I will start the next set as soon as I've picked up something to eat first.
A couple things as you continue:
1: Try to bring Hit down to 164, since we are considering the possibility of going over hit cap effecting crit returns as well.
2: Make sure you check the details tab in recount and record the amount of blocked damage. The percent of blocked damage shots will not show up in recount, so we have to do the math manually. If you are attacking a level 80 dummy, then you take the amount of blocked damage and divide by 67, and if you are attacking a level 83 dummy, then you divide by 76. This will always give you an integer equal to the amount of blocked damage shots. Then you derive a percentage using this number. Make sure that the amount of hits and crits in the details box is the same as the one for the individual break down, as sometimes it will bug out and show a frozen detail panel. If this occurs, close the detail and breakdown tab, click on your name again, and then go to details. This should bring up an updated panel.
3: There is no need to record DPS, speed, and range of crits and hits at this time.
4: Your crit bonus average needs to be to one decimal place at least.
5: Go ahead and report in the manner we are in the previous post. This makes the information easier to parse and breakdown, as well as setting it all up in neat little columns. I will translate your post for you.
20.12 crit, 169 hit rating
Level 80 dummy, from front
1: 19.1 crit return --- 1.02 crit loss --- 1.02 crit loss average --- 0 shots blocked (0.0 percent) --- 202.2 crit bonus
Thanks for helping gather info, this will expand the data we have to work with and get us to a solution faster.
I'll try to see if I can exchange some items to get rid of the last 4 hit rating through crafting or regemming.
I did check the details tab (and screenshotted both panes) for both the first, and my second run through, but thanks for the tip - it took me a while to find out how to find information on blocked shots.
I will edit me previous post to be in the format used consistently, and will add in my second set (as well as the decimal points of my crit damage bonus). I will keep the detailed info in my local .txt in case it becomes interesting later.
Yes Night Elves are the exception. They have an inherent racial that makes then 2 percent harder to hit. So it takes 7 percent hit to never miss when your target is a Night Elf. This has no bearing on our test on the dummy.
Hmm, then I wonder why my paper doll only went to 90.00 exactly, considering I had a static 42.5 or so crit. Paper doll may not reflect true crit after 90.00? But in your case it showed it, so I am not sure. I wonder if the the 95 percent was the highest capable crit as well. If only we could test at a 100 percent crit right....
I had 103% at one point on Loatheb. Now I'm sure I wasn't somehow critting 103% of the time but that's what my paper doll was
I just did an 1100 shot test. I know you want 1000, but I AFK'd at 900 and didn't get back before 1000 happened so I took it the nearest hundred to make it easier on the math.
Spec: 0/0/0
Crit Rating/Percentage: 402/18.99%
Hit Rating: 284
Number of Shots: 1100 total, 132 Critical (12 %)
Average Normal Hit - 633
Average Normal Crit - 1167
Blocked Damage: 4940 (4940/76=65 Shots Blocked if I understand it right)
Attack Location: Front of Heroic Dummy in Org
I attempted to get lower on hit but I did not have enough gear to swap without going under 263 hit rating. I turned off all aspects, and used the latest version of recount for my numbers.
Spec: 0/0/0
Crit Rating/Percentage: 402/18.99%
Hit Rating: 284
Number of Shots: 1100 total, 132 Critical (12 %)
Average Normal Hit - 633
Average Normal Crit - 1167
Blocked Damage: 4940 (4940/76=65 Shots Blocked if I understand it right)
Attack Location: Front of Heroic Dummy in Org
Go ahead and start a series at 5 percent hit cap, on a level 80 dummy from behind. This would lend itself to the base data we are trying to collect the most at this time. Try to take that to 10k shots. What I do to make sure I get 1000 shots is I buy ammo only 1000 at a time, and when my quiver runs out, it stops the test for me. If you put extra ammo in your backpack, it will continue shooting, just without quiver speed bonus.
I am going to go ahead and run each test of the two tests I am running to 12K total shots, and I will take out the high / low, since there seems to be an extremely long string of crits or non-crits every once in a while that is either inflating or deflating the average. Our aim at this time is to find what the median for each position is, and in order to work out these samples to reflect the more consistent returns we see in each sample, we will have to either test until it gets to a point where we have minimal return fluctuation to such a point as to where each new sample impacts the sample minimally, or we use the high / low exclusion method. Of course more then 12k shots is fine, and in the end this may be necessary.
