A better way to calculate Sens between games and zoom levels.

[Derpturtle]

Hi guys.

First of all, I don't mean to criticize any other method of doing it. The calculator tool is amazing and I greatly appreciate it's functions. It's just that none of the presets feel that good to me when I switch between games. I propose a better way of calculating sensitivity between games by comparing FOVs in a sense.

TL;DR  (I've attached a picture of what to do at the bottom!)

For a while now I've tried every possible combination of monitor distance and viewspeed v1 and v2 and just straight up 360sens etc and it has NEVER felt accurate between games.
Obviously this is because different games use different FOVs. The solution to this problem was supposed to be Viewspeed or Match monitor at 0% however none of these options feel right either.

So what gives?

Well, after playing BF4 for a couple of hours I started thinking about how they use the USA coefficient to dynamically change the sensitivity based on the zoom of your weapon and how it almost always feels accurate, regardless of zoom. We're used to thinking its viewspeed, but it doesn't feel right when using it for other games. So what I did was this.

TL;DR end

I looked at the ACTUAL Fov of hipfire (for me that was 106.52in BF4) and I looked at the actual FOV in Overwatch (for me that was 103). (use any combination of games with different FOVs that you'd like to try this with.)

Overwatch is 96.6954% of  BF4's FOV for me. So In your magical calculator I matched the sensitivity at 96.6954% monitor distance.
For the first time ever, my sensitivity in Overwatch actually feels accurate. My aim is on point comparable to BF4. I'm Division 1 in BF4 and 3.9kSR in OW  if that matters to anyone. 

Using the exact same method, I also calculated that Widow Zoom 42 was optimal and I can tell you after a few hours that it feels better than any other setting. Such as 38% or 44% that we hear so much about. 

I'm garbage at math so I had to use this website for this stuff.: https://percentagecalculator.net/
 

 51FOV is 49.5145% of 103FOV. Matching my hipfire sens to my ADS sens by 49.5145% results in an ADS sens of 42. This felt fantastic when paired with my new hipfire sens I got from BF4.

I also tried this with CS:GO and same thing there, now my sens feels great.
So for you to use this technique you just need to pick the game you are best at, where you feel like you perform at your best and compare the FOV and match monitor distances between that game and whatever other game you want to play. You can also use this for Zoom levels in the same game.
 

P.S: Setting the Battlefield Coefficient to 177% for 16:9 aspect ratio has the same effect. (set it to 1.7778 in the config file if you want to be extra picky).

Sorry if it was hard to follow, English is my 2nd language.

Edited May 9, 2018 by Derpturtle
More info

[DPI Wizard]

This method can actually be integrated quite easily. I'll play around with it a bit and see how it works.

[DPI Wizard]

Here's how it compares to Monitor Distance 75% (red line). Source FOV 106, target FOV on the bottom, 360 distance multiplier on the left.

[potato psoas]

15 hours ago, Derpturtle said:

First of all, I don't mean to criticize any other method of doing it. The calculator tool is amazing and I greatly appreciate it's functions. It's just that none of the presets feel that good to me when I switch between games. I propose a better way of calculating sensitivity between games by comparing FOVs in a sense.

TL;DR  (I've attached a picture of what to do at the bottom!)

For a while now I've tried every possible combination of monitor distance and viewspeed v1 and v2 and just straight up 360sens etc and it has NEVER felt accurate between games.
Obviously this is because different games use different FOVs. The solution to this problem was supposed to be Viewspeed or Match monitor at 0% however none of these options feel right either.

None of the presets will feel that good to you because none of them are perfect. There is always distortion on the screen, so muscle memory will never transfer equally for all points on the monitor, when monitor matching. Converting using cm/360 would be the perfect method but only if the distance between you and the monitor changed according to FOV, otherwise low FOVs feel too fast and high FOVs feel too slow.

15 hours ago, Derpturtle said:

So what gives?

Well, after playing BF4 for a couple of hours I started thinking about how they use the USA coefficient to dynamically change the sensitivity based on the zoom of your weapon and how it almost always feels accurate, regardless of zoom. We're used to thinking its viewspeed, but it doesn't feel right when using it for other games. So what I did was this.

BF4 uses 75% monitor match to convert between FOV.

