What's new
By:
  • Visit Rebornbuddy
  • Visit Resources
  • Visit API Documentation
  • Visit Downloads
  • Visit Portal
  • Visit Panda Profiles
  • Visit LLamamMagic

[InjectKeyEvent] Makes the bot move extremely slow.

Unknown Buddy

Unknown Buddy

Member
Joined
May 20, 2015
Messages
603
Reaction score
17
Any idea what is causing this?

Code: 
2016-04-15 02:11:34,881 [32] INFO  Logger (null) - [InjectKeyEvent] 56 ms sleep for 2.
2016-04-15 02:11:35,012 [32] INFO  Logger (null) - [Logic] Now moving towards {336, 340} because [dist (35) > MaxRangeRange (20)]
2016-04-15 02:11:35,038 [32] INFO  Logger (null) - [InjectKeyEvent] 13 ms sleep for 192.
2016-04-15 02:11:35,111 [32] INFO  Logger (null) - [Logic] Now moving towards {336, 340} because [dist (35) > MaxRangeRange (20)]
2016-04-15 02:11:35,130 [32] INFO  Logger (null) - [InjectKeyEvent] 54 ms sleep for 2.
2016-04-15 02:11:35,245 [32] INFO  Logger (null) - [Logic] Now moving towards {343, 347} because [dist (35) > MaxRangeRange (20)]
2016-04-15 02:11:35,282 [32] INFO  Logger (null) - [InjectKeyEvent] 19 ms sleep for 192.
2016-04-15 02:11:35,348 [32] INFO  Logger (null) - [Logic] Now moving towards {343, 347} because [dist (35) > MaxRangeRange (20)]
2016-04-15 02:11:35,364 [32] INFO  Logger (null) - [InjectKeyEvent] 70 ms sleep for 2.
2016-04-15 02:11:35,504 [32] DEBUG Logger (null) - [TrackMoreMobs] Now removing 1001.
2016-04-15 02:11:35,514 [32] INFO  Logger (null) - [Logic] Now moving towards {343, 347} because [dist (35) > MaxRangeRange (20)]
2016-04-15 02:11:35,544 [32] INFO  Logger (null) - [InjectKeyEvent] 7 ms sleep for 192.
2016-04-15 02:11:35,599 [32] INFO  Logger (null) - [Logic] Now moving towards {343, 347} because [dist (35) > MaxRangeRange (20)]
2016-04-15 02:11:35,620 [32] INFO  Logger (null) - [InjectKeyEvent] 65 ms sleep for 2.

Full log here: View attachment InjectKeyError.txt
 
Mine stay still!!! no Action att all

[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 103 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 106 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 108 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 111 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 103 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 107 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 106 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 110 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 107 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 113 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 107 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 107 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 108 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 107 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 107 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 108 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 104 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 108 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
[InjectKeyEvent] 109 ms sleep for 4.
[LatencyWait] Now sleeping 22 ms.
 
