from stabby 3+ years ago

I'm assuming a sensitivity of below this value with an incredibly high DPI, presumably you would reach a point where the human hand cant make a small enough movement to move under 2 pixels - I'm wondering if anyone knows roughly the cut off for "pixel perfect aim".

Depends on resolution and in game FOV and m_yaw(note that 90 fov in game isn't always 90FOV)
https://funender.com/quake/mouse/

Example

1920x1080
800DPI
m_yaw = 0.022 (default)

calculate the sensitivity and then the useful DPI

Basically you're looking for something that is equal to or less than the DPI you're using. *note pretty sure this calculation of useful DPI is based of the center of the screen and isn't the same for the far edge of your screen* using the calculator on this website give you a max sensitivity of about ~3.6 since 16:9 has a ~106FOV.

This should remain true regardless of DPI
400 DPI max of ~3.6 ingame
800 same
1600 same
etc

2560x1440
106FOV (max for resolution)
m_yaw 0.022

max sens = ~2.7

1024x768
90FOV
m_yaw 0.022

max sens = ~ 5.08

-edit-

so to take extreme examples here

640x480
90fov
m_yaw 0.022

12000 DPI
max sens = ~8.138

so unless you wanted to force a custom resolution that's lower than 640x480 or found a mouse that did more than 12000DPI without skipping /double counts this is your max sensitivity.

cm/360 = 0.4256127259322148
in/360 = 0.16756406532764362

As usual stabby's explanation is bullshit.

Pixel skipping is not a thing. It doesn't exist.

Video games are not some static pictures made up of pixels that can only be moved by a whole pixel in any direction.
Anything you see is rendered from scratch. You're not turning in pixels, you're turning in degrees (or fractions thereof).
That's the resolution the game is working with.
So one count means the game will turn by e.g. 0.001°. That's the resolution you're limited by. At no point does the game go "Oh you want to turn by 0.03°? Nah mate, can't do that with this resolution." Only much later does the game even care about resolution.
Also anti-aliasing takes multiple samples per pixel to figure out where exactly a line is to mix the colours properly. I mean you know how it looks without anti-aliasing? That's because any pixel's colour is just determined by which colour its center is. With AA it'll blend the colour depending on how much of the pixel is covered by which colour. Obviously that means nothing needs to be moved pixel by pixel, any step in between is fine.

So the problem isn't DPI/CPI, it's sensitivity. Because that's applied to the counts. Usually it's just counts*sensitivity but maybe 1 scaled count translates to 10 ingame counts or 0.003°, that's completely arbitrary. But no matter what there's a specific number of counts for a full 360° turn. Let's say it's 1 billion. Now if you set your sensitivity to 2.5 billion a single count sent from the mouse would lead to two and a half turns. That's a problem. Don't turn up your sensitivity to >100 and everything should be fine.

tl;dr
Pixel skipping doesn't exist.
2.7128 is a sensitivity wo1fwood used for an example and not some magic value beyond which the source engine will stop working. http://www.teamliquid.net/forum/tech-support/333648-an-overview-of-mouse-technology

So your question is wrong. The game will register single counts and move accordingly. That means with 1000 dpi (and sensitivity 1 to not further complicate things) you'd have to be able to move your mouse in 25.4µm increments. Can you do that? Of course that's less than one pixel. But as mentioned above limiting it to "one pixel" which doesn't work anyway because the sphere->plane mapping means each pixel got a different width/height and pretending that everything will be fine below that and that the sky will fall above it is incredibly pointless.

Even with this mindset the "2.7128" number is wrong. it assumes you have 90fov at 1920x1080 which is incorrect you actually have ~106fov. 2.7128 would be correct at 1920x1440 with 90fov in game though.

I think it's patented stabby math. 1080/400 = 2.7 but 2.7128 looks cooler. Or just random quoting.
Actual context from the article

It's about matching the degrees per pixel to the degrees per count and fairly pointless because like I said the game doesn't work in pixels anyway.

Thanks for the replies. When I said "one pixel" I was meaning the movement of one pixel of the crosshair on a 1920x1080p monitor, so apologies for any confusion surrounding the nature of my question. So, from what I've read here, high sens shouldn't be a problem then unless the numerical in game value is ridiculously high?

