Heya guys,
I am currently working on a guide to best show the engagement ranges from the point of view of a squad. In order to have this as good as possible I need some input from the community.
Adding all squads into a single graph is impossible, you won't be able to see much anymore. Similarly splitting it up into several graphs means the information for a unit is also split, which I personally dislike.
Even just focusing on the core squads (+ some upgrades) results in such a graph, which is far too big (click on the picture for bigger size):
Therefor I'd like from you a list of units to compare to from the Axis and the Allies side.
My personal list right now would be:
Axis:
Conscript
Conscript + PPSh41
Shock Troops
Guards + DP-28
Riflemen
Riflemen + 2 BAR
Riflemen + 2 M1919A6
Allies:
Grenadiers
Grenadiers + LMG
Panzergrenadiers
Assault Grenadier
Sturmpioneer
Volksgrenadier
Obersoldaten
And limit the chart to 5 times effectiveness. Shocks and AE have great short range DPS, but their long range DPS is abysmal. Other units tend to have up to 30 times their effectiveness on long range, which without this limitation would cause all other values to be almost indiscernible.
Which would look like this:
And respectively:
I'd like to know if you would add a unit or omit one. Please keep in mind that the more units added, the harder they get to discern. I found 7 is the maximum where you can still see the differences.
Some units I would always omit are either Shocktroops or Assault Engineers as the have really similar DPS( 71 and 68 close, .6 and .9 far).
An alternative I was thinking about would be norming the DPS, so that the best engagement range vs a squad is always 1 and the worst 0. The problem with that is it causes the graphs to overlap extremely often. That's why I did not use it.
If you have any other suggestions, feel free to state them!
Feedback required for a guide
13 Sep 2014, 14:37 PM
#1
Posts: 577
13 Sep 2014, 15:25 PM
#2
Posts: 604
Omit/add units: Right now it seems to contain all the normal combat targets. I don't know if I'd add more.
Can you test not color filling the areas below the graphs? Maybe that'd clean them up a little bit.
One thing often done in academic papers etc. is to assign symbols to the graphs. So you could add a little conscript icon (or just square/star/circle etc) every 5 range or so.
This is especially useful for when color might be changed/lost by copying/printing and when the reader can't discern the colors as well as the author (for example, some people can't discern between green/red).
Logarithmic scale would also be possible but I personally hate reading them (and many people read them wrong/don't understand them).
Can you test not color filling the areas below the graphs? Maybe that'd clean them up a little bit.
One thing often done in academic papers etc. is to assign symbols to the graphs. So you could add a little conscript icon (or just square/star/circle etc) every 5 range or so.
This is especially useful for when color might be changed/lost by copying/printing and when the reader can't discern the colors as well as the author (for example, some people can't discern between green/red).
Logarithmic scale would also be possible but I personally hate reading them (and many people read them wrong/don't understand them).
13 Sep 2014, 15:59 PM
#3
Posts: 577
Using line instead of area now.
Added dots to the Conscript one. That's the closest I can get to using symbols. AFAIK google does not allow symbols and since the graphs should be dynamic (directly linked to the spreadsheet so they get automatically updated when the sheet gets updated) I can't use any other tools.
Logarithmic would probably do more harm than good.
Added dots to the Conscript one. That's the closest I can get to using symbols. AFAIK google does not allow symbols and since the graphs should be dynamic (directly linked to the spreadsheet so they get automatically updated when the sheet gets updated) I can't use any other tools.
Logarithmic would probably do more harm than good.
13 Sep 2014, 21:55 PM
#4
Posts: 89
i'd do unit vs faction (conscript vs ostheer infantry etc)
and the point you try to make across, : the difference in dps at different ranges
gets reduced on a small scale mostly between 0 and 1, i'd use raw dps charts instead of relative
(also 0.25 being 4x less damage is not directly obvious, so making it needlesly complicated to make a point of these guys do 40 dps and you do 10 at range x)
and the point you try to make across, : the difference in dps at different ranges
gets reduced on a small scale mostly between 0 and 1, i'd use raw dps charts instead of relative
(also 0.25 being 4x less damage is not directly obvious, so making it needlesly complicated to make a point of these guys do 40 dps and you do 10 at range x)
13 Sep 2014, 22:08 PM
#5
Posts: 577
The guide contains an ample introduction and explanation of the graphs.
Regarding 0 and 1.. what about a -1 to 1 scale?
0 being the 'medium' where you receive pretty much as much damage as you deal, while -1 means you take more than you deal and 1 that you deal more than you take? That should make it pretty easy to find out which range is good!
Regarding 0 and 1.. what about a -1 to 1 scale?
0 being the 'medium' where you receive pretty much as much damage as you deal, while -1 means you take more than you deal and 1 that you deal more than you take? That should make it pretty easy to find out which range is good!
14 Sep 2014, 05:16 AM
#6
5
Posts: 1157 | Subs: 2
Comparing Sturmpio dps to shocks and AEs could be useful, although perhaps a bit misleading due to the recieved accuracy modifier they just got. I think the list you have at the moment is ideal.