i'm getting cucked also


Code: 
[Logic] Now moving towards {955, 1117} because [dist (44) > MaxMeleeRange (20)]
[FinishCurrentAction] Waited 0 ms for the action to finish Whirling Blades.
[FinishCurrentAction] Waited 111 ms for the action to finish Whirling Blades.
[LootItemTask] Now attempting to interact with the item. This is the #1 attempt.
[InteractWith] Now attempting to highlight 1001.
[InteractWith] Now attempting to interact with 1001.
[LatencyWait] Now sleeping 33 ms.
[AreaStateCache] The location 1001 [Portal Scroll] is being removed because the WorldItem does not exist.
[Logic] Now moving towards {947, 1072} because [dist (30) > MaxMeleeRange (20)]
[InjectKeyEvent] 148 ms sleep for 2.
[InjectKeyEvent] 138 ms sleep for 2.
[InjectKeyEvent] 132 ms sleep for 2.
[InjectKeyEvent] 140 ms sleep for 2.
[InjectKeyEvent] 148 ms sleep for 2.
[InjectKeyEvent] 99 ms sleep for 2.
[InjectKeyEvent] 141 ms sleep for 2.
[InjectKeyEvent] 143 ms sleep for 2.
[InjectKeyEvent] 104 ms sleep for 2.
[InjectKeyEvent] 146 ms sleep for 2.
[InjectKeyEvent] 147 ms sleep for 2.
[InjectKeyEvent] 145 ms sleep for 2.
[InjectKeyEvent] 153 ms sleep for 2.
[InjectKeyEvent] 152 ms sleep for 2.
[InjectKeyEvent] 142 ms sleep for 2.
[InjectKeyEvent] 148 ms sleep for 2.
[InjectKeyEvent] 152 ms sleep for 2.
[InjectKeyEvent] 143 ms sleep for 2.
[InjectKeyEvent] 158 ms sleep for 2.
[InjectKeyEvent] 153 ms sleep for 2.
[InjectKeyEvent] 135 ms sleep for 2.
[InjectKeyEvent] 155 ms sleep for 2.
[InjectKeyEvent] 144 ms sleep for 2.
[InjectKeyEvent] 158 ms sleep for 2.
[CommonEvents] core_player_died_event(2)

View attachment Exilebuddy 12088 2016-04-14 18.20.txt
 
Last edited:
Try without Alcor75PlayerMover and see if the issue still happens (it shouldn't).

replicate and pinduca please attach your own full log and not snippets only.
 
pushedx said:
Try without Alcor75PlayerMover and see if the issue still happens (it shouldn't).

replicate and pinduca please attach your own full log and not snippets only.
Click to expand...

I disabled Alcor75 and it still happened before uploading. It may even be in the log i uploaded where i disabled Alcor75 halfway through the run. I will try completely removing Alcor75 before starting the bot to see if the problem persists.

EDIT: Completely removed Alcor75 and the same error persists.

This error in movement is not only during combat, it also happens running around in town.

Full Log without Alcor75 installed: View attachment InjectKeyError2.txt
 
Last edited:
What about removing FlaskHelper, see if that helps.

There's a PlayerMover section on the Exilebuddy Ascendancy Major API/Bot Changes thread that goes over the design changes for the DefaultPlayerMover as well as the OldRoutine skill casting spam design.

In short, the default code will use fewer input actions to make the client cast for you, which results in faster movements and skill casting.

This means super fast moving characters are finally supported because the bot isn't spamming the client with new input events (which has resulted in the too many actions DC message in the past) and I've tested a 130%ms character and it worked amazing compared to before. I don't have the gear or accounts to go any higher, but the design of the default code is such that it should work leaps and bounds better.

InjectKeyEvent is basically the old PressKey/ReleaseKey but in one function to properly support some specific things without having to leave the frame twice. The delays you see being logged have always been there, I just accidentally left the sleep logging in the Release build.

As mentioned in that thread, there's been a change in the input rate limiting, from 100ms before (a bot like 10 clicks per second) now down to a more human amount around 4-5 clicks per second, which is 200-250ms. However, the design also changed to no longer be a global cool down.

When you see

2016-04-15 02:11:35,130 [32] INFO Logger (null) - [InjectKeyEvent] 54 ms sleep for 2.

or as others posted,

[InjectKeyEvent] 113 ms sleep for 4.
[InjectKeyEvent] 152 ms sleep for 2.

Those are your flask slots.

Any non-default code that uses skills or input in the old way will now be slower, because input actions are more limited, but will be safer because the bot isn't going to trigger too many actions in a short amount of time. However, all code needs to be updated to reflect this design.

Please try a stock bot without any other plugins to confirm this. Here's a video I just made using the current release on a fast moving character: https://www.dropbox.com/s/7vn7h8drz67crsl/bandicam 2016-04-14 22-50-42-612.avi?dl=0

That is what you should be seeing from stock EB itself now, which is day and night better than how it used to be (and I know because I tested the bot on Perandus before the changes and it was terrible). If you still have issues on the stock bot, then I'll need to know some info about the machine you're running on, and character setup and so on.
 