3.5 or lower + whatever DPI you like is about as close as a good recommendation anyone can give you in terms of "pixel" accuracy on a 1080p screen. Just don't do stupid shit like 200 DPI with 10 ingame sens.

No, your question is simply pointless.
The pixel directly in the center of the crosshair might move roughly one pixel per count at that sensitivity but all other pixels will not.
At any other sensitivity they will also not move "pixel perfect". At 25% slower sensitivity even that center pixel would move 0.8 pixels instead.
It's a completely arbitrary limit with no practical value.

Also it is related to the DPI of your mouse.

ehh this is a tadd misleading

if all you change is you dpi and you don't change any other settings including in game sensitivity the smallest angle you can move your mouse doesn't get any smaller. If you double your DPI and half you ingame sensitivity the smallest arc angle you can move is halved despite having the same inch/cm per 360. Whether or not you'll notice that kind of difference though depends on your FOV, resolution, where on the the screen you're looking for the difference, how far your screen is from you, the size of your screen, etc.

basically ingame sensitivity determines your minimum arc, your dpi determines the smallest real world movement that that count happens at.

Example.
Sensitivity = 1

200 DPI
Inch/360 = ~81.81818181
degrees per inch = ~4.400000
smallest in game arc movement = Degrees per inch / Dots per inch** = ~0.022000

800 DPI
Inch/360 = ~20.45454545
degrees per inch = ~17.600000
smallest in game arc movement = Degrees per inch / Dots per inch** = ~0.022000

~0.022000
distance to move in real world
200DPI = 0.005 inches
800DPI = 0.00125 inches

*some rounding was done in the formulas
**I'm fairly certain I've used the wrong formula for determining the minimum movement of arc but I'm certain that in this example 200DPI and 800DPI would have the same minimum

Why is it misleading?

It's obviously relevant for this.

Misunderstood the context. Thought you were replying to something different with that line.

Using cl_showpos and setpos it seems that 1 pixel is equivalent to around 0.05785 degrees for 1440p (I measured this by using a mega low sensitivity and looking at the rotation I have to make for a pixel to cross the width of my crosshair). For 1080p users I would expect this to be about .077 (0.05785 * 1440 / 1080).
With sensitivity 4 and 1600 DPI I can only move my view in increments of 0.09 or 0.08 - it isn't that I'm moving too fast it literally doesn't let you look at angles between these increments. You can try it for yourself by using cl_showpos 1 and trying to get specific values. For instance without changing my sensitivity I can look at 90.00 and 90.09 but no value in-between.
So if you want to be able to move by 1 pixel consistently (for whatever reason) then for 1600 DPI and 1080p your sensitivity should be able to be as high as ~3.6 (4 * .077 / .085).
It may be that other gamers get different results for their setups but hopefully this gives u an idea on how to work this stuff out in-game.

comanglia and setsul jerk off together thread

If a user has Raw Input disabled, then their mouse movement is being determined by the Windows Desktop mouse cursor, which operates pixel by pixel on a 2D plane. Pixel Skipping certainly exists for the Windows mouse cursor, and will occur if the user has their Windows sensitivity set higher than default. If Raw Input is disabled, and the Windows mouse cursor is experiencing pixel skipping, the problem will correspondingly manifest itself in TF2 (i.e. "exist").

Most users choose to enable Raw Input, which does not utilize the Windows cursor, so this is a bit of a moot point--but the question in the OP is a practical one with a definite answer, and the problem "pixel skipping" presents is relevant to it. I believe what he is asking is "What is the highest in-game sensitivity setting in TF2 that will not limit the granularity of my rotation?"

There has to be a concrete answer to this; like an actual number that can be given based on the user's DPI, display resolution, and settings like accel/FOV/m_rawinput. Saying simply "don't use anything stupid" strikes me as lazy and unhelpful. There's an actual answer to this actual question.

