Poll

3 features you want the most:

PBR Style material (Disney, Unreal Engine, etc..)
22 (12.8%)
Render-time booleans (cut/slice objects)
8 (4.7%)
Dedicated CarPaint Shader
1 (0.6%)
GPU/Hybrid rendering
20 (11.6%)
Speed improvements
17 (9.9%)
Cryptomatte
8 (4.7%)
Geopattern
14 (8.1%)
Sketch/Toon shader
5 (2.9%)
Reworking tone mapping (DSLR-style tonemapping)
34 (19.8%)
Interactive rendering in 3ds max viewport (with gizmos, object selection, manipulation, ...)
5 (2.9%)
Adding own materials to Corona Material Library
4 (2.3%)
Lightmix extended to materials, textures, ...
8 (4.7%)
Curvature map
6 (3.5%)
VR Scans compatibility
3 (1.7%)
Dedicated fabrics shader
8 (4.7%)
Decals workflow
9 (5.2%)

Total Members Voted: 63

Author Topic: The most wanted feature?  (Read 276706 times)

2019-01-06, 16:11:52
Reply #600

lupaz

  • Active Users
  • **
  • Posts: 513
    • View Profile
You forgot the improvements of SSS. It got much better and is a major plus for realism.

I have the impression that the vray compatibility and all the openvdb and Inside volume has to do with Chaos Group wanting to sell PhoenixFD.

Your comment was a little bit different before you've modified it. If you don't share your opinion things will never change, thats why I did it as well. I would like to find out if I am the only one out there with that opinion or not.
Nothing bad will happen if you tell them the things, nobody will ban you or cancel your subscription. I would like to make Corona better, thats all what I want.
Life is pretty simple, they sell something, we buy something.

Who else share their opinion with them if not us?

I felt I wasn't being constructive.

In any case, I think they get the picture:
We're frustrated with the recent development not being targeted towards archviz, or there's no development towards improving the quality (realism).

In general it seems that innovation stopped. Compare features in 1.5 and 1.6 with those of 2 and 3... It's pretty obvious to me.
« Last Edit: 2019-01-06, 16:25:19 by lupaz »

2019-01-07, 00:48:35
Reply #601

Jpjapers

  • Active Users
  • **
  • Posts: 1351
    • View Profile
I agree that there seems to be a fair chunk of development in areas that the majority of the user base (at least based on the cross section of this forum) would rarely use or notice in an already powerful and fast render engine and there seems to be less development recently on things that would make the renderer more competitive and useful to its archviz user base or at least the majority of it.

For example 2d displacement, clipper and pbr material have been asked for for a long time. Autobump and openvdb are cool but nowhere near as useful in the majority of cases. I completely understand when Tom says they have to be careful not to break the core features of corona we all rely on day to day to make money so smaller features are obviously going to be less risky than overhauling the way corona outputs colour or handles pbr. Having to take a 'roll your own' approach to pbr in the mat editor is just a little cumbersome however and there's no doubt in my mind that the corona team are considering it for a future release. Hopefully v4.

 Archviz is such a huge part of the user base here obviously and there are features that should have priority that have been pushed back release after release. It feels as though the thing that brought so many to corona was the fact that it fit so well in the Archviz pipeline but for taking that huge commitment of converting an entire production house to Corona there seems to be little payoff in terms of new archviz-centric features since v2. Perhaps we were spoiled with them initially.

I emplore the devs to be a little more open with the user base about what features are being seriously considered for each release and what are less likely to make it into the next version. its only recently more and more users here have began voicing these frustrations. I feel they are entirely justified considering how often large features from this poll are missing from major releases despite being highly voted for years.

Just as a closing note I said it before that memory optimisation and speed improvements shouldn't be on this poll at all as it should be expected at each release. When the poll was reset i mentioned this because they were dominating the poll and yet again they are scoring highly and it feels like a waste of a vote. It subtracts from what the thread should be about. New features.
I also still think a 'small annoying things' thread would be great for voting on smaller workflow adjustments that aren't new features or bugs.

This is all with the absolute highest respect for the team and the work you do. For. me, corona has made working in 3d so much more fun and creative again and for that I thank you!
« Last Edit: 2019-01-07, 18:07:56 by jpjapers »

2019-01-07, 02:31:48
Reply #602

zokni

  • Active Users
  • **
  • Posts: 8
    • View Profile
We definitely could use a more trustworthy and reliable development roadmap (trello).