15 hours ago, Derpturtle said:

I really don't think this method has the greatest assumptions. What is so special about finding a percentage? You'd have to explain why it works.
But I don't think having a dynamic formula is the best way to do things. Monitor matching relies on the gear ratio concept, which shows that you can sync different FOV at a single point on the screen. Dynamic formulas aren't in sync because the point keeps changing.

15 hours ago, DPI Wizard said:

Here's how it compares to Monitor Distance 75% (red line). Source FOV 106, target FOV on the bottom, 360 distance multiplier on the left.

Apparently it gives similar results to 75% monitor match. If you ask me, you probably prefer this method because it's so similar to BF4, the game you seem to have played a lot of. CS:GO also uses 75% monitor match. It's probably best that you just use 75% monitor match, for the sake of maintaining muscle memory.

[TTTTaAaTTTT]

and what FOVs in osu ?

[DPI Wizard]

5 minutes ago, TTTTaAaTTTT said:

and what FOVs in osu ?

No FOV in OSU, all 2D calculations are done based on movement to a point on the screen and this is matched to the same point in a 3D environment.

[iBerggman]

6 hours ago, potato psoas said:

I really don't think this method has the greatest assumptions. What is so special about finding a percentage? You'd have to explain why it works.
But I don't think having a dynamic formula is the best way to do things. Monitor matching relies on the gear ratio concept, which shows that you can sync different FOV at a single point on the screen. Dynamic formulas aren't in sync because the point keeps changing.

I always thought dynamic formulas made sense because the curve/arc of the 3d picture changes, so doesn't that mean that if you match the speed with gear ratio higher fovs are going to be perceived as moving faster because there's more of the curved projection on the screen unless you account for it. Or am I looking at this backwards? I don't really know enough  math so for all I know this is completely wrong, not sure about how rectilinear projection affects this either so comparing the 3d game to looking at the inside of a moving cylinder/sphere through a static 16:9 windows might be the wrong way to do it as well. 

[potato psoas]

49 minutes ago, iBerggman said:

I always thought dynamic formulas made sense because the curve/arc of the 3d picture changes, so doesn't that mean that if you match the speed with gear ratio higher fovs are going to be perceived as moving faster because there's more of the curved projection on the screen unless you account for it. Or am I looking at this backwards? I don't really know enough  math so for all I know this is completely wrong, not sure about how rectilinear projection affects this either so comparing the 3d game to looking at the inside of a moving cylinder/sphere through a static 16:9 windows might be the wrong way to do it as well. 

If the monitor match percentage changes for different FOV, then the sensitivity isn't properly synced, as per the gear ratio concept, assuming it is the best way to convert sensitivity. Keep in mind, the gear ratio method is simply 100% monitor match. And what you need to remember is that "100% monitor match" is actually arbitrary, as it is dependent on aspect ratio, as you can see below:

The same applies to the vertical. Vertically monitor matching is also aspect-ratio dependent.

[Drimzi]

That's why you use 76.19% on that 21:9 panel, and 100% on that 16:9 panel, to maintain the same match lol. Of course the percentages aren't going to line up when the calculator is using a percentage of the horizontal, which is dependent on the aspect ratio. If you think of it as a multiplier of the 1:1 aspect ratio, then both examples in your picture would have the same value and the same match point, which will be independent. In that example, it will be a 16/9 coefficient, 1.778. Any monitor match % results in the 'gear ratio', it just depends on what your measuring. Anything above 0%, it is a simple ratio between the angles, they all follow the 'gear ratio' principle. Compare 90/45 Hdeg 4:3 when matched to 4:3 and you will get 2 * the circumference, same thing when comparing 16:9 values when matched to 16:9, or comparing Hdeg or Vdeg if you are match to 1:1 (depends if landscape or portrait). For 0%, it's not just naiively dividing angles, it's the tan ratio, so compare 90 and whatever fov is exactly half of 90, (it is 53.13) and you will get the gear ratio. All monitor match percentages follow this principle. It just depends on where you measure.

Edited May 10, 2018 by Drimzi

[Ganja]

For the last weeks i tried to figure out whats the best sensitivity for me but every calculated sensitivity doesnt feel comfortable to me. Tried this method now in a few games and i love it, the calculated sensitivitys feel so much better now. Thanks for this Derpturtle ! (sry for my english im german)

[Thomas Claessens]

I've actually had a bunch of succes with this aswell. Especially in games with big FOV differences (like PUBG) the smaller FOVs always felt too fast. This fixes this, higher magnifications like 4x and 8x feel great now.

