What about the "Future" of Natron?

[giacomoalbe]

Hi all there!

The title of this topic is almost completely self explanatory, I’d like to know something about the future of Natron, with particular regard to topic such as the roadmap and development updates.

I introduced myself in the forum some time ago and I said I would have been available to help the devs to support this project and maybe improve it by connecting with other Open Source software in order to create a fully Open media video creation pipeline.

I also sent a PV to the founders and main developers of Natron, telling them that I would have been more than happy to start contributing to the project, but I haven’t received anything back in more than 3 weeks.

I’ve read many topics here on the forum with proposals of features/improvements to add to Natron, both for this release and for the future ones.

I also read a very interesting and impressive topic on UI enhancements that might leverage the usability and appeal of the software.

But I’m not able to get any piece of news regarding the future of Natron! No updates on the development status, almost anything that might betray that the project is still in an “active” state.
The only thing I was able to find is this topic that simply states that unless we are able to show that Natron is used also in commercial application and provide some showcase the core team cannot attract any funders to continue the development.

I think that the community is very interested in such a project, since it is a very good piece of software and it might be a very good replacement for other compositing suite out there that are not going to give the user such a level of freedom, customisability and community help. It is also free and open source, and I think we should defend this kind of project, both from a final user and a dev user point of view.

Is there anyone capable of proving such information on the future of the project? Is the roadmap still in progress? Is the main issue related with lack of proper financial aid or it is more about the core devs working on other projects?

I’d like to make this topic a way for the community to start thinking about how and wheater it is necessary or not to keep the project going and maybe help to achieve some interesting and still needed achievements, like a working 3D space, maybe some Deep Compositing capabilities, UI improvements and better integration with other FOSS projects, like Shotcut (or any other NLE), Blender, Audacity/Ardour and the Apertus/OpenCine project.

Let me know what you think, then! I can’t wait to see what we will come out with!

Best,
Giacomo

[ChanderTUTOS]

Hi giacomoalbe.

Before we enter on the previous topic we mentioned (a professional official showcase with all sort of good projects), I’ll wait at least until we get some news from de devs.

We need to know if their funding from Inria really ended and if it did, we need to know if the software is still being developed.

About the “proposals of features/improvements” I think this is a very delicate but also too heavy of a topic to extend on this topic. It’s very polemic.
It’s true that in order to match the basics to replace Nuke (and other softwares), Natron needs to have at least a 2.5D environment. 3D would be ideal. The UI is also super-relevant. But the devs made their point (on the original topic) and I think I agree with them: First Natron should become stable! The engine has to be unbreakable, it can not keep crashing! It has to become more reliable so companies can at least consider giving it a try.

So having all of this as a priority over those gorgeous enhancements you mentioned is the right thing to do (it’s just my opinion).

For example. I have a bit over 9k subscribers in my channel (I know it’s not a lot, but still is a considerable number of active viewers).
The majority of them keep asking for Natron tutorials and I intend to start from scratch and make a full list of Natron tuts. From basic motion graphics to heavy compositions. This is a good step at making it more usable and more popular and I think this could be my role at helping the community.

So I started experimenting with the new community plugins, and tried out making entirely procedural motion graphics in order to push the software and see how far I could go at teaching it. And guess what? Natron crashed so much I couldn’t continue. It was a logo animation with some combined displacements with some strategic alphas and glow filters in order to give a glass/glitch build up.

It wasn’t a heavy project but Natron couldn’t render it, coudn’t cash it properly (even with the cash to disk node) and I coudn’t continue the project. How could I start teaching something that can’t handle processing some basic procedural stuff?

The little it could process I could see that it was turning into something simple and yet amazing. I was going to use it as my channel new intro. But I had to give up on it and wait until Natron get more stable.

I can start the tutorials but I’ll have to stop it at a basic level until Natron can get stable enough to handle what it already has. That’s my point for having making it stable as a priority.

But as I said, this subject has it’s own topic and we souldn’t mix things on this one.

