Viewspeed v2

[Drimzi]

Edited January 28, 2018 by Drimzi

[seventhfrost]

On 12/27/2017 at 4:34 PM, Drimzi said:

Playing McOsu with the first person mod, which makes the whole screen move instead of the cursor, can show just how fast the 2D speed really is. Since most will be unaware of there true 2D speed, after translating that 2D speed correctly to rotation speed, one can quickly dismiss it as being too fast. Especially when they use the same physical movement that they are use to using, and judge by flicking distances rather than actual feel.

I had a feeling this was going on with my testing and when i tried to stop it, i started really thinking the faster ones made more sense. I'm glad the mcosu thing seemed to shed some insight, gotta try it myself soon. The thing about a 2:1 intrigues me, and i'd want to know how effective it might be versus just not matching at all and not worrying about it, but I'm not sure much can be done as far as telling how worthwhile that is (as opposed to just picking the ingame 360 you like and matching FOVs) without knowing the best 1:1 for sure.

On 12/27/2017 at 8:54 PM, Drimzi said:

You can use this formula, which is the formula I think is truly 1:1 with the desktop.

Gonna try this one out today. I'm at 300 dpi so the 360 is serviceable. 

[Cocyx Skeleton]

What's the best method for calculating CSGO's zoom sensitivity ratio using the new formula? I'm not sure what CSGO's zoom FOV is, or which one I should use.

[Drimzi]

Edited January 28, 2018 by Drimzi

[Cocyx Skeleton]

I got 0.4521473983201809 which doesn't seem right (1.77685339 divided by 3.92981005). I can't really see what I'm doing wrong though. Also, is the 40 FOV the first zoom or the second zoom of the awp?

[Skwuruhl]

I'd just like to point this out again:

https://imgur.com/a/05PqD

Imo matching physical mouse movements with physical screen distance should be the goal of ADS scaling. If 1:1 ratio diagonal match aka 70.71% match distance is your usual flick distance then that will be best for you. If your target is near your crosshair by the time you're in ADS then 0% or a small % will be better.

I had more typed but the forums editor reverted to an old version and I don't feel like typing it all again. Basically was just saying the important part is that you consistently use the same match %.

Edited December 31, 2017 by Skwuruhl

[potato psoas]

18 hours ago, Skwuruhl said:

I'd just like to point this out again:

https://imgur.com/a/05PqD

Imo matching physical mouse movements with physical screen distance should be the goal of ADS scaling. If 1:1 ratio diagonal match aka 70.71% match distance is your usual flick distance then that will be best for you. If your target is near your crosshair by the time you're in ADS then 0% or a small % will be better.

I had more typed but the forums editor reverted to an old version and I don't feel like typing it all again. Basically was just saying the important part is that you consistently use the same match %.

If you are trying to maintain consistency with the 2D image, you should be scaling your monitor match based on FOV/distortion. You can't just use one single monitor match. Using COS(FOV/2) I get a monitor match of 50% at 120FOV, 70.71% at 90FOV, 92.39% at 45 FOV and 99.14% at 15FOV.

That being said, I came up with a scaled (aspect ratio independent) monitor match formula and it ended up being the same as 0% monitor match, which honestly makes no sense.

Edited January 1, 2018 by potato psoas

[Skwuruhl]

2 hours ago, potato psoas said:

If you are trying to maintain consistency with the 2D image, you should be scaling your monitor match based on FOV/distortion. Using COS(FOV/2) I get a monitor match of 50% at 120FOV, 70.71% at 90FOV, 92.39% at 45 FOV and 99.14% at 15FOV.

Why would I do this? Monitor match is already taking into account this distortion. It's literally the entire point of using it over adsFOV/hipFOV scaling. The amount of distortion is modeled by tan(x/2). Your FOV at any given point is given by 2*arctan(d*tan(x/2)). These equations didn't come out of nowhere. I mean in my last post there's empirical evidence showing that it works. It ties screen distance to mouse distance. If you change the screen distance by throwing it into some arbitrary equation you completely undo that.

2 hours ago, potato psoas said:

That being said, I came up with a scaled (aspect ratio independent) monitor match formula and it ended up being the same as 0% monitor match, which honestly makes no sense.

Can you link to the post with the math for this?

Edited January 1, 2018 by Skwuruhl

[potato psoas]

3 hours ago, Skwuruhl said:

Why would I do this? Monitor match is already taking into account this distortion. It's literally the entire point of using it over adsFOV/hipFOV scaling. The amount of distortion is modeled by tan(x/2). Your FOV at any given point is given by 2*arctan(d*tan(x/2)). These equations didn't come out of nowhere. I mean in my last post there's empirical evidence showing that it works. It ties screen distance to mouse distance. If you change the screen distance by throwing it into some arbitrary equation you completely undo that.

