Edit by @unknown, updated by @unknown:
Original test world was unreliable because it had the observer looking at lit redstone dust across a chunk border. Test world and description updated to include other blocks on November 2, 2021.
Observers looking at several kinds of blocks emit a pulse on world reloading. The attached test world demonstrates the bug. The following blocks were tested in 1.17.41 Hotfix:
Block | Triggers observer pulse on relog |
|---|---|
Lit redstone dust | ✔ |
unlit redstone dust | ❌ |
Redstone torch (lit or unlit) | ❌ |
Repeater (powered or unpowered) | ❌ |
Comparator (powered or unpowered) | ❌ |
Empty single chest / barrel | ❌ |
Empty double chest | ✔ |
Chest / barrel with item in it | ✔ |
Dropper (empty or with item) | ❌ |
Item frame | ✔ |
Glow item frame | ✔ |
Tripwire with mob on it | ✔ |
Pressure plate with mob on it | inconsistent |
Mob head | ✔ |
Campfire | ✔ |
Cauldron | ✔ |
Steps to reproduce:
Open the attached world.
Close and open it again.
See what the dispenser does.
Strangely, the pulse is too short (0 ticks maybe?) for dispensers to trigger so you need a repeater to make the pulse a bit longer.
Linked issues
is duplicated by 29
relates to 4
Attachments
Comments 38
I can confirm in stable 1.6.1 that it is `redstone_wire` that is powered (error message of `/testforblock <pos> redstone_wire 0` says about a different data value).
Edit: Using a normal Piston for the output can also detect the result easily (put a pushable block on its face)
(Tested before reported - from 'Technical Bugs List' https://docs.google.com/document/d/1nziKYzCr4pBdCRj2gIMlfr9MVdWHd6HyafdiBHx9F0c/edit)
This issue happens in 1.15.0.53 also.
I'm guessing this probably also relates to MCPE-180268 (found it through mutually related bug MCPE-179848)
Still a issue
Also observers pulse on relogging when facing redstone powered by a storage facing comparator
This bug makes it so much harder to use the copper bulb. Real shame. Copper golem interaction is less than useful because of this as well. Hope it can be fixed.
Thanks PHO! I didn't perform enough testing to realise that it was only due to the observer, and I'm glad you've deduced that.