I suggest we focus on 2 things:

• Feedback from devs, what is the development status?

• The showcase/ Open Movie/ Any other ideas for making an official reel for Natron’s capabilities so far.

This are my suggestions, what do you guys think?

Let’s do our part to keep Natron going!

[giacomoalbe]

Hi ruzzi,

Thanks for your answer!

First Natron should become stable! The engine has to be unbreakable, it can not keep crashing! It has to become more reliable so companies can at least consider giving it a try.

I totally agree with you! I simply wanted to collect ideas on the future on Natron in one place, in order for us to find a simpler way to coordinate this effort!

The majority of them keep asking for Natron tutorials and I intend to start from scratch and make a full list of Natron tuts.

This is also very interesting! It would be very nice if we were able to create a team capable of working for the project in different areas, like development, website/adv and, of course, learning stuff.

We need to wait for other (especially @MrKepzie and @frederic_devernay, the core devs) to join this conversation and at least give some basic information on the development status!

As long as I saw, though, many people here on the forum are interested in collaborating to the project, we simply need a way to structure this in order to make changes possibile.

I’m very excited about this, I hope to be able to start contributing to the code (stability first, needless to say) as soon as possibile, but there are 525k lines of code between me and that target ahah!

@ChanderTUTOS: we should in some way extend the visibility of this topic. Maybe some reposting on the FB group or in other Natron forums? Do you know any?

[Thomas-Morley]

Ruzzi,

I’m not a programmer for Natron but I’m glad you are taking a second look at it. I too have been creating some tutorials and part of why I left Natron a year ago was it’s unstable core. I feel that the core is way more stable now. Most of the crashing issues I’ve had are user related issues. The big issue for me was memory management. The default settings do not work. It allows way to much system ram to be used and locks up the workstation. I’d love to hear what problems you are having so that we could narrow down these issues and kill them once and for all.

FOSS365 - Thomas Morley

[ChanderTUTOS]

Hi ruzzi,

Thanks for your answer!

The majority of them keep asking for Natron tutorials and I intend to start from scratch and make a full list of Natron tuts.

This is also very interesting! It would be very nice if we were able to create a team capable of working for the project in different areas, like development, website/adv and, of course, learning stuff.

Hi @giacomoalbe you are most welcome, let’s keep engaging on this topic until it’s as active as we need!

Yes, maybe assembling different teams/small groups, each one working on it’s own area is the way to start this journey.

Speaking for myself, I can be responsible for video advertising, explainer videos, tutorials and I will always credit Natron on my videos and incentive my viewers to get it a try.

I had an idea I think it may be very effective at getting users engaged and also make non-users at least curious, informed and excited about what’s happening and what’s new on the Natron world.

The idea is to make small Motion Graphics videos for each new release/sub-release or new feature that gets on the new builds (at least the ones cool enough to make a video of it). One small, easy to understand, right to the point, Video Copilot/HitFilm style update for each new update/ feature/ release and such.

If we can manage to do this consistently, at least we can keep old users on board and even excited to joining
the cause, and also possible new users will start paying more attention at Natron, as they will constantly see that things are happening and evolving.

Of course that in order to do this we will need the developers to work with us in order to set the script for each small video update. And it would be highly essential to publish this videos on Natron’s website.

I know it’s a big thing but please (everyone) let me know what you think of this idea and how it could be improved.

We need to wait for other (especially @MrKepzie and @frederic_devernay, the core devs) to join this conversation and at least give some basic information on the development status!

As long as I saw, though, many people here on the forum are interested in collaborating to the project, we simply need a way to structure this in order to make changes possibile.

Yes, I agree. Like the idea I mentioned before, almost everything we decide to do from now on to keep Natron going, be it in the development area, website and advertising, official courses and so on, we will need the developers to evaluate almost all of it so they would have to get on board with us on this. I know they are very busy but there will be things they will have to evaluate, approve, inform us, update/upload to the official website (if they agree with anything we’re suggesting) in order to be possible to get this things done.

