BodyPosition's power is suddenly less

Bug Reports Engine Bugs
#14

I noticed some changes, yes!

It seems that for all three stepping methods (Fixed - Default - Adaptive), parts with BodyPosition now all feature the same physics.

quick mention:
-The subtle, natural / ordinary dip (0:55 on the clip of first post) always occured to a part with BodyPosition when the power wasn’t on the extreme - even before the change to ‘Default’
-The exaggerated, abnormal / extraordinary dip is the big lack of power (0:16 on clip; until an NPC/Player/Humanoid touches) which this thread was created around

It appears that this subtle dip is now also gone, on every stepping method.
This means for example the platform no longer slightly lowers itself when power isn’t sufficient.
This also means BodyPosition with this insufficient power added to a Player keeps them set in place.

There is this instance where BodyPosition is added to them by this kind of Sorcerer. The force of this BodyMover tries to keep the NPC or Player in place, but you can resist and slowly move away from the point set as its position.
image

Right now however, after this change from a day or two (?) ago, you’re just stuck and can’t move away or resist whatsoever. This is because the subtle dip (it allowed you to slowly but surely move away) has been gotten rid off on every Stepping Method. When I wrote my latest message (October 17th), changing the Stepping Method to ‘Fixed’ removed the exaggerated dip, but nicely kept this subtle one!

I thought tweaking the Power, Damping and MaxForce would bring back the ability for the Player to resist this force, which after a lot of trying seemed to sort of work.
Something to the movement or force put towards it seemed off however. It felt as if the force wasn’t constant and that with every movement direction I took, I was able to manipulate the force to a point where sudden movements affected the force too much.

.
I set up a basic test with a part to check on this:

You can clearly see the part doesn’t move whatsover, even when the Force is lower. No more subtle drop…
Only when it’s reached a Force low enough - changes occur.
At 0:31, putting in 1100 wasn’t enough to get the part back into the air, whereas on 0:43, 1100 was enough to keep it afloat somehow.
It seems as if it’s “the one or the other” now and the forces impact each other.

.
I greatly appreciate the thoughts on how to improve the methods, but with the abnormally loss of power gone and the subtle dip still there (message from October 17) the interactions and physics worked well from there!

Fast forward to today, the abnormal loss is gone, but so is the subtle dip - essentially causing another complication :sweat_smile:

Was it intended for either Stepping Method to now give the same results regardless, or can one method allow no dips whatsoever and another keep just the subtle one?

Essentially:
~ A few months ago : Subtle dip, worked fine :+1:
~ September 14th / Post creation : Abnormal drop :-1:
~ October 17th (I updated the stepping method to ‘Fixed’) : No abnormal drop, but still subtle one :+1:
~ November 9th : No abnormal drop, but no subtle either :-1:

1 Like
PhysicsSteppedMethod behaviour is the same regardless of setting
PhysicsSteppedMethod behaviour is the same regardless of setting
#15

You and I are having similar issues with BodyPositions. I have a game where players fly around using BodyPositions, and I experienced an issue similar to yours where flying players would slowly fall down while idlling when they should have been in a set position. What fixed this was reverting the instance to a previous version history then importing all my newer models to the old place. This fix was great, until sometime around November 7-8 my players report they suddenly are either falling or floating upward but nobody is staying in one place. I will work to find a solution to bring the physics back to what they were and post it here.

Edit: It appears primarily the Y-axis is what is affected like you stated previously. There are little to no differences in regular flight except the slow falling or floating upward.

1 Like
#16

This update essentually ruined the main mechanic surrounding my game. I use BodyPositions to have characters move around while being pulled towards the players. BodyPositions seem to have been broken in away that prevents said characers from move at all except for the force applied by the BodyPosition. There doesn’t seem to be any Class that behaves similarly to BodyPosition before it broke, so this leaves me with no alternative.

1 Like
#17

Apologies for the issue. Can you send me a placefile so I can further look into any problems?

2 Likes
#18

You can use the non-deprecated constraints to obtain the same behavior. BodyMovers are deprecated and janky, and are on minimum maintainence at the moment :sad: Using AlignPosition is a much better choice in 99% of cases.

We will still look into this further though! Apologies for the issues.

