Bat position shift after contact with table | Eleven Table Tennis | Page 1

river roost Apr 9, 2026, 8:24 AM

#

Hello,

yesterday I have noticed an issue while playing, I wanted to make a FH flick and made a contact with table - so my hand physically went through the table and the bat rotated upwards - since it doesn't follow my hand through the table - and after it didn't return to the original position, only partly and it stayed partly in that rotation upward. I tried then multiple times and the same, so i had to calibrate again. And also after starting the game again, the position slighly shifts, especially the rotation of closing angle, but I'm not sure, if it's a Quest tracking inaccuracy, could easily be. I have to calibrate all the time the closing angle, the FH angle gets more closed after closing and re opening the game. I tried to upload to cloud and download, but it downloads the setting with that incorrect angle, not a big difference, but noticeable when playing. I can notice it, since I play in passthrough, so I see the relative position of bat to my physical hand.

Thank you for your response

burnt blade Apr 9, 2026, 1:08 PM

#

Woah, that's weird. Are you saying that right now if you start the game and then the bat is at a certain angle relative to your hands but if you move it into the table and through the table then when it snaps back to your hand it's now at a different angle than before?

river roost Apr 9, 2026, 2:12 PM

#

burnt blade Woah, that's weird. Are you saying that right now if you start the game and then...

Yes, exactly that was happening yesterday. I'll try it again todat, just to be sure

river roost Apr 9, 2026, 7:39 PM

#

@burnt blade Okay, I am confirming, the situatuon is the same today as well, I touched the table and had to recalibrate the Z axis, it shifted cca. 2-3cm's...(at least I think it was the Z axis, the bat got further away from me)

burnt blade Apr 9, 2026, 7:39 PM

#

strange...will check thanks for confirming

river roost Apr 9, 2026, 7:40 PM

#

burnt blade strange...will check thanks for confirming

Thank you very much 😉

burnt blade Apr 10, 2026, 12:53 AM

#

@river roost i can't replicate it. I have tried a bunch of ways.
Do you haev time to replicate and share logs and chat about it
and i'd also like ot see a video of the result - just so i can understand the effect completely

river roost Apr 10, 2026, 4:48 PM

#

@burnt blade Of course, I thought about capturing it, but I'm not sure if it will be visible. The most evident change is visible, if you compare the virtual bat with physical one in the followimg manner : you hold the bat in front of you and you peek through that little hole that is in between your nose and the headset (i use the stock headstrap), so below that "dividing line" is the real bat and above the virtual one, that is how I precisely calibrate the bat and this way any misalignment can be spotted most accurately. But an important note to consider in terms of debugging is, that if the position shifts and I download the previously uploaded position from cloud, after downloading the position is exactly misaligned in a way, it was after hitting the table. So I a. just thinking out loud -> the data must be untouched and the issue will be somewhere else, because the cloud data cannot be affected by hitting the table, if I understand it correctly, the only way from user's perspective to change cloud data is by uploading. So that's why I don't where is the issue, but unfortunately i don't exactly know how quest tracking works and the codebase communicates with it, I cannot debug further

burnt blade Apr 10, 2026, 4:52 PM

#

hmmmm there are other ways to debug

are you using an adapter or a controller (sounds like you are?).
2, you can use the butt of the other controller as a sort of probing spot...so if you put it at the base of the handle, you can see how it lines up after the event
also, instead of looking through the nose hole, you can try to look in mixed reality..though there is skew there anyway..but if something is suddenly off by a few cm then it woudl be visible there as long as you hold it in the same relative spot
where the headset thinks the controller is is pretty stable... so if you go to the custom positioning menu in the game, it will render the controller and you can check if the controller is in teh same spot relative to the virtual racket. what might be the case is that it is in the same spot, but you feel the virtual controller itself is in the wrong spot after some event