@ChanderTUTOS: we should in some way extend the visibility of this topic. Maybe some reposting on the FB group or in other Natron forums? Do you know any?

About FB groups and forums, I only know this forum. The FB groups that I know are the offical one and also my group (Natron / Fusion Brasil). We should extend the visibility of what we are planning to do, but in order to do this we will have to reveal the news that Natron is going through a rough time and it may even stop being developed.
I know some people will want to take action like we are trying to do now, but also we take the risk of scaring the hell of new and possible new users. So I don’t know a good way to approach this. Any ideas?

[ChanderTUTOS]

Ruzzi,

I’m not a programmer for Natron but I’m glad you are taking a second look at it. I too have been creating some tutorials and part of why I left Natron a year ago was it’s unstable core. I feel that the core is way more stable now. Most of the crashing issues I’ve had are user related issues. The big issue for me was memory management. The default settings do not work. It allows way to much system ram to be used and locks up the workstation. I’d love to hear what problems you are having so that we could narrow down these issues and kill them once and for all.

FOSS365 - Thomas Morley

Hi @Thomas-Morley ! Thank for joining the topic!

About the problems I had, it was exactly what you just described! I couldn’t have said it better myself!

The default settings doesn’t manage the ram and CPU as it should and it ends up locking up the entire system!

I tried lots of different custom configurations for this settings but nothing worked and I would always have to reboot the system.

I was running only Natron 2.3.10 and nothing else.

I use Ubuntu 17.10 (but I also tried on Windows 10) and my specs are:

I7 2600
28 GB of Ram
GeForce GTX 460

If some of this issues could be solved I would already have enough to get back making tutorials.

I already have some scripts and ideas on projects I could teach for new and more advanced users, so it would be great to put it in action.

Thank you and look forward to know your thoughts about it all.

[Thomas-Morley]

[Thomas-Morley]

I think leaving 30% free is the ticket. (let me know if you find something better ) It feels like there are some memory leak issues when building your node tree as well sometimes. If things start to bog down I find saving, closing Natron and then re-opening the scene helps make it snappy again. ( This also feels like a memory leak issue to me ) I use SSD for cache drive too.

Open a Task Manager and view the memory usage as you work in Natron, it helps to trouble shoot any memory issues. let me know how you make out.

My setup is very close to yours.

Natron 2.3.10

AMD FX 8350 8 core
Ubuntu Studio 17.10
16Gb Ram
2 X GTX 970’s
128GB SSD for disk cache

[ChanderTUTOS]

Thank you @Thomas-Morley, I’ll try this settings and if I get any better results at all I’ll mention you here!

[Thomas-Morley]

To address the original issue on the thread. Yes development is still going on, that’s why we have version 2.3.10 It’s down to one developer and he has to share his time with other priorities. I do understand the frustration in spotty collaboration and maybe Omar and I can help with that. I think you’ll see more transparent communication in the near future. There appears to be a new batch of people trying to wrap their arms around the current state of the Natron project and help it flow better with the community.

[Stephenthomas0]

I’m a motion designer/animator and I mainly use Blender for my work. I started using Natron as a compositor because it’s more powerful than the built in compositor in Blender. I agree with the stability issues, had a project recently where outputting the render would crash Natron on Ubuntu, but it rendered out ok on Windows.

The main reason I’m posting though is because of the possible funding issues mentioned. This is a bigger concern than stability and new features, as Natron can’t be stabilised without the funds to pay developers to work on the code, obviously. I think a good sustainable solution would be to open a Natron store to sell digital assets. Or, even better in my opinion, a subscription model like the Blender Cloud. I’ve found their collection of models and textures invaluable in the work that I do and I know there’s new content being added all the time. Well worth the €9 a month subscription fee. If the content is high quality and useful, I’m sure many compositors (not even limited to Natron) would sign up to such a service.

An open movie project is a good idea too. A collaboration between Natron and Blender on this would be amazing, as I know one of their goals for the 2.8 project is to improve pipeline support. Plus, they have a lot of experience in this area.

