outoff

What I'm glad to see in near future with using Max5 - is possibility to "burn patch" to "blank" microscheme, may be it's sounds futuristic ) But it will be great to create audio or video interface, declare inlets and outlets and get it in hardware!

What are you know about this or the same technologies that we can use right now? 

Thanks for your time and sorry for bad english )

Hiya!
What I'm glad to see in near future with using Max5 - is possibility to "burn patch" to "blank" microscheme, may be it's sounds futuristic ) But it will be great to create audio or video interface, declare inlets and outlets and get it in hardware!
What are you know about this or the same technologies that we can use right now? 
Thanks for your time and sorry for bad english )
Cheers!


max-perspectives-hope-in-the-near-future

What are you know about this or the same technologies that we can use right now?

That's a pretty interesting and complex question...hardware can be so tricky and costly to produce, and generally doesn't have much flexibility unless you program that in from the start. But a lot of that is changing (and has been) with FPGA chips and other examples that are similar. I know only a little about these, but they're very cool: the general idea is that one can program certain kinds of logic on them (which could be simple math processes chained together), then they can process data extremely fast as they have tons of parallelism for the grunt work. So you can really optimize certain things, and reprogram them as many times as you want (or at another level, they can be programmed to reprogram themselves, including on-the-fly or based on conditions.) They can also be simulated, benchmarked, and optimized before you ever put your program on the chip itself.

So it's a long ways from simply making a Max patch and clicking "Burn".... which would be insanely complicated to do for a lot of reasons... but there's definitely potential in these chips to handle some of the more intensive raw processing that you routinely do in Max, like with signal or matrix data. Typical computers can handle most of what one might conceive of doing with audio stuff, except for tons of channels or really intensive processes...though we of course push the envelope a lot when we have the horsepower.

Multiple videos of appreciable resolution is a lot more difficult at this point, and would really benefit from this---like being able to send a matrix and have each row (or block of pixels) processed in the same time as each pixel was before, due to the parallelism. Or if your process relied on information from the whole matrix, you could send the raw data sequentially, but with a bunch of streams at once, which are then all processed at once... so with video, you could have a frame buffer of the # of parallel threads, thus processing video in these "chunks" of frames (perhaps better in a non-realtime application). Or you could have multiple input streams (many videos) all going through the same processing, on the same chip.

Plus there are fairly inexpensive FPGA-type chips out there already which do a lot of this, or manage things like synchronizing different frame rates or resampling the matrices to other resolutions, so that the matrix calculations will match up. And the re-programmability (some of which can be done very quickly, by the chip itself) allows you to use one chip for a lot of different things at different times.

The chips use their own languages, which have their particular learning curves, but there's also a number of C compilers for them, which vary considerably in price (including free) as well as how robust they are. I imagine you could get quite a ways with this free one

if you were doing relatively simple processes on your data and knew some C. From there, you can keep building complexity, and route your information from your input sources (audio/video/data in) to your outputs, provided you have a way to do this "around" the main chip itself (using a computer's hard drive or RAM, or building a hardware unit with inputs and outputs).

Seems pretty interesting and a different way of looking at programming and data processing. Max could definitely be a good "top layer" which manages the logic and data streams, pumping them into and out of these high-powered chips which can do certain things very very quickly.

outoff wrote on Tue, 26 May 2009 06:30Hiya!
What I'm glad to see in near future with using Max5 - is possibility to "burn patch" to "blank" microscheme, may be it's sounds futuristic ) But it will be great to create audio or video interface, declare inlets and outlets and get it in hardware!
What are you know about this or the same technologies that we can use right now? 

So it's a long ways from simply making a Max patch and clicking "Burn".... which would be insanely complicated to do for a lot of reasons... but there's definitely potential in these chips to handle some of the more intensive raw processing that you routinely do in Max, like with signal or matrix data. Typical computers can handle most of what one might conceive of doing with audio stuff, except for tons of channels or really intensive processes...though we of course push the envelope a lot when we have the horsepower. 

The chips use their own languages, which have their particular learning curves, but there's also a number of C compilers for them, which vary considerably in price (including free) as well as how robust they are. I imagine you could get quite a ways with this free one 

if you were doing relatively simple processes on your data and knew some C. From there, you can keep building complexity, and route your information from your input sources (audio/video/data in) to your outputs, provided you have a way to do this "around" the main chip itself (using a computer's hard drive or RAM, or building a hardware unit with inputs and outputs). 