I am pretty sure I am not the only one who checks trello time by time for the planned new features. It even worse (it happened to someone else I don't remember who was it) if you make some project plans according to the promised features and then that feature cancelled later.

Again, I know its easy to say, much harder to do. More or less I know why it is like that, sometimes you change something and 20 other features will stop working and the deadline is a deadline for everybody. But try to be just a little bit more accurate with it if possible.

It is very disappointing when you promise a feature (caustics is a good example) in every major release and then it just pushed forward and forward and forward.

I am not calling you to account I am begging, please work together a bit more closely with your customers and try to listen more to their requests.


PS.: One more thing. Probably you have already noticed that from my grammar that english is not my native language. I was born in eastern Europe and that time Czechoslovakia still existed.
I exactly know what meaningless polls mean.
« Last Edit: 2019-01-07, 04:20:05 by zokni »

2019-01-07, 08:51:17
Reply #603

Jpjapers

  • Active Users
  • **
  • Posts: 1351
    • View Profile
We definitely could use a more trustworthy and reliable development roadmap (trello).

I am pretty sure I am not the only one who checks trello time by time for the planned new features. It even worse (it happened to someone else I don't remember who was it) if you make some project plans according to the promised features and then that feature cancelled later.

Again, I know its easy to say, much harder to do. More or less I know why it is like that, sometimes you change something and 20 other features will stop working and the deadline is a deadline for everybody. But try to be just a little bit more accurate with it if possible.

It is very disappointing when you promise a feature (caustics is a good example) in every major release and then it just pushed forward and forward and forward.

I am not calling you to account I am begging, please work together a bit more closely with your customers and try to listen more to their requests.


PS.: One more thing. Probably you have already noticed that from my grammar that english is not my native language. I was born in eastern Europe and that time Czechoslovakia still existed.
I exactly know what meaningless polls mean.

I'd probably recommend not making project plans based on an arbitrary road map of software you don't control.
« Last Edit: 2019-01-07, 09:18:36 by jpjapers »

2019-01-07, 12:10:42
Reply #604

bluebox

  • Active Users
  • **
  • Posts: 238
    • View Profile
Agree on most things in at least 10 previous posts.

We are the ones tat got backlashed by this trello-promise of clipper/slicer. The thing is, at the time everything that was on the trello road map was pushed into stable releases so we had no reasons to restrain ourselves from taking it serious :)

Maybe the team wasn't aware of how serious people take this kind of stuff.

Anyway, our fears/doubts are voiced. I encourage everyone that feels the same to speak up, and in the meantime I'm eagerly waiting for the update of the road map or at least some kind of real debate on what features the majority of users would want implemented in upcoming releases.

2019-01-07, 12:33:54
Reply #605

TomG

  • Corona Team
  • Active Users
  • ****
  • Posts: 3316
    • View Profile
Howdy all!

Great to hear your opinions, ty for sharing those! A few thoughts in reply :)

- Software developers have two options really - say nothing about what they might be working on to save disappointment when things change, but it leaves all the users in the dark; or be open and share what we are / might / would like to work on, which keeps the development process open to users but then does have disappointment when things change (and it's always WHEN and not IF, things always change with software development).

We chose the second, but of course that has the catch that then people can get upset if something gets moved back. We still think that is the "lesser of two evils" and so we'll continue to share. We do our best to keep Trello updated, we'll see if we can improve how we do that.

No matter how much we improve it though, Trello or any other announcement of things we'd like to work on - or even things we are working on - should never, ever be taken as a promise, as someone called it. Even a long way into developing something, it can all be found to be unworkable and have to be abandoned. So, we will share what we are doing, but do please remember the realities of development when share. We can't stress enough that it is usually impossible to promise anything in development, and so we never do, until it's actually in a release :)

- On features and their priorities, everyone always has their own opinion. For example, there was an opinion here that the NVIDIA Denoiser was really nothing major, just something generic. Well first, it still takes a lot to implement it, just because it is NVIDIA's Denoiser doesn't mean making it work well with Corona is trivial. Next, in fact, our implementation is better than most ;) We get good denoising in reflections and refractions, that is not always achieved by others using the same NVIDIA denoiser.

