Jump to content

Question

Posted

Hi guys,

 

So recently i did a new IL server for our small group, and i noticed that there is a bug regarding speed. If the characters base runspeed is standard, and you toggle/buff with a certain buff that reduces the speed temporarily, the runspeed goes down to 0 until you cancel that skill. I recompiled my server, checked the skills settings for the runspeed reduction, checked the core settings for the formula calculations, the problem still persists. The only solution i found that works is if i increase the base runspeed by 30, but this isn't a fix. Any ideas why?

 

Here is the link to a short gif to see whats happening:

SpeedBug.gif 

7 answers to this question

Recommended Posts

  • 0
Posted

So, regarding your question I did some testing with a Human Fighter (no speed boost). Base run speed 126.

If i buff with WW lv2, thats +33 speed, totaling at 159 (ok).

Adding Quick Step passive lv2, thats +11 speed, totaling at 170.

Equipping the DC robe set that gives +7 speed, totaling at 177.

As you can see, I can increase the speed as much as I want, and its correct.

Now, with 177 speed, if i activate Parry Stance which is -10% off the basic runspeed (126-10% = 12.6)  (<mul order="0x30" stat="runSpd" val="0.9"/>)  it gives me 164 run speed which is correct again. But, after several tests, i came to the conclusion that the speed decrease is working until the runspeed reaches 115. If the value goes below, it turns to 0. 

I just don't get it why...In the SQL table at the char_templates, the MOVE_SPEED is 115.

This is one weird bug to be honest...

  • 0
Posted

So, "run to walk" toggle also triggers that 0 behavior ?

 

And if you disconnect while being 0, you get 0 at player store ?

 

Buffing you enough (ww2 + run speed totem + bers + ...) + parry stance debuff, for a run speed total higher than 115, don't trigger the behavior ?

 

----

 

You should check about custom config related to speed.

 

If you're sure "115" value triggers it, search for this value on core and  properties.

  • 0
Posted

So after a long and painful testing and searching the core what might cause this bug, i finally caught the little bastard.

My knowledge in coding is mediocre at best, but i still managed to fix the problem.

In the java file gameserver/model/actor/stat/CharStat.java is where the speed is calculated. I compared this file with an older core i have and noticed a small difference:

 

This is the bugged code:

 

	public final float getMovementSpeedMultiplier()
	{
		if (_activeChar == null)
		{
			return 1;
		}
		
		return getRunSpeed() / _activeChar.getTemplate().baseRunSpd;
	}

And this is the fixed code:

	public final float getMovementSpeedMultiplier()
	{
		if (_activeChar == null)
		{
			return 1;
		}
		
		return (getRunSpeed() * 1f) / _activeChar.getTemplate().baseRunSpd;
	}

As you can see, the getRunSpeed function was missing this character * 1f. Not sure what it does, but my guess would be it rounds up the final value. I might be wrong. Either way, its fixed now. Thanks for the advice.

 

  • 0
Posted

If both getRunSpeed() and _activeChar.getTemplate().baseRunSpd are int, then you lose the division accuracy. Introducing 1f add back the float behavior.

 

But normally those values aren't supposed to be both int. Not sure what is your getRunSpeed(), but it should return float normally.

  • 0
Posted (edited)
21 hours ago, Tryskell said:

If both getRunSpeed() and _activeChar.getTemplate().baseRunSpd are int, then you lose the division accuracy. Introducing 1f add back the float behavior.

 

But normally those values aren't supposed to be both int. Not sure what is your getRunSpeed(), but it should return float normally.

Actually, L2J has been using both as integers until Freya (in the case of getRunSpeed, it's probably because packets write speed stats as integers and it seemed an easy solution to them).

Here is the method he's probably using.

	/**
	 * Return the RunSpeed (base+modifier) of the L2Character in function of the Armour Expertise Penalty.
	 */
	public int getRunSpeed()
	{
		if (_activeChar == null)
		{
			return 1;
		}
		
		// err we should be adding TO the persons run speed
		// not making it a constant
		double baseRunSpd = _activeChar.getTemplate().baseRunSpd;
		
		if (baseRunSpd == 0)
		{
			return 0;
		}
		
		return (int) Math.round(calcStat(Stats.RUN_SPEED, baseRunSpd, null, null));
	}

They even preferred rounding value and casting it to int instead of returning it as it is.

Edited by DnR
  • 0
Posted

The first example was not returning a float result as the second one seems like it's the correct one however when you want to do a division between 2 floats and you want the exact result you have to type something like "2/.1f”

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...