Simulation accuracy: AMR matches real combat logs


I don’t know about the other bosses.

In the result, at the top it shows:
Fight Length

The version, rotation, and boss/script are all links to snap-shots of those at the time of the simulation, so that you can recreate any simulation you want. Just click on Krosus and it’ll take you to the script.


Click this link for the “DW no miss”

You must be logged in. When you are, at the top center of that page there will be an option to save that to your scripts. Let me know if you don’t see that option.


Sorry for the delayed response. I got Mo’s bracers and tried them last week on krosus, I didn’t like the way they played. But here is another heroic Krosus kill from this week, using the same talent build as we discussed 2 weeks ago - the ‘simulate this setup’ tool shows me that I should be doing ~850K dps using the configuration that you detailed in this thread. But I did 950K and I screwed up… I neglected to use a Potion of the Old War a second time. So based on the yield of the first PotOW (~4.3 million additional damage) that would mean that I should have done ~973K dps. So the tool seems to be off by 973-850 = ~123K dps or just under 15%.

I’m not sure if this is helpful at all to you guys or if this is just argumentative - but given the title of the thread and the seemingly large degree of inaccuracy. I was hoping to help, not aggitate :smiley:


Here is the log for the reference above:


I’m not sure what you simmed to get a result that much lower. Without seeing your simulation report, I can’t see how you set it up.

I imported your character directly from the log you linked and used the DW no miss script that I referenced. The simulation comes out at 931,500 DPS, close to the 949,000 from your log. The standard deviation is 57,000, so that is will within the standard deviation and nothing really stands out. Even without the extra potion of the old war.

I see you are in the 99th percentile for your item level, so having the simulation come out a bit lower is expected -
top ranks benefit from lucky streaks and lucky procs that get smoothed out in the simulator. Coming back to that standard deviation again - without changing your skill or even a single thing about your rotation, you can expect any given boss attempt to range between 865k and about 1 million damage.

Here’s a link to the sim that is based off your log:


So I am trying to work out why my real world DPS is coming out a fair amount lower than what I am simming. I get it will vary but I am seeing results way off (Like 100k lower). It probably has everything to do with me just failing at being efficient but is there a way I can see where I am going wrong? I have logs in warcraftlogs but I can’t seem to import them using import logs:

Here is the log I am trying to import


Simple problem: It’s a private log, which means it is only available if you are logged in to your Warcraftlogs account.

You need to change the visibility of that log to public so AMR is allowed to access it.


Thanks, didn’t even notice that. Will try again.


Okay so I get around a 70k difference (my actual being that much lower). This seems to be the trend for me I get significantly lower than what is being simmed. This must be due to my fault so I am trying to make sense of what I am doing wrong.

Log (653k):
Sim (724.5k):

From what I can see is what I did wrong was I got was less casts of Annihilation and Death Sweep in and that seems to be the biggest difference, some other minor differences. Perhaps I am moving badly and wasting time on the boss? I also see my crit % on Annihilation is way lower than the sim. It almost looks like the sim is adding 10% cit chance and the game isn’t OR its probably just really bad rng on my side