Monitor match doesn't really take distortion into account. E.g. if you use 100% at 179 FOV, it will feel so incredibly slow at the center. It would be matched to the edge of the screen but the closer you get to the center the more incorrect it feels. Same applies to 0% except now the edges are too fast. But what scaling does is it finds the "middle ground" where the edge and the center are equally incorrect, therefore making it much more consistent between FOVs at all points on the monitor.

Here is a diagram showing what I mean, where the red lines depict the distribution of points on the monitor:

The monitor match will scale according to the distortion/FOV. And I thought the best way to scale it would be COS(FOV/2).

If you were going to use a single monitor match, the most average of them would be 20%. You can use integrals to calculate that.

But I don't think it's a good idea because as you approach 180FOV and 0FOV you want to approach 0% and 100% respectively.

If the distortion was corrected then we could just use 100% the whole time.

3 hours ago, Skwuruhl said:

Can you link to the post with the math for this?

I linked an excel spreadsheet before, but I don't think you should bother looking at it cause it's probably wrong. You can use COS(FOV/2) to scale your monitor match but the hard part is making it aspect ratio independent. Although COS(FOV/2) is so close to aspect ratio independent that you probably wouldn't be able to tell the difference.

Edited January 1, 2018 by potato psoas

[Skwuruhl]

1 hour ago, potato psoas said:

Monitor match doesn't really take distortion into account. E.g. if you use 100% at 179 FOV, it will feel so incredibly slow at the center. It would be matched to the edge of the screen but the closer you get to the center the more incorrect it feels. Same applies to 0% except now the edges are too fast. But what scaling does is it finds the "middle ground" where the edge and the center are equally incorrect, therefore making it much more consistent between FOVs at all points on the monitor.

Well yeah that's kinda how the distortion works. If you match at the very edge of your screen then the center isn't going to be matched and vice versa. You can't get around this. Even if you match at 50% that'll just mean it's perfectly matched at 50% but still off at the center and edges of the screen. It's not possible for there to be a method where the entire screen is matched at once.

However

1 hour ago, potato psoas said:

Here is a diagram showing what I mean, where the red lines depict the distribution of points on the monitor:

This is a valid point that comes down to personal preference. Yes it will be "most correct" for every point on the screen but in my opinion the area around your crosshair is the most important to match. Mostly because that's where you're doing micro flicks and tracking.

1 hour ago, potato psoas said:

The monitor match will scale according to the distortion/FOV. And I thought the best way to scale it would be COS(FOV/2).

If you were going to use a single monitor match, the most average of them would be 20%. You can use integrals to calculate that.

But I don't think it's a good idea because as you approach 180FOV and 0FOV you want to approach 0% and 100% respectively.

If the distortion was corrected then we could just use 100% the whole time.

Here I disagree though. I don't see how taking the cosine of half of the higher FOV is supposed to correctly give an 'average' monitor match. You are right that you can't just linearly scale it because monitor match doesn't scale linearly, it has more "matches" at low screen distances. https://www.desmos.com/calculator/48ico4emry

you can use integrals to average all this: https://www.desmos.com/calculator/wnbs9nuyeo

In this case it comes out to be about 62% match distance http://www.wolframalpha.com/input/?i=0.434508085152%3Darctan(xtan(51­°%2F2))%2Farctan(xtan(103°%2F2)).

I do want to reiterate that even though on average this will be the closest to correct, it will still be somewhat significantly off at the crosshair. Also that having different match distances for different zoom levels may impact your aim. I still believe that 0% is the best for lots of reasons.

[potato psoas]

17 minutes ago, Skwuruhl said:

Here I disagree though. I don't see how taking the cosine of half of the higher FOV is supposed to correctly give an 'average' monitor match. You are right that you can't just linearly scale it because monitor match doesn't scale linearly, it has more "matches" at low screen distances. https://www.desmos.com/calculator/48ico4emry

Well scaling using cos(FOV/2) made the most sense to me. MM would scale along the unit circle, with radius 1 (representing 100% MM) and through trig I end up with MM(FOV)=radius cos(FOV/2).

But even though it makes sense it's not aspect ratio independent.

The other idea I had was to use a linear method: (180-FOV)/180

Even though it makes less sense, the 360 distances are tighter than using cos(FOV/2).

 

Ultimately, to make it aspect ratio independent we need the monitor match formula for both horizontal and vertical to give the same 360 distance. I.e.:

HCL((π HMM)/(ATAN(HMM TAN((π HFOV)/360)))) = VCL((π VMM)/(ATAN(VMM TAN((π VFOV)/360)))) (CL = chord length, MM = monitor match percentage)

but I don't know how to get HMM/VMM on its own.