1 Like
#19

AlignPosition does not appear to behave the same as BodyPosition as it does not have a Dampening property. If they are supposed to behave the same then AlignPosition is suffering from the same issue that’s affecting BodyPosition. After this change came out, additional forces that can be applied on a character such as walking is effectively nullified when a BodyPosition or AlignPosition with the same properties post and prior to the update is present in the character.

The properties that I have applied to the BodyPosition are
MaxForce is 8000 on all axis
Power is 25
Dampening is 7
The purple character has a Walkspeed of 30 yet it is unable to move at all.

As demonstrated by the video below, you can see that the purple character is trying to run away from the player, however the force applied by the BodyMover entirely takes over all additional forces. Prior to this update, the purple character was able to run around while being pulled towards the player by the BodyMover. I’ve tried changing it to AlignPosition, however the strange behavior continues to persist.

2 Likes
#20

Is there a way I can provide a place file privately?

1 Like
#21

Can you try playing around with the “responsiveness” and “max force” settings to see if you can get what you want with AlignPosition? It will automatically assign a damping constant that will be close to the number required for the system to be critcally damped. I’m pretty confident that this should cover 99% of use cases. If you need a system that is somehow noticably over/underdamped, using a AlignPosition or BodyPosition might not be the right approach for you.

You can DM me your placefile, I can take a look to see if anything isn’t working properly :slight_smile:

2 Likes
#22

At about October 17th, when I changed the Stepping Method from ‘Default’ to ‘Fixed’, the abnormal drop was gone and the subtle natural one was still there - this brought back the desired behaviour, fixing the issue!

Now (as of around 8/9 November) this subtle one is also gone, causing these new complications
Some Test - 10 Nov.rbxl (73.4 KB)

In this placefile, the Sorcerer and a random NPC can be found. When playing, you used to be able to resist this force - essentially allowing the player to slowly push through, out of the range - instead of being stuck in one place.

Quoted from the previous message:

“I thought tweaking the Power, Damping and MaxForce would bring back the ability for the Player to resist this force, which after a lot of trying seemed to sort of work.
Something to the movement or force put towards it seemed off however. It felt as if the force wasn’t constant and that with every movement direction I took, I was able to manipulate the force to a point where sudden movements affected the force too much.”

The new behaviour appear identical on every Stepping Method. Perhaps one method could provide these new changes, whilst the other retains what the ‘Fixed’ method used to at 16/17th of October :thinking:

1 Like
#23

Any heads up on the new situation?

1 Like
#24

It’s a month later and I’m still experiencing issues due to this new change.

The initial issue with the abnormal dropping/sinking was essentially fixed when on October 17th, I changed the PhysicsSteppingMethod to Fixed, thanks to RadioGamer’s helpful reply.

.
As of the changes November ~9th, the behaviour is left in a state where yes, the initial issue is gone, BUT another issue arose. I cannot alter between these methods at all to resolve this new issue as described in my previous messages.

Both the ‘Fixed’ and ‘Adaptive’ SteppingMethod result in the same BodyPosition physics.
Seeing how Adaptive is the new Default, it’d be great - and all issues would be fixed - if Adaptive continues with the new, current physics (~ November 9th) and Fixed kept the physics of how they used to be before this recent change.
If this may be done, the entire issue would be resolved and creators would have the option to pick between ‘Fixed’ Stepping and this new responsive ‘Adaptive’ method to better suit their work.

1 Like
#25

I still haven’t heard back on this and I’m still experiencing issues.
Feeling left in the dark as there hasn’t been a response on any of my messages regarding this newly arisen issue :frowning_face:

1 Like
#26

I am still experiencing this issue.

1 Like
#27

@Hajimalago Any news? I still haven’t heard back.

1 Like
#28

Hey! Sorry for the late reply. I looked at the placefile you posted above but not sure what the procedure is to reproduce the issue. Lets start a DM so we can figure this out.

1 Like
#29

image
image840×139 9.81 KB

Ever since this DM was started, I haven’t heard anything back at all.

Sep. 14, '23
When initially getting this thread listed under Bug Reports I was optimistic about it getting attention and hopefully being resolved rather soon.

|
Sep. 15, '23

Being thanked for the report and told:

|
Oct. 13, '23

A ~month went by and no response…

|
Oct. 15, '23

Then Radiogamer comes out with a way (around it?) that fixes the issue. (thanks :D)
Case closed, right?

