PDA

View Full Version : Future Direction



DynamoBen
12-12-2011, 01:59 PM
I'm currently converting my SX relay project (http://doityourselfchristmas.com/forums/showthread.php?4723-8-Channel-DMX-Relay-Pack) to the Prop. Which got me thinking about the future direction of the PropController. It seems to me adoption of the PropController is hampered by its flexibility. I'm wondering if I we should start creating single purpose boards like a DMX/Renard node, or AC dimmer, DC Dimmer or a Pixel controller. We could even leverage the power of the Propeller and make combination boards like DMX, AC Dimmer, and DC dimmer all on the same board.

I still like the idea and flexibility of the current design using daughterboards and would like to see that continue, but the standalone boards might make things more approachable for those they aren't into writing their own code. (BTW one of the big reasons I created the PropController was so people could learn how to code)

Thoughts?

lightman
12-12-2011, 02:13 PM
I am in. The power of the Propeller with multiple processors on one chip make it the ideal candidate for future, higher powered dimming/lighting systems.

It is a perfect candidate for DMX controlled systems, as one cog can manage addresses while the other cogs manage dimming or pixel nodes.

Daughter boards or full up boards?....whatever the masses prefer. Personally, I like single board dimmers with 8, 16, or 32 channel output drivers that directly control the lighting effect.

My 2 cents....
Lightman

DynamoBen
12-12-2011, 02:16 PM
Personally I need a controller that I can reconfigure since I use them one way for Halloween and another for Christmas, so that concept will live on. But I see other Prop based projects here that are being quickly adopted by people but only a handful of folks use the PropController. So I see two groups, devs and users. Devs want flexibility but users want build and play. I guess I'm wondering if we should cater to the users since there is a high learning curve between the two groups?

BTW regardless of direction all of the code a schematics will continue to be open source.



Daughter boards or full up boards?....whatever the masses prefer. Personally, I like single board dimmers with 8, 16, or 32 channel output drivers that directly control the lighting effect.


I could go as high as 128 w/o breaking a sweat.

Matt_Edwards
12-12-2011, 05:11 PM
Ben, to be honest, I purchased the PropController as it is a great development platform with a very very strong lighting controller bent.
I think the flexibility makes it too difficult to use as a dedicated controller.

DynamoBen
12-12-2011, 05:26 PM
I think the flexibility makes it too difficult to use as a dedicated controller.

So are you suggesting that dedicate purpose boards would make more sense moving forward?

chilloutdocdoc
12-14-2011, 12:55 AM
Ben, to be honest, I purchased the PropController as it is a great development platform with a very very strong lighting controller bent.
I think the flexibility makes it too difficult to use as a dedicated controller.

+1. I have both the DMX and Ethernet version, but I think purpose built things like the E1.31 to DMX/Ren bridge is what people really want to see.

Matt_Edwards
12-14-2011, 06:24 AM
So are you suggesting that dedicate purpose boards would make more sense moving forward?

Yes.

DynamoBen
12-14-2011, 12:03 PM
Any thoughts on which dedicated boards we should create?

DynamoBen
01-03-2012, 10:38 PM
I guess what I'm trying to figure out is if there is a project that needs a Propeller to get it done. The Prop is very powerful and often is over kill (especially at $7 each). I have to believe there is some project or idea out there that would benefit from a Prop.

In another thread I suggested a DMX/Renard bridge but that has been done before by RPM. With that said I am building my own bridge/node which will have an additional output and could include a WIFI interface. Not sure if anyone would be interested in something like that.

Thoughts?

