Bug: Submaster + cue list in a frame does not work properly

The issues found when using the Virtual Console
Post Reply
nkaminski
Posts: 2
Joined: Tue Oct 31, 2017 6:31 am
Real Name: Nash Kaminski

On QLC+ versions 4.11.x (and possibly earlier), if a submaster slider is placed into a frame with a cue list, two fairly significant unwanted behaviors occur:

# If the submaster is set to zero, the submaster ceases to have any effect on the intensity of the cue list and when cues are advanced, all future cues display (incorrectly) at full intensity. This appears to be due to an overly restrictive check in Function::adjustAttribute where -1 is returned and the override not applied if the value of the override is equal to the initial value of AttributeOverride::m_value, which in this case is zero.

# If the same cue/scene is listed twice in succession, or if a random order chaser within a cue list transitions to the same scene that was previously active, such scene is always displayed at full intensity regardless of the value of the submaster. This appears to be due to the logic inside of MasterTimer::timerTickFunctions where when a function is to be restarted, postRun and then preRun are called in order. However this has the unwanted side effect of clearing the override set earlier by the submaster causing the scene to be incorrectly displayed at full intensity.

I have created a pull request with fixed for these two issues as well:
camwolfe
Posts: 9
Joined: Sat Oct 08, 2016 12:56 pm
Real Name:

#1 also applies when you have buttons inside a frame. I attached a workspace where you can see this. As far as I know it worked as expected in QLC+ 4.10.
Attachments
submaster_problem.qxw
(15.39 KiB) Downloaded 102 times
snorreflorre
Posts: 2
Joined: Sat Nov 26, 2016 6:26 pm
Real Name:

camwolfe wrote: Fri Nov 03, 2017 1:19 pm #1 also applies when you have buttons inside a frame. I attached a workspace where you can see this. As far as I know it worked as expected in QLC+ 4.10.
I can also confirm this issue. It first occured in version 4.11,1, but works as intended in version 4.11.0.
  • Operating System: Ubuntu 16.04.1 LTS x64
  • Kernel: Linux 4.10.0-38-generic
  • QLC-Versions: 4.11.0/4.11.1
  • Interface: Enttec DMX USB PRO
Attachments
example.qxw
(7.15 KiB) Downloaded 82 times
snorreflorre
Posts: 2
Joined: Sat Nov 26, 2016 6:26 pm
Real Name:

nkaminski wrote: Tue Oct 31, 2017 7:20 am I have created a pull request with fixed for these two issues as well:
Unforunately I can't find any pull request on github. What I only found, is a commit on your own fork. Please submit again.
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

Fixed on GIT: https://github.com/mcallegari/qlcplus/c ... f48ea79e71
One line of code was missing.
Thanks for reporting
nkaminski
Posts: 2
Joined: Tue Oct 31, 2017 6:31 am
Real Name: Nash Kaminski

It doesn't appear that this issue is 100% fixed though. In the same scenario, where a cue list and submaster are both placed into a frame, or alternatively when a VC slider is used to control the intensity of a chaser, functions jump to full intensity during fade outs. It seems like the fade begins at full function intensity, as opposed to the overridden value set by the submaster.
dino.loss
Posts: 16
Joined: Sat Feb 06, 2016 2:22 pm
Real Name: Dino Loss

Hi,
the latest version of QLC+ 4.11.2.42 still has this problem with the submaster in the frame.
in this example start the chaser and set the submaster to 100, you can see from the monitor that the fade starts from the maximum value 255 ignoring the value of the submaster.
with the old version everything works perfectly.

Happy holidays to all, Dino
Attachments
test-submaster.qxw
(6.16 KiB) Downloaded 86 times
AKAI APC mini
Behringer Eurolight LC2412
Enttec DMX USB PRO MK2
Windows 7 64
debbio
Posts: 40
Joined: Wed May 06, 2015 10:18 am
Real Name:

Did someone try new git version? Does it work?
Why Massimo doesn't release a fix to this officially?
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

debbio wrote: Sun Feb 25, 2018 7:08 am Did someone try new git version? Does it work?
Why Massimo doesn't release a fix to this officially?
Because Massimo doesn't release a new version of QLC+ for every fix he does.
TEST versions are there for that exact reason:
- users report an issue
- the issue gets fixed on GIT and a TEST version is built
- users test the build and report if it works

When a number of issues are fixed, a new QLC+ version is released.
debbio
Posts: 40
Joined: Wed May 06, 2015 10:18 am
Real Name:

So I imagine that the posts of users in february are fake ones.. cause it's a problem since release 4.11. I didn't know it's fixed
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

debbio wrote: Sun Feb 25, 2018 11:29 am So I imagine that the posts of users in february are fake ones.. cause it's a problem since release 4.11. I didn't know it's fixed
I don't quite get how your being sarcastic (in this and other posts) can help in any way the project or to resolve an issue.
Since you have no idea of how many things I am working on in parallel, please spare us from your poor comments.

Today I found some time to work on this, and I think now the behavior should be restored as what seen in 4.10.
Test versions available. Please report if everything works as expected now.
debbio
Posts: 40
Joined: Wed May 06, 2015 10:18 am
Real Name:

I'm not sarcastic. I thank you for your work but in the last year I feel you take distance from the forum and the user's needs.
Just an example: there is a post few time ago in which a user change for mistake your name.. how do you say your answer doesn't help in any way the project.

However I'll try this day the new test version, thank you
User avatar
mcallegari
Posts: 4446
Joined: Sun Apr 12, 2015 9:09 am
Location: Italy
Real Name: Massimo Callegari
Contact:

debbio wrote: Sun Feb 25, 2018 2:51 pm I'm not sarcastic. I thank you for your work but in the last year I feel you take distance from the forum and the user's needs.
Yes I did. First because this forum stress me out. Second because I concentrated on QLC+ 5. Otherwise it will never see the light. Third read this.
Also, can I be free to do what I prefer in my spare time or should I always be judged for my choices by people that did nothing for this project ?
QLC+ is not led by a company and I'm not paid for what I do, nor I am willing to take shit from random people.
Try to submit an issue to Autodesk regarding Autocad, and see what happens. Most likely they will never reply or a fix will be delivered 6 months later. And you paid thousand $$ for it...
debbio wrote: Sun Feb 25, 2018 2:51 pm Just an example: there is a post few time ago in which a user change for mistake your name.. how do you say your answer doesn't help in any way the project.
I hardly reply to feature requests. Cause they are what they are: requests.
That doesn't mean A) they will be implemented the day after B) they are good ideas C) it is even possible to implement them. (the one you're referring to is C)

But you can always discuss the request with the user who asked for it, and explain to us your plan to implement it in QLC+.
debbio
Posts: 40
Joined: Wed May 06, 2015 10:18 am
Real Name:

Never mind.. you answered to that post just to say: "Who is Marco" ..
Requests are only requests ok but they are done by people who are working with this program and find out something that can be improved. It's not just a waste of time

However we're going OT.

Edit: my personal request of 16 bit is done in this moment in which you're working on a new qlc cause last year you told me that it's hard to implement because it was created in that way.. if I don't make a request now when should I,?
debbio
Posts: 40
Joined: Wed May 06, 2015 10:18 am
Real Name:

Tried on Ubuntu 16.04.. new git version works how it should. Fixed!
Thanks
Post Reply