The "2.7182" number comes from me doing the following with the equations I found courtesy of this site and its "Useful DPI Calculator":
https://funender.com/quake/mouse/index.html

Here are the relevant equations:
Real Sensitivity ("i"): 360 / (m_yaw * DPI * TF2sensitivity)
Useful DPI: (pi * Horizontal Resolution) / (i * tan(FOV / 2))

Here's what I did with them:
1) I inputed my settings into the "Real Sensitivity" calculator:
360/(.022 m_yaw * 9050 DPI * 1.3 TF2sensitivity) = 1.3908743190511148

2) I calculated the useful DPI for a 1920 resolution, 90 degree FOV, and ~1.3" Real Sensitivity
(pi * 1920) / (1.391 * tan(90/2)) = 4336.738274819446

3) I calculated the TF2sensitivity that I would need to use to keep my "Real Sensitivity" the same if using "Useful DPI":

360 / (y * d * s) = 1.3908743190511148
360 / (.022 * 4336.738274819446 * s) = 1.3908743190511148
s = 2.71286834

2.7128 is the sensitivity that corresponds to a maximum useful DPI with a horizontal resolution of 1920 and an FOV of 90.

With an FOV of 106 (instead of 90), using the same math, that number changes to ~3.6, which it would seem is corroborated by HOI's findings:

With my own setup of a 2560 horizontal resolution and 106 FOV, the number does in fact happen to come out to ~2.70.

By all means, critique my math, the equations/the site I'm working with, or how I'm approaching the question. My hope is that the OP's question can actually be answered, not that I'm right.

higher sensitivities will definitely make your tracking suffer but if you aim by leading/flicking then that doesn't really matter unless your definition of "aim skill ceiling" is tracking.

It doesn’t “cap your aim ceiling” it will just take longer to get better since a lower sens is more forgiving

dpi is irrelevant in this question, it's simply the number of "ticks" a mouse gives to the computer for every inch you move the mouse. if you had 1dpi and something normal like 1.5 ingame then your movement would still be smooth, just VEERRRY sloooowwww

if you had 1dpi and something like 100 ingame, then you'd get large amounts of "pixel skipping"

if you had 1 dpi and something like 0.1 ingame, every pixel on your screen would change when you moved your mouse an inch, and you'd technically be skipping an angle that you can't aim at, just a very small one.

there isn't a definitive answer to this, literally just "dont use anything stupid" and you won't notice any skipping.

Correct--and *because* DPI is irrelevant to the question, we should be able to provide a definite numerical answer (if we know resolution and FOV).

Once we've established what the "maximum in-game sensitivity" for a given FOV/resolution is, an individual can then adjust their DPI around whatever that sensitivity value may be to achieve their preferred inches/360.

There has to be a definitive point at which "anything" becomes "stupid".

I’m getting a headache

For pure interest: with fov_desired 90 & 1920x1080, trying out a custom map that looks like this, and trying out setang values with until the center line was shifted exactly 1px (had to use max antialiasing settings to compare better), I got an angular movement of ~0.062 giving me the best results. Stabby/wo1fwood's ~0.0596 degrees (from 2.71286834) and HOI's ~0.077 were both demonstrably off when I tried it. This would imply that the "pixel perfect" sens should be ~2.818.

Setsul is entirely right that it's nearly meaningless and a very poor way to pick your sens though. You will notice every other factor that goes into a sens (whether it's too high or too low for your aiming, whether it malfunctions your mouse sensor) and every other factor that causes stuttering / inconsistency (monitor hz, framerate, antialiasing settings) long before you notice your sens being slightly mismatched with the pixels directly in the center of your screen. You can easily play with a higher value than that and never notice anything wrong, because we're talking about such a specific and minor criteria that 99% of players would not notice if you pointed it out to them.

That point is "as long as it's not bothering you." If a sens works for you, it's not stupid. If a sens doesn't work for you for any reason whatsoever, it's stupid.

You could also decide that you want no pixel skipping anywhere on your screen (completely reasonable, arguably more valuable anyway), and require a ~0.03 degree angular movement, and now ~1.36 is your max sens. Maybe you want granularity of half a pixel on the edge of your screen (fair, definitely an improvement), and with a ~0.015 degree movement ~0.68 is your max sens. You can come up with whatever cutoff for "definitively acceptable" you want, and it's all nonsense because no one actually notices this shit ingame.

Wrong.
TF2 does not use windows desktop pixels for the rotation counts (no game should).
Depending on the ingame sensitivity the minimum step that corresponds to the number of counts in one desktop pixel could be 1 ingame pixel or 13.245 ingame pixels or 0.0045212 ingame pixels.

