Welcome to Fractal Forums

How good of an idea would it be to set up a distributed network for some awesome fractal pictures, and especially animations? Okay, it probably won't cure cancer or find aliens, but we'd see some jaw dropping results to show for it if enough people joined in.

Just thought I'd throw the idea out there...

i had the idea of managing an ultrafractal5 server list of users, so that each user can register when he is online,
and every user can load such a dynamic ip list to set up its server list, i believe at least for large render projects
this would be great if more than 100 people are constantly in ;)

greets

Count me in.....once I have bought a decent PC :):)

It would be ace if we could get projects outside of ultra fractal too (e.g. nice custom 3D raytracing stuff), though that might be tricky. Also afaik, each computer in the UF network would have to buy the $140 version.

We could have a forum/thread containing 'candidate' projects, and the most popular (through voting or however else) can have priority. Sort of a queue system could be developed, where a judgement can be made according to a: how good a particular render will turn out and b: how quick it will take. Short-time, more interesting projects would go to the front...

I'd pitch in too :)

as far as i know does an uf5 server not need an registered uf5, just the user who is using the server must have it ... it is simple watermarking of the image, but i am not sure about it

perhaps i should ask on the uf-programmers list ... i will do it right away ...

 O0

The trial version will work, one does not need a licensed version to run the server application.
 

So there won't be a "Unregistered copy" watermark problem with the trial UF5? If not, I will gladly join in with my little Toshiba notebook. Every last clock cycle counts, right?  :P

yes, it would work, but we would need a technical concept
- dynamic updated server list that is downloaded by hand before starting uf5
- bug the programmer for using an http xml url to retrieve an updated server list

do not know really how this could be done easily ...

i have had a small farm with 4 computers running last time, and the deeper you zoom, the more you rely on many servers to support the actual image...

 :'(

This reminds me by the way of the Electric Sheep screensaver... No more is necessary to say, I guess :P

That would be SO COOL. I think, if it happened, the first projects should all be some totally "uber" zooms of the mandelbulb; maybe some really high res/high iteration images of the different "species" of phenomena of the mandelbrot, trying to build a "zoo", and then some of the fractals that go up "quarticly" (as opposed to quadratically) with resolution, such as the Supermandelbrot and Superjulia.

My question is (I don't know to much about this) - why would we need a server through UltraFractal? If enough people joined, maybe some kind of bitTorrent would work, with tasks just listed on Google Sites or something... then, as the images were processed, they could be stored on PhotoBucket, and then compiled together and saved on YouTube if it's a movie. Would that be at all possible?

I like this idea. A lot.

I would be happy to provide some CPU time and be involved in some coding.
Right now though, I'm balls deep in a nother project.

Yes it would be possible.  But setting it up, managing it, getting people involved (and having them stay involved), and all the other organizational aspects is another story.

Having something established that uses Internet-connected computers, like in the SETI@home scientific experiment, is a much easier way to generate hundreds of images.
 

Could it be done through BOINC? Would probably make things a lot easier. BOINC is a platform distributed computing; anyone who downloads one piece of software can then choose any multitude of projects to work on, which it will then cycle through.

it would be the first BOINC projekt i add on all my friends computers having boinc installed :D

i think it is an open standard and, the programmer of uf frederik slijkerman should have to set it up to make use
of it ....

and i do not know how efficently ultrafractal would handle those nonconform global tables formulas which
are necessary for global illumination purposes on single images .... that would be hard, but it would be enough
if the uf network could render frames of an animation easily ...

btw. i always think that i am rendering some stupid test parameters of students sitting somewhere testing
some theories when running boinc ( from the filenames of the tasks )