This is a great success story of Blender - being able to make changes within an organisation to have a competetive advantage. I don’t think that having freelance developers paid for their work and contributing that code back into Blender would have any negative impact on this.
I can help you do the build for OSX. You can also run OSX in virtual machine or on Intel PC too.
I think @smilebags does bring up a good point, namely that if the development is an active effort to get into Cycles ASAP, then people are more likely to pay.
As it stands now, @bliblubli intends on contributing his code to Blender once the year is up. However, there’s no guarantee what he has will make it into master. We don’t know if what he does conflicts majorly with what Blender devs want for Cycles and if the E-Cycles as it stands by the end of the year would be ultimately rejected.
However, if E-Cycles is worked on actively with direct feedback from Blender devs right off the bat, that seems more promising. And I would think that would lead to more people funding the development.
About Patreon: See my answers E-Cycles - Faster cuda rendering and here E-Cycles - Faster cuda rendering : Short version, already tried. Patreon is good for lot of small updates, not big projects. For example, I’m working on a light cache, it’s not something that can be done in one month on free time and I can’t quit my job and hope Patreon will be enough.
About releasing after a year: With my funding attempt, last year, as soon as the patch was committed, and the buildbots had the patch, all donations stopped immediately. With as I said a very poor donator/downloader ratio.
So a render farm keeping all it’s improvement secret, with no plan to releasing it to public and making a good amount of money with feature films is a success story. Me working on my free time for years for free and now trying to fund my work and teaching how to do is a thing to fix?
@0o00o0oo Why would it work better this year than last year? It’s not like I never tried funding with freely available source and builds. It’s win-win in the beautiful theoretical world I also lived in for 2 years, but it’s win-loose in reality.
And see https://armory3d.org/fund for example, very cool engine, they struggle to get funded.
You mean this developer: Improved OpenCL build beta,
https://developer.blender.org/D2254, https://developer.blender.org/D2254#53699, Patch for free view during transform operations :), Cycles 1.5 to 2x faster interior rendering for GPU and CPU ? I don’t remember another dev who tried to work on Blender based on direct donations, but I may be wrong.
If you are ok to learn to do your own build, take the course, you can then do the builds as often as you want, you even get more perf as some more fine tuning can be done for your specific GPU and I also teach how to find optimal performance. And there are also other thing like new modifiers, using branches, so you can even build E-Cycles on top of mantaflow or fracture modifier
I’m happy for you, but unless you’ve been specifically authorized, you shouldn’t be revealing any such thing on a public forum. It may not be illegal, but it’s always good form to keep your customers confidential.
Of course, I asked and they allowed. It’s actually normal to put the name of your clients when they allow. Just an example: https://juliacomputing.com/ full of clients logos on the front page.
Thanks for the links. Remember me again how many cool things got lost and how many good devs left.
Say no more, colored wireframes. People wanted but so what
By the way, a lot of people battled against the blendermarket addons at the beginning, saying addons should stay free or saying that GPL would make everyone to get free copies. They also are using the Blender API thus Blender’s code made by other and are behind a pay wall. But in the end, it works pretty well for many devs and I think the addon ecosystem got much better.
Oh, I didn’t realize you tried last year. I thought development like significantly speeding up Cycles would be something people would want to fund.
BTW, I think you have every right to make money off your efforts. I was just postulating, since your goal is to ultimately have it accepted into master, if it wouldn’t be better to actively work on integrating it from the get-go, and if that wouldn’t encourage people to fund it more.
Yeap, i would propably already left Blender to other dcc if there was no movement in addons like Hops, MM, uvpacking etc.
Yea, it would be better if they were part of blender or at least free. They are behind grey-area of paywall, but without money those addons wouldn’t be created at all.
Ok, question if I may, company pays dev for optimizing renderpipeline who contributes back to OSS afterwards is somehow more advanced than Freelancers paying freelance Dev for optimizing renderpipeline who contributes back to OSS afterwards?
I think Piotr has a point here, advertize with your client´s list on your homepage and perhaps link to the list.
Only very few companies have a deep desire to be displayed in a free vs paid debate
I agree, again it grew to many pages pretty fast. I modified it. @smilebags thanks for the suggestion, but I think it would be good to look for alternatives in another thread. We all agree free is better, the question is how make it viable. This thread is about E-Cycles.
Edit: opened a thread to continue the discussion about better ways of funding coders How to fund specific features/enhancements in a viable way?
I guess this lovely project here is also meant with this
They are entirely different. A studio making changes and merging those changes into master is equivalent to an individual making changes by themselves, for free, then merging those changes into master. Here, people are paying for the promise that these changes will be merged into master (among other things). If the public was paying the studio to do the work under the assumption those changes will end up in Blender, then they would be similar, and that would be a similarly bad idea.
No one here is debating whether developers should get paid. If you would like to understand my reasoning for my view on this and why it isn’t a good idea, feel free to head over and take a read.
I’m open to discussion, but I’ll speak with my customers before taking any decision. I think we all agree there is room for improvements to the working conditions of external developers and in general about the coordination of external development. Anyway, for this discussion the thread is here. And in the mean time, I will now concentrate on my work.
What is the problem here - that things may not end up in master? Since anyone who’s paying gets access to the source code, those changes aren’t locked away. Anyone who wants those changes in master can bring them there, right now.
Is the big difference really just about that where that open and free source code is stored?