Seems pretty interesting and a different way of looking at programming and data processing. Max could definitely be a good "top layer" which manages the logic and data streams, pumping them into and out of these high-powered chips which can do certain things very very quickly.


Thanks for the information and reflexion. I hope technology moves in this direction faster than I think!

Thanks for the information and reflexion. I hope technology moves in this direction faster than I think!
It can will be next revolution !


until actual hardware can have its circuits reporgrammed to adapt to system designs built in max, there's always the lemur:

http://www.jazzmutant.com/lemur_overview.php

yep, nice device! have some fun with using it in Max...

Actually, Interface-Z is working on an hardware which can receive small Max patches. This device isn't on the web site now, but, if you need it, you may contact them.

This device as analogic inputs, numeric imputs, transitor outputs, relay ouput and servomotor outputs.

Thanks, Michel! I'm never hear about Z, it's very interesting!

I'm works only with Make Controller Kit...

Around us many things that's very interesting, but needs some more knowledge that we(I'm) have...

I'm spend with Max more time - over 9 years, and now do all my work on it... and all my thing of this post - that we can do more and more purposes with lovely environment!

Thanks, Michel! I'm never hear about Z, it's very interesting!
I'm works only with Make Controller Kit...

There are only a few options. Max relies on an operating system. Either Windows or OS X. I got the older Max 4 runtime running with wine on Linux. Have only tried an early Max 5, which didn't work...

One solution: get a Mac mini, an interface, some glue and connect and glue them together. Place the whole into another box. Now simply configure the mini to automatically start your patch. If you need a screen get one (or a Lemur) and glue it into the same box...

All hardware boxes I know, have normal computers inside. All these VST hosting expensive solutions are still normal computers, but a little less configurable (and less to mess around and break things as well...) I would not pay extra for this - a little discipline, and a laptop setup can be as trustable as any hardware box, with the advantage that you can fix it yourself more easily and also configure it to your personal needs, as you do it already with your software... I have all my patches hosted at getdropbox.com. If my computer gets stolen or breaks, I can get back to my patches with a running MaxPlay on any computer within a pretty short time.

Recently my old PowerBook died a day before the concert. I got the new computer 2 hours before the concert. That was sufficient to setup everything...

All hardware boxes I know, have normal computers inside. All these VST hosting expensive solutions are still normal computers, but a little less configurable (and less to mess around and break things as well...) I would not pay extra for this - a little discipline, and a laptop setup can be as trustable as any hardware box, with the advantage that you can fix it yourself more easily and also configure it to your personal needs, as you do it already with your software... I have all my patches hosted at getdropbox.com. If my computer gets stolen or breaks, I can get back to my patches with a running MaxPlay on any computer within a pretty short time.
Recently my old PowerBook died a day before the concert. I got the new computer 2 hours before the concert. That was sufficient to setup everything...

, which has the footprint of 4 cd stacked (10x10x3cm). it runs atom and it costs 3-400$. the hardware is the same as found on those eee's but without the screen and with giganet

i have max running on the fitpc, which has the footprint of 4 cd stacked (10x10x3cm). it runs atom and it costs 3-400$. the hardware is the same as found on those eee's but without the screen and with giganet

is there any more info on this interface z?? sounds really interesting

is there any more info on this interface z??  sounds really interesting


In my opinion, the best solution to put Max running on an "piece of hardware" is more simple than create an specific machine to upload patches. 

Max could have another option on "build collective menu".

A great feature on future Max is to put an patch running like OS.

Build collective > select OS option > select hardware drivers folder > and "burn" bootable CD or USB, or even instalation CD to make premanent applications.

It could also have an translator. And you ask, an what??? 

An translator capable to translate an patch file to an executable one, WITHOUT the need of runtime, like other regular programs. Pure machine code runing.

I think the next step in Max is to allow the build of an independent software.

Other thing is an option to select the cpu core in patch ispector... that could be great too!

In my opinion, the best solution to put Max running on an "piece of hardware" is more simple than create an specific machine to upload patches. 
Max could have another option on "build collective menu". 

A great feature on future Max is to put an patch running like OS.
something like:
Build collective > select OS option > select hardware drivers folder > and "burn" bootable CD or USB, or even instalation CD to make premanent applications.
It could also have an translator. And you ask, an what??? 
An translator capable to translate an patch file to an executable one, WITHOUT the need of runtime, like other regular programs. Pure machine code runing.

I think the next step in Max is to allow the build of an independent software. 

brush up your french and go to:http://interface-z.com

Max Perspectives, hope in the near future...