34 minutes ago, Skwuruhl said:

I do want to reiterate that even though on average this will be the closest to correct, it will still be somewhat significantly off at the crosshair. Also that having different match distances for different zoom levels may impact your aim. I still believe that 0% is the best for lots of reasons.

I actually disagree that it's best to match only at the crosshair. Yes you do a lot of aiming there but it's not the only place on the screen you need muscle memory for. Snapping to targets closer to the edge happens all the time, especially in games like osu!. Scaling the monitor match minimizes the flaws of both 0% and 100% monitor match.

[potato psoas]

3 hours ago, potato psoas said:

I actually disagree that it's best to match only at the crosshair. Yes you do a lot of aiming there but it's not the only place on the screen you need muscle memory for. Snapping to targets closer to the edge happens all the time, especially in games like osu!. Scaling the monitor match minimizes the flaws of both 0% and 100% monitor match.

Elaborating on this, if you do want to have the best muscle memory at the crosshair then you shouldn't use 0% you should use 20% or something. When it comes to tracking, you are reacting to how the enemy moves. By the time you have reacted they are already off your crosshair. 20% is actually pretty good because the center hardly feels any different between FOV but it makes a world of difference with regards to the rest of the screen.

[Drimzi]

Edited January 28, 2018 by Drimzi

[WhoCares?]

26 minutes ago, Drimzi said:

I'm back from a week holiday. Hope you all had a happy new year. After using the computer after a brief hiatus, my desktop speed actually felt pretty fast. After testing multiple games, I still think the 3D speed feels identical to the 2D speed using the diagonal monitor match. I don't think I will be working on this any further, as I am extremely happy with this method.

Thanks, hope you enjoyed your holiday as well !

Which formular are you referring to be exact ? Between all the going back and forward between different formulars I just want to make sure I am certain what formular you are talking about 

This one?

http://www.wolframalpha.com/input/?i=x%3D90;+y%3D3%2F4;+h%3D1080;+m%3D2400;+w%3D1%2F8;+f%3Darctan(sqrt(2)+y+tan((π+x)%2F360));+d%3Dπ+h%2F(m+w);+d+%2F+f

Or maybe this one? http://www.wolframalpha.com/input/?i=x%3D90;+y%3D3%2F4;+h%3D1080;+m%3D2400;+w%3D1%2F8;+f%3Darctan(sqrt(2)+y+tan((π+x)%2F360));+d%3Dπ+sqrt(2)+h%2F(m+w);+d+%2F+f

Edited January 5, 2018 by WhoCares?

[Drimzi]

Edited January 28, 2018 by Drimzi

[WhoCares?]

9 minutes ago, Drimzi said:

This one.

Alright, thank you

 

I think I going to use 300 DPI with WPS at 5 for an effectiv DPI of 225 (before I used 300 WPS 6). Thats pretty close to my old 360° distance at 106.26 Hfov.

This formular is close to a 80% monitor match distance (for comparison), right? 

Edited January 5, 2018 by WhoCares?

[Drimzi]

Edited January 28, 2018 by Drimzi

[Cocyx Skeleton]

2 hours ago, Drimzi said:

I'm back from a week holiday. Hope you all had a happy new year. After using the computer after a brief hiatus, my desktop speed actually felt pretty fast. After testing multiple games, I still think the 3D speed feels identical to the 2D speed using the diagonal monitor match. I don't think I will be working on this any further, as I am extremely happy with this method.

I actually like it a lot too after using it for a while at a lower DPI. What's the zoom sensitivity ratio that you use? I couldn't seem to figure out how to calculate mine (it gave me an absurdly low sensitivity)

[KandiVan]

3 hours ago, Drimzi said:

This one.

I change y to "1" if im using vFOV on a 16:9 resolution correct?

 

 Is this correct for 1920x1080, 67 vFOV, 400 DPI 4/11 WPS

Edited January 5, 2018 by KandiVan

[Drimzi]

Edited January 28, 2018 by Drimzi

[KandiVan]

Just now, Drimzi said:

1.017332

 

Yes that is correct. 1 is vertical.

Roger that. Significantly faster, will give it a try and report back.

[WhoCares?]

Is this formular valid for all aspect ratios? 21:9 as well ? (the  y value would be 27/64 for 2560*1080 resolution ?)

[Drimzi]

Edited January 28, 2018 by Drimzi

[seventhfrost]

Was gone for a bit, but I've been trying out the formula you settled on and i don't have any complaints with it. It's fast but it makes sense to me, having tested it. And despite how fast it is, I seemed accurate enough right off the bat that I'd buy it. I am still interested in how viable people think using a ratio when going 2d to 3d might be, or if that might provide too little tangible benefit to be worth it over finding a preferred ingame sensitivity.

Either way, I'm very happy this seems to have found a solution. Of course, thanks for all the hard work drimzi.

[Drimzi]

Edited January 28, 2018 by Drimzi