caffeine
Full time elf
CopperCreekLights said:
Copper -> This stuff is on my list of apps to write for the community.
If you were willing to share I'd happily finish them off for you, save me starting from scratch. No big deal either way.
Frustrated with all sequencing software
- Thread starter Gilrock
- Start date
Copper -> This stuff is on my list of apps to write for the community.
If you were willing to share I'd happily finish them off for you, save me starting from scratch. No big deal either way.
Gilrock
Full time elf
- Thread starter
- #17
caffeine...sure let me take a look at what I have so I can write up how they work in the current state and what limitations there were. I could probably upload them to my github account. The projects are currently created in Visual Studio 2008 I believe as C# .Net Gui's.
Gil
Gil
Valentinorx
Apprentice elf
Andy.....great write up. Is that all it takes?! Lol. For newbies this is great info when trying to decipher which direction to go. Ultimate everyone will find their own path and sequencer. It is a painstaking process.
smeighan
Dedicated elf
when i wrote nutcracker in feb 2012 it was because i didnt see a sequencer that would make it easy to put effects on rgb megatrees. I really hope 2014 sees some synergy between all of the developers. Vixen3 and Vixen+ have both been in contact. They want us to make xlights/nutcracker into an engine or dll that can be called to create nutcracker effects. Vixen3 is currently 22 versions behind us. This is because when they ported nutcracker in april of last year it took them one month to convert our c++ code into c#. It worked, but it is not supportable in the long range.
this year we will be adding new stuff to xlights, maybe redo the grid we use to be more like everyone else. Think of a video editor. timing done horizontally vs vertically.
we will be allowing for a grid display of cells like vixen or lor does now to allow individual setting of cells.
lots more things.
I consider nutcracker creating effects to then be brought into another sequencer a broken process. export out, add effects, import in. this really slows down things. i think having nutcracker inside a sequencer is the way to go. unfortunately there is no sequencer out there that has the performance of xlights. you can currently generate a butterfly effect on a 64K model and it works sub second. our code is not multithreaded so that performance is based on a single cpu.
here is clights playing my entire 24K channel show. note we are a 32bit app, single threaded. We use 1/3 of a cpu and < 100mbytes of memory. Also note, i had absolutely nothing to do with achieving this. Matt Brown is the genius who rewrote nutcracker into C++. I think the numbers show just the level of programmer that matt is. He writes fast, efficient code in his day job. He did the same thing when he wrote xlights. Along the way Matt set up xlights to be multiplatform. Windos XP,7,8 , MAC and linux.
matt,frankr, don julien, dave pitts,dan kulp and myself will be having a developers summit this month to talk about what we will do with xlights for 2014. will keep everyone posted.
I agree with the OP:
LSP, LOR, Vixen 2.1, Vixen+, Vixen3, HLS, Xlights/NC, Madrix, LightFactory
so many software choices, none gets it right.
My wish list?
i would want a gui that looks like a video editor
one track for each model. Drop a spiral effect onto a megatree track, you can see it expand and match what you want. Drop the same spiral effect and have it effect all models (arches, eaves, stars, megatrees). Effects should auto scale to models.
click an effect stretch it. double click and then the sliders come up that allow you to modify the effect.
well, one can dream
this year we will be adding new stuff to xlights, maybe redo the grid we use to be more like everyone else. Think of a video editor. timing done horizontally vs vertically.
we will be allowing for a grid display of cells like vixen or lor does now to allow individual setting of cells.
lots more things.
