Welcome to Fractal Forums

This poll has one and only one option: Yes
This is because the point of this poll is to see what support the proposal below has.

---

Mandelbulb3D is currently a 32 bit Delphi application, which will sooner than later become unsupported.
I really enjoy using Mandelbulb3D and fear that it will sunset before it's usefulness comes to an end.

To expect the fractal forums community, to come to it's rescue in terms to brute force coding is something I fell is unlikely.

However, there are people who for, a fee, will spend  the effort  required to get this task done.
The main blocking point is money - this could be overcome by crowd funding (e.g. Kickstarter).

If you are willing to contribute your hard earned cash to a Kickstarter project whose sole purpose is to recode Mandelbulb3D into a more supported language (like 64bit C) - vote Yes

Not to be discouraging; this project would be simply impossible or very close to impossible :(
Why don't simply start a new similar raymarcher similar to Fragmentarium but using 64 bits in C, that simulates MB3d without being identical?
And even if I deeply love that proggy. And huge respect to Jesse. Honestly it has become outdated and too limited with his hardcoded stuff. There are some new awesome projects out there with realistic mindblowing effects (the magical knighty cloudy shader) and kali's xray.
Also the code is intrinsecally "rigid" and full of limits that could never be removed (assembly everywhere). Great for speed and I love speed. But is everyone really happy with that???
Modestly I am constantly trying to bring new content and I constantly hit limits ...even for "simple" formulas it is really hard to port them - like say a normal Perlin noise.
Do a flexible raytracer do not resurrect MB again... look at the future :)

With respect to Darkbeam's position, I am not sure I agree 100%.

The hardcoded bits help performance, but are invisible to the end user.

What the end user sees is an interface and the results they get from the program.

And, for me, no other 3D Fractal Generator even comes close to MB3D's interface and way of getting results!  I get around the program and have a lot of fun with minimal or no difficulty.

With literally every other program of it's type, I find myself spending more time figuring out how to do what I want to do than actually doing anything.

I have never once gotten an image I liked from any other Mandelbulb renderer.  And in some cases I never even got an image at all!

Obviously, this is a shortcoming on my part.  I admit that!

But regardless of fault, I get what I consider to be great results from MB3D and would love to see it updated!  Jesse didn't make the code available so we could give it a decent funeral.  I think he'd approve of this.

That said - I can't vote in the poll since I am so broke I can barely afford most of the essentials lately.  So I support this project in every way except financial   O0

That said - who accepts such commissions?  And what sort of support do they offer for bug fixes, new feature requests, and that sort of thing?  Or would the code be public?  If so, do the people you pay expect to retain any rights to the finished product?

A lot of tricky stuff like that needs to be addressed...

I agree in the ease of use - In fact I meant that creating a "similar" program the programmer can (and should) preserve the look and interface; the code should be just ... new to allow new extensions impossible now. :beer:
In every case any extension of mb3d would make necessary a new m3f format with all the annoying consequences!!!
As for the closed source hmm... Should not be allowed if you keep mb open source code as a base.

The goal of the Kickstarter project would be to provide funds for a phased recode by people other than myself (I do not have the skills for something like this).
The whole idea is keep preserve the features of Mandelbulb3D by having them rewritten into a language that has a large following (e.g. C) and allow people to play with the internals to make those features even more awesome at a later stage.

If there is sufficient interest, my next step to would be to essentially put the recode out for expression of interest for an Agile rewrite.
Freelance programmers can then give their input into what they personally can deliver and how much they would charge to produce something of value.
Note: 'something of value' does not mean that they need to rewrite the whole thing - just provide something that works,  is of value (meets expectation) and can be incrementally improved on.

Regarding questions of ownership of the source code:
The rewrite source code will be open-source, public, free and allowed for inclusion of any other free unrestricted application.

Regarding bugs:
For the paid recode work - there will be a warranty period where Freelancers will be expected to support their work.

Regarding new features:
Out of scope for the rewrite - new technology that preserves the features is fine, new features are noise and not wanted for the rewrite.