So now with a completely fresh stock install, i am still seeing the [InjectKeyEvent] but not nearly as often, both in town and while out fighting.

With that said, i am not experiencing any of the stutter stepping or crawling / back and forth behavior i was experiencing with other plugins installed. With stock its moving / fighting pretty smooth.


So i guess now its time to install plugins one at a time, and see which one is breaking my EB!
 
Unknown Buddy said:
So now with a completely fresh stock install, i am still seeing the [InjectKeyEvent] but not nearly as often, both in town and while out fighting.
Click to expand...

Correct, we still have to send input events, but we just send significantly less now. :)

Better CR design would allow for even fewer action switches by not throwing traps or summon totems in the middle of longer spam casting when the screen is full of mobs, but the defaults should be good enough for now.
 
I disabled flask helper and immediately noticed a huge reduction in the stutter stepping, but it's not entirely gone. Looks like tonyx needs to review his coding for Flask Helper, when he gets a chance.

EDIT: Further testing, noticed that gem leveler pro was also causing various key injecting events as it has basically stopped stuttering now.

It is still doing some weird back and forth stuff with the "[MoveTowards] Adjustments being made!" until it resolves itself.

The other issue with this event is the [OpenChestTask (Leash -1)] is causing it to move backwards and forwards a few paces, stutter, backwards/forwards, stutter until it resolves a path. I'll post up a log later when I finish the run.

Edit2: Log attached. Was a long play session so hopefully it can bring up some good results. Also, the start has a few the plugins running and towards the end they had been turned off.

View attachment Exilebuddy 1684 2016-04-15 09.54.zip

Hope that helps guys.
 
Last edited:
lyvewyre said:
I disabled flask helper and immediately noticed a huge reduction in the stutter stepping, but it's not entirely gone. Looks like tonyx needs to review his coding for Flask Helper, when he gets a chance.
Click to expand...

Its not only Flask Helper atm. Since i am going though and testing 1 plugin at a time, every single plugin i add that sends keys (Alcor, FlaskHelper, etc) tremendously adds to the input lag / movement lag. With no plugins running, i walk through town fine. As soon as i enable just one plugin, you can tell im a bot because it starts to stutter abit and you see more of the [InjectKeyEvent]s, thus more micromovements.

When you use Alcor75+FlaskHelper+GemLevelerPro together, you better hope you make it out of town quick because the amount of stutter stepping looks like its quadrupled. Its like watching the bot move in slow motion a few pixels at a time. This gets even worse when in combat, its like the bot doesnt know what to do anymore. It will slowly move one direction (talking microsteps), turn around, do a circle, get itself back on track for a few, then repeat. Turning off all plugins, the bot behaves fine.

Another thing i noticed is that even when i stop the bot and disable Alcor75, it doesnt actually disable. Its unchecked, but its still whirling blading around after starting the bot back up. Need to close EB and turn it back on to actually disable Alcor75. Im sure this is a seperate issue, but still something i noticed and will report to Alcor.


Anyways, im sure the devs will be up and about soon and look into this. It really just seems like every plugin is fighting for priority against each other.
 
Last edited:
lyvewyre said:
It is still doing some weird back and forth stuff with the "[MoveTowards] Adjustments being made!" until it resolves itself.

The other issue with this event is the [OpenChestTask (Leash -1)] is causing it to move backwards and forwards a few paces, stutter, backwards/forwards, stutter until it resolves a path. I'll post up a log later when I finish the run.
Click to expand...

Global adjustments was added as an experimental way to try and avoid some bot stuck issues with moving close to walls and stuff, so there will be some side effects. Overall, to me it feels like the bot is doing better now than before, but some oddities will still exist.

Perfect movement isn't possible in this game because of how it's coded; there are very specific game issues that exist that make flawless movement impossible. You can try this though, in your CR or another plugin that is enabled, if you execute this code in Start:

PlayerMover.Instance.Execute("SetDoAdjustments", false);

Then those adjustments will stopped being made. Of course, some areas like City of Sarn or Docks will be back to having a lot of issues with walls and whatnot, but other areas that might not need the adjustments as often might be better.