Right now we seem to be seeing some patterns occurring at hit capped performance on both the level 80 and 83 dummies, and so far position on the level 83 dummy is leading to higher crit returns which fall in line with the hypothesis. Once we have a satisfactorily large enough data base, we will conclude the current findings and end the testing phase for non FA hit capping.
I know doing so many test seems tedious, but it's the only way we will get away from RNR tainting our findings. Thank you all for your time! Keep posting away and I will pool the data when we are at a good point to make some conclusions.
I am going to go ahead and run each test of the two tests I am running to 12K total shots, and I will take out the high / low, since there seems to be an extremely long string of crits or non-crits every once in a while that is either inflating or deflating the average. Our aim at this time is to find what the median for each position is, and in order to work out these samples to reflect the more consistent returns we see in each sample, we will have to either test until it gets to a point where we have minimal return fluctuation to such a point as to where each new sample impacts the sample minimally, or we use the high / low exclusion method. Of course more then 12k shots is fine, and in the end this may be necessary.
This is where calculating the standard deviation comes in handy -- it lets you identify which numbers are outliers.
Spec: 0/0/0
Crit Rating/Percentage: 339/14.78%
Hit Rating: 164
Number of Shots: 1000 total, 142 Critical (14.2 %)
Average Normal Hit - 636
Average Normal Crit - 1107
Blocked Damage: 0 (Shocked, at that result)
Test 2
Spec: 0/0/0
Crit Rating/Percentage: 339/14.78%
Hit Rating: 164
Number of Shots: 1000 total, 144 Critical (14.4 %)
Average Normal Hit - 632
Average Normal Crit - 1106
Blocked Damage: 0 (Shocked, at that result)
In both of these tests I used Terrorshaft Arrows to make sure it was all WotLK gear that I am using.
Test 3
Spec: 0/0/0
Crit Rating/Percentage: 339/14.78%
Hit Rating: 164
Number of Shots: 1000 total, 147 Critical (14.7 %)
Average Normal Hit - 627
Average Normal Crit - 1105
Blocked Damage: 2881
This was from the front of the lvl 80 target dummy. Enough testing for the day, need to get raid ready.
Observation: In my test at 5 percent hit from gear, along with 3 percent from FA, I was getting average hits of 515, and average crits of 1036, or about 201.165 percent. Yet another significant discrepancy. Since I did not screen shot my 1000 block test and observe this difference, I will have to add that into my observation log as well.
So I will clarify the information I am looking for, if anyone else wants to provide test data as well.
1: Spec is limited to points in FA, or points in Lethal shots, and this is in combination with gear to reach the Hit Cap.
2: Screen shot your recount for each test, showing both the auto shot break down, and the detail panel to show the amount of blocked damage. When selecting the details panel for blocked damage, make sure to click on your name first in order to reset the details panel to the current spot. If you had looked at it at some point during the test, then it may freeze on that point and give an inaccurate reading.
If posting Screen Shots is not an option, then record the following values.
- Crit rating (percentage, specifying percentage from Lethal shots as a secondary number)
- Hit rating (numerical / percentage from gear, percentage from Focused Aim)
- Number of shots fired, number of crits, and the numerical averages for both of these (include crit percentage of course)
- Amount of blocked damage, which can be divided by 76 in order to determine the number of shots with blocked damage
- Specify if you are attacking from the front of the dummy, or the back.
Thank you everyone for the information gathering you are doing here.
I'll take the time to run some tests myself here soon, get you some more data. You can check my armory for my hit/crit, I'll let you know my results here in a few days.
I am currently testing:
0/0/0, hitcapped (5%) from gear (actual rating 169), crit of 20,12% on a lvl 80 dummy in SMC from the front.
I will update this post as I get results from my subsequent sets:
1: 19,1 crit return --- 1,02 crit loss --- 1,02 crit loss average --- 0 shots blocked (0,0 percent) --- 202,2 crit bonus
2: 21,9 crit return --- -1,78 crit loss --- -0,38 crit loss average --- 0 shots blocked (0,0 percent) --- 201,8 crit bonus
3: 20,9 crit return --- -0,78 crit loss --- -0,51 crit loss average --- 0 shots blocked (0,0 percent) --- 201,8 crit bonus
I will continue to update this post with the results from this setup.
Edit: Set data changed to be compliant with previous reports.
Edit2: Forgot the crit loss average (This is the crit loss if all sets are gathered together and taken as 1, right?)
Thanks for joining in the data-gathering!
On your test log post above, you say you are testing from the front of a lvl 80 dummy, yet you are recording zero blocks. Are you actually testing from the front or back?