I consider nutcracker creating effects to then be brought into another sequencer a broken process. export out, add effects, import in. this really slows down things. i think having nutcracker inside a sequencer is the way to go. unfortunately there is no sequencer out there that has the performance of xlights. you can currently generate a butterfly effect on a 64K model and it works sub second. our code is not multithreaded so that performance is based on a single cpu.
here is clights playing my entire 24K channel show. note we are a 32bit app, single threaded. We use 1/3 of a cpu and < 100mbytes of memory. Also note, i had absolutely nothing to do with achieving this. Matt Brown is the genius who rewrote nutcracker into C++. I think the numbers show just the level of programmer that matt is. He writes fast, efficient code in his day job. He did the same thing when he wrote xlights. Along the way Matt set up xlights to be multiplatform. Windos XP,7,8 , MAC and linux.
matt,frankr, don julien, dave pitts,dan kulp and myself will be having a developers summit this month to talk about what we will do with xlights for 2014. will keep everyone posted.
I agree with the OP:
LSP, LOR, Vixen 2.1, Vixen+, Vixen3, HLS, Xlights/NC, Madrix, LightFactory
so many software choices, none gets it right.
My wish list?
i would want a gui that looks like a video editor
one track for each model. Drop a spiral effect onto a megatree track, you can see it expand and match what you want. Drop the same spiral effect and have it effect all models (arches, eaves, stars, megatrees). Effects should auto scale to models.
click an effect stretch it. double click and then the sliders come up that allow you to modify the effect.
well, one can dream
Valentinorx
Apprentice elf
Sean, that sounds great!! For people new to RGB like myself, with simple computer skills there are a few major criteria to think about.
How easy is it to time effects to the music? How well does a sequencer enter timing marks?
What is the process to create awesome effects on large RGB matrices, trees, etc?
Are those effects transposable to multiple elements? Like you said.....Scalability would be key.
Making that easy leaves the mind clear to create rather than frustrated. There is a steep learning curve with all sequencers when it comes to RGB I am finding. I want it to be fun.
Thank you all for your efforts.
How easy is it to time effects to the music? How well does a sequencer enter timing marks?
What is the process to create awesome effects on large RGB matrices, trees, etc?
Are those effects transposable to multiple elements? Like you said.....Scalability would be key.
Making that easy leaves the mind clear to create rather than frustrated. There is a steep learning curve with all sequencers when it comes to RGB I am finding. I want it to be fun.
Thank you all for your efforts.
Gilrock
Full time elf
- Thread starter
- #21
Sean....I like a lot of what I'm hearing in your past and like I originally stated I thought xLights to be an excellent program for the tasks it was doing. What I would like is a combination of different types of sequencing:
- ability draw traditional on/off, ramps, chases, etc. on a horizontal grid layout that is easy to see the audio for synchronizing the effects.
- ability to do macro effects like nutcracker that can map effects to a model or the entire display
- ability to create effects like you can in SuperStar where you can draw morphs on a pixel string or strip or morphs that move across a model. The nice thing about the SuperStar method is you can fine tune the start and end points to perfectly coincide with the audio and you can fine tune colors and head and tail lengths of the morphs. You can also draw images and specify a start and end position so you can draw animations that are in sync with the audio. The other nice thing about the SuperStar method is that effects can be created that move across each other. Like I have an effect where I make it look like my ribbons are twinkling while at the same time I have morphs that look like comets shooting across the strips. The software is able to sum the effects when they occur on the same pixel concurrently.
I wouldn't mind helping the development team mentioned. I'm an embedded programmer and I work on real-time software for my day job. C++ is my main development language but I have used C# quite a bit when developing GUI's on Windows. I've been holding myself back from creating a new sequencer program starting from scratch.
Gil
- ability draw traditional on/off, ramps, chases, etc. on a horizontal grid layout that is easy to see the audio for synchronizing the effects.
- ability to do macro effects like nutcracker that can map effects to a model or the entire display
- ability to create effects like you can in SuperStar where you can draw morphs on a pixel string or strip or morphs that move across a model. The nice thing about the SuperStar method is you can fine tune the start and end points to perfectly coincide with the audio and you can fine tune colors and head and tail lengths of the morphs. You can also draw images and specify a start and end position so you can draw animations that are in sync with the audio. The other nice thing about the SuperStar method is that effects can be created that move across each other. Like I have an effect where I make it look like my ribbons are twinkling while at the same time I have morphs that look like comets shooting across the strips. The software is able to sum the effects when they occur on the same pixel concurrently.