Since the areas in this game are randomly generated, there's no nice way to handle everything in a generic way. Some tilesets have conflicting designs where some random layouts work great while others give the bot issues.

As for the other issue, if you go to Settings -> Global and enable Debug Last Task, you should be able to see which two tasks are causing the back and forth. A number were found and fixed during the downtime, but once again due to the random area generation, certain scenarios arise that breaks generic logic and results in those things. The bot should eventually recover though as it will blacklist the chest its stuck on after a minute or so I think.

If you catch the bot looping in a back to forth again, enable that setting, and shoot me the log and I'll see if it's a new case that needs to be worked around or try and find a better way to avoid it.
 
After some more testing with 100% stock, with Auto Flask (the default flask tool) disabled, i still see some long stuttering strings, but its much much less common.

Full Log View attachment InjectKeyEventErrors3.txt

Everything is running fine until line 5447. From here, the bot starts taking very small steps, waiting, small step, wait, small step, wait, etc. Im not sure what the bot is trying to do here. Its definitely not trying to flask, no plugin is enabled to use a flask. Its not trying to loot or fight, because there are no mobs around.

EDIT: Pushedx mentioned he may need PC specs if the problem persisted. While i dont want to share my exact specs because tinfoil hat! I dont mind PMing the specs if needed.

The rig is roughly Win 10, 32g ram, new model i7, titan-ish graphics card, large SSD. Just running from my desktop, no VMs, proxies, or anything funky. Standard Firestorm build. Using firestorm as aoe, single target, and fallback. Few auras in there as well.
 
Last edited:
I have this isses too when running bot in VM, i botting in my PC still normal but in VM bot move so slow like Unknown Buddy
 
not tested yet updating

So, couldnt test today is out of date lol
any news about update?

Cheers
 
Unknown Buddy said:
Everything is running fine until line 5447. From here, the bot starts taking very small steps, waiting, small step, wait, small step, wait, etc. Im not sure what the bot is trying to do here. Its definitely not trying to flask, no plugin is enabled to use a flask. Its not trying to loot or fight, because there are no mobs around.
Click to expand...

That log looked normal to be up until the point you mentioned. I didn't see any other extra plugins that could be contributing to the issue now, so I'm not sure the core issues behind that specific instance. I just ask for general specs to make sure obvious issues aren't happening like pathfinding on the CPU is eating up too much processing and stuff like that. Your rig looks fine, better than mine by a lot. :p

Anyways, the new design of the player mover is centered around moving like you normally play, start a move action, hold the button, then continue to move the mouse. Now, when the key is released and a new movement happens, it's because either the movement action finished completely (maybe something got stuck in the client so the character stopped) or other code in the bot executed to cancel the movement (which is a possibility, but I've not seen any huge issues myself).

Movement for me when testing, both in VM and limited FPS seems extremely well compared to before and the issues that could happen with faster moving characters. However, nothing is perfect, and the game does have its own issues that can contribute to cases where movement doesn't appear perfect, which is unavoidable really. For now, there's nothing to change really because the old method of movement by spamming clicks is not viable (and worrying in terms of potential detection) so I'll keep my eye out on any other possible improvements that can be made.

Also, if you're in lockstep, any sort of client/server issues will be noticeable, as your game stops, and movement breaks, so try switching to predictive to see if that helps if you've not tried already.
 
I'm investigating FlaskHelper's stuttering issue, I don't use any waiting but it might be related to the GCD timer, I keep you informed :)
 
i haev this error pls can some1 help me?
[LatencyWait] Now sleeping 58 ms.
[UseAt] 1 False {694, 1108}
[LatencyWait] Now sleeping 58 ms.
[UseAt] 1 False {694, 1108}
[LatencyWait] Now sleeping 58 ms.
[UseAt] 1 False {694, 1108}
[LatencyWait] Now sleeping 58 ms.
[UseAt] 1 False {694, 1108}
[LatencyWait] Now sleeping 58 ms.
 
You must log in or register to reply here.
Back
Top