If from the front, are you sure you are looking at the "summary report" for your whole 1000 shots when you look for your blocked damage results? Sometimes you have to close the "summary report" window and reopen the "detail window" after the test is over, and go back to the "summary report" again; it will often show only the information that was available when you first went to the summary report, rather than updating it as the test goes on as the "detail window" does.
Reason I am asking: I did a 12K shot test on a lvl 80 dummy and got 4.89% of my shots blocked, so if you are getting zero blocks from the front, this will require some investigation!
PS Yes, your interpretation of the crit loss average is correct.
Just finished my 4th set, which was split in 2 500 shot tests (had to pick up my kid in the middle) and will update my post in a second.
Yes I am testing from the front, I have attached a screenshot of my testing position along with the detail summary so you can doublecheck its not just me doing something wrong (or missing something, in whcih case I'll just redo the test so far)
Adding while photoshoping my image that I just noticed blocked shots in my 4b set, so will relook at the previous screenshots as I might have glanced at the wrong place - unfortunately I'm not sure if I have the 3rd set (which I will then simply discard)
While attempting to figure out what ahs gone wrong, I did indeed find blocked damage in my other sets that I Screenshotted (set 3 has no screenshots, so thats is invalid)
I was looking at the yellow circle in the details report, somehow I completely missed the red one.
The red circled target is my target and I am in front (but the missing blocks were discovered)
To extrapolate the block % I need to divide the blocked damage by 76, correct? Or do I need to find how much my shot gets blocked by?
(to correct myself, it seems that my block value is 67)
Just finished my 4th set, which was split in 2 500 shot tests (had to pick up my kid in the middle) and will update my post in a second.
Yes I am testing from the front, I have attached a screenshot of my testing position along with the detail summary so you can doublecheck its not just me doing something wrong (or missing something, in whcih case I'll just redo the test so far)
Adding while photoshoping my image that I just noticed blocked shots in my 4b set, so will relook at the previous screenshots as I might have glanced at the wrong place - unfortunately I'm not sure if I have the 3rd set (which I will then simply discard)
While attempting to figure out what has gone wrong, I did indeed find blocked damage in my other sets that I Screenshotted (set 3 has no screenshots, so thats is invalid)
I was looking at the yellow circle in the details report, somehow I completely missed the red one.
The red circled target is my target and I am in front (but the missing blocks were discovered)
To extrapolate the block % I need to divide the blocked damage by 76, correct? Or do I need to find how much my shot gets blocked by?
Ah, that explains it, then! On the lvl 80 dummy, you need to divide by 67 -- you can double check that it is 67 by looking in your combat log, but it should be the same for everyone. 76 applies to the Heroic (lvl 83) dummy.
VS Level 80 Dummy from Behind in Orgimmar
Test 1
Average Normal Hit - 636
Average Normal Crit - 1107
Test 2
Average Normal Hit - 632
Average Normal Crit - 1106
Test 3
Average Normal Hit - 627
Average Normal Crit - 1105
Does anyone else look at these numbers and notice that there is a ~10% reduction in crit damage, to go along with the apparent 5% reduction in crit rate?
Does anyone else look at these numbers and notice that there is a ~10% reduction in crit damage, to go along with the apparent 5% reduction in crit rate?
It looks like bosses have resilience.
On the tests I have been doing vs the lvl 80 dummy, the crit bonus is actually slightly higher than 200% -- see this post earlier in the thread for full details. I haven't been recording my average hit and crit damage in the forums, though I have a log I have been keeping. My average hit damage is about 576 and average crit damage 1198. My log is at home and I am at work, so I can't give you the exact numbers, but those are off by no more than 2 points or so.
Thayer has been recording his test results in a previous post too, and he is also showing a crit bonus of over 200%. I am not sure why Kolkka's results are in the 175% range, unless it would have to do with his crit rate being lower. He is testing at 14.7%, while I am testing at 20.15%. Thayer is testing at 27.8% on the heroic dummy. I don't know if this would make a difference, but it's the only variable I see from what Kolkka has posted.
Last edited by Rosamonde : 01/15/09 at 2:39 PM.
Reason: clarification
Do either of you have the +3% crit damage meta (Relentless Earthsiege) in your helm? If you two do, and Kokkla doesn't, that could be a factor in the different crit damage bonuses we're seeing.
It'd be terribly interesting if we found out that resilience didn't affect damage from people who had certain gems. I know it's reaching, but I'm just brainstorming on possible causes of the crit rate reduction and the discrepancies for the crit bonus damage.