river roost Apr 10, 2026, 5:09 PM

#

I am using SolidSlime v3
I am not sure how do you mean this exactly
I used to do it that way, but I found out that it distorts the real position quite a lot, so I rather rely on physical reality 🙂
I have thoroughly checked the position yesterday in the following manner :
While I was training/playing I checked multiple times the position through that little hole, and the position was perfect every time. Although after closing and re opening the game the position shifts a tiny bit, mostly the closing angle, but it is negligible - again after downloading the position from cloud, the angle is the same. And after only touching the table even mildly, the position shift is immediately obvious. I'm not sure if this answered you question #4🙂

burnt blade Apr 10, 2026, 5:16 PM

#

what i mean is... there are 3 of data transfer

real world mapped to controller position
controller position mapped to paddle slot position
paddle slot position holding the virtual paddle through a system that allows the virtual paddle to be pushed around by the table

i'm trying to assess which one of these is wrong

when you download a cloud settings (why are you downloading a cloud setting and not using a preset for ssv3?)
that only affects 2

(2) is something that can be logged and observed very easily programmatically, and can be observed by eye, in game, when you go into the paddle position menu and try to edit your position . in this mode, the virtual controller is rendered along with the virtual paddle. So if there is a shift, you can see this, adn also easily take a screenshot and it would be easy to compare A-B

(1) is a bit harder to measure and does infact change a tiny bit i assume at application/headset initialization/wake. There isn't much you can do about this shift...though recalibrating through the calibration menu at least lets you offset any offset (but taking the headset off and putting it back on might introduce a new offset)

(3) could be a game bug, but that too would be visible similar to 2

river roost Apr 10, 2026, 5:28 PM

#

Why am I not using the default ssv3 setting : I had been using it for a long time and I have not played IRL also for a long time. After a long time I decided to try IRL and missed literally every single ball - every single stroke landed above the ball IRL. So I started to investigate where is the issue and at this point I found out, the default ssv3 position places the racker - not going to exaggrate - 5cm below my hand, so I get used to make the strokes higher and that resulted in missing every single ball IRL. Since i set my own position and lifted the racket 5cm higher and got accustomed, I can play again IRL as before and now it matches the position...so this is the reason...but now I am not sure if my Quest isn't having some tracking issue, since it seems to me I am the only one noticing this issue...

burnt blade Apr 10, 2026, 5:30 PM

#

@river roost wow... i wish you reached out to me first instead of going through all of that

#

there could be a bug, there could be so many things

#

perhaps it's a bug that effects other people, i would love to have known and to fix it promptly

#

but yea...a lot of people use (i am assuming that is the latest adapter) and no one has had this issues, so it might be something breaking on your side

river roost Apr 10, 2026, 5:34 PM

#

One thing though to mention, I turned off the boundary setting in developer mode, sice the room I am playing isn't very big and whenever I stepped back the table disappeared since I was close to the wall...could that be the issue?

burnt blade Apr 10, 2026, 5:34 PM

#

oh no hah

#

sounds like you are in statoinary mode

#

that's not great for many reasons..but if you are turning that off aaaaaanyway..it is actually much better to just define a big roomscale boundary far beyond your room

river roost Apr 10, 2026, 5:38 PM

#

I will answer a bit later to your questions I did't answer, now I am going to play a bit and try to turn on the boundary beyond the room...curious 🙂
Thank you very much so far for your time, I really appreciate it 🙏

river roost Apr 10, 2026, 11:05 PM

#

I wanted to answer to you questions after play, but today, as if by magic it wasn't happening. The only thing I changed was, that I turned back on the physical space features in my Quest developer options and I created a boundary manually, as you advised, bigger than my room.

But if I would have to bet, I would place my bet on the option 2, because I remember that all the time if the position shifted I noticed during calibration that the controller was in a different position in relation to virtual racket. It was easy to notice, since I use Stiga Cybershape and the controller's edge is almost exactly at the lower-near edge of the Cybershape. But I don't want to claim this for 100%, but I am 90% positive, that the shift happened in relation controller-virtual bat, so option 2.

burnt blade Apr 10, 2026, 11:42 PM

#

ok, well if two, then that's something that can be confirmed programmaticaly so if you encounter a repeatable case, i'd love to resolve that

#Bat position shift after contact with table