CaptKirk
01-03-2012, 10:54 PM
If a more dedicated design were done, you could still allow expansion capabilites with optional inexpensive dual Berg style connectors. Think Lynx Express 16 where you plug in the wireless board if desired, but in this case it could be anything. Please also consider having as much "on board"/in-field managability as possible. Using a simple 3 or 4 7 segment display, you could allow for status displays (the name of the board, firmware revision, etc), configuring simple things like DMX start address, and for kicking off test modes and/or set-up modes (all on, all off, ramp all up/down/repeat, cycle through one at a time, blink all, etc.). Although it adds a little bit of cost, the extra capability and usability it adds is seriously nice and really makes the controller (I cannot get anymore) one of the best DMX DIY setups out there. It would be nice if we could also standardize on the enclosure (like a CG-1500 which I still have quite a few of, or the CG-2000 for mounting) as that makes life really easy also. I had started a concept thread (Renard RE16) where I dicuss doing this with PIC because we can get free samples (thank you Microchip), and the programming is straight forward via the PicKits, and lots of people are familiar with them. Switching to a prop means ANOTHER set of programming tools I need to get stuff for and remember how to use. Oh- on board programming capability is nice also (why people refuse to put a penny single 6 pin Berg on their pic designs is beyond me...)

Well that is my $.02 whatever that is worth these days...

<rant over> ;-)

DynamoBen
01-03-2012, 11:31 PM
If a more dedicated design were done, you could still allow expansion capabilities with optional inexpensive dual Berg style connectors. Think Lynx Express 16 where you plug in the wireless board if desired, but in this case it could be anything.

I have no issue with this, but I'm starting to discover the community isn't as interested in learning as it once was. So too much flexibility is a bad thing because you have to write code, there is a balance between functions and capabilities for most they want functions.


Please also consider having as much "on board"/in-field manageability as possible. Using a simple 3 or 4 7 segment display, you could allow for status displays (the name of the board, firmware revision, etc), configuring simple things like DMX start address, and for kicking off test modes and/or set-up modes (all on, all off, ramp all up/down/repeat, cycle through one at a time, blink all, etc.). Although it adds a little bit of cost, the extra capability and usability it adds is seriously nice and really makes the controller (I cannot get anymore) one of the best DMX DIY setups out there.

For my node project I had considered adding an LCD and switches to configure it and allow for those sorts of things and scrapped it for a webpage interface. Personally I prefer the LCD but it requires an enclosure of some sort to work well and adds some to the cost. The risk here is I design it the way I like it and no one is interested. Which is what happened when I designed the PropDimmer (http://doityourselfchristmas.com/forums/showthread.php?7863-Propeller-%28Prop%29-Dimmer), its got a lot of high end features that I love but no one was really into it.


It would be nice if we could also standardize on the enclosure (like a CG-1500 which I still have quite a few of, or the CG-2000 for mounting) as that makes life really easy also.

We could/should be doing this for anything that is likely to be outdoors.


Switching to a prop means ANOTHER set of programming tools I need to get stuff for and remember how to use. Oh- on board programming capability is nice also

This is a double edged sword. If we add on-board programming via USB it means SMT parts or expensive modules. If we go the serial route it means you need a serial port, which is hard to find on PCs these days. If we provide a programming header you need to purchase a $20 programmer. So then what does one do? I'm indifferent I just want to know what people prefer.

chilloutdocdoc
01-03-2012, 11:35 PM
I'm not going to bash the design of "easy to configure" but really... we should only be configuring our controllers once or twice per season. I don't see the point in spending $10-30 per board in adding a LCD, or buttons, or all of this stuff, when you can simply just reflash via ICSP or Set it up with a computer.

Maybe it's the cheapskate in me.

DynamoBen
01-06-2012, 11:41 PM
What about a DMX/Renard/sACN (or whatever else) tester? I've been situations where I would rather use a handheld controller instead of booting my laptop or I just want to isolate a problem, this could fill both problems.

chilloutdocdoc
01-06-2012, 11:43 PM
What about a DMX/Renard/sACN (or whatever else) tester? I've been situations where I would rather use a handheld controller instead of booting my laptop or I just want to isolate a problem, this could fill both problems.

