[0:05:45]: [Reflection] Trying to solve: Beekeeper
[0:05:47]: [Reflection] Random solving failed for Beekeeper
Was fishing in Karamja, nothing out of the ordinary happening.
Wanted
2011-01-06 23:07:58
How is it possible to get a false positive with reflection???
mormonman
2011-01-06 23:09:10
From: IceFire908
How is it possible to get a false positive with reflection???
The detection functions aren't good.
Hobbit
2011-01-06 23:18:34
Its happened to be before with reflection
Edit: lol perfect timing
[Reflection] Trying to solve: Beekeeper
[Reflection] Random solving failed for Beekeeper
cooking lobsters in rogues den
Overtime
2011-01-07 01:01:28
This was brought up months ago.
When MSI first got to use Reflection when Cstrike started helping with reflection randoms.
Happened while testing the seers barbarian fishing walking from the bank to fishing spot, it detected for an odd reason the beekeeper random when it was just walking.
Looks like things haven't changed.
Bad Boy JH
2011-01-07 04:00:04
I keep saying that is the NPC poping up to grab someone else into a random, its a 1 in a trillion chance, but it happens..
In other words, we need to get a better random detection.
Edit, could we add a Interacting with me double check, most randoms make you interact with a character. if it does the basic detect, then it waits for 2 seconds for the character to interact with an NPC, that seems to be the simplest solution.
From: Bad Boy JH
I keep saying that is the NPC poping up to grab someone else into a random, its a 1 in a trillion chance, but it happens..
In other words, we need to get a better random detection.
Edit, could we add a Interacting with me double check, most randoms make you interact with a character. if it does the basic detect, then it waits for 2 seconds for the character to interact with an NPC, that seems to be the simplest solution.
How about if that NPC is found to pop up, wait a couple seconds and see if you actually
in BeeKeeper before trying to solve?
Iamadam
2011-01-07 05:39:16
Or you could not even bother watching for the NPC to pop up. Just check your location, and if the tile is within the beekeeper random, solve it. That way you can't get a false positive, and the beekeeper showing up for someone else won't pause the script at all.
KingKong
2011-01-07 06:27:57
^ That would be the best way for randoms to be solved since there is no chance of going wrong(hasn't happened to me anyway)
Bad Boy JH
2011-01-07 06:59:12
Guys, I am manually superheating over the next while, do you devs want me to get the Tile Box, and perhaps even the plane, of different randoms?
edit: I have one for MILES, don't know if its going to be different for Giles and the other one…
Bottom Left (8595, 4498)
Top Right (8604, 4508)
Plane Index - 0
mormonman
2011-01-07 10:03:27
From: Bad Boy JH
Guys, I am manually superheating over the next while, do you devs want me to get the Tile Box, and perhaps even the plane, of different randoms?
edit: I have one for MILES, don't know if its going to be different for Giles and the other one…
Bottom Left (8595, 4498)
Top Right (8604, 4508)
Plane Index - 0
Some randoms have random tileboxes. It would be better to have an object that is in the random.
Naike
2011-01-07 10:16:37
As mormonman said..
In the future we'll have NPC + Object checking, it's a honest mistake on my behalf :(
pyroryan
2011-01-07 11:22:06
From: Naike
As mormonman said..
In the future we'll have NPC + Object checking, it's a honest mistake on my behalf :(
Partially working fix > No working fix.
3Garrett3
2011-01-07 11:30:57
How about if it detects the npc, quickly grab our location, then wait a couple seconds (for the teleport) then use a DistanceFrom(newcurrenttile). If the distance is more than a few tiles we've been teleported. Would be a quick, dirty, workable fix so people would stop complaining about it while a new, better fix was developed.
Wanted
2011-01-07 15:17:19
From: pyroryan
Partially working fix > No working fix.
I disagree. :-P
No working > False positives.
pyroryan
2011-01-07 15:19:03
From: IceFire908
I disagree. :-P
No working > False positives.
No account flag from random > Account flag from random.
Assuming Jagex does flag the account if it fails to complete a random in a specific amount of time.
Wanted
2011-01-07 15:28:32
From: pyroryan
No account flag from random > Account flag from random.
Assuming Jagex does flag the account if it fails to complete a random in a specific amount of time.
The script will malfunction for no reason randomly, seems like a bad trade off.
Becks
2011-01-07 15:35:05
From: NCDS
[0:05:45]: [Reflection] Trying to solve: Beekeeper
[0:05:47]: [Reflection] Random solving failed for Beekeeper
Was fishing in Karamja, nothing out of the ordinary happening.
I get this at-least 3 times a day in rogues den
[Reflection] Trying to solve: Beekeeper
[Reflection] Random solving failed for Beekeeper
3rd time today
3Garrett3
2011-01-07 16:42:48
From: IceFire908
I disagree. :-P
No working > False positives.
You'd rather not have a working random than one that would work and have false positives? That's kinda off the general beating path for the solutions mentioned, but I don't see why anyway.
It was said that a fix that works but is messy and not very efficient is better than no fix which leaves behind false positives all the time. Perhaps one of us misunderstood, but it seems that you're saying to break the solver (no working random solver if I understood correctly) is better than false positives - which are not really a hindrance to the script, it just debugs that it's failed the random a lot.
Wanted
2011-01-07 16:46:39
From: 3Garrett3
You'd rather not have a working random than one that would work and have false positives? That's kinda off the general beating path for the solutions mentioned, but I don't see why anyway.
It was said that a fix that works but is messy and not very efficient is better than no fix which leaves behind false positives all the time. Perhaps one of us misunderstood, but it seems that you're saying to break the solver (no working random solver if I understood correctly) is better than false positives - which are not really a hindrance to the script, it just debugs that it's failed the random a lot.
If the entire purpose of a random solver to extend the longevity of a script run, then what good is it if it thinks you're in the random before you actually are. The hindrance is it will terminate the script early, at the exchange price of the chance that it will pick it up at the right time.
Bad Boy JH
2011-01-07 18:54:47
From: mormonman
Some randoms have random tileboxes. It would be better to have an object that is in the random.
should I start gathering Object IDs from randoms then?
mormonman
2011-01-08 01:29:02
From: Bad Boy JH
should I start gathering Object IDs from randoms then?
That and/or other npc ids/names that could be used to double check.