This has been happening to a few friends I know alongside players of my games. I have a microprofile log attached.
I’ve observed this happening in a few games:
https://www.roblox.com/games/4111023553/Deepwoken
https://www.roblox.com/games/9872472334/Evade
https://www.roblox.com/games/122421323102335/literally-just-an-empty-baseplate
System info:
microprofile-20250213-235315.html (6.9 MB)
I have been getting reports of this issue in our game Notoriety as well. Some users are reporting lag spikes when there’s many guns being fired, and after checking the Microprofile logs they have sent, it’s the same "SchedulerHostConfig.default” causing it.
It might be related to Sounds being created, since that’s what seems to be what takes most time of that CoreScript. I’m attaching the microprofile log we got in here as well:
microprofile-20250214-135921.html (9.8 MB)
From what I gathered, this has started today - before there were no reports about it beforehand.
I’ve been getting reports of FPS issues as well. I’ve attached screenshots of some of the reports I observed while investigating if other games had the same issue
https://www.roblox.com/games/4282985734/Fixes-Combat-Warriors
https://www.roblox.com/games/4588604953/Criminality
https://www.roblox.com/games/4111023553/Deepwoken
https://www.roblox.com/games/9391468976/HEIAN-Jujutsu-Shenanigans
The React scheduler has always been taking up huge frame times and causing huge lag spikes for me. I wonder if it’s severity only got worse to the point where it’s more noticable for everyone. I’ve reported about it earlier in different threads.
I am not so sure it’s a react thing. For some reason, it’s calling GetPropertyChangedSignal and indexing AncestryChanged a lot.
That’s interesting because as far as I know SchdulerHostConfig.default is the React Scheduler. That’s unless my information is wrong.
Eitherways, it would be cool if Roblox would document what these Corescript tags are all about.
Hey folks, the feature related to this has been rolled back as we investigate! You should no longer observe this behavior.
Had the issue too yesterday when the music finder was released.
My fps was averaging at 6 while this tool was not even enabled.
I’d like to report this is still a problem today.
It seems to be causing huge FPS spikes from normal framerates to zero for a couple of frames randomly every 4-10 seconds.
Hi, would you be able to provide the microprofile file that you are taking screenshots from? The original issue was identified and rolled back; no additional changes since that time. It could be that the FPS issue for you has a different root cause, so having the microprofiler would be helpful for our own debugging.
Thanks in advance!
Here is one from one of my players which has very similar spikes.
The microprofile file is over the 10MB file limit so I’ve uploaded it to a file sharing service.
https://limewire.com/d/sRUbb#RZgvMztsFu (it didn’t embed correctly)
Hi, I have troubleshooted the root cause of the issue allowing you and your team to swiftly work on a fix.
Edit:
I believe I have mistaken Heartbeat;ugc for Heartbeat::LuaApp.
Regardless I believe this is a low hanging fruit for your team to resolve.
To add to the original thread’s topic…
I recall having major frametime spikes on the 22nd March 2025. It may be that this bug has been resolved as I have not been experiencing spikes of this nature via Heartbeat;ugc since then.
@Crazyblox is right. Even simply unequipping/equipping an item on your avatar causes frame-drops.
Sorry about the poor quality - had to compress…
Hi folks, we’ve root caused the initial cause of this bug report to the Music window (opened via the topbar dropdown → Music menu). A fix has been rolled out to all platforms, so keeping the Music window open should no longer have any performance implications.
That said, as noted by others in the thread there are parts of the app flow that experience a similar issue attributed to SchedulerHostConfig.default. If you could file separate bug reports for those flows so we can disambiguate them and get them to the teams that will be able to address these surfaces it would be a big help!
I’m going to close out this report, but please feel free to reopen if any further issues arise from the Music window.
I have filed a bug report about the performance issues of the avatar editor on Mobile long ago and there was another one that includes the Mobile and UWP apps (Universal app) here:-
None of the above reports include any information about SchedulerHostConfig.default spiking, but nonetheless the causes of these lag spikes is of course that script as can be seen through the Microprofiler. None of the above reports got any staff response unfortunately.
Are these two bug reports sufficient or do we need to file another one that specifically includes the issue of SchedulerHostConfig.default spiking causing lag spikes?
Edit:-
I already filed a bug report that is broader about the topic of the issue. Essentially, after further testing, it actually appeared that every interaction with the Universal app results in the Corescript spiking, and it’s visible in the entire app excluding website based elements, and not just the avatar editor. The bug report in question:-
the performance of the avatar editor is not exclusive to any apps, its also been notoriously sluggish on the website for as long as I can remember, including long before UGC / high-poly models were common. It is way too performance intensive especially when many accessories have been changed within the window session (presumably a memory leak).
What you said is true and I agree with it. Though, I think that the causes of the performance issues in the avatar editor on the website differs from that of the Universal app.
Let me explain this further, what I can see causing the spikes through the Microprofiler in the avatar editor on the Universal app is Script_SchedulerHostConfig.default, the same script also does spike when doing anything else in the Universal app.
