3dcoat edits, features.. (demonstrated with a presentation and videos)

[NinjaTaco]

I really didn't want to have to create a bunch of videos and toss it into a presentation. But having used this tool for so long it pains me to see that some aspects of the tool are overlooking the obvious. Simply put, 3dcoat is still feeling rough and unrefined in a variety of ways. So this an attempt to patch it up and speed up user interactions by showing good examples. This is also an attempt at fixing some bugs, and adding some features. I shouldn't have spent this much time making these videos but I did.

https://docs.google.com/presentation/d/1HPiMFMrXek7-OFDTcF2fvBPQO3MIdam4uB2HfzCjdxg/edit?usp=sharing

[Sandu]

 Thank you for putting forward a lot of good suggestions with great care. For developers, I believe this is very interesting suggestion. But after absorbing the features and advantages of so many software, I feel that this is a very difficult task for developers. For the whole software, it needs long-term thinking and top-level design

[Boozer]

A mod or two will say thank you and nothing will change.

[zf3d]

6 hours ago, NinjaTaco said:

我真的不想创建一堆视频并将其扔到演示文稿中。但使用这个工具这么长时间后，我发现该工具的某些方面忽视了显而易见的事情，这让我很痛苦。简而言之，3dcoat 在很多方面仍然给人粗糙和不精致的感觉。因此，这是一次尝试通过展示好的示例来修补它并加速用户交互。这也是修复一些错误并添加一些功能的尝试。我不应该花这么多时间制作这些视频，但我还是这么做了。

https://docs.google.com/presentation/d/1HPiMFMrXek7-OFDTcF2fvBPQO3MIdam4uB2HfzCjdxg/edit?usp=sharing

Some suggestions are great

[Gorbatovsky]

13 hours ago, NinjaTaco said:

I really didn't want to have to create a bunch of videos and toss it into a presentation.

I'm still on vacation this week, but nevertheless I looked at everything.
Thank you very much for your cooperation.
I haven't digested everything yet.
For me, the most painful issue is a user-friendly interface.
Moreover, we are starting a completely new stage.
Watch video here  

We get a lot of criticism for the development of low-poly modeling. But nevertheless, we are starting to introduce surface and solid parametric modeling.
Naturally, I have a desire to make everything as convenient as possible.
I already understood (thanks to our users) that it is better to make it more convenient than just creating many new tools.

[NinjaTaco]

I think what is painful to see with 3dcoat is seeing how a feature is implemented and how close a feature is to feeling somewhat complete. But it has a lot of minor blemishes that detract from the user experience. How minor are these things is somewhat subjective. I'll give an example..... Going to the modeling tab click 3d primitives and it defaults to a geodesic sphere. In my 29 years of doing 3d animation I cannot recall a time when I needed a geodesic sphere. Maybe I will one day. But if you are going to have a default primitive shape. A cube seems like a pretty safe default shape to have. This is a very minor blemish and a lot of these blemishes add up overtime. Imagine every user that has used the 3d primitive tool and immediately changed to a cube or a cylinder immediately. This needs some context sensitivity because there are hotkeys that are setup that serve no purpose in some tabs.
~ is somewhat pointless in modeling.
I feel like you could have something like a marking menu to quickly pick the primitive shape that is desired.
A better thing would be to just default to the last primitive that was picked and a hotkey to reset the primitive to its default settings. Also the 3d manipulator giving the user everything is sometimes worse than just allowing the user to use standardized hotkeys of W, E, R for translation, rotation, scaling because the user has to slow down their actions.
Look at manipulator. There is a strong chance I am going to occasionally mis-click due to the close proximity of these icons on the manipulator. When using the primitive tool, W, E, R aren't doing anything useful. But for some reason, G is bound to move and uses the blender hotkeys z x y... But you didn't bother to include the usage of S for scaling or R for rotation. (which I've already explained why those are terrible hotkeys)
Like I said. Minor blemishes. That hits harder and harder the more the user runs into these. It is like getting hit with a spoon. I imagine a lot of these blemishes could be fixed in a week.

[rubeos]

17 ore fa, NinjaTaco ha detto:

Non volevo davvero dover creare un mucchio di video e inserirli in una presentazione. Ma avendo usato questo strumento per così tanto tempo mi addolora vedere che alcuni aspetti dello strumento trascurano l'ovvio. In poche parole, 3dcoat si sente ancora grezzo e grezzo in vari modi. Quindi questo è un tentativo di correggerlo e velocizzare le interazioni degli utenti mostrando buoni esempi. Questo è anche un tentativo di correggere alcuni bug e aggiungere alcune funzionalità. Non avrei dovuto dedicare così tanto tempo a realizzare questi video, ma l'ho fatto.

https://docs.google.com/presentation/d/1HPiMFMrXek7-OFDTcF2fvBPQO3MIdam4uB2HfzCjdxg/edit?usp=sharing

Thank you so much for this, I think it will be very useful to me.
In my opinion, another guide on workflow modeling, how to make voxel, surface, low poly and now also mesh to nurbs interact fluently would also be useful..
I see that 3DCoat is a very powerful software, with unique advantages in the 3D landscape and yet this abundance of tools and possibilities disorients the beginner.
There are too many overlaps of concepts...for example the concept of resolution...or the tab/rooms.
Many things, tools and functions should be merged to make all the software more streamlined and intuitive to understand. Kind of like Plasticity. I am aware that changing the layout and the current logic of 3DCoat involves a great deal of organization and thinking.

