Remove Advertisements

PHP WoW Tanking Simulator

All things /script related

Moderators: Fridmarr, Worldie, Aergis, _Chloe

Postby solina » Thu Aug 28, 2008 9:50 am

moduspwnens wrote:
solina wrote:Also: "Click to repeat" doesn't retain mob values.


This should now be fixed. I also fixed a few other similar errors that I noticed from seeing this one. Thanks!


Looks good. My test was in IE6.

I put in my unbuffed stats, and 3 fast-swining mobs (1.2), and was surprised to learn I would eat more crushings than get redoubt blocks. Is that right?

(My link: http://www.bennlinger.com/tanksimulator/index.php?class=paladin&miss_chance=11.84&dodge_chance=26.35&parry_chance=16.84&block_chance=17.93&crushing_blows_on=1&number_of_mobs=3&average_attack_speed=1.2)


Edit: Something's definitely wrong.
Tanking Simulator wrote:Mob 1: DODGE
Mob 2: BLOCK (PASSIVE) - Redoubt proc.
Holy Shield fades.
Redoubt fades.
Mob 3: HIT (CRUSHING)
Image
Image
solina
 
Posts: 374
Joined: Thu Mar 20, 2008 1:24 pm

Postby moduspwnens » Thu Aug 28, 2008 9:59 am

solina wrote:
moduspwnens wrote:
solina wrote:Also: "Click to repeat" doesn't retain mob values.


This should now be fixed. I also fixed a few other similar errors that I noticed from seeing this one. Thanks!


Looks good. My test was in IE6.

I put in my unbuffed stats, and 3 fast-swining mobs (1.2), and was surprised to learn I would eat more crushings than get redoubt blocks. Is that right?

(My link: http://www.bennlinger.com/tanksimulator/index.php?class=paladin&miss_chance=11.84&dodge_chance=26.35&parry_chance=16.84&block_chance=17.93&crushing_blows_on=1&number_of_mobs=3&average_attack_speed=1.2)


Edit: Something's definitely wrong.
Tanking Simulator wrote:Mob 1: DODGE
Mob 2: BLOCK (PASSIVE) - Redoubt proc.
Holy Shield fades.
Redoubt fades.
Mob 3: HIT (CRUSHING)


Checkin that out atm.
I rule.
moduspwnens
Moderator
 
Posts: 6211
Joined: Tue Nov 06, 2007 12:28 pm
Location: Shattered Hand

Postby moduspwnens » Thu Aug 28, 2008 11:13 pm

For some reason, one of the variables wasn't getting reset. Anyway, it's making sure it's reset now, so this issue should be resolved. Good catch, and thanks. :)
I rule.
moduspwnens
Moderator
 
Posts: 6211
Joined: Tue Nov 06, 2007 12:28 pm
Location: Shattered Hand

Postby solina » Fri Aug 29, 2008 6:53 am

moduspwnens wrote:For some reason, one of the variables wasn't getting reset. Anyway, it's making sure it's reset now, so this issue should be resolved. Good catch, and thanks. :)


Cool! Testing's (part of) what I do for a living.. this comes naturally.

One suggestion: can you have an option to stagger the start times of multiple mobs? I think my numbers are getting skewed slightly by having continual simultaneous strikes (where in-game I'd be able to put up HS between mobs).