And next, this "just a generic feature" to one person is a huge game changer to others, as we've heard from other users. So what is the greatest thing since sliced bread to one person, is just some generic feature to someone else. This is true for all features of course, the NVIDIA one is just an example. We could pick the material library too, also said by someone in the post to be not very important except for newbies, while we've heard from major archviz studios that it is an awesome addition.

So in that regard, the last few releases have had their features that many DO regard as game changing, and critical to their workflow (and yep, that's archviz workflows!) Everyone will always see a feature differently. You can't please all of the people all of the time, as the saying goes :)

- Don't forget that we also mentioned a lot of work went into things behind the scenes to improve stability, and clean up the code base. This becomes necessary once the software grows past a certain point (if you want to keep the code stable, easy to add new features to with a minimum of risk of breaking things, and so on). Sure, we could avoid doing such "busy work" as it isn't very visible or exciting, and pile new feature on top of new feature, but this brings a lot of pain and hardship for users down the line (I am sure everyone can think of software where this has happened, and is aware of how that plays out!)

This kind of work may not be that exciting in that release, but it sure is exciting further down the line when innovation can continue!

- We ain't done being revolutionary and groundbreaking :) Whether the last few releases have got you excited or not, there will be things coming in the future that will. That will be made possible in part because of keeping the code lean and mean. Also, some of these things take time, they involve pure research first - for example, our planned implementation of caustics (planned, not promised ;) ). Our plan is to make it easier, and faster, than what we are used to for caustics. Whether that could have been done a year ago is open to question, it took research, and testing, to see if this was even feasible. So, sometimes the overnight revolution to rendering takes a while to happen :)

2019-01-07, 16:11:15
Reply #606

Jpjapers

  • Active Users
  • **
  • Posts: 1351
    • View Profile
Howdy all!

Great to hear your opinions, ty for sharing those! A few thoughts in reply :)

- Software developers have two options really - say nothing about what they might be working on to save disappointment when things change, but it leaves all the users in the dark; or be open and share what we are / might / would like to work on, which keeps the development process open to users but then does have disappointment when things change (and it's always WHEN and not IF, things always change with software development).

We chose the second, but of course that has the catch that then people can get upset if something gets moved back. We still think that is the "lesser of two evils" and so we'll continue to share. We do our best to keep Trello updated, we'll see if we can improve how we do that.

No matter how much we improve it though, Trello or any other announcement of things we'd like to work on - or even things we are working on - should never, ever be taken as a promise, as someone called it. Even a long way into developing something, it can all be found to be unworkable and have to be abandoned. So, we will share what we are doing, but do please remember the realities of development when share. We can't stress enough that it is usually impossible to promise anything in development, and so we never do, until it's actually in a release :)

- On features and their priorities, everyone always has their own opinion. For example, there was an opinion here that the NVIDIA Denoiser was really nothing major, just something generic. Well first, it still takes a lot to implement it, just because it is NVIDIA's Denoiser doesn't mean making it work well with Corona is trivial. Next, in fact, our implementation is better than most ;) We get good denoising in reflections and refractions, that is not always achieved by others using the same NVIDIA denoiser.

And next, this "just a generic feature" to one person is a huge game changer to others, as we've heard from other users. So what is the greatest thing since sliced bread to one person, is just some generic feature to someone else. This is true for all features of course, the NVIDIA one is just an example. We could pick the material library too, also said by someone in the post to be not very important except for newbies, while we've heard from major archviz studios that it is an awesome addition.

So in that regard, the last few releases have had their features that many DO regard as game changing, and critical to their workflow (and yep, that's archviz workflows!) Everyone will always see a feature differently. You can't please all of the people all of the time, as the saying goes :)

- Don't forget that we also mentioned a lot of work went into things behind the scenes to improve stability, and clean up the code base. This becomes necessary once the software grows past a certain point (if you want to keep the code stable, easy to add new features to with a minimum of risk of breaking things, and so on). Sure, we could avoid doing such "busy work" as it isn't very visible or exciting, and pile new feature on top of new feature, but this brings a lot of pain and hardship for users down the line (I am sure everyone can think of software where this has happened, and is aware of how that plays out!)

This kind of work may not be that exciting in that release, but it sure is exciting further down the line when innovation can continue!