Edited May 10, 2018 by Thomas Claessens

[DPI Wizard]

Just published a small "hidden" feature: Type auto into the "Match At" box, and it will automatically do what we discuss here. The output will show you the calculated monitor distance.

Does not work with 2D conversions as of now!

Try it out!

[Thomas Claessens]

Seems to work well when going from higher to lower FOV. Messes up when going from low to high FOVs I think.

E.g.:

BF zoom sens to pubg hipfire gives 28 with auto feature (screenshot 1). When trying to 'reverse engineer' the value you get 37. ( see screenshot 2)

[DPI Wizard]

44 minutes ago, Thomas Claessens said:

Seems to work well when going from higher to lower FOV. Messes up when going from low to high FOVs I think.

E.g.:

BF zoom sens to pubg hipfire gives 28 with auto feature (screenshot 1). When trying to 'reverse engineer' the value you get 37. ( see screenshot 2)

Fixed now, try again.

[Thomas Claessens]

works perfectly now, thanks :)! With this method I get a zoom sens of 0.9 in cs, which is exactly what I used during my competitive days.

[potato psoas]

14 hours ago, Drimzi said:

That's why you use 76.19% on that 21:9 panel, and 100% on that 16:9 panel, to maintain the same match lol. Of course the percentages aren't going to line up when the calculator is using a percentage of the horizontal, which is dependent on the aspect ratio. If you think of it as a multiplier of the 1:1 aspect ratio, then both examples in your picture would have the same value and the same match point, which will be independent. In that example, it will be a 16/9 coefficient, 1.778. Any monitor match % results in the 'gear ratio', it just depends on what your measuring. Anything above 0%, it is a simple ratio between the angles, they all follow the 'gear ratio' principle. Compare 90/45 Hdeg 4:3 when matched to 4:3 and you will get 2 * the circumference, same thing when comparing 16:9 values when matched to 16:9, or comparing Hdeg or Vdeg if you are match to 1:1 (depends if landscape or portrait). For 0%, it's not just naiively dividing angles, it's the tan ratio, so compare 90 and whatever fov is exactly half of 90, (it is 53.13) and you will get the gear ratio. All monitor match percentages follow this principle. It just depends on where you measure.

You are assuming that all games use the same FOV Type. If you maintain vertical FOV then the horizontal FOV is cropped and added but if you maintain horizontal FOV then the vertical FOV is cropped and added. The diagram below shows "1:1" monitor matching and how it is not maintained with different aspect ratios because we are maintaining the horizontal FOV rather than the vertical FOV.

But it doesn't even matter because what exactly is "X% monitor match" is arbitrary.

[Drimzi]

When this method is turned into a formula, it looks like this:

s = input sensitivity
a = input pitch/yaw
A = input fov
b = output pitch/yaw
B = output fov

The fov you put in can be any measurement.

Example:

Edited May 11, 2018 by Drimzi

[Derpturtle]

potato psoas - It might not make sense to you mathematically. I'm sure it wouldn't make sense to me either if I understood math higher than a 4th grader (literally havent had to use math my entire life) but..
my whole point is that this actually FEELS perfect. I perceive it as being perfect between every game. And ultimately, what feels right is going to work better than what math tells you is right. Let's also not forget that it is impossible to actually match sensetivity perfectly for all the reasons stated in this forum, ultimately what's going to be the winner is what feels right. Muscle memory isnt just about travel distance and mouse weight, it's about how fast the camera moves in relation to your hand. So when you use 75% it will be close but not perfect because you will actually notice the difference in perceived sensitivity if its too fast or too slow even by just a couple of %.

You mention what type of FOV matters, yes of course it does, that's why I explained that it has to be the true horizontal FOV for the best result. If I used the vertical or if I used the FOV that is in the config files, this simply would not work. In the BF config file for example, it says I have 90FOV, thats only for 4:3 monitors. On a 16:9 its actually 106.52. So it's important to have the true FOV. Most games are just left to right movements with minimal up and down so it makes sens to just focus on getting the left and right correct  I see what you're saying though and I'm not trying to dismiss it. 75% simply does not work for me because it always feels off. Too fast or too slow. Hence why using the exact % works better I think. Thank you for your point of view, I appreciate you taking the time to analyze it further. I'm by no means a smart guy, just going by feel and instinct

Even if the vertical FOV gets cropped or added it doesn't matter because with this method it's all about the perceived sensitivity. I.e the speed at which the camera moves in relation to your hand.