[NinjaTaco]

5 hours ago, Gorbatovsky said:

I'm still on vacation this week, but nevertheless I looked at everything.
Thank you very much for your cooperation.
I haven't digested everything yet.
For me, the most painful issue is a user-friendly interface.
Moreover, we are starting a completely new stage.
Watch video here  

We get a lot of criticism for the development of low-poly modeling. But nevertheless, we are starting to introduce surface and solid parametric modeling.
Naturally, I have a desire to make everything as convenient as possible.
I already understood (thanks to our users) that it is better to make it more convenient than just creating many new tools.

I'm afraid you will receive similar criticism for other new features that aren't well refined. This is what I think is happening.

You implement new feature. It has problems. People just avoid the feature because it is too much trouble to report every single problem they find. Then the developer assumes the feature is sufficient and the user is satisfied.

Given that I am not looking for glitches in 3dcoat but I keep finding them with ease suggests that the users are either not using these tools or simply not reporting the glitches. I think in the past users (or at least I was 10 years ago) were forgiving of 3dcoat's shortcomings because Andrew seemed to be developing the tool by himself and so users felt that they were lucky that 3dcoat existed in the first place.

But some of these problems have lingered on for far too long. These minor blemishes. Glitches. Strange problems that are still here. 

For example. 2d primitive glitch where I can't go up to 60 sectors. It won't allow it.

The transform tool seems to inherit the brush settings of the previous tool but then initializes those settings during selection. For some reason the transform tool wants to reset itself to align to the normals of the selection. Not something I want it to do but there isn't any type of setting like, "use local, use normals, use world"... These are the type of refinements that 3dc is in desperate need of. It needs some kind of continuity to its transform manipulator which it lacks. It is weird that some transform manipulators use W,E,R hotkeys and other transform manipulators are using blender hotkeys. That is obtuse functionality that turns users away.

I sincerely believe any new features you add without fixing even just the basic nature of the transform manipulator is just going to be ignored by users and tossed into a pile of features to avoid. I imagine fixing these features isn't particularly fun as a developer. Adding support for tabbing between numerical inputs is more of a chore but that is the level of refinement that is expected of a professional tool. I don't even know why a separate number field pops up to input a numerical values, that seems hacky and unnecessary. Why not just use the numerical field itself?

 

[NinjaTaco]

The Quick Access menu is awful. The buttons move around. Just gray them out. Please don't make the menu rebuild itself. If an option is gone just make it dim! Why are the buttons moving around so much?! This is just another anti common sense thing that 3dcoat does.

Moving the buttons around defeats the purpose of "quick access". Just dim the buttons. Yes this will make the menu even more convoluted unfortunately but at least everything will be stationary. Hence, the presentation explaining. "Quick Access" is convoluted.

Why is "clear selection" at the top and then at the bottom?! Even better question. Why is there a clear selection button?!? Can't you just click away and drop the selection. I think most of us know how to use the shift key to add to a selection or ctrl to remove from the selection. That would help.

Edited August 11, 2023 by NinjaTaco

[Andrew Shpagin]

Thanks for presentation! First step - I re-made gizmo plane controls.

[NinjaTaco]

On 8/11/2023 at 12:16 AM, Andrew Shpagin said:

Thanks for presentation! First step - I re-made gizmo plane controls.

I guess I'll have to check it out soon. I have updated the presentation with more videos. And more glitches I found along the way. If you need more context. See the presentation..  Note these are very small changes.

 

-That will help to refine 3dcoat just a little bit more. Locking the cursor's position down while editing a numerical value is a common thing.

-Maintaining the position of a primitive is important because typically the user is focused at the location of the preview primitive shape and not at the world coordinates of (0,0,0). This would be helpful with sculpting primitives as well.
 