[giacomoalbe]

Hi Thomas!

Yes development is still going on, that’s why we have version 2.3.10 It’s down to one developer and he has to share his time with other priorities.

It seems that you have more information about the project for the future of Natron, this looks very interesting and useful!

So, as you said, now Natron is being actively developed by only one guy, good to know but not good enough for the project to continue and get better over time!

There appears to be a new batch of people trying to wrap their arms around the current state of the Natron project and help it flow better with the community.

Who are you talking about? Are there new developers working on the code? If so, can they join this conversation in order to give more insights on the development and overall status?

The roadmap is stille the one present on the website and here depicted?

As previousely said, I’d like to take the journey of start contributing to the code, I’m currently doing all the steps needed for compiling it, studying and trying to solve some issues and many more things. I think this is going to take some time, but I’m very determined to fill the gap and start contributing soon.

Nevertheless, I think we (as community) need to get involved a bit more. The community needs to step up, I am very interested in making my part, I’d really like to see Natron used by many more pros out there and at the same time help less specialized user to get into the VFX world.

In order to do that, in my opinion, we need to create a better “structure” or at least to define some goals we as community togheter with @frederic_devernay for the next releases.

Maybe we should find some main arguments/topics that need some love () and create dedicated teams to address those issues/parts.

The main area I see are:

• development (choose new feature to work on, improve stability, improve UI, etc)
• website (create a more useful and informative website or add content to the present one in order to improve visibility
• marketing/advertising (spread the word about Natron and maybe create some commercial initiative in order to keep the project alive)

After that we might find a way to decide which task have which priority, ask the community what they think about them, make some polls or whatever and then start working on it, updating at a constant rate the community and getting them involved.

For instance, from a merely developer standing point I miss a lot the “quick fix list” that might get some devs onboard quickly and easily, for example, or a dedicated system to track development and coordinate devs in general.

I think a good sustainable solution would be to open a Natron store to sell digital assets. Or, even better in my opinion, a subscription model like the Blender Cloud. I’ve found their collection of models and textures invaluable in the work that I do and I know there’s new content being added all the time.

I totally agree with you and I was thinking the same! This would be a great way to keep things rolling and having lots of fun at the same time!

I’d also like to add the idea of adding a professional support system capable of addressing the needs of professionals. One of the reason Blender (which is by far one of the most interesting, useful and powerful 3D suite out there, needless to say, and with the 2.8 release we might get even further) is not as widely accepted as it deserves in the VFX/Animation/3D world the lack of professional support (they are working hard on it, so we’ll see something beautiful in the near future!). This might be a step into the VFX world in a useful and sustainable way, in my opinion.

The main thing here, for me, is to have a positive attitude towards this initiative, to see this as a community effort to bring this software to a new and more sustainable level

In the end: @Thomas-Morley it seems that you are one of the admins of the NatronNation FB group. Would it be possibile to make a post about this topic/initiative on the group in order to attract more people here to share their thoughts? This would be nice! Thanks a lot in case

Just my two cents! I hope to give back something more than this to Natron in the near future

[Stephenthomas0]

Doesn’t Natron already have a professional support service? https://natron.fr/support/

This has been mentioned about Blender before, large studios don’t want to jump on board a new platform without the safety net of a support service. Ton’s response is that he doesn’t want the Blender Insititute (or animation studio as it’s now called) to be a support helpline. I’m sure it takes a lot of resources to do well.

[giacomoalbe]

Doesn’t Natron already have a professional support service?

Yes, I think they have something, but I also read that many big studios are not interested in getting involved into Natron “until it becomes better than…”.

We obviousely need to target another market in order to get to that point, my idea would be a cloud based support service with different customer levels (and thus different price tags) with progressively demanding services (ticket based support, email support, 24h support, installation, maintainance, feature request, feature voting, plugin development, etc.) and custom support where needed. This seems to be a very common business model nowadays, and it might work for Natron too. In the end what we all need is a way to pay people working behind Natron and selling it as a service rather than as a product might be a good compromise and an overall interesting solution. Just my 2 cents, again, but I hope to start a conversation about this soon, too!