No.

|

Then, coincidentally the same day a staff response.

( mind it was fixed by now, despite some concerns and raised questions )

|
Nov. 9, '23

Issue… fixed?

…
Well, turns out this ‘fix’ did indeed fix the weakened power (again, this was already fixed by changing PhysicsSteppedMethod to ‘Fixed’…)
HOWEVER, it brought with it a new bug… So technically it fixed the problem, but as a result it created a new issue.

|
Nov. 10, '23

The day after the fix (that then broke something else) others and I who now face this newly create issue.
Being asked for the placefile, this was given the same day.

|
Nov. 28, '23

Nothing back yet.

|
Dec. 12, '23

Still nothing. “Suddenly no more responses?”

|
Jan. 11, '24

|
Mar. 28, '24

|
May. 1, '24

|
May. 14, '24

Finally a response half a year after seeming so involved in the matter back in November 2023.
Excited to finally hear back again, I eagerly respond the same day with a full explanation from beginning to end.

image
image825×727 75.4 KB

(full text) https://devforum.roblox.com/t/re-bodypositions-power-is-suddenly-less/2970483/2?u=milanomaster

|
Jul. 5, '24

|
Sep. 16, '24

image
image930×683 49.1 KB

|
Nov. 12, '24

And I haven’t heard back ever since.

It’s been a year ever since this second issue started, and it’s tormented and troubled ever since.
Every time I open the DevForum again I’m hoping to finally see a response, but all I’m getting is:
image

I had hopes it’d finally started getting fixed as soon as I read this

It’s been half a year since and I’m feeling – just as I said back in January – left in the dark and forgotten about. It makes me regret sending a message ever after Radiogamer helped me out.
“If only I hadn’t sent another message, then maybe it would’ve been done and fixed in October 2023.”

Thanks to their solution it worked again, but then the ‘proper/official’ fix came which broke something else :confused:

I don’t even know what to write at this point, as any call appears to get ignored and it seems that the issue will never be fixed.

@Hajimalago, would you please respond and help fix the issue at last, as again, the newly created bug as well as the thought of ‘being responsible for the second bug’ have been haunting me in my head ever since.

All the information on the issue is summarised in my DM message from May (https://devforum.roblox.com/t/re-bodypositions-power-is-suddenly-less/2970483/2?u=milanomaster), along with the suggested solution of keeping the ‘Fixed’ steppingmethod the way it was in October 2023, and Adaptive ( / the new default) the way it is now, as right now all methods are somehow identical.

Kind regards,

milanomaster

2 Likes
#31

Unfortunately a single engineer cannot always champion the issue to completion, there’s a lot of competing priorities that a single engineer will end up facing, and they may simply not have time for an involved back-and-forth anymore. Not to mention, you’re the only one having a large issue, and there’s not really any code changes that need to happen on the Roblox side. In terms of value created versus time investment, you’re better helped by someone else.

What’s the summary of the remaining issues?

1 Like
#32

You appear to have missed the point.
If there hadn’t been a ‘fix’ in the first place, Radiogamer’s solution would’ve solved it.

But since they did follow up, essentially causing a another problem to arrise and then go radio silent, leaving this in a broken state whilst still promising to figure it out. Not hearing back again feels like a broken promise - especially for something that wouldn’t have been necessary if it hadn’t been broken.

About these issues, I would highly suggest not to comment on it without properly reading up; my latest message in particular. If you had done so, you would’ve found this “summary of the remaining issues”.

image
image700×48 10.9 KB

1 Like
#33

That is a private DM. You are not helping yourself by being snarky. You’re not going to get further help from staff at this point without a new bug report. It’s clear this issue is out of scope of the original topic and has been dropped. Recommend having a little respect for the people who choose to engage with you, if there’s anything non staff can do to help mitigate your issues, they sure don’t want to try with a messy reply chain and that attitude.

1 Like
#34

Hi @milanomaster

We are closing this thread as the initial issue is fixed, and the ticket is closed!
But regarding the ongoing issues you currently have, I approved your bug report here, and we will have an engineer assigned to the ticket to take a look!

Thank you!