Viewspeed v2

[potato psoas]

2 minutes ago, CaptaPraelium said:

That's...mspaint.............. You're a God. Sorry I lack the skill and patience for that jazz lol

Yeah I've spent hours making diagrams these past two weeks. (Lucky I don't need to go to work)

[potato psoas]

31 minutes ago, CaptaPraelium said:

IMO the vertical and horizontal sens should be the same since 1 degree up and 1 degree left are the same angle of 1 degree no matter how that 1 degree is represented on screen. Consider how your suggestion would effect a player who can rotate around the Z axis (example an airplane which can roll). Now that you bring it up, "would this work in a plane?" is probably a good checklist item for any formula...

I'm not so sure I can wrap my head around the z-axis...

But since CS:GO was converted in Green we need to have a different vertical sensitivity because it isn't supposed to share the same cm/360 value as the horizontal part. I don't think it really matters which method was used though as long as there is a way to adjust the vertical sensitivity the way we like in the settings. I'm thinking I may have to use the m_yaw value in CS:GO now (changing DPI can only be so accurate).

[Drimzi]

Edited January 28, 2018 by Drimzi

[Drimzi]

Edited January 28, 2018 by Drimzi

[WhoCares?]

2 minutes ago, Drimzi said:

I'm going to test diagonal FOV soon. The numbers are very desirable, but it is going to take a long time to calculate all of the FOVs for CSGO to test it out.

Testing a formular for 2D to 3D convesion?

Or from 3D to 3D?

In your posted video Viewspeed v2 looked on point! (And I always felt it was really good, but maybe because I always used 75% match distance before, which is very Close)

 

I am sorry that I ask, but there were so many new ideas in the last weeks...and most of them got knocked over that I am confused which formular is in progress and which not 

 

[Drimzi]

Edited January 28, 2018 by Drimzi

[WhoCares?]

1 minute ago, Drimzi said:

It's the exact same viewspeed v2 formula, but you give it the diagonal fov instead of the vertical. Scaling through every FOV incrementally does look and feel completely on point, but jumping straight from a high FOV to a low FOV does not.

All right

And for 2D to 3D? We are mostly sure that no method is completely correct atm, right?

[Drimzi]

Edited January 28, 2018 by Drimzi

[WhoCares?]

1 minute ago, Drimzi said:

Remember those previous formulas where I thought the 2D to 3D was better? The diagonal results are very close to those formula.

So viewspeed v2 could be correct for 2D to 3D as well?

Or do you mean one of the newer formulars but with diagonal fov like:

 (360 × vertical-resolution)/(dpi × fov)

?

[Drimzi]

Edited January 28, 2018 by Drimzi

[KandiVan]

On 12/9/2017 at 3:57 AM, Drimzi said:

I still haven't settled on the desktop conversion. The formula does already use the arc length scaling, so applying it again to the desktop conversion probably doesn't make sense. Using the diagonal length like I did in viewspeed v2, results in 90 degrees having the 180 degrees arc length after the scaling, and thus a much slower sensitivity overall. At 90 4:3 fov, you do end up with a sensitivity that is slower than viewspeed v2, but the fov scaling for aim down sights and scopes should make up for this. I am going to test the slower alternatives for the desktop conversion shortly.

 

This is the formula where the desktop length is the diagonal, and thus the 180 degrees arc length is remapped to 90 degrees.

Horizontal Deg. | 4:3 Base

Horizontal Deg. | Res Base

Vertical Degrees

 

I definitely encourage this one over the previous.

^ For the record, this is the one that feels the best to me. Feels like I can consistently two tap and track with this formula, others feel too fast and I just bounce around the head.

Edited December 10, 2017 by KandiVan

[WhoCares?]

9 minutes ago, Drimzi said:

Viewspeed v2. It could have been correct this whole time, but required the diagonal fov and not the vertical.

That would be great!

Tho,..I could swear someone already tested diagonal fov some time ago. And we came to the conclusion is wasn't the right direction. Not sure if it was this thread or one of there others were we were discussing viewspeed. 

Maybe I remeber wrong, I read some much information about this topic today (again), maybe my brain just trick me 

Edited December 10, 2017 by WhoCares?

[Drimzi]

Edited January 28, 2018 by Drimzi

[KandiVan]

1 hour ago, Drimzi said:

With the original viewspeed v2 formula using the diagonal fov, comparing 4:3 90 FOV and 45 FOV, the ratio is 0.52, the exact same as that formula that feels best to you. So that looks very promising. Except this 2D to 3D conversion is the same as the ones that I thought was best.

:thinking:

[Drimzi]

Edited January 28, 2018 by Drimzi

[CaptaPraelium]

potato psoas what I'm saying is that horizontal and vertical sensitivity should always be the same, because 1/360th of a full rotation is 1 degree no matter which way we rotate, and we always want the mouse to have the same counts per degree.

Thinking this through, I feel like in the past we're usually focussed on the changing FOV as though the rendered world on our screen is changing. This is evident by pretty much every diagram on these forums where the FOV is represented by a circle, and we have different sized circles for different FOV. Which seems to make sense. But let's remember this image:

The amount of distortion at any point in the world is actually never changing. We could imagine that everything is like the BF3 180FOV video, all the time. We just have a limited window of that exact image. The FOV setting in game controls the size of that window, and then whatever is within it is scaled up to fill our monitor.

 

[CaptaPraelium]

food for thought.

[Drimzi]

Edited January 28, 2018 by Drimzi

[Cocyx Skeleton]

The Diagonal 90 Arc Length feels fantastic to me, personally. The 90 Arc Length feels way too fast.

[Drimzi]

Edited January 28, 2018 by Drimzi

[Cocyx Skeleton]

3 minutes ago, Drimzi said:

Thanks, what FOV did you test at? Shoot the Beat at 100 vertical just felt too slow with the  Diagonal 90 Arc Length formula.

Played CSGO on the regular 90 FOV (HDeg 4:3).

I might just like it better because the Diagonal 90 Arc Length formula is closer to the Viewspeed v2 calculator conversion for Windows / Desktop, but it felt just the right speed to me.

[Drimzi]

Edited January 28, 2018 by Drimzi

[KandiVan]

My sleep schedule is still messed up from sweden and i need to sleep or else id give both a try right now. These are adequate 2D->3D conversions though correct? I convert from 4/11 windows 400 dpi at 1600x900  and have been using that basically since the start of the summer. so i should have a decent idea of what feels right, ill give them a go in the morning and report back. Thanks again for all your hard work fellas!

[Drimzi]

Edited January 28, 2018 by Drimzi

[potato psoas]

8 hours ago, CaptaPraelium said:

potato psoas what I'm saying is that horizontal and vertical sensitivity should always be the same, because 1/360th of a full rotation is 1 degree no matter which way we rotate, and we always want the mouse to have the same counts per degree.

Well if you want the same cm/360 for different FOVs then you need something like this:

And if either FOV axis shares the same circumference then they must share the same radius. So you can pick either to do the calculation.

cm/360 = 2*pi()*radius

radius = (0.5*monitor distance)/(0.5*FOV)

However when I looked at CS:GO it was more like this:

If the converted Vertical FOV was lower then they could share the same cm/360 but it is not.

As an example, at 90 config FOV the horizontal FOV is 100.39 and the vertical FOV is 73.74 but really if you want to use the same cm/360 the vertical FOV should be 57.39.

The correct way to convert is using this formula:

VFOV = 2*DEGREES(ASIN((0.5*vertical monitor distance)/horizontal radius))

Edited December 11, 2017 by potato psoas