Sunday 29 June 2014

OpenTx "classic" - still the best for canned setups?

Like many Taranis users, I've upgraded to OpenTx v. 2. And I like it, it's got lots of little enhancements. Nevertheless, for whatever reason, some flyers are still using the "classic" version.

I wondered about this when developing my latest F3J setup. Would it be better to provide the EEPE file in "classic" format, thereby reaching the widest possible audience? Or should I target version 2 to maximise the feature set?

Oh and by the way, what exactly are the extra features provided in OpenTx version 2?

To answer these questions I did a little test: I modified my F3F setup to target OpenTx version 2, using all the programming tricks possible. The result was compared with the original "classic" version in terms of (a) functionality and (b) maintainability.

"classic" v. version 2

As expected, the v. 2 setup needed fewer mixers. But - and here's the rub - the missing mixers didn't simply vanish, they reappeared in different places- as new Inputs, or as a curve in the Servos menu. The result really wasn't really any easier to maintain, in fact in some ways it was more difficult as the 4-character limit for Input names made the setup less readable. However, you can continue to use the old style if you wish. Verdict for mixers: version 2 or "classic" would work equally well for the F3J setup.

Turning now to Logical Switches. Version 2 has some nice little tricks up its sleeve here, especially when it comes to creating 'sticky' switches e.g. for arming motors. However I had already managed to implement a sticky switch in the "classic" version, albeit with some effort. Verdict: v. 2 better, but "classic" still fine for the F3J setup.

Lastly, there's the enhanced Custom Functions. One small change makes it much more useful now - you can trigger functions directly on change of flight mode (by specifying an 'FMx' switch). This meant that I could get rid of six rather tricky logical switches. Verdict: Version 2 has the edge here, for a neater more manageable setup.

So Version 2 won in terms of neat code and maintainability. However version 2 offered little in terms of extra raw functionality. Everything I wanted could already be done in "classic". Sure, there is the new relative-trims feature (where a trim setting in one flight mode can be relative to another). However I figured that the classic way of doing trims would be more appropriate for a canned setup.

We have a decision!

The outcome may seem surprising. After weighing up the pros and cons, I decided that it would be better to provide my F3J setup in "classic" format. Sure, it gains a few extra lines, but it runs on both old and new versions of OpenTx, with identical setup procedures, and it doesn't lose out on functionality. 

Does this mean that version 2 is not worth the upgrade? Not a bit of it! Even if you only intend to use your old "classic" setups, there are a two good reasons for upgrading to version 2: first, no more pesky GVAR popups; and secondly, version 2 can remember the position of all your sliders and knobs.

Lua changes all

It looks like Lua scripting is well on its way to being the game changer that many people had hoped. All that's missing is some definitive documentation of the API, but I'm sure the overworked developers will address this.

I think I've identified a possible application for F3F - a snapflap mix, with deadband adjustable in flight using a rotary knob or slider. Even better - imagine receiving haptic feedback through your hands, as the snapflap kicks in - instead of guessing what your snapflap is doing, you could feel it!

3 comments:

Anonymous said...

So, when will the F3J setup be available?

RC Soar said...

Package, documentation and web page all ready to go, just waiting for feedback from a couple of testers before pulling the trigger. If you'd like to check it out please PM me via http://rc-soar.com/email.htm.

RC Soar said...

F3J setup now available, see http://rc-soar.com/opentx/setups/f3j/index.htm