I think there is actually no need for this. When I maintained the MB3D project, many people said "Thanks for what you are doing, but I do not need it, I'm just pleased with the version as it works". It was a little surprise, but I accepted it, and now I even agree. It works good as it is for most of the users.

Cheers!

While I have great respect for all the work you did and love the improvements you introduced, I have to disagree that there is no need for a 64 bit version.

Even if no other changes are made (actually - Especially if no other changes are made!), that one thing should be enough to justify the effort.

At least in my view.

Many things have changed since Mandelbulb3D was originally conceived. I think a better idea would be to take the best features from the top 3 or 4 fractal programs and bring them together with fresh code based on the current (and future) state of hardware, software, programming and interfaces, rather than trying to patch and fudge pascal into another language.

Don't get me wrong, I don't think this is a bad idea but I also think that it won't bring the desired result. There is pascal to C translating softwares, why not crunch it with that and then clean up the generated code? well, because it's not a straight over conversion, would have to be done in pieces and would take some effort to make it all work. Not to mention the logistics when more than 2 or 3 people get involved.

There is also only one incentive for investment and that is the love of fractals, that's why I put time into Fragmentarium, no other reason.

I deeply agree, and I see here the main problem: "to recode the app in C" is just a huge bunch of work, but not an interesting work at all. To create the same program again needs a lot of dedication because it lacks challenge, and in case of MB3D the expectations are infinite.

Regarding the pascal-to-C-translation: the worst problem is that major parts of the code are in assembler, not pascal.


Cheers!

I attack this problem indirectly from a different angle, since many years: teaching as many people who will listen about ray tracing.

In my opinion the problem isn't really one of C/C++ versus other languages, for example something like C# would be great. GLSL / Fragmentarium is something of a dead end for various long-winded technical reasons, Patryk Kizny bumped into them long ago.

If there is some kind of bounty on the problem (with clear feature requirements) it would most certainly attract my attention, and I'm sure several other people as well.

Indeed it does work  well.
As long as 32 bit applications are supported, it will continue to work.

However, in it's current state, any bright ideas for improvements reply on a very limited pool of people.

I deal with offshore resources (e.g. India) and have been involved in some crowd-sourcing projects where people are paid based on being better/faster at the task than others (i.e. only those that get their first get paid).
It is startling how much effort those people will put in ... just for the possibility of a small reward.
It doesn't put them off that it is tedious.

yes, not a straight over conversion, would have to be done in pieces and would take some effort to make it all work. I actually tried converting some of it but quickly realized what a huge task it would be.

PK (I think) was looking for something other than a program to explore math and fractals with.

Name your bounty, how much would it take to get you seriously interested? That could set the kickstarter minimum goal ;)
Feature requirement? convert to a more widely used language. C,C++,C#

I understand the idea and have no doubt that it works, but it will still require benefactors (investors require a return, benefactors don't) that are willing to support the effort for the love of fractals.

That is the magic of Kickstarter - there are no investors, just people interested in financially helping an concept to become reality.
There is  a requirement for a pledge to be rewarded, but the definition of reward is very broad.

As simple user I love and adore MB3D.
Where this program really shines are its stability, abilities for realistic colouring and lighting and its absolutely brilliant intuitive UI (have yet to see a better one).

But I also see its drawbacks: as the program is now, deficient memory-use, cpu-only, inaccessible code and as soon as 32-bits support is gone it'll be dead.

As far as my wishes go, all I really want is a black box with the same UI and the same or better outcome  :D

My point: to generate large enough sums I think there would have to be something more than a "token" reward.

No so, having looked at other successful software based kickstarter projects - it is quite feasible to have recognition based rewards.
People are willing make small contributions just because of the 'WOW I helped make that happen' factor.
Also we are not talking about big sums here ;)

We can easily add a "Medal" or something underneath a user name indicating they contributed!

It would only be a reward for forum members, but I've seen other forums use that type of graphic for various things, so it's one of many "intangible" rewards that could be offered.

Hard coding a list of names of contributors in a splash screen or elsewhere in MB3D is another idea, but if it's open source that may not work.

But the point is that intangible rewards can be done and there are many options for them.