I wouldn't mind helping the development team mentioned. I'm an embedded programmer and I work on real-time software for my day job. C++ is my main development language but I have used C# quite a bit when developing GUI's on Windows. I've been holding myself back from creating a new sequencer program starting from scratch.
Gil
ShellNZ
Senior elf
CCL - Your tool VisualizeThis sounds exactly what I want. Thats my main problem with LOR at this stage.
vincerules
Full time elf
- Joined
- Aug 3, 2012
- Messages
- 100
CopperCreekLights said:
While I agree it does have a learning curve (every software does) and is not free, LSP does all of this.
I will say that the built in effects are not nearly as nice as the ones in SuperStar, hopefully the new version will adress this.
With the tidbits from Sean, I think xLights/SuperStar is going to be very awesome this upcoming year. Not that it already isn't, but even awesomer.
Gilrock
Full time elf
- Thread starter
- #24
I bought the most expensive version of LSP so I know what it does. My list above is not a complete list of what I want but I have spent a lot of hours in LSP and I'm past the learning curve. There are things I like about LSP but the things I don't like are keeping me from using it right now. Since I paid for it I'd be glad to switch back if it improves. I'm really big into the visualization/preview aspect of sequencing and I ran into a lot of issues trying to model my layout. Even the shape of the bulbs in the preview was annoying. How about some nice round circles instead of the tiny L shapes I ended up with.
caffeine
Full time elf
OK.. so now I'm starting to be encouraged by this thread.
There seem to be people here with a similar mindset/concept of where we should be heading.
I had a good chat in chat with Eddy yesterday on this topic.
My thoughts were that the main software packages should be focusing on the complexities of the core, being light/sound control and output, scheduling and most importantly providing a framework for add-ons.
Then leave it to the community to provide that functionality for visualisation and effects.
The reason I say that is that it reduces some of the issues being:
1. Viability of the software investment for very complex software in a small market by the companies.
2. Ability to focus on memory/CPU optimisation and output optimisation which are becoming big issues.
3. Flexibility for the community, which already has a large DIY component.
We are NOT commercial lighting people. We don't run that kind of show, and we don't have that kind of funds, but we DO expect that sort of capability rightly or wrongly but there's no way we could pay for it.
There seems to be supported add-on capability coming to LSP we hope, which is a great step forward.
I echo the thoughts of others though that for pixel displays we need to alter the paradigm into a drill-down model. Effects for the display (Full display sweeps for example). Common effects for each element in the display (chases, shockwaves, fades etc.) that are 1 'type' of effect but manifest in different ways depending on the element (tree/arch/matrix/set of mini trees etc.).
Drill down further into each element for custom things (images/animations on mega trees etc.) or down to each pixel if you so desire in traditional sequencing fashion.
It's a visual medium and we need to transition to a visual sequencing methodology.
I for one will get involved wherever and whenever I can, as like Copper I am trying to hold myself back from writing one from scratch as that seems ludicrous.
There seem to be people here with a similar mindset/concept of where we should be heading.
I had a good chat in chat with Eddy yesterday on this topic.
My thoughts were that the main software packages should be focusing on the complexities of the core, being light/sound control and output, scheduling and most importantly providing a framework for add-ons.
Then leave it to the community to provide that functionality for visualisation and effects.
The reason I say that is that it reduces some of the issues being:
1. Viability of the software investment for very complex software in a small market by the companies.
2. Ability to focus on memory/CPU optimisation and output optimisation which are becoming big issues.
3. Flexibility for the community, which already has a large DIY component.
We are NOT commercial lighting people. We don't run that kind of show, and we don't have that kind of funds, but we DO expect that sort of capability rightly or wrongly but there's no way we could pay for it.
