Having another Sound Cue breakage problem. In the attached screenshots you’ll see my sound cue and the BP that calls it. I’ve got two variables that need to be set depending on what is present in the level. If both bools are true, everything completes without a problem, but if either of them are false, the sound cue stops processing altogether at the false branch and doesn’t go any farther. I’ve tried plugging in an empty wave file, a non-empty wave file, and leaving the pins empty, but it doesn’t matter. If the branch returns false, regardless of what’s plugged (or not plugged) in there, the sound cue stops.
Here’s the cue:
And here’s my BP:
Hey Mbublitz,
I just ran a test and was able to confirm what you are reporting in the preview release. I am going to run a few more tests before entering the report, but I will return once I have completed my testing.
Thanks for taking the time to write up the bug report, and let me know if you have further questions.
Cheers,
Hey Mbublitz,
I think your problem here is how you are calling the bGuardians and bHunters to be checked in blueprints whether they are true or false.
By keeping the ‘In Bool’ option checked on the ‘Set Boolean Parameter’ node, you are always saying the instance is ‘True’. If you uncheck the ‘In Bool’ it should allow you to correctly play the false boolean sound.
I just ran another simple test and was able to get the false sound to play as expected.
Let me know if you have further questions or need additional assistance.
Cheers,
The problem here is that if either of the bools are false, then the sound does not process any further beyond that point. The rest of the pins on the concatenators never fire and stop at the first false response. The “and_reach_the_portal” and “before_time_runs_out” are never played.
For additional clarification: if I have a sound attached to the False pin, it WILL play. It’s the REST of the audio chain that doesn’t get played.
This seems to be related to the other concatenator bug as the inherited behavior of the Concatenator node is meant to play the individual inputs in succession.
I will add this information to that bug, and if it is returned that they are unrelated, I will create a new bug and update this post.
Thanks,
Sounds good to me. I was thinking that the Concatenator was probably the real culprit here, too, but wanted to create this one separately since I really didn’t have a way to test this behavior against other types.
I’m curious to know if there’s any more information on this particular problem as this is something that is very important to the aesthetics of my game.
I have an introductory audio script that is played at the beginning of the levels and uses these branches to play appropriate audio based on which obstacles are present in the level. I’ve been faking it at the moment by creating a bunch of individual sound cues for each possible outcome, but this gets exponently more difficult to do and maintain as things get added.
If I could work out a way to find the length of a sound-cue at runtime, I could at least make a sound-chain and use delays to prevent them from stepping over themselves when playing, but I don’t see any way to access the length of a sound cue from within a blueprint.
Just to clarify, have you updated your project to the most recent 4.11.2 version, and tested to see if this issue still occurs?
First, I don’t see a 4.11.2 update yet, but second, I haven’t tried to re-implement the original code for the 4.11.1 update yet (been busy in other areas of the game) so I’ll try that now and let you know. Crossing my fingers, because this is very important to the way I want to handle many of my sound elements. All of my content is generated on-the-fly at runtime, so I need to be able to piece together my audio on-the-fly also based on how the rest of the game was generated.
Apologies, the 4.11.2 update is not out yet. I meant to say the 4.11.1 version. Yes please let me know if this is still an issue, as our audio engineer made some major changes in 4.11 and minor fixes in 4.11.1.
same problem. Here’s my BP:
and my sound cue:
When I play the first level after the “find_the_orbs” wave, it plays the “20” wave (a test wave I plugged in there, so I could tell which pin was activated) which is attached to the “false” which I would expect to play since the bool is in fact false at this point, but nothing beyond this plays. If I play through level 3 (when the bHunter bool would be true), then the last sound that is played is “avoid_the_hunters”. It doesn’t play ANY of the bGuardian branch (which it SHOULD at least be playing the wave 18, because bGuardian is false at this point) and still not going past to the last two waves. Now if I play to level 6, when both of them are true, it still doesn’t play the bGuardian wave or go past it.
So, essentially, I’m now not getting anywhere past the bHunter bool.
Hey guys,
We found an issue with the sound-concat nodes that happens when one of its children nodes results in no generated wave instances.
In that case the concat node won’t progress its own node index. If any of you can attempt a local patch to confirm the issue, try the following code change in SoundNodeConcatenator.cpp:
I was able to reproduce the issue you guys seem to be running into with a much simpler sound cue, and this code fixed it. If you can, please let me know if it fixes it for you guys. This is a long thread w/ some super-complex sound-cues, so not 100% sure its the same thing you’re all running into.
This is the simple cue where I was able to reproduce the concat node halting if a child node fails to play a wave instance (i.e. the blank in the random node).
Would love to try this, but have never been able to get the source. I’m linked up like it says in the instructions (My github account login is in my UE profile information) but it doesn’t allow me access and only wants to show me how to create a github account (which I already have) and put my github account name into my profile (which I also have). I’ve never really bothered with any of that up until now, though, since I didn’t really have a need before.
Just got the invite to the github page, so I’ll download the source and give this a try today.
YES! I made the code changes you suggested and recompiled the engine and can confirm that this fixed the problem I was having with my sound cue! The sound cue is now behaving as I would expect it to.
Here’s a shot of my desktop. The left monitor has the code changes you suggested, the center monitor has my blueprint, and my right monitor has my sound cue.
