Lbp2 bugs list

[Tygers]

I'm surprised this isn't on the list....

Timer duration input is inaccurate.

There are 3 parts to this...

1. Timers of certain duration will be 1 frame shorter than what they should be. For example, a timer set for 1.0s should be 30 frames, but it only runs for 29 frames.

2. The same is true of the current value setting; if you set a 1.1s timer for 1.0s, it should be at 30/33 frames, but it will actually be at 29/33 frames, and run for 4 more frames.

3. The default timer you place has a value of 4.0s. Normally if you select 4.0s, it is really 3.96s (3 seconds+29 frames), but the default setting is a true 4.0s.

Testing this:
The third is the easiest to test. Place a fresh timer, hook up a 0% battery to its input (To disable it running), then tweak it and set the current time to 0s, being careful not to change the total time. Now press right on the D-pad 4 times bringing the current time to 4.0s. At this point, the output should be active, but it is not. Press it one more time, and the output becomes active. Alternatively, create two timers. Hook a toggle up to both and turn it off. Set both to current time 0s, but tweak one to change it's total time to something else, then back to 4.0s. Wire each timer up to its own reset. Now let them run. They should both run together in sync, both being set for 4.0s, but they will start tun get out of sync.

To test the first two, I created a circuit which can mesure which signal comes first, down to frame accuracy. This tells me if one timer fired fist, or both at the same time. Then I set up the timers with some parameters which SHOULD fire at the same time, but which don't. For example, one timer set to 0.4s with 0.3s current time, and the other set to 0.5s with 0.4s current time. (The 0.5s one fires first.) Or 0.6s with 0.5s current time vs 0.7s with 0.6s current time (The 0.6s fires first.) As a control, I did tests like 0.4s with 0.3s current time vs 0.7s with 0.6s current time (Fired simultaneously). Or for another test, I hooked timers up in series, and set them all to 0.0s, with different times that should add up to the same value. (For example two 0.5s timers vs a 0.4s and 0.6s, the two 0.5s fire first.)

The same inaccuracy in settings exists in sequencers, but NOT in randomizers. (A randomizer, it turns out, is the most accurate way to time exactly 1 second.)

[mistervista]

Thanks Tygers. I had this problem with timers the other day when trying to use one for a health bar and work out how many hits equated to each second but I couldn't because it just didn't seem to tally. I can't remember the figures offhand but say for example I set it to 10 seconds and it took 20 hits, I assumed then that setting it to 20 seconds would take 40 hits but when I tried it, it was nowhere near. It was 5 or 6 hits different.

I didn't realise it was a bug, I thought it was just me not being able to work it out properly. I had no idea what the timings should actually be. I'll add it to the list now.

[Shifty Geezer]

Tag sensor's range is inaccurate. There's a similar bug with distances, which aren't accurate. I've had Tags inside a sensor's range not trigger the sensor, but most noticeably you can see the sensor switch off when you gradually move a tagged object out of range, and not switch on again until it's much nearer - the detection range changes between losing a signal and gaining.

Also, battery signals and similar don't get transmitted through a music sequncer, and maybe other nested sequencers. That is, to change behaviour of a sackbot on the sequencer I ran a battery from the sequencer, through the music sequencer, and out to an SBot chip. If I ran the sequencer, the battery signal worked, but if I ran the music sequencer which triggers the sequncer, the battery never worked. Tried replacing it with a one-shot counter but same problem. So I used a labelled tag to wirelessly trigger changes. This is actually the better solution, as you can set up every actor in a scene to resopnd to the one signal; each tag is labelled scene 1a, 1b, 2a etc. in the sequencer and a full-range sensor is added to SBots and whatever in the scene needs to be triggered. this makes 'wiring' a lot easier and more intuitive, just reuiring a sensor of the right label for that part of the vid to trigger whatever you want.

[ll_ye]

is it just me or is barbed wire using the wrong texture?