Do either of you have the +3% crit damage meta (Relentless Earthsiege) in your helm? If you two do, and Kokkla doesn't, that could be a factor in the different crit damage bonuses we're seeing.
It'd be terribly interesting if we found out that resilience didn't affect damage from people who had certain gems. I know it's reaching, but I'm just brainstorming on possible causes of the crit rate reduction and the discrepancies for the crit bonus damage.
Yes, I do have that gem as a matter of fact and I was using it during the test. It would account for my crit bonus being 208ish% instead of 200% (and probably Thayer's as well). Good catch -- I had forgotten about the meta! The more people that look at this data, the better, since no one can think of everything.
I agree it would be very interesting if the gem overrode resilience as well, although I suppose that would be an unintended side affect from Blizzard's perspective. I will finish the series of tests I am doing WITH the meta for consistency's sake, and then I suppose we will need some rounds without it for comparison. Doing anything tonight?
Important talents: 5/5 Lethal Shots, 2/5 Mortal Shots 0/3 Focused Aim 3% Crit Dmg Meta
Test 1:
Heroic Training Dummy (Lvl 83)
Number of shots: 1000
264 Hit rating - 8.05%
25.78% Paperdoll Crit
Avg hit: 304
Avg Crit: 630
Hit %: 80.8
Crit % 19.2
Blocked: 3496 damage / 46 blocks
Test 2:
Heroic Training Dummy (Lvl 83)
Number of shots: 1000
264 Hit rating - 8.05%
25.78% Paperdoll Crit
Avg hit: 300
Avg Crit: 627
Hit %: 80.0
Crit % 20.0
Blocked: 3116 damage / 41 blocks
Test 3:
Heroic Training Dummy (Lvl 83)
Number of shots: 1000
264 Hit rating - 8.05%
25.78% Paperdoll Crit
Avg hit: 298
Avg Crit: 620
Hit %: 81.2
Crit % 18.8
Blocked: 3496 damage / 48 blocks
I noticed something odd. I was doing my 4th test and a ret pally showed up and put heart of the Crusader(3% crit) debuff on the mob. I did about 400 shots with the debuff on the mob and was seeing way more than the 3% more crit. Graned it was a very small sample but I was getting up to 27% crit. Maybe that's something that could be looked into later. I ended up giving up on testing because every time I'd try to do another someone would show up and debuff the mob.
The Dalaran bow made the tests go alot faster.
Hope it helps you guys. If I missed anything, I have screenshots of both the detail window, Summary window and Character sheet so I provide any additional information.
I don't know if it would be a good idea to do, so before I even attempt I wanted to ask you all first.
Last night, the [Cult's Chestguard] dropped and I saved it from DE'ing since no shaman in our raid wanted it.
Expertise is supposed to reduce blocks, parries, etc if I understand the mechanic right and so to test and see if that is where our crit may be hiding, would you want me to run a test or two with it on?
I know your wanting 10k shot blocks, but I just don't have the time to do that. Family and work prevent me from doing that many, but I could do a 2k shot block today.
I have no idea how much the piece of loot would help or if it would be a negligible amount that is not worth testing.
I have no idea how much the piece of loot would help or if it would be a negligible amount that is not worth testing.
Actually, we're looking for clarification on some of the numbers you posted. When you did those 1000 shot blocks, did you have the 3% crit damage metagem in your helm & active? Your autoshot crits were about 10% less than the normal double damage, and are trying to figure out why.
Any more 1000 shot runs are helpful. With or without the meta gem.
I notice that Nooska has added the information that she is not using the meta gem --and her crit bonus is right around 200% on the lvl 80 dummy, just what we would expect. So something is going on with Kolkka's test if he is getting ~175% crit bonus. Nooska is also testing with ~20% crit vs Kolkka's ~15% -- again, I don't know if or how that would affect the crit bonus, but it is a variable.
To lower my hit, I wasn't running a helm at all during my tests to get rid of the hit rating from the helm so no meta gem should be involved.
Sorry, someone else answered the question so I didn't think I needed to confirm it, but I will do that next time.
The more and more I read this thread it seems like the training dummies are the problem. My WWS shows that my crit % was in range of what I was sporting for a crit rating at that time. I know I have no evidence of it, but we have people who can stand in front and get no blocks, yet some are. Some are getting high crit percentages, and I just showed that I am getting lowered ones. It just seems like the more we do with this, the more wierd errors we find. I am heading home soon from work and will do another test.
01/14/09, 4:46 PM
Similar Threads