- We ain't done being revolutionary and groundbreaking :) Whether the last few releases have got you excited or not, there will be things coming in the future that will. That will be made possible in part because of keeping the code lean and mean. Also, some of these things take time, they involve pure research first - for example, our planned implementation of caustics (planned, not promised ;) ). Our plan is to make it easier, and faster, than what we are used to for caustics. Whether that could have been done a year ago is open to question, it took research, and testing, to see if this was even feasible. So, sometimes the overnight revolution to rendering takes a while to happen :)

Thanks tom. Just as an appendix to my two cents above im not saying what you guys have achieved thus far isnt awesome and game changing and im certain everyone finds use in different areas of the software otherwise the features wouldnt be there. I think the issue arises when you have this poll here that is massively user centric as you guys have always been, and there ends up being features included in the newest releases that werent on the poll and only a few have use for vs the larger and obviously harder to implement features. That seems to be the disconnect if you break it down. Again thanks for everything you guys have done so far! I dont want to sound ungrateful for the sheer amount of user interaction yourself and all the Corona team do. Its really valuable to have this level of contact. If only more companies were like this . Looking at you Autodesk.
« Last Edit: 2019-01-07, 16:18:07 by jpjapers »

2019-01-07, 16:25:47
Reply #607

TomG

  • Corona Team
  • Active Users
  • ****
  • Posts: 3316
    • View Profile
No ingratitude perceived! Just wanted to give a pretty detailed answer to some of the points, and they were from a mix of posts and people. Even that fairly long answer isn't the full story; as noted, the range of factors that have to be taken into consideration in deciding what is being worked on next is very complex, and we always do our best to try and balance between all of that. Requested features is indeed a big part of it, but there's a whole mix of other stuff too - and we'll keep doing the best we can, and see where we can do better, and of course always be making Corona better as we go :)

2019-01-07, 17:04:22
Reply #608

Ondra

  • Administrator
  • Active Users
  • *****
  • Posts: 8923
  • Turning coffee to features since 2009
    • View Profile
Thank you Tom for long answer. I wanted to answer few times, but I every time I deleted my post because it was too impolite.

I will just say that everyone is entitled to their opinion. However you must realize, that even features "that should have been already done" or "are not implemented entirely from scratch" or "are just optimizations that I expect to get for free with every release" nonetheless still take time to develop, test, and debug. If you decide that something adds 0 value just because it falls to one of these categories, then that is your opinion and you are entitled to it (and I am entitled to not take it into account).

This is why we are forcing you to choose between optimizations and new features - optimizing something takes time, same as implementing something new. This is just a fact of life. Also lot of decisions we take are based on feedback outside of forum. For example during the last tour in London, I got almost no requests for faster rendering, but most people want better performance in huge scenes (RAM, startup and shutdown times), so that is going to be optimization focus in v4. Only fraction of our users post here. In fact, many of the biggest powerusers simply do not have time to post here.
Rendering is magic.
Private scene uploader | How to get minidumps for crashed/frozen 3ds Max | Sorry for short replies, brief responses = more time to develop Corona ;)

2019-01-07, 18:05:02
Reply #609

Jpjapers

  • Active Users
  • **
  • Posts: 1351
    • View Profile
Thank you Tom for long answer. I wanted to answer few times, but I every time I deleted my post because it was too impolite.

I will just say that everyone is entitled to their opinion. However you must realize, that even features "that should have been already done" or "are not implemented entirely from scratch" or "are just optimizations that I expect to get for free with every release" nonetheless still take time to develop, test, and debug. If you decide that something adds 0 value just because it falls to one of these categories, then that is your opinion and you are entitled to it (and I am entitled to not take it into account).

This is why we are forcing you to choose between optimizations and new features - optimizing something takes time, same as implementing something new. This is just a fact of life. Also lot of decisions we take are based on feedback outside of forum. For example during the last tour in London, I got almost no requests for faster rendering, but most people want better performance in huge scenes (RAM, startup and shutdown times), so that is going to be optimization focus in v4. Only fraction of our users post here. In fact, many of the biggest powerusers simply do not have time to post here.