There seems to be supported add-on capability coming to LSP we hope, which is a great step forward.
I echo the thoughts of others though that for pixel displays we need to alter the paradigm into a drill-down model. Effects for the display (Full display sweeps for example). Common effects for each element in the display (chases, shockwaves, fades etc.) that are 1 'type' of effect but manifest in different ways depending on the element (tree/arch/matrix/set of mini trees etc.).
Drill down further into each element for custom things (images/animations on mega trees etc.) or down to each pixel if you so desire in traditional sequencing fashion.
It's a visual medium and we need to transition to a visual sequencing methodology.
I for one will get involved wherever and whenever I can, as like Copper I am trying to hold myself back from writing one from scratch as that seems ludicrous.
kane
Dedicated elf
I echo your comments - I couldn't have done my sequences with sanity intact if I hadn't used my own app to read in the LSP .msq files, and insert my own effects before re-saving. Saved me a huge amount of time, and I do hope that LSP does embrace the community in this regard.
I'd love to be able to sequence a show as a "conductor" of sorts. Each display element is configured via drilldown, perhaps a bit like Vixen where elements can be in multiple groups etc.
You would then configure the types of effects that can be applied to each element.
Then, you get to the fun bit - choose the element(s) you want to apply effects to, and press play. Music plays, and based on the music, you hit certain keys to apply effects at certain times. Then perhaps you're shown a list of effects that will be applied, at which you can adjust the finer details of those effects (choice of colours, timing, etc), and then click "OK" to get the grunt work to start..
It would certainly open up the software to beginners, as well as those that want to expand on the types of effects etc..
And leaving the grunt work to be done in a batch would also be good for sanity - I'd love if LSP could do this, as waiting for anywhere from 20 seconds to 5 minutes for it to apply an effect is very annoying - would prefer to queue 10-20 effects and then leave it to go..
anyway, will be interesting to see how the various packages proceed!
I'd love to be able to sequence a show as a "conductor" of sorts. Each display element is configured via drilldown, perhaps a bit like Vixen where elements can be in multiple groups etc.
You would then configure the types of effects that can be applied to each element.
Then, you get to the fun bit - choose the element(s) you want to apply effects to, and press play. Music plays, and based on the music, you hit certain keys to apply effects at certain times. Then perhaps you're shown a list of effects that will be applied, at which you can adjust the finer details of those effects (choice of colours, timing, etc), and then click "OK" to get the grunt work to start..
It would certainly open up the software to beginners, as well as those that want to expand on the types of effects etc..
And leaving the grunt work to be done in a batch would also be good for sanity - I'd love if LSP could do this, as waiting for anywhere from 20 seconds to 5 minutes for it to apply an effect is very annoying - would prefer to queue 10-20 effects and then leave it to go..
anyway, will be interesting to see how the various packages proceed!
Kane said:
LSP already allows for this through layers and is one of the strengths of LSP for managing large RGB displays
My inputs are not as detailed & based on experience as others, but here goes. This was my first year to sequence. I actually decided to use LOR in 2012 and had everything ready to go but had to go out of town for a job so I was not able to setup anything whatsoever in 2012. I decided to add non-LOR RGB-based arches in 2013. I echo other inputs that LOR is a real pain when you are not using their stuff. My guess is that they want you to use their stuff, then it will work really well, and their users will be HAPPY!! I like LOR's software & hardware, but not what I noted above. I setup my RGB-arch effects in xlights this year, then exported them into LOR. That worked really well. Based on what I have read and based on inputs from people here at ACL, I am migrating over to HLS this year before I get up to a large number of channel counts. My 2 cents. I hope this helps!
arw01
Full time elf
Kane said:
On the roadmap, that one is, along with multi-threaded rendering, which should make it massively faster. I also suggested they figure out some gpu use for rendering, as that's what those are for!
kane
Dedicated elf
arw01 said:
Nice..
Really look forward to seeing what they come up with with the "rewrite"