[flamingemu]

The grabinators sometimes push an object through a layer. for example:

----------------------------------................................O
-----------------------------------------------...................I

O
I is sackboy's height.
---is the floor/ ceiling

if sackboy grabbed an object with the grabinators while he was between the floor and ceiling (provided the object was there too) sometimes the object instead of just being dragged around, will pop up striaght through the top layer.

[Tygers]

Tag sensor's range is inaccurate. There's a similar bug with distances, which aren't accurate. I've had Tags inside a sensor's range not trigger the sensor, but most noticeably you can see the sensor switch off when you gradually move a tagged object out of range, and not switch on again until it's much nearer - the detection range changes between losing a signal and gaining.

I think that's intentional (Player sensors are the same) - it's a sort of "debounce" mechanism, so a tag doesn't start rapidly triggering something on and off; that way it has to cross one threshold to turn on, but another to turn off. Otherwise something right on the edge might some times trigger the sensor, sometimes not, especially if they are both moving "together" but not exactly sticking together.

[Aya042]

I think that's intentional (Player sensors are the same) - it's a sort of "debounce" mechanism, so a tag doesn't start rapidly triggering something on and off; that way it has to cross one threshold to turn on, but another to turn off. Otherwise something right on the edge might some times trigger the sensor, sometimes not, especially if they are both moving "together" but not exactly sticking together.

Yep. It's been the same since LBP1 - check out this thread.

[Chazprime]

is it just me or is barbed wire using the wrong texture?

I think it's an artifact. I reported that on their support site and they've logged it, hopefully a fix is incoming.

[Blue Helmet]

Another bug: You know those Story objects (the ones that aren't lights) that you can tweak to change their color, such as the LBP1 balloons, the shoe, or the electric razor? Well, like most things with tweakable colors, they have two color/brightness settings, one for if a switch that's wired up to them is turned off. Guess what - that second color/brightness setting is entirely useless, since even though you can wire switches up to them, they have no effect (other than just transferring player color data).

On a related note, I once had a piece of hologram refuse to let its color be tweaked: I could still use the tweak menu, but the holo wouldn't update its appearance. Replacing it with a fresh piece of holo fixed that problem, though.

[LOLinator v1.02]

All i know is that if you place a sackbot with avalon or da vinci costume 2 medium grid spaces above the ground in pause, the legs stretch to the ground.

[Matimoo]

All i know is that if you place a sackbot with avalon or da vinci costume 2 medium grid spaces above the ground in pause, the legs stretch to the ground.

This happens with all sackbots, and sackboy too (I think it happens if you stand on the edge of ever-so-slightly uneven ground, he may extend one of his legs to touch the lower area)

[ll_ye]

I think it's an artifact. I reported that on their support site and they've logged it, hopefully a fix is incoming.

http://i0.lbp.me/img/ft/c00505dad96f...4186aa7086.jpg

Looks like the entire border is bugged.

[Shifty Geezer]

Yep. It's been the same since LBP1 - check out this thread.

Okay, I see what they're doing and it's by intent. But the hysteresis zone is a percentage of the full range, resulting in huge none-changing spaces when using long-range sensors. The notation that an object can move several units into or out of range of a sensor without changing its state can lead to unexpected behaviour (at least until educated), and I think the hysteresis zone should be of a fixed range, such as one small grid square, so the standard behaviour is more in line with the typical user's expectations.

[Ungreth•]

Sackbots that use movers or followers to fly will get squished any time they fly into an angle in the scenery, where two pieces of material meet. Normally a sackboy would need to get wedged in a very tight angle to get squished, but flying sackbots seem unusually and annoyingly squishable. I'm guessing that's a bug.

[qsksw]

My friend and I were in Create Mode getting pins for him. While he was trying to get the Puppet Master trophy I brought out a Sackbot to help him figure out the Tweak Menu. Then the game diverged because his internet sucks, and my unplaced Sackbot stayed where it was. After I rejoined him, we switched to Playmode, and it was still there. It cannot be reseletected with the popit. http://lbp.me/p/zts-35

This game has so many glitches. So many awesome glitces.