Ranged Bow Rate of Fire Testing

[Bengalih]

I was sitting around pondering about how ranged builds have continuously been getting ignored in recent updates. I find that playing a ranged toon is ALOT of fun and as long as one can remember to not forget that they can also pick up a set of melee weapons (especially end-game) they can always be an effective contributor to a party.

Unfortunately it seems (to me) that instead of addressing the dps issues that ranged builds face, Turbine will be introducing items to the game that serve to only bridge the gap making those that aren't ranged builds more competitive with ranged dps than those that specialize in it. While I don't want to start a ranged vs. melee dps discussion, it would seem to me that a class that is defined as specializing in ranged combat should have considerable benefits when it comes to using ranged weapons. I think these are things to keep in mind as we will most likely see some more changes to the AA and Deepwood sniper paths within the ranger class.

Of course, this discussion is not limited to Rangers, as other classes should be able to invest in ranged combat skills as well. But, of course, the level of return should be proportionate to the investment. For example a Wizard build specced to use THF should not be putting out the same type of numbers that a THF Barbarian is -as the Wizard has a whole additional skill set to work with (namely their spells). In the same fashion a FvS or Rogue should probably not be as proficient with a ranged weapon as a Ranger who specializes in a ranged path. This perspective is, of course, up for discussion, but also not the main point of the thread (feel free to begin another thread to discuss ).

Instead of crying Dooooooom (although it will soon cost me fewer spell points to cry it), I thought I would do something constructive as a basis for future discussions. I know there have been other threads with measurements of RoF but partly because this information is a bit spread out, and partly because I wanted to perform my own measurements, I have come up with the data shown below.

My measurements were done using my Level 20 Ranger both with and without the 25% capstone enhancement. I utilized the gear I had which has been the most traditional/original gear available that impacts Rate of Fire (RoF). I did not have a suit of Black Dragon Armor to test with but I believe the items I tested will cover all the available "types" of bonuses and can be used with future data collection and calculations. Although I don't believe there should be any difference based on the Bow used, for sake of full disclosure I was using a GS Lit II bow.

As this is a rather lengthy thread, for those that would prefer to skip to the "main event", I believe Chart #3 (and the immediately preceding text) as well as Chart #8 are the main meat of this experiment.

For those that would like the full info, please begin here:

Below are the various items that can effect ranged alacrity/RoF with their stated % benefit increase as well as the type.

Chart #1 - Stated Item Effects

Item		Stated Increase		Stated Type								
Base (B)	N/A			N/A								
Capstone (C) 	25%			Competence								
Quiver (Q)	10%			Untyped								
AA Set (AA)	10%			Competence								
Haste (H)	15%			Enhancement

My first set of tests involved my 20 Ranger using Manyshot. I was completely sans equipment (minus what is stated) and free of any buffs. My sample set varied, but my numbers are based on the MODE of my sample sets and not the MEDIAN. Generally my sample set was at least 5, but increased to as many as 15 when there seemed to be a large disparity between tests.

The following chart shows the # of Arrows fired with a Manyshot with various combinations of the items above:

Chart #2 - Capstoned Manyshot

				B	C	C,Q	C,AA	C,Q,AA	C,H	C,Q,H	C,AA,H	C,Q,AA,H	
Arrows Fired			76	84	88	88	92	92	92	96	96	
% Increase from Base		N/A	10.53%	4.76%	4.76%	9.52%	9.52%	9.52%	14.29%	14.29%	
% Increase from Previous					4.55%			4.35%		
										
Volleys (Arrows/4)		19	21		22	23	23	23	24	24

Based on the above chart, we can see that the stated effects from Chart #1 are not accurate, but instead the numbers seem more inline with the below chart. Also, based on how the effects stack, we see the following:

1) The Quiver's benefits are overridden by Haste.
2) The AA Set stacks with the Capstone

Based on those points, it would be safe to say that:
1) The Quiver offers an Enhancement type bonus like Haste.
2) The Capstone and AA Set cannot both be Competence types (assuming we go with the well known adage of bonuses of the same type do not stack).

Therefore, i have reclassified the Capstone bonus as a new type "Class" instead of the original classification of Competence.