I would SERIOUSLY get behind this.

critic2
01-07-2012, 12:07 AM
Personally I need a controller that I can reconfigure since I use them one way for Halloween and another for Christmas, so that concept will live on. But I see other Prop based projects here that are being quickly adopted by people but only a handful of folks use the PropController. So I see two groups, devs and users. Devs want flexibility but users want build and play. I guess I'm wondering if we should cater to the users since there is a high learning curve between the two groups?

BTW regardless of direction all of the code a schematics will continue to be open source.



I could go as high as 128 w/o breaking a sweat.

I am currently a user, but want to be a devs. I can't read this site quick enought!! Learning something new EVERY DAY.
I love DIYC. Yea, I'm hooked...

Chuck P.

DynamoBen
01-07-2012, 12:19 AM
I am currently a user, but want to be a devs.

Spin is a pretty friendly language to learn, nice to see another Propeller head in training. ;)

dirknerkle
01-07-2012, 12:37 AM
Spin is a pretty friendly language to learn, nice to see another Propeller head in training. ;)

But don't you get dizzy going 'round and 'round like that??? I fall down enough as it is... that would make it even worse for me!!! :lol:

Materdaddy
01-08-2012, 02:20 AM
What about a DMX/Renard/sACN (or whatever else) tester? I've been situations where I would rather use a handheld controller instead of booting my laptop or I just want to isolate a problem, this could fill both problems.

I would love this. One of my goals for the year is to design a DIYC tester (unless I can get somebody else to do it... Tom Sawyer anyone?). I don't know what all I want, but I do know of a few things:
* Send out Renard data (stand-alone). Test controllers via RS485 without the PC. Maybe simple ramps for a configurable amount of channels?
* Receive Renard Data - to see if a controller is repeating data and give some sort of diagnostics
* Test SSRs - Loop through your standard 1-4 channels of an SSR.
* Act as an SSR - This would be LEDs driven by a controller to test the SSR output from your controllers
* All of the aforementioned functions, but replace "Renard" with DMX/sACN (or whatever else).

DynamoBen
01-08-2012, 04:38 PM
That's a big list, I think for this tester I'm going to limit it to data only. I will most likely model it after this (http://www.goddarddesign.com/a.html).

I think everything will be through hole, but I will likely include the SMT USB parts on the board for those that don't want to buy a prop-plug. Initially I will focus on DMX and sACN support since those are easily testable for me and then widen the data type options down the road.

Zeph
01-26-2012, 09:17 PM
I just want to say that as a newbie to DIY animation and these microcontrollers (but not to programming), I'm really glad to have an option which supports user programmability. I've just discovered the PropController, and do not yet know what it's capable of, but I'm interested in learning development. Everybody using Spin seems to say it's fun and powerful, so that intrigues me. (I do want to get some simple and supported controllers like Ren 24ss working for this year before going too deep into development myself - of something which might not work).

I'm speaking up in response to the question of whether some in the community are still interested in development - interest is not dead!

DynamoBen
01-26-2012, 09:20 PM
I've just discovered the PropController, and do not yet know what it's capable of, but I'm interested in learning development. Everybody using Spin seems to say it's fun and powerful, so that intrigues me.

If and when you get to the point where you want to jump in I have some boards left over, just PM me.

Jeff Millard
04-30-2012, 05:27 PM
I just got a couple, thank you very much Ben. It seems to me from reading this entire section, that the prop is an extremely powerful processor and a wonderful match to the hobby. It also appears that a few of the devices that were born from the prop have become some of the hottest new things in the hobby since Phil enabled us to dim. Somehow, I always seem to jump aboard the boat as it's pulling into the dock and tying up for the winter. I currently have 4 E681's, 2 bare RPM E1.31 to DMX bridge PCBs and one BOM, along with 2 of your Ethernet Propcontroller PCBs. (Further, I have 3 ELORs due to my choice of Software) I'll use the Sandevices gear and one RPM board for my display this year. I purchased a Prop Ed kit and your PCBs to try and learn.

