I’ve been mulling over an idea for a while now, and wanted to present it to you guys to tear apart.
There’s a lot of interesting papers that come out of SIGGRAPH about new concepts in modelling, simulation, optimisation and so forth. I was wondering if maybe once every month or so we take an idea from one of these papers and try and implement it in maxscript.
For example, a recent paper I was reading was on creating cities based on population density maps and L-Systems. This would be fairly easy to do in maxscript I believe, with the right people on the job.
Now obviously there is a limit to what can be done (maxscript is not known for high-speed analysis of large amounts of data), but i’m sure the community would find a few gems that could be useful, if not just for learning new ways to tackle problems with maxscript.
I see this as seperate from the challenges, as it would require scripters who are comfortable handling larger projects.
What do people think?
I think it is an good idea.
But projecting the participation in challenges to this eager project it propably will end as “club of the dead maxscripters” (never seen this movie so I can be wrong).
I guess this is due to the fact that most most of the hardcore scripters in this forum either have a full-time job and/or a life. Personally I try to avoid both but somehow don’t find the time to participate in every challenge.
Beein that said, I think we shouldn’t get distrubed by the lack of “high-speed analysis of large amounts of data”. Sure everybody her wrangles daily 1000000000 iterations of nested execute for-loops, which slows down production. But what you are proposing here is more a theoretical approach in pushing the boundaries.
I would be interested in knowing if the real life studios haven’t already implemented a lot of this stuff, but just don’t tell because it is their capital?
Personaly I would be interested in such a project for two/three reasons:
1: Building and researching a robust pattern language for maxscript. There are some good attempts in this forum like the structs thread.
2: Researching and developing interface standards for maxscript, as I see a nightmare of ‘.net candy’ in my dreams. This is sure a part of the pattern language thing.
3: Maybe attracting some cool c++ coders and developing a common language to interact effiecently. Bobo explained many times the maxscript/c++ workflow at frantic. There are times when “usability” beats “speed optimising”. But even if you capitulate to the lordship of the coders it is hard to find one who wants to implement a library.
Summarized I think this is a good idea to communicate. There have been some interesting approaches in implementing papers and concepts in this forum. But most of them silt up, because of the fact that the coders are under NDA and could only give hints of what they are trying to do. Or they were maybe studends who dropped their assignments before comming to the point. I think I dropped a few myself, caused by change in priorities or lack of interest in the forum.
A Index of Siggraph Papers, maintained by Tim Rowley:
http://www.cs.brown.edu/~tor/
Georg
Yes, we do. Plus the fact that most of the stuff im looking into goes way way over peoples heads for instance, getting people to understand the basics of peicewise polynormial curves took a whole thread pretty much. And thats a relatively simple summation with a basis function.Not even actual maxscript rollout stuff, variables etc – im talking algorithms.
The paper im currently reading are ‘wiggle splines using space time constraints’ and this alone would be too much for alot of people here, in the paper they essentially develop a new cubic curve type that allows realtime interactive oscillation and frequence base on tangent handles and complex numbers generating phase shifts. It needs understanding of sin and cosine waves, fourier synthesis and basis functions of cubic curves.
Its the biggest issue here i think not making the maxscripts but understanding core ideas of math, summation, derivatives etc etc So im always up for lending a hand here and there but currently a little too busy at work.
Thanks for the reply Georg! And you’re right, it would be a very small group of interest, but that may even work in it’s favour.
And yes, i’d like to drag Bobo into this if possible too.
Sounds like an interesting idea, but what exactly would the purpose of a ‘project’ like this be?
I mean, what do you want to do with the scripts that come out of it all? Release them for free, sell them, give them to autodesk to implement in max, etc etc?
I don’t know how other think about this, but developing scripts and giving them away for free just because you can is not something I’d see myself doing. There has got to be a bit more incentive for people to collaborate I think.
When I look at the collaborative section here on cgtalk I think that there are mainly two reasons for people to work for free:
-Because they want to get better at what they’re doing (to find a (better) job later). When you’re looking for (semi-) pros, this will be less the case.
-Because they have clear purpose or goal to reach for themselves. For example, they want a tool to do something easily.
My personal view on this is: why would I do something for free when developers at Autodesk do it for money? I would have to gain something substantial with it as well.
ps. this is not ment as a rant, more as a consideration.
My personal view on this is: why would I do something for free when developers at Autodesk do it for money? I would have to gain something substantial with it as well.
Nobody would be willing or even able to pay for any application that offers all possible features.
The purpose of such project could be to push the limits or find new frontiers. A maybe welcome relief form answering the same faqs over and over again.
The ‘problem’ of possessing knowledge is always is is worth nothing if it is not communicated and/or shared. The more you give the more you get. I am sure if I cannot offer gems like Bobo. But sharing it teaches everybody. And maybe you can get something back too – something that is not just another feature request.