The only thing I have to add is that it could be useful to include a screen shot that displays the range values in game, so that people can have a visual aid to go along with the data that displays comparative dps at specific ranges. For example, 2 squads facing off at 35 range, 30, 25, 20, 15...etc. I don't know the best way to display this, but having an image could really help cement the data from the graph to the actual engagement ranges in the game.
Thanks for putting in the work for this!
The only thing I have to add is that it could be useful to include a screen shot that displays the range values in game, so that people can have a visual aid to go along with the data that displays comparative dps at specific ranges. For example, 2 squads facing off at 35 range, 30, 25, 20, 15...etc. I don't know the best way to display this, but having an image could really help cement the data from the graph to the actual engagement ranges in the game.
Thanks for putting in the work for this!
14 Sep 2014, 05:35 AM
#7
Posts: 896
This is great work, but I suggest a different approach.
Instead of doing this manually I think you should develop a web application that automates the process (just like the new feature on coh2-stats website). This way you dont have to keep updating the charts and people can select the units to compare as they choose.
Another suggestion is to change relative dps to actual, because right now its giving confusing results like the higher dps unit getting a lower graph for example.
Instead of doing this manually I think you should develop a web application that automates the process (just like the new feature on coh2-stats website). This way you dont have to keep updating the charts and people can select the units to compare as they choose.
Another suggestion is to change relative dps to actual, because right now its giving confusing results like the higher dps unit getting a lower graph for example.
14 Sep 2014, 07:21 AM
#8
Posts: 752
You are using the figures that include aimtimes and reloads?
Mention of whether they are static or moving derived accuracies would also be helpful.
Mention of whether they are static or moving derived accuracies would also be helpful.
14 Sep 2014, 07:46 AM
#9
Posts: 24
The guide contains an ample introduction and explanation of the graphs.
Regarding 0 and 1.. what about a -1 to 1 scale?
0 being the 'medium' where you receive pretty much as much damage as you deal, while -1 means you take more than you deal and 1 that you deal more than you take? That should make it pretty easy to find out which range is good!
I feel this would be the most useful and easy way to read the graph.
You can at a quick glance see whether a unit is the favourite in the selected engagement over a selected range while also comparing everything on the same scale and keeping it relative.
There are scaling and readability issues with using the "total" DPS as a measure.
14 Sep 2014, 10:35 AM
#10
Posts: 577
This is great work, but I suggest a different approach.
Instead of doing this manually I think you should develop a web application that automates the process (just like the new feature on coh2-stats website). This way you dont have to keep updating the charts and people can select the units to compare as they choose.
No. TensaiOni has his page, I'll keep up a spreadsheet as I use that to experiment and test new values.
Another suggestion is to change relative dps to actual, because right now its giving confusing results like the higher dps unit getting a lower graph for example.
This is relative DPS, because I do not think that a lot of people really understand the other DPS graphs. They look where their DPS is higher and then try to get to that range. These Graphs should help people to show where the ideal engagement range for an enemy unit is. CoH2 contains so many different weapon types and the engagement range is different depending on whatever unit type you fight.
You are using the figures that include aimtimes and reloads?
Mention of whether they are static or moving derived accuracies would also be helpful.
Yes. That's full DPS as usual. As I said earlier, the guide contains already sufficient explanation. I just want feedback on which way you'd like the graphs
I feel this would be the most useful and easy way to read the graph.
You can at a quick glance see whether a unit is the favourite in the selected engagement over a selected range while also comparing everything on the same scale and keeping it relative.
There are scaling and readability issues with using the "total" DPS as a measure.
Having it normed, no matter how makes it IMO harder to read to be honest :/
I just tried several ways, for example:
I find the first ones best to read
14 Sep 2014, 11:05 AM
#11
2
Posts: 2181
I think you should split up the graph per faction. 1 gren vs US and 1 gren vs SU
14 Sep 2014, 11:44 AM
#12
Posts: 577
And one Gren to rule the world? (:
PAGES (1)
1 user is browsing this thread:
1 guest
Livestreams
84 | |||||
43 | |||||
26 | |||||
15 | |||||
8 | |||||
7 | |||||
1 | |||||
652 | |||||
24 | |||||
18 |
Ladders Top 10
-
#Steam AliasWL%Streak
- 1.653231.739+13
- 2.838223.790+1
- 3.35057.860+15
- 4.590233.717+6
- 5.278108.720+29
- 6.306114.729+2
- 7.645.928+5
- 8.922406.694+1
- 9.1118621.643-1
- 10.265138.658+2
Replay highlight
VS
- cblanco ★
- 보드카 중대
- VonManteuffel
- Heartless Jäger
Einhoven Country
Honor it
10
Download
1243
Board Info
772 users are online:
772 guests
0 post in the last 24h
2 posts in the last week
28 posts in the last month
2 posts in the last week
28 posts in the last month
Registered members: 49388
Welcome our newest member, KETTA
Most online: 2043 users on 29 Oct 2023, 01:04 AM
Welcome our newest member, KETTA
Most online: 2043 users on 29 Oct 2023, 01:04 AM