What I don't see is much audience participation. I believe that's because the things that came from Ben, Jim and Robert were what was needed for access to RGB pixels and E1.31 to DMX/Renard. They fit well into the plans people had for the big "WOW" that comes from pixel addition to their display. Sean Meighan's work making a tool that can create sequence data is a perfect addition to that WOW. LOR has a programmer "ItsMeBobO" who's making tools to help users create some fantastic stuff for pixels. The availability of the hardware, which IMHO is very inexpensive for what it can achieve coupled with the work these programmers are sharing is making the use of high channel count RGB displays full of LEDs and Pixel devices very available and afforable. With no need for any development in their plans to create DIY displays.

This is why I think all the work Ben did to create this section and these tools has recentlybecome stale. I can think of 30 things I'd like it to do, but I'm just learning it. At the same time I have an entire display to build and sequence (using a substandard software that is struggling to keep up with the times) Ben has shown how willing he is to help with the code and it's all available for anyone to tweak and adjust to fit their needs. But for as powerful as the Prop is, it seems to be a little too much for the DIYer who's here to make blinky flashy.

Here's my point (I know, long winded...) IMO, the future direction would be much clearer if more people knew and understood more about the code. Mainly because it would help them to understand how powerful the prop is. To guys like me it's a mystical beast people like Ben, Robert and Jim have learned to tame. I see where Ben says learn Spin language. More than anything, I would like direction on where to begin. As it stands, it's just a foreign language I need to learn. I want to order Shrimp in Lobster sauce, and the propeller chip is looking at me like I have two heads. It thinks it's supposed to be dimming pixels and I'm ordering Chinese take-out in a language meant for sentient beings not a silicon chip. But who has time to learn a new a new language? I had a few ideas to improve on a product currently available but I have no idea how to make it happen, that comes from the genius of the guy who made the thing. But I can't gamble my display on getting someone else to code some firmware for me. And the work you gentlemen do to write this firmware and debug it is plainly beyond the ability of people like me. So, what do I need to do to learn enough to even ask you for assistance? It's clear to me that all of you are willing to help. Jim has been wonderful while I helped to test a board he made. This entire section indicates how willing you are to spend time matching firmware to people's needs.

...and what can I do to help? I firmly believe if I jump on board and can acheive enough skills to make something someone needs... then maybe I can help the guys down here at my laymen's level to make some hardware they'd like to have.

Jeff

DynamoBen
04-30-2012, 05:47 PM
Jeff your situation isn't a unique, I was there myself at the beginning. For me hardware and firmware development is a means to an end. I started with the Prop for a work project that needed video, then to continue my learning I created a "fully featured" (http://doityourselfchristmas.com/forums/showthread.php?7863-Propeller-%28Prop%29-Dimmer)dimmer for myself. And while no one but me was interested in it I figured I could break that design into pieces later, which is where the PropController came from. It's part development board and part swiss army knife and is meant as a tool for learning and making cool stuff. It gets you over the "I need hardware to make firmware" hurdle.

So if you are new where do you start? For me I need to decide what I want to make. From there it's a matter of stitching together and creating new code to make it happen. For me if I have a purpose in mind its easier to learn how to code and ask good questions when I get stuck.

The interesting thing about the PropController project is much of the hard work is done. There are objects and drivers for most of the things we do here that are tested and ready for use. I think the reason things are stale here is because people aren't creating new projects and I don't really have a need for anything specific at the moment. With that said I have started work on a DMX/ACN tester and was thinking about making a 4-6 universe DMX sACN node. In both cases I'm likely going to be the only person that uses them, since they will be feature rich as opposed to inexpensive, but the code will be available for others to use on their own projects.

As always I'm happy to help with any projects and who knows you could be creating the next big thing...but you need to start with and idea.