Also if if corresponds to 1.1 ingame pixels then you will simply turn in 1.1 pixel increments, without skipping any pixels in between.
Pixel skipping doesn't exist.

Again, irrelevant.
Iirc the source engine uses float to represent angles. That means if we pretend that exponents and signs don't exist just the full precision of the 24 bit mantissa requires 16,677,216 counts per 360. At 12,000 DPI (and that's not really a thing, just multiplication and some fancy DSP footwork) that's ~1398.1 inches per 360. So about 116 and a half feet or about 35.5 meters for a full turn. And that's the lower bound, it's actually much higher.

See above. There is an actual answer and for the source engine it's utterly useless.

I won't check the actual numbers.
But setting the sensitivity via DPI? Come on.

So are you using 4336.738274819446 DPI now? If yes I'd like to know how you set that.
If not then it's not "pixel perfect". With 9050 DPI and 1.3 ingame sensitivity you'll instead turn by 0.47919759942756309392265193370166 center pixels per count (all other pixels are different). Literally unplayable.
Now imagine what would happen with 2.8 ingame sensitivity. 1.0321179064593666638334041648959 pixels per count. LITERALLY UNPLAYABLE.

The question is pointless and the answer is useless.

No mouse will allow non-integer DPI steps anyway, but if you can't set that then one count will translate to something like 1.0004 pixels of 0.9997 pixels, making the limit even more arbitrary and pointless.
It just has to opposite effect because non-native DPI steps fuck up everything.

setsuls a facking genius 0_o

this threads sick its like geniuses collide wat u would imagine a Harvard dorm room sounds like

but also there ISN'T a definitive answer. it's a sliding scale and it'll differ depending on what an individual prefers (if they can even perceive changes properly at that extreme). and if there was, why on earth would you want to be right on the edge of what's technically the highest ingame you can have? it makes no sense

if you want to give yourself the highest fidelity in a perfect world you would use a tiny ingame and a big DPI, but as setsul said big DPIs are just fudged numbers and some mice become less accurate at higher dpis

literally, just keep your dpi within reasonable bounds (400-1600 is pretty normal) and set your sens to whatever you want. if you change dpi from 400 to 1600 and change your sens to keep inches/360 the same, I guarantee you wont notice any difference

and if you're using non-raw input or 0.1"/360 or fov toggles or a fidget spinner for a mouse or some shit you've got bigger things to worry about

https://i.imgur.com/bx59Yzu.jpg

I think you misunderstand me. Pixel skipping exists on the Windows Desktop, and the mouse input in TF2 is based on the Windows Desktop cursor if m_rawinput is '0'.

No, I use 9050 DPI. I was showing my math with what you're referring to, in order to demonstrate where the 2.7128 number was derived from.

The question has been answered, I think. You wouldn't *want* to have the highest in-game sensitivity, you might simply want to know whether to use your mouse's lowest or highest native DPI step, and lowest might put you above that "highest" sens.

No, I don't.
Read it again.
Moving one pixel on the desktop does not mean it'll move one pixel ingame.

You still don't understand.
2.71286834 is the sensitivity where one count will cause a turn of one center pixel if and only if the DPI are set to 4336.738274819446.

The point is that this

is completely wrong.
No pixels will be skipped.
If the sensitivity is anything other than ~2.7 you will also be unable to move your view by one pixel at a time. For example at 2.1 ingame sensitivity each count will move your view by 7/9 center pixels. You are utterly unable to move your view by one pixel at a time.

No it's simply utterly irrelevant.
The 1 pixel limit is completely arbitrary. It doesn't matter if one count results in 1 pixel, 0.7 pixels, 2.3 pixels or 0.03 pixels of center movement. You don't have enough control or enough time to move the mouse in single counts because those would be <32µm steps. No one does that.
Literally the only thing you have to do is not go full retard and set your mouse to 1 DPI.
No one gives a shit if 800 DPI results in 1.5 pixels and 1600 DPI in 0.75 pixels. At the edges both will move more than one pixel anyway. Choose the setting which isn't broken on your mouse. If both are working flip a coin.