Chart #3 - Actual Item Effects for Capstone

Item		Actual Increase		Actual Type								
Capstone (C) 	10.53%			Class								
Quiver (Q)	4.76%			Enhancement								
AA Set (AA)	4.76%			Competence								
Haste (H)	9.52%			Enhancement

If we take the numbers from Chart #3 we can utilize them with the following well-known methodology:

1) Bonuses of the same Type do not stack
2) Bonuses of different Types stack
3) When bonuses of the same type are applied, the highest one takes precedence
4) Bonuses are applied in sequence, with each bonus multiplying the result of the previous total (if anyone knows what the mathematical term for this is, please let me know...BTW this is the same way that healing amp is applied).

The below examples show the combination of various items with the stated forumlas plugged in and their results calculated in terms of # of arrows:

Chart #4

Combination	Base + Capstone (Class)									
Formula		76 + 10.53%									
# of Arrows	84.00									
										
Combination	Base + Capstone (Class) + Quiver (Enhancement)									
Formula		76 + 10.53% +  4.76%									
# of Arrows	88.00									
										
Combination	Base + Competence (Class) + Quiver (Enhancement) + AA Set (Competence)									
Formula		76 + 10.53% + 4.76% + 4.76%									
# of Arrows	92.19									
										
Combination	Base + Competence (Class) + Quiver (Enhancement) + AA Set (Competence) + Haste (Enhancement)							
Formula		76 + 10.53% + 0% (overridden by Haste) + 4.76% + 9.52%									
# of Arrows	96.38

We can see from Chart #4 that the # of Arrows calculated falls almost perfectly in line with the actual results garnered in Chart #2.

The above then describes an accurate representation of the various item type classifications and demonstrates solid examples of how they apply in practice. I have continued with this methodology to show a similar breakdown with a Non Capstoned toon (in this case my 20 ranger with the Capstone removed).

The chart below demonstrates the same data as Chart #2, but using the Non Capstoned build:

Chart #5 - Non Capstoned Manyshot

Non Capstoned										
Manyshot			B	Q	AA		Q,AA	H	Q,H	AA,H	Q,AA,H	
Arrows Fired			76	80	80		84	84	84	88	88	
% Increase from Base		N/A	5.26%	5.26%		10.53%	10.53%	10.53%	15.79%	15.79%	
% Increase from Previous					5.00%			4.76%		
										
Volleys (Arrows/4)		19		20		21	21	21	22	22

As a comparison to Chart #3 for the Capstoned build, the below chart represents the actual observed increases for the Non Capstoned build.

Chart #6 - Actual Item Effects for Non Capstone

Item		Actual Increase		Actual Type								
Quiver (Q)	5.26%			Enhancement								
AA Set (AA)	5.26%			Competence								
Haste (H)	10.53%			Enhancement

The below examples show the combination of various items with the stated forumlas plugged in and their results calculated in terms of # of arrows for the Non Capstoned build:

Chart #7

Combination	Base + Quiver (Enhancement)									
Formula		76 + 5.26%									
# of Arrows	80.00									
										
Combination	Base + Quiver (Enhancement) + AA Set (Competence)									
Formula		76 + 5.26% + 5.26%									
# of Arrows	84.21									
										
Combination	Base + Quiver (Enhancement) + AA Set (Competence) + Haste (Enhancement)									
Formula		76 + 0% (overridden by Haste) + 5.26% + 10.53%									
# of Arrows	88.42

Once again, we can see how these values fall in line with what is shown in Chart #5.

The below chart shows a summary comparison between the bonuses available to both Capstone and Non Capstone builds.

Chart #8 - Summary Comparison of Actual Item Effects

Item		% for Capstoned		% for Non Capstoned		Actual Type						
Capstone (C) 	10.53%			N/A				Class						
Quiver (Q)	4.76%			5.26%				Enhancement						
AA Set (AA)	4.76%			5.26%				Competence						
Haste (H)	9.52%			10.53%				Enhancement

We can see from the above that a Non Capstone build actually has slightly higher % increases to their base RoF speeds compared to a Capstone build. Obviously, since the Capstone "Class" type is not available to them, they can still not reach as high a RoF as a Capstoned build.

