Vixen 3.0 Introduction Series - Part 4

[justinj]

I was hoping the most common outputs would be available at release (And tuned before the season) so that we can start testing, sequencing and debugging!

Well I'll keep my fingers crossed that Renard, Generic Serial, LEDTriks and E1.31 plugins are available near release

[sallz0r]

I was hoping the most common outputs would be available at release (And tuned before the season) so that we can start testing, sequencing and debugging!

Well I'll keep my fingers crossed that Renard, Generic Serial, LEDTriks and E1.31 plugins are available near release

Yeah; hopefully they'll be not far behind, I would imagine most plugins wouldn't be that hard to rewrite. (I haven't played with any of them though, I haven't done anything in that area). But I wouldn't expect to be able to use an alpha release for anything approximating a real display; it's more for early feedback than anything else.

[justinj]

Yeah; hopefully they'll be not far behind, I would imagine most plugins wouldn't be that hard to rewrite. (I haven't played with any of them though, I haven't done anything in that area). But I wouldn't expect to be able to use an alpha release for anything approximating a real display; it's more for early feedback than anything else.

I was hoping that Alpha would be out - We can all start sequencing while they work on Beta. We continue to provide feedback and by November we'd hopefully have a stable release.

[sallz0r]

I was hoping that Alpha would be out - We can all start sequencing while they work on Beta. We continue to provide feedback and by November we'd hopefully have a stable release.

Well, that's true, but you don't need the output plugins to be able to sequence. :-) That's why we hope to have a display preview as well; it will give people something to work with for preparation.

But, again, by the very nature of an alpha software release, it's not intended to be stable or production-ready. I would suggest having a contingency plan available if needed.

[dirknerkle]

I would rather have the development team take whatever time needed and not release an alpha version at all. My experience has been that the amount of support necessary to "fix" and alpha problem for a user often detracts from working on the base code. It's fine it you have an alpha version for internal use, but don't release it. This will allow you more time to work on the important things instead of try to analyze the screwy little problems that most certainly will ensue...

Bottom line: take the time to do it right. Just my humble opinion...

[budude]

I would rather have the development team take whatever time needed and not release an alpha version at all. My experience has been that the amount of support necessary to "fix" and alpha problem for a user often detracts from working on the base code. It's fine it you have an alpha version for internal use, but don't release it. This will allow you more time to work on the important things instead of try to analyze the screwy little problems that most certainly will ensue...

Bottom line: take the time to do it right. Just my humble opinion...

However... while it may not be ready for a 2011 release, if an alpha was released while everyone has their stuff up, they could do a full scale test for a week or so and get back crucial feedback to the development team. I tear my stuff down the first weekend of January - and frankly at that point I don't/won't feel like touching anything blinky-flashy for a while... All new code will need some "burn-in" time and this would be a good opportunity to kill off the really major bugs.

[justinj]

However... while it may not be ready for a 2011 release, if an alpha was released while everyone has their stuff up, they could do a full scale test for a week or so and get back crucial feedback to the development team. I tear my stuff down the first weekend of January - and frankly at that point I don't/won't feel like touching anything blinky-flashy for a while... All new code will need some "burn-in" time and this would be a good opportunity to kill off the really major bugs.

I didn't even think of that, but that's a fantastic point. There's no point in having V3 work perfectly stable with a preview, and the moment it starts outputting data to controllers it crashes.

But either way I would like to see a useful V3 released rather than the headache that V2.5 caused a lot of people (Although I still respect the KC and the team for their hard work and dedication for the 2.5 release, if it worked well it would of been a great intermediate release between 2.1 and 3!)

[rjchu]

As much as I'd like to have a new version of Vixen to play with, note I said play with not run my 2011 show with, the former software developer and manager in me has to agree with both Dirk and Brian. Skip the alpha and instead release a beta in early December so folks can run it with their live display/controllers before teardown after Christmas (which for us is the first weekend after New Years Day). This will avoid the inevitable distraction as folks rush to implement sequences for their 2011 shows and flood the dev team with bugs or requests and allow for some "on the ground" integration testing with everyone's hardware and setup.

If the dev team feels the need for feedback on UI or layouts or things of that nature, perhaps release screenshots and scenarios/test cases and ask for feedback that way instead of releasing binaries.

Just my 2 cents to go along with Dirk's humble opinion. ;)

Thanks for all your hard work!

-Joni-

[sallz0r]

Interesting comments and thoughts. Again, I can't speak for KC, so will let him comment when (and if) he has the time; I can but share my own opinions. :-)

The reason I was making sure to stress that a first release would be an alpha release was, more or less, to manage community expectations -- to mitigate any hype, and ensure people don't think it's a fully finished product out of the gate. However, for all we know, it might be fairly good, and after a week or two of having people play around with it, test it with their controllers, making a few sequences, etc. we might call those bits/modules "stable", and people can use it for 2011. Or, it might go completely the opposite way, with everything crashing and burning. :-) I'd hope not, and I wouldn't think so, but that's the point of the early versions: to get feedback on it all as a whole, and utilize the 'community test base', if you will. :-)

Unfortunately, by the very nature of the new framework (being completely modular), it won't be a cut-and-dried switch from alpha to beta to release -- not in my mind, at least. So it's important that we get something to the community to test with so we can get those transitions happening; the only problem is that we can't schedule or guarantee when those milestones will occur. However, Christmas is a fixed point in time, and the rest of it is flexible: so I don't know if it all squishes in before Christmas, or after.

I know my few previous posts have been pretty doom-and-gloom about it all, and I apologize: I just know there's been a lot of anticipation regarding 3.x -- heck, I'm excited as hell about it too -- but I don't want everyone to assume it's good to go out of the box. It very well might be (fingers crossed), but to be sure, we need everyone's help to roll up their sleeves and test it out with us. :-)

OK, back to it... I'm working on a new sequencer at the moment, which will make sequencing a lot easier -- being able to work with effects, rather than a bunch of discrete values. :-)

[squid]

Have you thought about adding a beat detection to make design a lot easier such as:

http://www.lightjams.com/#descriptionSection
http://www.venuemagic.com/index.php/...nuemagic-dmxav

Just a thought as log as you are releasing a new version.