In addition, something like a “Natron Cloud” for anything related to learning and asset marketing seems a perfect fit for the project and, as you correctly spotted in your last post, as long as you provide quality content many people will be trusting it and in the end paying for it!

[Stephenthomas0]

Yes, I think they have something, but I also read that many big studios are not interested in getting involved into Natron “until it becomes better than…”.

Took the words out of my mouth there. I think a pro support structure is generally a good way to go, but as you’ve already said, the software needs to be rock solid before VFX houses even consider using it, let alone paying for a support service for it. It’s a bit of a chicken and egg situation. You need money to pay developers to stabalise the software, but VFX houses won’t be willing to pay until the software is more stable. If this is the main revenue stream for Natron at the moment, I can see why there’s only one developer who seems to be working on it part time. While the hard work of this developer is admirable and I’m very grateful as a user for it, I’d like to see a support structure in place so that he can fully concentrate on the project while having fellow developers to balance the load.

This is the way I see the current situation:

• Professional compositors need production ready software to get jobs done on time. Deadlines don’t often accommodate software crashing all the time or not producing the required results. It needs to be rock solid. A good benchmark would be Fusion. Fusion 9 is free to use, it’s being used in the VFX industry on huge productions and is completely cross platform. Natron needs to be rock solid and offer all of the features of the free version of Fusion at the very least to be a viable alternative for professional compositors.
• Hobbyists/professionals who aren’t specifically compositors (such as myself) are more forgiving of these problems and often willing to help improve the software if they feel they are part of a community. This is one of Blenders strengths. The community is insanely good, from bug reporting, creating tutorials, users helping users with problems, developers dropping in and creating new features (Animation Nodes is a prime example. Excellent feature that isn’t currently supported by the Blender Foundation, but it’s a feature that’s very valuable and the developer has created a revenue stream to help further development). Without the strong community behind Blender, I very much doubt it would have reached the heights it has.
• From what you’re saying, it’s not currently possible/very easy as a developer to drop in and work on the project. This seems to be ignoring the greatest strength of open source software and limiting the potential of the software and the developers who’d like to work on it.

For me, the solutions are as follows:

• Create a more robust developer framework that easily allows developers to drop in and contribute. Obviously their contributions should be checked and signed off by a trusted developer, but the system should allow a new developer to drop in, fix a bug and submit for review at the very least.
• Let the community help. It looks like the only way of supporting Natron at the minute is to be a company willing to pay for professional support. It seems as though this isn’t working too well at the minute. Allow for donations. Sell assets. Create a cloud service that offers worthwhile assets, training and a place for users to commune. Canonical offers a similar professional support service for Ubuntu, but they also allow users to donate to the project if they desire. If people want to give you money, let them…
• With the above in place, chances are that Natron will have more developers and more funding to make the software more stable. At this point, it’ll at least be an option for professional compositors and VFX houses who may be willing to pay for the professional support service.

[Thomas-Morley]

There appears to be a new batch of people trying to wrap their arms around the current state of the Natron project and help it flow better with the community.

Who are you talking about? Are there new developers working on the code? If so, can they join this conversation in order to give more insights on the development and overall status?

The roadmap is stille the one present on the website and here depicted?

image.png1235x456 64.5 KB

As previousely said, I’d like to take the journey of start contributing to the code, I’m currently doing all the steps needed for compiling it, studying and trying to solve some issues and many more things. I think this is going to take some time, but I’m very determined to fill the gap and start contributing soon.

I’m referring to everyone that is actively trying to help Natron move forward, and that includes you.

We can all agree that the website hasn’t been actively maintained for a while. It’s impossible to update that part of Natron without the current owner giving permission to do so. Updating the “official” website is a dead issue until access has been granted to someone. For the website, I don’t think it’s really about the money, there are free alternatives if funding is not available. I think Frederic Devernay holds the rights to the website.

