Cue List faders problem

The issues found when using the Virtual Console
Post Reply
maximortal
Posts: 64
Joined: Mon Dec 19, 2016 7:07 pm
Real Name: Iro Suraci

Vers. 4.10.5.246

Virtual console bug
Faders of Cue list works incorrecly if linked
If you got 2 cue named A and B both with 2 generic dimmers
A: CH 1 @ FF / CH 2 @ 0
B CH 1 @ FF / CH 2 @ FF

If you put fader each to 50% the DMX monitor show this values:
CH 1 @ 51 %
CH 2 @ 51 %

expected behaviour is
CH 1 @ FF
CH 2 @ 50 %


I'm not sure but this bug there is also in previous versions
This bug is really important to been fixed for theatrical use of QLC+
Thank for you grat job
Cheers
janosvitok
Posts: 1265
Joined: Mon Apr 13, 2015 7:05 am
Location: Bratislava, Slovakia
Real Name: Jano Svitok
Contact:

Maximortal,

I've moved your topic to a better place.

Jano
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

Can you please post a sample workspace to reproduce the behaviour you're seeing ? Thanks.

By the way, the description you provided seems like the correct behaviour to me.
When linked, the side faders go in opposite directions, so it is correct that @ 50% they are both @ 50%
maximortal
Posts: 64
Joined: Mon Dec 19, 2016 7:07 pm
Real Name: Iro Suraci

yes, from a particulart poin of wiev you got right, this behaviuour is correct, but no if you lock the sliders each other. I mean that the expected behaviour is something like Spark consoles: connected faders become a linear crossfader between two scenes. In other words DMX values of scene A and scene B booth @ 50% must add each others.
to recreate bug remember to lock fades each other.
Attachments
bug report.qxw
(2.8 KiB) Downloaded 1102 times
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

I think I've found a solution to this without impacting the code too much.

The code is in the "morphblend" branch on GitHub.
Otherwise, Windows build available here (4.10.6.54): https://dl.bintray.com/mcallegari/qlcplus/

Please test and let me know if it works and if it causes regressions somewhere else
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

@maximortal, can you please help with this by testing the development version ?
The issue resolution is now in progress and I'd appreciate to hear feedbacks from different users. Thanks
maximortal
Posts: 64
Joined: Mon Dec 19, 2016 7:07 pm
Real Name: Iro Suraci

yes I can help if youneed it. What I have to do exactly?
plugz
Posts: 637
Joined: Sun Apr 12, 2015 6:30 pm
Real Name: David

Hello,
Windows build available here (4.10.6.54): https://dl.bintray.com/mcallegari/qlcplus/
You go here, you install the test version, you test it.
maximortal
Posts: 64
Joined: Mon Dec 19, 2016 7:07 pm
Real Name: Iro Suraci

tested on 4.10.6.54 and it has finally expected behaviour.. great!
now i'm investigating about another issue on chaser inside an another chaser.
i'll keep you udpated.
User avatar
edogawa
Posts: 565
Joined: Thu May 07, 2015 10:34 am
Real Name: Edgar Aichinger

I just found this thread, and as it seemed to address the most severe long standing problem we theater guys have had with the cuelist crossfade, which has been discussed in several threads over the past 2-3 years (intensity dropping when moving both faders in parallel/linked mode), I immediately had to build from git and test it :)

Well, I currently don't have my conventional dimmers at home, and the two 24 channel DMX LED Controllers I ordered from AliExpress haven't arrived yet, so I cannot test with physical lights right now. But from watching the dmx monitor it indeed seems to work, yay!!!

Can't wait to test this in a realworld environment, this will save me a lot of time and hassle with directors in future productions. Many theater lighting techs prefer to perform manual in favour of timed crossfades, because the fade time often depends on what actors do on stage, which can vary in time a lot from one to another evening.

One question though, is it on purpose that this new code only works on linked faders? Each and every lighting desk with a two fader crossfade playback that i've had the chance to work with operates with this "crossfade law" while still being able to move each fader individually to separately control fade in/out ...

Cheers, Edgar
maximortal
Posts: 64
Joined: Mon Dec 19, 2016 7:07 pm
Real Name: Iro Suraci

I've testet GIT vers in real world (venue with 36 dimmers channels) so you can trust me, but this only if you lock faders.
IMHO can be useful if users can decide behaviours one as before and another like this one but working also with faders unlocked.
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

I have separated the blending into a dedicated check box (above the link check box), so now the logic is not implicit anymore in the linked faders case.
Also, the check boxes status is now saved into the project QXW file, so it is restored upon reloading.

Please check a test version and report if it works for you.
anorod
Posts: 15
Joined: Mon May 04, 2015 6:22 pm
Real Name: Antonio

Just checked the test version and seems perfect. It's a great idea to have an independent checkbox to control the behavior of the blending at anytime during the show.
Awesome work!
User avatar
edogawa
Posts: 565
Joined: Thu May 07, 2015 10:34 am
Real Name: Edgar Aichinger

I agree (after another test with just the dmx monitor to control resulting values), this is an elegant solution and will work very well for me.

Thanks a lot, Massimo!
maximortal
Posts: 64
Joined: Mon Dec 19, 2016 7:07 pm
Real Name: Iro Suraci

tested 4.10.6.82 and it works fine
Nice job
User avatar
edogawa
Posts: 565
Joined: Thu May 07, 2015 10:34 am
Real Name: Edgar Aichinger

There are two remaining cosmetic issues i still have and this might file under nit-picking, but i'm slightly annoyed everytime i come across them:

- the layout of the new checkbuttons: each is centered horizontally, and due to their different text lengths they don't line up in a vertical row. I played with the aignment settings and using "Qt::AlignLeft" instead of "Qt::AlignCenter" made them look pretty and in order. (vccuelist.cpp, lines 108 and 113)

- whenever one of the two sliders reaches the 100% position the geometry changes so that the sliders and main cuelist widget move a bit. I've read up Qt docs for hours now but I don't seem to be able to find how to set them to a static width that honours eventual font or fontsize changes etc. I suppose most of you devs have a good understanding of Qt and will be able to do this in no time, so I ask you kindly to do so :)

Thanks, Edgar
User avatar
edogawa
Posts: 565
Joined: Thu May 07, 2015 10:34 am
Real Name: Edgar Aichinger

Hello again,

I've forked and created a github pull request for a patch that deals with the first point in my previous post.

(

https://github.com/mcallegari/qlcplus/pull/941

I've only recently started to actively use git and github beyond merely cloning and pulling, so I hope I'm doing things correctly there.

Edgar
janosvitok
Posts: 1265
Joined: Mon Apr 13, 2015 7:05 am
Location: Bratislava, Slovakia
Real Name: Jano Svitok
Contact:

I've merged your PR. Thanks!

Jano
User avatar
edogawa
Posts: 565
Joined: Thu May 07, 2015 10:34 am
Real Name: Edgar Aichinger

Hi Jano,

thank you too :)

And here's another PR, to address my second issue above: https://github.com/mcallegari/qlcplus/pull/944

I gave it another go, read a bit about QLayouts and started to grasp how it works, not very difficult in fact.

Cheers, Edgar
Post Reply