One other thing: seeing some RNG weirdness on Redoubt. Went over 40 seconds without a proc.. is this just RNG, or did something misfire?
Tanking Simulator wrote:19.2s | Redoubt fades.
Mob 3: 19.2s | BLOCK (PASSIVE)
19.2s | Holy Shield fades.
20s | Holy Shield cast.
Mob 1: 20.4s | MISS
Mob 2: 20.4s | MISS
Mob 3: 20.4s | DODGE
Mob 1: 21.6s | MISS
Mob 2: 21.6s | DODGE
Mob 3: 21.6s | BLOCK (HOLY SHIELD)
Mob 1: 22.8s | DODGE
Mob 2: 22.8s | PARRY
Mob 3: 22.8s | BLOCK (PASSIVE)
Mob 1: 24s | BLOCK (HOLY SHIELD)
Mob 2: 24s | BLOCK (HOLY SHIELD)
Mob 3: 24s | BLOCK (PASSIVE)
Mob 1: 25.2s | BLOCK (HOLY SHIELD)
Mob 2: 25.2s | DODGE
Mob 3: 25.2s | MISS
Mob 1: 26.4s | BLOCK (HOLY SHIELD)
Mob 2: 26.4s | PARRY
Mob 3: 26.4s | DODGE
Mob 1: 27.6s | BLOCK (PASSIVE)
27.6s | Holy Shield fades.
Mob 2: 27.6s | DODGE
Mob 3: 27.6s | DODGE
Mob 1: 28.8s | BLOCK (PASSIVE)
Mob 2: 28.8s | DODGE
Mob 3: 28.8s | BLOCK (PASSIVE)
Mob 1: 30s | DODGE
Mob 2: 30s | HIT
Mob 3: 30s | BLOCK (PASSIVE)
30s | Holy Shield cast.
Mob 1: 31.2s | PARRY
Mob 2: 31.2s | DODGE
Mob 3: 31.2s | DODGE
Mob 1: 32.4s | DODGE
Mob 2: 32.4s | PARRY
Mob 3: 32.4s | DODGE
Mob 1: 33.6s | BLOCK (PASSIVE)
Mob 2: 33.6s | DODGE
Mob 3: 33.6s | BLOCK (PASSIVE)
Mob 1: 34.8s | DODGE
Mob 2: 34.8s | BLOCK (PASSIVE)
Mob 3: 34.8s | PARRY
Mob 1: 36s | MISS
Mob 2: 36s | MISS
Mob 3: 36s | MISS
Mob 1: 37.2s | BLOCK (HOLY SHIELD)
Mob 2: 37.2s | DODGE
Mob 3: 37.2s | PARRY
Mob 1: 38.4s | MISS
Mob 2: 38.4s | BLOCK (PASSIVE)
Mob 3: 38.4s | DODGE
Mob 1: 39.6s | BLOCK (HOLY SHIELD)
Mob 2: 39.6s | MISS
Mob 3: 39.6s | PARRY
40s | Holy Shield fades.
40s | Holy Shield cast.
Mob 1: 40.8s | DODGE
Mob 2: 40.8s | PARRY
Mob 3: 40.8s | BLOCK (HOLY SHIELD)
Mob 1: 42s | BLOCK (HOLY SHIELD)
Mob 2: 42s | BLOCK (HOLY SHIELD)
Mob 3: 42s | BLOCK (PASSIVE)
Mob 1: 43.2s | MISS
Mob 2: 43.2s | PARRY
Mob 3: 43.2s | DODGE
Mob 1: 44.4s | BLOCK (PASSIVE)
Mob 2: 44.4s | BLOCK (HOLY SHIELD)
Mob 3: 44.4s | DODGE
Mob 1: 45.6s | BLOCK (HOLY SHIELD)
Mob 2: 45.6s | PARRY
Mob 3: 45.6s | DODGE
Mob 1: 46.8s | DODGE
Mob 2: 46.8s | BLOCK (PASSIVE)
46.8s | Holy Shield fades.
Mob 3: 46.8s | DODGE
Mob 1: 48s | BLOCK (PASSIVE)
Mob 2: 48s | PARRY
Mob 3: 48s | HIT (CRUSHING)
Mob 1: 49.2s | HIT (CRUSHING)
Mob 2: 49.2s | MISS
Mob 3: 49.2s | PARRY
50s | Holy Shield cast.
Mob 1: 50.4s | DODGE
Mob 2: 50.4s | PARRY
Mob 3: 50.4s | PARRY
Mob 1: 51.6s | BLOCK (PASSIVE)
Mob 2: 51.6s | DODGE
Mob 3: 51.6s | BLOCK (PASSIVE)
Mob 1: 52.8s | BLOCK (HOLY SHIELD)
Mob 2: 52.8s | BLOCK (HOLY SHIELD)
Mob 3: 52.8s | DODGE
Mob 1: 54s | DODGE
Mob 2: 54s | BLOCK (PASSIVE)
Mob 3: 54s | BLOCK (HOLY SHIELD)
Mob 1: 55.2s | MISS
Mob 2: 55.2s | DODGE
Mob 3: 55.2s | BLOCK (HOLY SHIELD)
Mob 1: 56.4s | MISS
Mob 2: 56.4s | PARRY
Mob 3: 56.4s | BLOCK (HOLY SHIELD)
56.4s | Holy Shield fades.
Mob 1: 57.6s | BLOCK (PASSIVE)
Mob 2: 57.6s | DODGE
Mob 3: 57.6s | HIT (CRUSHING)
Mob 1: 58.8s | DODGE
Mob 2: 58.8s | DODGE
Mob 3: 58.8s | HIT (CRUSHING)
60s | Holy Shield cast.
Mob 1: 60s | PARRY
Mob 2: 60s | MISS
Mob 3: 60s | BLOCK (PASSIVE) - Redoubt proc
Image
Image
solina
 