Just on the 'expected with every release' perhaps my wording was off. I meant its something that really doesnt need voting for because realistically its something im sure you and we all want. I didnt mean for it to sound like we didnt find it useful. Just it doesnt seem to be that useful in terms of a voting option. Who doesnt want it faster? :)
Again the work you, Ondra, and your team has done has for me made it much more fun to work in 3d again and i cant thank you guys enough. Ive converted so many people to corona over the years and im not likely going to stop!
« Last Edit: 2019-01-07, 18:08:48 by jpjapers »

2019-01-07, 18:09:02
Reply #610

lupaz

  • Active Users
  • **
  • Posts: 513
    • View Profile
Ondra, I think there's a miscommunication and a disconnect between the developers and the users.

I believe the users are questioning the usefulness of the developement of features in versions 2 and 3, and they are suggesting that the efforts of your team didn't go to the right place, and that as a result some of the really useful features were released incomplete.

Being transparent:

Corona 2:
1-PhoenixFD (Did the user base really ask for this?)
2- OpenVDB and CoronaVolumeGrid (Did the user base really ask for this?)(I can't use it without a VDB file)
3-3D Volume Materials (Not usable as global atmospherics in my opinion. I can expand if necessary)
4-Compatibility with vray (Did the user base really ask for this?)
5-Bokeh, Center bias and vignetting (nice to have, did the user base really ask for this?)
6-Material Library (Did the user base really ask for this?)
7-Corona toolbar (I can't imagine this being a big task, with all due respect)

Corona 3:
1-IR feedback improvements (Honestly don't notice a difference)
2-Nvidia denoising (excellent!)
3-Improved displacement (as others mentioned it's not ideal the way it works. In the video it's shown as if it works amazing, but it was used in a particularly easy case -large separation between stones, and surface detail. In real archviz cases at least I didn't get good results with setting at 3, 4 or 5 pixels)
4-Multimap (What's wrong with the free multitexture map?)
5-New united lister (How long did this take really? It seems to be incomplete too, no cameras or proxies for example)

Edit: And most of all, users are voicing their opinion so Corona focuses efforts in the future
Edit 2: OR to make sure they understand where corona wants to go so they can make an informed desicion to whether they stay with corona or not and what hardware they need to get if they decide not to.

In my opinion less is more. 
« Last Edit: 2019-01-07, 18:27:53 by lupaz »

2019-01-07, 18:41:38
Reply #611

PROH

  • Active Users
  • **
  • Posts: 1138
    • View Profile
@lupaz: I've seen user requests for every single point on your list on this forum!

2019-01-07, 18:52:09
Reply #612

lupaz

  • Active Users
  • **
  • Posts: 513
    • View Profile
@lupaz: I've seen user requests for every single point on your list on this forum!

I'm sure you did. I'm talking about a majority. The user base as someone else called it.

2019-01-07, 19:43:21
Reply #613

Ondra

  • Administrator
  • Active Users
  • *****
  • Posts: 8923
  • Turning coffee to features since 2009
    • View Profile
Just on the 'expected with every release' perhaps my wording was off. I meant its something that really doesnt need voting for because realistically its something im sure you and we all want. I didnt mean for it to sound like we didnt find it useful. Just it doesnt seem to be that useful in terms of a voting option. Who doesnt want it faster? :)
Again the work you, Ondra, and your team has done has for me made it much more fun to work in 3d again and i cant thank you guys enough. Ive converted so many people to corona over the years and im not likely going to stop!
yes, it is something everyone wants... but it still takes time to implement, so the question, how much exactly do you want it? Would you be happy with speed improvements at cost of dropping 1 potential big feature? 2 big features? 3? ;) Same with memory optimizations...
Rendering is magic.
Private scene uploader | How to get minidumps for crashed/frozen 3ds Max | Sorry for short replies, brief responses = more time to develop Corona ;)

2019-01-07, 19:47:05
Reply #614

Jpjapers

  • Active Users
  • **
  • Posts: 1351
    • View Profile
Personally I'd like to see standards like aces and pbr implemented sooner rather than later but that's just my opinion as you said. But with those being high on the poll I'd guess alot of others feel the same too

Edit: I completely understand it takes time to optimise and I hope you and Tom and the rest of the team in turn aren't seeing this as any sort of attack on your skills or anything. For me its more about understanding the teams priorities because currently there is no reliable indicator.
« Last Edit: 2019-01-07, 21:02:56 by jpjapers »