 

Tomas Claessens - WOO Im so happy it worked for more people!!

Ganja - YEAH!

Drimzi - I have no idea what you're typing out but I trust you are correct. Thanks for figuring out the complicated parts

DPI Wizard - THANK YOU!

Edited May 11, 2018 by Derpturtle

[DPI Wizard]

10 hours ago, Drimzi said:

When this method is turned into a formula, it looks like this:

s = input sensitivity
a = input pitch/yaw
A = input fov
b = output pitch/yaw
B = output fov

Not exactly how it's done in the calculator since this method doesn't work with every game. It uses input counts and output counts instead. Also the formula needs to check which FOV is biggest to make reverse calculations work.

Now, how would we apply this method to 2d to 3d calculations I wonder?

[Derpturtle]

Wouldn't it just be match monitor distance 100%? 
It wouldn't make sens to use the same method from 2D to 3D since what you're working out is view speed in relation to mouse movement. The perceived sens.

2D doesn't have camera movement, just cursor movement. So in my mind it makes sens that the same distance I use to move from the middle of my screen to the edge should be the same in both 2d and 3d. Yes, no?

This actually makes a bit more sense when I compare the results with pro CS and Overwatch players.
Most of them use 400dpi and ~42cm/360 in game. That's almost exactly match monitor distance 100% from 2d to 3d.
I would argue that they use 400dpi in windows just because it feels more natural to them having to move their arms about the same amount in both 2d and 3d.

[DPI Wizard]

3 minutes ago, Derpturtle said:

Wouldn't it just be match monitor distance 100%? 

That's what I'm thinking as well. This is also what 360 distance does when converting between 2d and 3d.

[iBerggman]

I don't know how you calculate it but for each fov there's a match % that gives the least average deviation compared to desktop, right? Could that work?

[Thomas Claessens]

(delete)

Edited May 13, 2018 by Thomas Claessens

[potato psoas]

On 12/05/2018 at 03:06, Derpturtle said:

potato psoas - It might not make sense to you mathematically. I'm sure it wouldn't make sense to me either if I understood math higher than a 4th grader (literally havent had to use math my entire life) but..
my whole point is that this actually FEELS perfect. I perceive it as being perfect between every game. And ultimately, what feels right is going to work better than what math tells you is right. Let's also not forget that it is impossible to actually match sensetivity perfectly for all the reasons stated in this forum, ultimately what's going to be the winner is what feels right. Muscle memory isnt just about travel distance and mouse weight, it's about how fast the camera moves in relation to your hand. So when you use 75% it will be close but not perfect because you will actually notice the difference in perceived sensitivity if its too fast or too slow even by just a couple of %.

That's why I said you probably prefer 75% because it is what you are used to. Of course it would feel perfect for you because you've developed so much muscle memory using 75% from playing BF4.

On 12/05/2018 at 03:06, Derpturtle said:

75% simply does not work for me because it always feels off. Too fast or too slow. Hence why using the exact % works better I think. Thank you for your point of view, I appreciate you taking the time to analyze it further. I'm by no means a smart guy, just going by feel and instinct

There really shouldn't be any difference in feel between your percentage method and 75% at all. The problem with going by feel is that you ignore what the math says. And the math says that there is no such thing as a perfect method. Something will always feel a little bit off. And knowing what feels off is the key to choosing what conversion method to use.

[Drimzi]

The best way to test really is to compare a very low and a very high fov. There is very little difference within the 'normal' fov ranges with these methods (except 0%) as they produce very similar results. When you compare low and high, the scaling becomes very obvious. You can then extrapolate that data and assume how 2D (0 FOV) is going to behave, or any other FOV for that matter. If it feels off then something is *wrong*, but I guess as long as you get use to it and you know how the sensitivity is going to behave, then it doesn't matter what method you use.

On a sidenote, I haven't tested this method. Does it even hold up with any aspect ratio other than 16:9?