to comment
Share on other sites

  • Member

I wouldn't spend too much time reading C++ books... maybe look for some short tutorials on basic C++ syntax and usage. If you know Javascript at all, you'll already be part-way there. Based on a quick look at the AngelScript site, writing AngelScript code will look something like original C++, and unlike many scripting implementations is statically typed, but like other interpreted languages handles memory management for you (no raw pointers or heap allocation).

The problem with Python is that it has a big footprint. I have heard of AngelScript being used in game engines before to provide their scriptability, and this is probably because of its familiar syntax, small footprint without external dependencies, and easy integration with app/libraries built in C/C++. If its primary purpose is to create macro-like automation within 3DC, it should do fine. Where Python really shines is where it is exposed as an external interface, letting TDs build glue/workflow applications in Python that drive e.g. Maya, Motion Builder, LightWave, etc. in production. With some apps such as Maya, you can effectively write plugins in Python as well. Neither of those may be 3DC objectives though.

Link to comment
Share on other sites

  • Advanced Member

Go here to download and test the Lag fix and report your finding in the notes section. Windows version only at the moment.

The download file is in the notes section.

http://3d-coat.com/m...view.php?id=909

The lag is like 97% fixed in this new build! (I say 97% because it still has an ever-so-slight lag), but it is still very usable now! Back to how it used to be. Thanks for the fix!

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

×
×
  • Create New...