Posts: 374
Joined: Thu Mar 20, 2008 1:24 pm

Postby moduspwnens » Fri Aug 29, 2008 9:17 am

solina wrote:
moduspwnens wrote:For some reason, one of the variables wasn't getting reset. Anyway, it's making sure it's reset now, so this issue should be resolved. Good catch, and thanks. :)


Cool! Testing's (part of) what I do for a living.. this comes naturally.

Great! I can't possibly think of enough test cases to do it myself, so I need people to find out where I messed up. Any time you spend testing and giving feedback is greatly appreciated.

solina wrote:One suggestion: can you have an option to stagger the start times of multiple mobs? I think my numbers are getting skewed slightly by having continual simultaneous strikes (where in-game I'd be able to put up HS between mobs).


A feature like this isn't currently available, but could be. Your numbers shouldn't be skewed, though. Staggering the mobs would (as far as I know) lead to very similar results, only they would be more randomly timed (like in-game) and easier to differentiate.

solina wrote:One other thing: seeing some RNG weirdness on Redoubt. Went over 40 seconds without a proc.. is this just RNG, or did something misfire?


It's for situations like this that I added the ability to actually show the numbers generated. Click "Yes" for "Show Rolls" and it will show you all the random numbers generated throughout the simulation. With it on, you'll see something like this:

The Simulator wrote:Mob 1: 84.84 | HIT No proc. (5)
Mob 1: 49.35 | BLOCK (PASSIVE) - Redoubt proc. (1)


The first random number is the number generated to tell what the attack comes in as (Miss, Dodge, Parry, Block, etc.), but at the end of all melee swings that do damage (the ones that could proc Redoubt), it will either say "No proc" or "Redoubt proc" followed by a number in parentheses. That number is the random number generated. It should be between 1 and 10, and a 1 means Redoubt proc'd (thus the 10% chance).

The situation you posted is likely just a collection of bad RNG rolls. Thanks for the feedback, though, and let me know if you have anything else. :)
I rule.
moduspwnens
Moderator
 
Posts: 6211
Joined: Tue Nov 06, 2007 12:28 pm
Location: Shattered Hand

Postby Neuron » Fri Aug 29, 2008 11:09 am

neat

eeeeewwwwwwww at php though
http://www.ardentdefender.com - A combat log parsing API
User avatar
Neuron
 
Posts: 716
Joined: Mon Aug 27, 2007 8:26 am
Location: Nail Salon

Postby moduspwnens » Fri Aug 29, 2008 12:34 pm

Neuron wrote:neat

eeeeewwwwwwww at php though


Yeah, but I don't know Javascript. On the upside, you don't have to install anything, it's very easy to use, and effective, but on the downside, the simulations are done server-side and it's inefficient. For now, though, it works, and I can deal with that.
I rule.
moduspwnens
Moderator
 
Posts: 6211
Joined: Tue Nov 06, 2007 12:28 pm
Location: Shattered Hand

Postby solina » Fri Aug 29, 2008 4:38 pm

moduspwnens wrote:
solina wrote:One suggestion: can you have an option to stagger the start times of multiple mobs? I think my numbers are getting skewed slightly by having continual simultaneous strikes (where in-game I'd be able to put up HS between mobs).


A feature like this isn't currently available, but could be. Your numbers shouldn't be skewed, though. Staggering the mobs would (as far as I know) lead to very similar results, only they would be more randomly timed (like in-game) and easier to differentiate.