I haven't researched software based kickstarter projects yet I am enthusiastic to see it work :)

I am pondering about the next steps to take.
With other pulls on my time - it will be a while.
However, we have to at least Windows 12 before anything needs to be done.

That's easy!

The next steps should be - when you are ready   O0

You only get one unveiling of a new project.  Make it the best you can, even if it takes longer than you'd prefer.

I also had the kickstarter idea some days ago, so it must be a good idea  :tongue1:  ^-^
Quite the same interface but a clean code as base should be the first target to keep that project alive. Result should be open source.
When we mastered this step there will be improvements. Maybe we can use the donation amount as a kind of voting score. The people that donate the most can give most point for a new feature, so they will reach "their" goal faster.
But also without that I would donate something (>50€) just to keep that wonderful program alive and not frozen.
So Bryan, please go on and start the thing  :thanks1:

Greetings,

I am mandelbulber collaborator seeking perspective.
Entered the game around mid-2016.
Upon initial inspection, I operated Mandelbulb3D for about 10 mins, then about 10 mins of Mandelbubler and chose.

What is the main diff between the two code-bases?
Given adequate LICENSE status of Mandelbulb3D, why don't we just absorb the language and cherry pick the stuff you seek?
Mandelbulb3D is not future proof, perhaps we can just absorb the missing features into next couple releases of Mandelbubler?

Is the team of Mandelbulb3D still available?
It is my opinion that these codes require merge for community unity.

Wow.

Lots to address there.

In reverse order, I don't think the codes unifying will help.

The big distinction between the two programs has nothing to do with code at all.  It's the User Interface that makes people choose one over the other.

I personally have issues getting along with Mandelbulber's UI and have yet to make a single image I like from Mandelbulber, but have had a lot of fun exploring and playing with MB3D.

The original MB3D "Team" was 1 person (our member Jesse) and he has moved on, leaving the code for us to pound our heads against.

Thargor6 has had the most success with that and is responsible for the latest release.

It's a mashed up Delphi / Assembler thing and from what I hear it is excessively difficult to modify.

As for absorbing what is there, I don't see why not.  But as stated above, it's not what MB3D does, it's how it does it.  There will be no way to unify the two interfaces that I can think of.

I still go back to 'Bulber periodically, just to see if my tiny little brain can "make it go" - but still no luck.

I keep meaning to do a screen cap of my efforts and pinpoint exactly where I get lost, but that's work.  I use fractal generators for play, so it will take the right combination of free time and mindset to make that happen.

If anyone wants to work on dressing Mandelbulber up in a Mandelbulb3D suit, and make it work via a similar interface, that would provide the answer to your question.  No code integrating needed.  Unless there are features that are extremely different!  The lighting comes to mind, with image based lighting in MB3D being a real plus and a feature I use a lot.

Good luck with your effort, and I hope what little insight I offered helps.

Very interesting information.
Mandelbulber uses QT5 for GUI, so anything is possible and we have freedom of infinite revisions.
We are simply limited by the number of heartbeats we are willing to allocate,  ;D

There exists a planned GUI overhaul for Mobile support.
Perhaps this is the opportunity to mold the bulber into a more user friendly window.

If the only issue is GUI, consider it solved.

Please think hard and list other features that we are missing.

Thanks!

Simple formula: actual coding >> talking about it :D

This is moving off the original topic. But here is my two cents:

Optimistic :)

To update Mandelbulber UI , we need advice  from the new users. For me it is all straightforward to use  ;D  My complaints with the UI are not easily solved as I have only a certain amount of laptop screenspace, not enough to cover all I would wish to have on the main UI.

.  The  main work is formulas, and a lot of it is way beyond my abilities and understanding (as some of you will know by now  :dink:), and I am restricted by time.

I would love  to see M3D updated into c, then I would be able to understand a lot of the early formulas that are not documented, and I would be able to check my implementations.

Tabs?  They can work better than different windows for everything in some cases.

Wow, the MB3D formulas are literally written in ASM nested in Pascal.
https://github.com/thargor6/mb3d/blob/master/formulas.pas

This is a major rework to convert to C.