Plus the fact that most of the stuff im looking into goes way way over peoples heads […] ts the biggest issue here i think not making the maxscripts but understanding core ideas of math.
I am not sure I should agree.
But I am currently preparing to open a ‘school’ addressing exactly that kind of stuff.
One of the first lessons will be: “The Sine and how it can help”.
There is a big gap between the applications and their users. And its getting wider day every day: Let the big button fix it. Apps like MAX, and Photoshop are interfaces to manipulate numbers after all. No button can hide this.
But what to you suggest?
Georg
The time it takes to answer those same faqs is peanuts compared to developing cutting edge stuff.
But still, I personally do not see much point in doing something just because you can. I could try walking on my hands each day until I’ve pushed my limits so far that I can run a 100 meter steeplechase on my hands. It would have been very entertaining to the general public, but it wouldn’t gain me anything.
I agree with the fact that knowledge is worth nothing if you keep it to yourself, but it can be worth a lot if you share it in the right way. And then it’s a win-win situation, rather than a serious win for the general public and a small win for yourself.
hmm… i dont know. All i know is basically i spent months just understanding basic math and math notation to start understanding stuff. Breaking an equation down into little chunks, understanding it bit by bit.
“… rather than a serious win for the general public and a small win for yourself.”
and
“Plus the fact that most of the stuff im looking into goes way way over peoples heads”
is an interesting combination.
I will need to think about this.
Could take some time.
Georg
I should elaborate that a bit I think. With ‘a small win for yourself’ I mean a smaller win relative to the ‘win’ you would have in the first situation.
Even though I’m generally very pro open source and ‘free’ software and all that, I don’t think that it should be persued to the limits. People should recognise the quality of products and reward the makers for it, whether it is made by a large company or a single person.
After all it’s not just someones time of the making that goes into it, but also all the time he spent building up his knowledge.
Understand math notation.
First I have to apologize to you – maybe there were some unlucky phrases.
I do appreciate your efforts in communicating your progress towards the use of (higher) math in digital content creation for several reasons.
I were lucky enough to go to school. It took me several years to see the need for what I learned 10 years ago, which was a pure theoretical pain back then. Now I spend alot of time reanimating this ‘knowlegge’ and transfering it to my everyday work.
This forum claims to be “For scripters and enthusiasts alike”.
If there is no hope in finding a way to communicate between ‘us’, they should change to claim to “For those who cannot read or whose direct download didn’t contain the helpfile”.
I am sure I can find a more abstruse usage for math and maxscript than you. But this is of no help. I am also sure you came to this forum with the same questions like everybody else.
A ‘club of the dead maxscripters’ would be a start to try to communicate about problems and finding a way to communicate aside of copy/pasting wikipedia and wolfram articles. It will be a slow, long and painfull way – but in the end there could be a paper that holds our names.
I sometimes get the impression that digital artist have alot to learn from those disciplines they left behind when starting their journey into cyberspace. There is Math and Physics, communication culture, the art of citing, how to write an abstract, how to build thesis and hypothesis … you name it.
Stuff waiting to be ‘discovered’ again without beeing lost.
A ‘club of the dead maxscripters’ wouldn’t offer solutions but could try to find answers.
And if there is something that I learned in the last 10 years:
If you aren’t ready an answer cannot help you – no matter how hard you try.
Georg
I think that it is a great idea, but as others have said I wonder if it will not get great support. Like most people I work, well in my case I run a company. I currently have three major clients on the go, all require scripting to be done, one is a full new pipe line for a series. Nothing as glorious sounding as what Eek is dabling in but it is what I do for a living. I wish that I had the time to just learn more stuff, I have been reading the papers that Eek as but I think that I’m much further behind just due to time.
Here is my current list of personal projects.
More Math
More C++
More mountain biking
More time with the kids
This list was in reverse order by the way.
I think that it works well here just the way that it is. Some one has a problem that they need to work on and we all jump in and try to help. What would be cool is if we could just discuss here some the ideas that come up in Siggraph papers. Have a look at their applications and what areas could bennifit from using those ideas. At some point I’m sure some one will try to impliment the ideas in code that they need to write and we can all find out how it works.
Im certainly up for taking a paper im interested in and going through it bit by bit for people to understand. When i started getting the hang of this stuff I got a lot of help from programmers explaining the notation and how the structure of the equation works. Im still way off understanding the really hard stuff, as one character could mean a whole function. But slowly getting there.
A ‘club of the dead maxscripters’ would be a start to try to communicate about problems and finding a way to communicate aside of copy/pasting wikipedia and wolfram articles. It will be a slow, long and painfull way – but in the end there could be a paper that holds our names.
Yes it is painful, and no I dont have any sort of math degree whatsoever (i was terrible at math in school, but kind of gained a new aphinity with it since coming here) I’d be more than happy to show people how i started understanding it but it would be very slow and frustrating to start but would get better.