There is an open source, cross-platform sequencer and scheduler already in development. See the xLights project on SourceForge. It will run on Windows, Mac, and Linux. We're starting on the scheduler first and then will work on the sequencer. An alpha version of the scheduler is a few weeks away. It will be able to read LOR and Vixen sequences and output to LOR/dlight, Renard, and DMX networks.
If you are an experienced C++ coder, or just willing to help with testing, please send me a PM.
Matt Brown
Spring, TX
Just took a look at that Matt.. looks good but i'm dubious about the idea of running it on a cross platform system, sounds like for something this intensive that you may have far too many overheads for it to run smoothly on every system.
I'll be interested to see the final product though 🙂 good luck with your development
John.
ElDiablo wrote:Just took a look at that Matt.. looks good but i'm dubious about the idea of running it on a cross platform system, sounds like for something this intensive that you may have far too many overheads for it to run smoothly on every system.I'll be interested to see the final product though 🙂 good luck with your development
John.
With proper design, it is quite feasible to make code cross-platform - Typically there's a cross-platform core and then small OS-specific abstraction layers. Thanks to cross-platform libraries that take care of large amounts of OS-specific stuff, the OS-specific layers within an app are often VERY small.
I can't see where you get your idea that something this intensive can have trouble running cross-platform. It's nothing compared to, say, a 3D first person shooter in terms of performance requirements, and many of those have a well designed cross platform codebase.
In general, in terms of hardware design, I strongly suggest making sure any lighting app has a "plugin" architecture like Vixen does for control mechanisms. That allows support to be added for whatever lighting control methods there are. Probably top priority for users here would be Renard, DMX, and one of the DMX-over-Ethernet protocols.
As a hardware guy, the order of design difficulty of the interfaces dithese days is:
1) USB
2) Ethernet
3) Firewire
There's a huge jump in design complexity from 2) to 3)
To quote my programming lecturer at uni, "Just because something is hard to design, it doesn't mean we shouldn't design it."
This is a concept design for something thats not been done before, purely because everything so far has been made for windows. Just because everyone brought and drove a Ford Taurus, does that mean we can't design and build other cars?
The Apple generation has grown since the introduction of the iPhone and the iPad and since the price of a Mac Mini has come down alot, yet designers and programmers haven't been bothered to produce anything for it because its not windows and it means having to recode the way things are done.
Everyone will have their ideas about how to design the best system and how it should all work, but you're not going to please everyone. Sure, there's gonna be some hard problems to overcome, but for future expansion and design i think Firewire is going to be the best thing to do with the Apple OS and hardware.
The problem i have with Cross-Platform anything Entropy is that when you design something for cross platform, you have to make it very very generic. The big problem with generic is that if you want to cater for something other than the "normal windows environment" you'll have to make sure there's something there to cope with it, or write your own. I've seen so many programs like Azureus/Vuse that chose to build cross-platform, struggle and get slower and slower in order to cope with everything they need to do on all platforms. I'd rather make three seperate programs, written specifically for each OS and designed to maximise the benefits of all OS's and get around each individual shortcoming.
A great friend of mine who is now no longer with us once told me something that i stick to even today. "Why try and please everyone with something that works 60% to design, when you can please the people you're aiming at 100% by making it work where they need it."
Well I spent my lunch break looking through what Firewire takes to impliment (my wife thinks I'm crazy doing this kind of stuff to relax :p) From brief, and I stress brief, look I had at this it seems that a Firewire breakout box would be very very challenging to create and quite possibly out of the reach of diy folks.
Honestly, I would love a Firewire based breakout box but I think it would be really time and resource intensive to impliment if it could be done at all. I'm probably talking out my @$$ (it wouldn't be the first time!!!) but I think your time would be better spent trying to tie into existing protocols.
To that end there was a quite interesting thread started here describing a new hardware solution that would do exactly what I was thinking about i.e. a breakout box to multiple DMX universes. It uses the E1.31 protocol which just so happens to also be supported by the Open Lighting Architecture which is a framework that supports E1.31 amongst several others that is written for OSX and Linux. I don't know if this helps but it seems like it would be a very good fit. Also, if your software was plugin based as Entropy suggested this would allow it to function with existing hardware and still have ability to add a Firewire breakout box in the future.
It amazes me how so much stuff just seems to come together at just the right time!