Hmm.. ok. Assuming the results are accurate, it's hard for me to make a case for keeping Redoubt.. at best it seems to be preventing half the crushing blows in a 3-mob situation fast-swinging situation where HS is likely to run out of charges.

But crushing chance can never be >15%, right?

TankSim wrote:Mob 3: 66s | 77.03 | HIT (CRUSHING) No proc. (6)


Is that right?


Oh.. and clicking the Paladin icon while my results were up gave me the form with my values filled in, and the # of mobs, but not crushing, show rolls, show time, or attack speed. Safari this time.
Image
Image
solina
 
Posts: 374
Joined: Thu Mar 20, 2008 1:24 pm

Postby moduspwnens » Fri Aug 29, 2008 6:38 pm

solina wrote:Hmm.. ok. Assuming the results are accurate, it's hard for me to make a case for keeping Redoubt.. at best it seems to be preventing half the crushing blows in a 3-mob situation fast-swinging situation where HS is likely to run out of charges.


Yes, but I can't think of a time where you're tanking 3 mobs that swing quickly and crush.

solina wrote:But crushing chance can never be >15%, right?


Yep (assuming normal circumstances, and this simulator does). What you were probably seeing in what you quoted is when Redoubt and Holy Shield were down, so, maybe 70-85 were a Crushing blow, while 85-100 is a regular hit. I don't know what avoidance/block levels you were using, though, so I'm only guessing. That result alone does not mean there's an error, though.

Solina wrote:Oh.. and clicking the Paladin icon while my results were up gave me the form with my values filled in, and the # of mobs, but not crushing, show rolls, show time, or attack speed. Safari this time.


Yeah, that's something I think the browser is doing automatically. The code says only to fill in the values for you if it's listed in the URL (like from the "Link to your results" links), so the browser is filling in what it knows, but it doesn't catch on to the "show rolls" or other fields. I'll have to fix this to be more consistent.

Thanks for your feedback!
I rule.
moduspwnens
Moderator
 
Posts: 6211
Joined: Tue Nov 06, 2007 12:28 pm
Location: Shattered Hand

Postby moduspwnens » Sun Aug 31, 2008 12:12 am

Solina wrote:Oh.. and clicking the Paladin icon while my results were up gave me the form with my values filled in, and the # of mobs, but not crushing, show rolls, show time, or attack speed. Safari this time.


What I told you before wasn't true. I had, for whatever reason, coded it so part of the values were filled in automatically and others weren't.

Anyway, here's the bump for Version 1.7! I fixed the above issue, and some others:

- Took out a small debugging tool I accidentally left in for a warrior's Shield Block. It didn't affect functionality.
- Shield Block was recently changed in beta to last 10 seconds instead of 5. This is reflected in the simulator now.
- Updated the tooltip on the input page to reflect the new Shield Block.

- Even when no points were in Critical Block, the simulator was still printing the Critical Block rolls. It never proc'd, though. This change doesn't affect the functionality of the simulator.
- There was not a check in place to recast the WotLK style Shield Block if its cooldown was up between swings, leading to Shield Block not being recast if the duration didn't divide cleanly by the swing time. This has been fixed.
- Two problems: 1. When clicking the drop-down menu after simulating, only some values were filled in from the last simulation correctly, while others were defaults. 2. I was having trouble getting it to choose between what's in the URL and what's been submitted as to what goes in the box. I solved both and added a little functionality by making the Re-do URL the URL of the simulation. I removed the link to "repeat this simulation" because it's no longer necessary. The current URL is accurate.
- When showing the attack table, it now puts Shield Block and Holy Shield values in italics, to denote that it's only the case when they're active.
I rule.
moduspwnens
Moderator
 
Posts: 6211
Joined: Tue Nov 06, 2007 12:28 pm
Location: Shattered Hand

Postby solina » Tue Sep 02, 2008 8:07 am

moduspwnens wrote:
solina wrote:But crushing chance can never be >15%, right?


Yep (assuming normal circumstances, and this simulator does). What you were probably seeing in what you quoted is when Redoubt and Holy Shield were down, so, maybe 70-85 were a Crushing blow, while 85-100 is a regular hit. I don't know what avoidance/block levels you were using, though, so I'm only guessing. That result alone does not mean there's an error, though.