I feel money for developers is not really a concern at this point. We have developers that are willing to help fix core issues, including you, for free. ( that’s what FOSS is all about ) Funding campaigns are great and will help speedup development but that’s a mute point without better collaboration. The elephant in the room is the apparent lack of communication, collaboration and management on the project. The rest of the ideas for funding can come quickly after this part is smoothed out.

It’s very apparent that Frederic Devernay is the key to making this change happen. He’s the “Master Control” in all of Natron. Someone that has contact with him, and I don’t, should ask him to appoint someone as the coordinator/project manager for Natron. Myself along with others are willing to help but that help can’t happen without him giving permission to do so.

[giacomoalbe]

I’m referring to everyone that is actively trying to help Natron move forward, and that includes you.

For an instant I thought you were pulling a rabbit from the hat in the form of new strong and capable developers, but nevermind

Someone that has contact with him, and I don’t, should ask him to appoint someone as the coordinator/project manager for Natron. Myself along with others are willing to help but that help can’t happen without him giving permission to do so.

I’ll be going through this, I’ll try to contact him and ask somethign about this “ownership” shift or at least enlargement.

I know for sure that core devs want the community to step up in some way, but without permission or right to do so it’s almost impossibile to make things happen!

What about extending this conversation to the FB group also?

[Thomas-Morley]

I know for sure that core devs want the community to step up in some way, but without permission or right to do so it’s almost impossibile to make things happen!

What about extending this conversation to the FB group also?

I can’t stop you from posting this topic on FB pages but overall I think it would just have a negative impact on Natron.

I’m discussing this topic on my Natron Programmers Group page as it’s a closed group. Many of the users in that group have already contributed great things to Natron. Lens Flare Suite, GLSL Shaders, Shadertoy effects etc. We are moving ahead in areas that we currently have access to and are looking at more ways contribute.

https://www.facebook.com/groups/1620803784859824/

[Thomas-Morley]

Frederic jumped on FB NPG and said the website and forums are looking like they will be paid for 1 more year. He will know for sure next week. There was also talk about him needing help with coding an optical flow node for Natron. If we have anyone with the skills to help in that area PLEASE step up and help.

“Frédéric Devernay I took a look at the slomoVideo source code a while ago, and it seemed to be doing image warping the wrong way (using backward warping instead of forward warping). See eg https://www.cs.unc.edu/.../research/fall08/lec08_faces.pdf”

" Frédéric Devernay the algorithm used by slowmoVideo (backward warping) is not good. it should really use forward warping, which is more complicated to code. If anyone wants to dive in it, it’s a few weeks of coding."

“Frédéric Devernay What the VFX world calls optical flow is really image morphing and uses what is called optical flow in Computer Vision to estimate the displacement between two images (t and t+1 or left and right).”

“Frédéric Devernay I took a look at the slomoVideo source code a while ago, and it seemed to be doing image warping the wrong way (using backward warping instead of forward warping). See eg https://www.cs.unc.edu/.../research/fall08/lec08_faces.pdf”

[fabiof17]

I personally feel very confident about Natron’s future, although it seems to be going through some turbulences.
They’re not to me. I was sure ,when I jump to Natron something like 2 years ago, that at some point its dev model would have to be drastically changed.
Let’s say that some emphasis should have been put on some very specific aspects of the app to make it interesting and valuable for studios.

At the time it would have been quite difficult to me to convince that it was the way to follow (assuming that i was right) and restrospectively, i think things could hardly have been different. The dev team did (and still does) everything they could, and the amount of work/time that has been put so far is quite impressive.
Now the past is the past and, so far, the quality of this discussion makes me think that Natron’s future could be a very bright one.

My personal little strategy was to first start to convert Nuke Gizmos ( as much as possible ), and then a few months ago jumping to GLSL, to give Natron as much visibility as possible. (we’ve almost reached 250 PyPlugs so far). It’s starting to give some results but took 2 years to get there. But it may be a very good timing in the end.

So, let’s feel confident, i think we’re doing a not too bad job