 

Edited August 12, 2023 by NinjaTaco

[NinjaTaco]

Like I've already said. 3dcoat just needs to sort out the minor blemishes. I used to overlook these things because I assumed rendering artifacts weren't actually affecting the model or at least I hope they weren't. And some rendering artifacts at times seem unavoidable. Zbrush still today has rendering artifacts and those users seem to accept it as a cost of using zbrush. (and honestly I've seen the zbrush rendering artifacts appear in other apps on occasion) But some of these 3dcoat rendering artifacts seem to only exist for 3dcoat's rendering method. I am assuming that 3dcoat is refreshing the screen continuously even when nothing is happening because the only time this seems to occur though is when 3dcoat is the current active window and in retopology mode.
You'll see in the video the lines begin to march when there is no activity. And then stop and then begin again when there is some activity. Sometimes it will happen continuously even when there is no activity.  I had to zoom in to make this apparent but it happens when zoomed out as well.
I imagine this is one the difficulties of rendering lines to a computer screen. The visual representation is an approximation between two floating point vectors.

 

Edited August 12, 2023 by NinjaTaco

[L'Ancien Regime]

I just discovered your mesh to NURBS functionality. Very interesting. I think I was one of if not the first to recommend that you go the NURBS route to make 3D Coat the full spectrum modelling software. There are problems in topology that are best served by voxels. Others are best served by polygons, or your variable density surface sculpting. Hopefully your NURBS package is in its infancy. Perhaps eventually it can even morph from NURBS to BREP.  The same with your cloth. Once you go down that road all sorts of  opportunities to use the physics of cloth as a modeling tool in its own right for phenomena that are best served by that technique.  Whatever the case, thanks to the developers and Andrew for not only being workaholics but also for being forward thinkers at the cutting edge of 3D Graphics. Remember, Andrew was the first to come up with autoretopology and he did it after reading a white paper from SIGGRAPH and then developing and implementing it in 3D coat in about a month and a half. It was a remarkable achievement. I was at a Vancouver Siggraph a few years back and one of the top developpers for Zbrush was at their booth. I overheard him having a private conversation with someone and saying that they'd hired 27 math PhD's to develop their autoretopology tools.  No such luxury for Andrew.

 

Я щойно відкрив для себе вашу функцію перетворення сітки в NURBS. Дуже цікаво. Думаю, я був одним з тих, хто рекомендував вам піти шляхом NURBS, щоб зробити 3D Coat програмним забезпеченням для моделювання повного спектру. Є проблеми в топології, які найкраще вирішуються за допомогою вокселів. Інші найкраще вирішуються за допомогою полігонів або ліплення поверхні зі змінною щільністю. Сподіваємось, що ваш пакет NURBS знаходиться у зародковому стані. Можливо, з часом він може навіть перетворитися з NURBS на BREP.  Так само і з вашою тканиною. Як тільки ви підете цим шляхом, у вас з'являться всілякі можливості використовувати фізику тканини як інструмент для моделювання явищ, які найкраще підходять для цієї техніки.  Як би там не було, дякуємо розробникам і Ендрю за те, що вони не тільки трудоголіки, але й далекоглядні мислителі, що знаходяться на передньому краї 3D-графіки. Пам'ятайте, Ендрю був першим, хто придумав авторетопологію, і зробив він це після того, як прочитав технічний документ від SIGGRAPH, а потім розробив і реалізував її в 3D-платформі приблизно за півтора місяці. Це було видатне досягнення. Кілька років тому я був на виставці SIGGRAPH у Ванкувері, і на їхньому стенді був один з найкращих розробників Zbrush. Я підслухав його приватну розмову з кимось і почув, що вони найняли 27 докторів наук з математики для розробки своїх інструментів авторетопології.  Ендрю не може дозволити собі такої розкоші.

 

 

[img][/img]

And wow, this is really good news. 

[quote]

I'm still on vacation this week, but nevertheless I looked at everything.
Thank you very much for your cooperation.
I haven't digested everything yet.
For me, the most painful issue is a user-friendly interface.
Moreover, we are starting a completely new stage.
Watch video here  

We get a lot of criticism for the development of low-poly modeling. But nevertheless, we are starting to introduce surface and solid parametric modeling.
Naturally, I have a desire to make everything as convenient as possible.
I already understood (thanks to our users) that it is better to make it more convenient than just creating many new tools.[/quote]

Edited November 21, 2023 by L'Ancien Regime

[L'Ancien Regime]

On 8/10/2023 at 8:52 AM, Gorbatovsky said:

I'm still on vacation this week, but nevertheless I looked at everything.
Thank you very much for your cooperation.
I haven't digested everything yet.
For me, the most painful issue is a user-friendly interface.
Moreover, we are starting a completely new stage.
Watch video here  

We get a lot of criticism for the development of low-poly modeling. But nevertheless, we are starting to introduce surface and solid parametric modeling.
Naturally, I have a desire to make everything as convenient as possible.
I already understood (thanks to our users) that it is better to make it more convenient than just creating many new tools.

 

Edited November 21, 2023 by L'Ancien Regime
Accidental clumsiness