Got it. My passive avoidance is ~72, and knowing that the extra hit chance is at the end of the table, this makes sense.

moduspwnens wrote:
Solina wrote: wrote:
Oh.. and clicking the Paladin icon while my results were up gave me the form with my values filled in, and the # of mobs, but not crushing, show rolls, show time, or attack speed. Safari this time.



What I told you before wasn't true. I had, for whatever reason, coded it so part of the values were filled in automatically and others weren't.

Anyway, here's the bump for Version 1.7! I fixed the above issue


Retested and confirmed. ;-)
Image
Image
solina
 
Posts: 374
Joined: Thu Mar 20, 2008 1:24 pm

Postby Mithos » Wed Sep 03, 2008 6:08 pm

Nice work on it modus.

Just played with it and noticed Crushing blows are broken - I select "no crush" but your nasty script crushes me anyway :P. This happened every time as far as I saw. I selected "crushing" and "no crushing" and got the same result, i.e. I got crushed.

Edit: I ran the warrior one twice as well and got crushed with the option set to "no" I think.
Image
Mithos
 
Posts: 728
Joined: Thu Aug 30, 2007 7:54 am

Postby moduspwnens » Thu Sep 04, 2008 8:47 am

Mithos wrote:Nice work on it modus.

Just played with it and noticed Crushing blows are broken - I select "no crush" but your nasty script crushes me anyway :P. This happened every time as far as I saw. I selected "crushing" and "no crushing" and got the same result, i.e. I got crushed.

Edit: I ran the warrior one twice as well and got crushed with the option set to "no" I think.


I couldn't recreate your situation, but I did have the exact opposite problem (I selected "crushing" and no crushes) and found what I believe to have been causing it. It's fixed with the new version.

Version 1.8 Update
Shield Block costs no rage in WotLK. This is now reflected in the tooltip.
Crushing Blows being toggled on or off wasn't registering correctly due to a mistype on my part. This is now corrected.
Small error with Libram of Repentance fixed.
I must have removed the data validation and forgotten to replace it. It is fixed now.
Updated files: index.php, initialize_variables.php, simulate.php, readme.txt.
I rule.
moduspwnens
Moderator
 
Posts: 6211
Joined: Tue Nov 06, 2007 12:28 pm
Location: Shattered Hand

Postby Selinaria » Mon Oct 06, 2008 2:24 am

Great work on this Modus! Been messing around with it for a while but only now thought to comment.

Thought I would maybe help you out by showing all the things you will at some point want to include when you decide to flesh out the DK side of things. These two are basically avoidance CDs similar to Holy Shield or Shield Block but with lower uptime obviously.

Unbreakable armor + presence/absence of Glyph of Unbreakable Armor
Lichborne
Frigid Dreadplate (or just tell the user to add 6% miss because there really is no downtime for this currently as a hidden debuff)


Other areas of interest:

-Indicate how many hits over the course of the simulation Bone Shield is catching. (assuming DK has it of course) Also 4/5 charges unglyphed/glyphed. Or perhaps more accurately how long it remained up since obviously the number of hits mitigated will be proportional to duration of test.
-Add a notation at the start that the user should just put their parry value as their base + 10% assuming they have 100% uptime on Blade Barrier or could add some options to assume lower than 100% uptime (but since it is easy to keep up....yea).

I'm sure I am missing some things too but those are the bigger ones for avoidance/mitigation besides their version of shield wall Icebound Fortitude

For when you get around to the DK I thought you might like a sorta laundry list of things to add that directly effect avoidance and/or damage taken. Of course I may have just wasted my time and you know this already but so it goes haha. And of course DK is the class that will still probably fluctuate the most even now so we shall see.

Edit: Forgot one factor, maybe give an option of whether the DK will use the abilities on CD or stagger them so as to always have something up for extended periods of time.
Selinaria was suddenly surrounded by a gang of sharks, a cracked bat as her only protection!
Image
User avatar
Selinaria
 
Posts: 84
Joined: Mon Mar 23, 2009 7:55 am
Location: Burlington,Ontario, Canada

Previous

Return to Add-ons, UI, and Macros

Who is online

Users browsing this forum: No registered users and 1 guest


Remove Advertisements

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest
?php } else { ?