All of the above calculations were performed using only 20 second bursts of Manyshot. To attempt a more diverse experiment, I expanded to 3 more tests:

1) 20 seconds of fire (Non Manyshot)
2) 60 seconds of fire
3) 120 seconds of fire

To save on time, the tests are not as comprehensive as the previous but show relatively the same bonus percentages being applied. In addition, the data continues to back up the various type classifications and their stacking methods.

The below charts show the data from the 3 new tests. No tests were done using Haste, primarily because I was unable to get someone to help out at the time I conducted the tests to cast a Haste longer then the 30 seconds my potions would allow. I would be supremely confident that the bonuses from Haste would apply in a fashion similar to what has already been shown above.

Chart #9 - Standard Firing Tests

				B	C	Q	C,Q	AA	C,AA	Q,AA	C,Q,AA		
20 Seconds										
# of Arrows			19	22	20	23	20	23				
% Increase from Base*		N/A	15.79%	5.26%	4.55%	5.26%	4.55%				
										
60 Seconds										
# of Arrows			56	63	59	66	59	66				
% Increase from Base*		N/A	12.50%	5.36%	4.76%	5.36%	4.76%				
										
120 Seconds										
# of Arrows			111	125	117	131	117	131	122	138		
% Increase from Base*		N/A	12.61%	5.41%	4.80%	5.41%	4.80%				
										
Avg % Increase from Base*	N/A	13.63%	5.34%	4.70%	5.34%	4.70%

* The % increase from Base above is the percentage increase from either the Base (non-Capstone) or the Capstone, depending on the combination being tested (e.g. For 120 seconds the 4.80% percent listed for C,Q means that C,Q was 4.80% faster then just C - since the Capstone is used in both measurements)

While there are variations between the % changes from the 3 tests, we see that the Average increase from all 3 durational tests is roughly equivalent to those shown in Chart #8. The 13.63% shown for the increase from the Capstone seems to indicate that the Capstone bonus may have a slightly higher RoF increase for standard shooting (as opposed to Manyshot).

For the final columns in Chart #9 (those that with a combination of both the Quiver and AA Set) I decided to use my formula results to derive what I should be seeing in terms of RoF of a 120 second test prior to performing the tests.

Based on the data gathered from the previous columns, I could work out what we should expect to see:

Chart #10

Combination	(Capstone + Quiver (Enhancement)) + AA Set (Competence)									
Formula		(131) + 4.80%
# of Arrows	137.29									
										
Combination	(Base + Quiver (Enhancement)) + AA Set (Competence)									
Formula		(117) + 5.41%
# of Arrows	123.32

Our results are very close to what we would expect to find:

For the C+Q+AA (Capstone) test we compute 137.29 arrows. As Chart #9 shows we actually observed 138.

For the Q+AA (Non Capstone) test we compute 123.32 arrows. As Chart #9 shows we actually observed 122.

I believe the above sample sets along with the corresponding formulas and their applied use to predict results serve as a pretty good basis for comparing the bonuses and type classifications of the listed items. Additional items bonuses and classes should be able to be determined relatively quickly by adding their results into this gathered data.

Again, I realize that this is not the first test done of this type and perhaps there may have been one as comprehensive already performed (or maybe even more comprehensive). However as I wanted to perform these tests for myself, I thought I would share the data with the community.

I hope it comes in useful for someone else.

[DustTheWind]

Nice thanks for chiping in.

[Glenalth]

Numbers are close to expected, but you get a large margin of error by only firing for 20 seconds. Your initial volleys could have been 19.0 or 19.999999 shots fired, which can throw things off by almost 5%. Further numbers can suffer the same issues making the speed increases jump slightly up or down from actual.

It would be really nice if the speed increases actually worked as advertised.

[RandomKeypress]

Thanks for the report - a good read. Chart 2 had me confused though - why is 88 shots referred to as 4.76% increase from base - should that read 'Increase from previous' or have I drunk too much to be able to read this sort of thing?

Speed boosts not working to the numbers advertised seems to be universal, as I'm sure you know. Not having played a ranger I'm not sure - do feats like rapid-reload impact manyshot RoF? Which feats did you have active?

I'd argue the point regarding commitment to ranged - if my mechanic rogue spends every feat to improve his ranged combat, and has class feats focussed on ranged combat, why should he be worse ranged DPS than a ranger?

[xberto]

This was a good write up. Has anyone seen any current test reults?