I dunno if it'd be something that you'd be interested in but I'd love to see a comparison across different Doom source-ports. There seems to be quite a bit of variation between them and it's an aspect that's not really talked about much in the Doom community.
At one point I was under the impression that we were moving away from aim assist with both Killzone Shadowfall and R6:S. It's a shame that never happened. Valorant isn't my type of game but seeing it release for console is pretty massive. That forced aim smoothing must be rough!
@@purebaldness It is a shame more devs didn't try games without assists. I think the competitive aspect would have been cool to see, but I have a feeling AA is one of the bigger reasons stick improvements have been so slow. I was worried this would control and flub like CS:GO did, but it seems like it's doing pretty well.
@@EternalDahaka Yeah, double A studios taking us back to the early-mid 2000's! Yeah to be fair it looks like the devs here are taking it seriously. The options here are impressive from what I saw (didn't watch whole video). Ah man, that CS:GO port seemed rough. I didn't play it myself, mind.
Thanks as always for putting the work into this. I am hugely appreciative of your work and fully understand the frustration. This port is so close, yet so far away in terms of being an amazing product for controller users. I can accept that these will never be fixed in Q1. I am -coping- hopeful that we'll get an 'Update 2', following 'Update 1' with Quake II Enhanced, perhaps addressing these issues. Either way, if Quake III ever got the same Nightdive treatment, there is absolutely no way that these controls would fly in that game. I can accept everything else but the years long deadzone issue is unbearable. I _am_ hopeful that at least DOOM + DOOM II will be fixed, since there's not even aim assist to mask the problems. I've seen more people express concern about it this time round. I have, as always, shared your video on doomworld.
@@purebaldness I imagine support is done for the Quake titles, but I can dream. I'm dreading a Q3 remaster for that reason. It would have to be among the best implementations to play ideally. If not, it'll just be the example for the next decade to show how Quake can't work on console like Arena Arcade has been.
@@EternalDahaka and that's the ultimate problem: the misconception that thumbsticks cannot work for games like Q3 or CS. As much as I don't like Quake Champions' Champion mechanics, I'd have loved to have seen a console port (though maybe not in hindsight).
This is one of those cases where I ask myself why they put so much effort in something so bad. Good aiming isn’t rocket sience. Hot/Vert sens. Circular adjustable center deadzones. Optional axial deadzones. Adjustable sensitivity curve steepness from linear to exponential. Optional acceleration spike. A setting that lets you adjust aim acceleration in milliseconds, 0-1000. That’s everything you need to match 99% of „aim styles“ out there. Thanks for your analyzing this for years now, it’s just frustrating to see there isn’t really a step forward to best practices for better analog aiming. Everyone does its own thing.
Could be negated somewhat if console manufacturers allowed its consumers to adjust it natively, like Steam, a fricking PC storefront primarily focused on MOUSE AND KEYBOARD gaming, does. It is actually ridiculous. And don´t even mention the "settings" locked behind pro controllers like the Elite controllers by MS, they are pathetic.
Was going to ask the same question. Why is it this complex? There must be better solutions which are much simpler. I'd like to say Destiny 2's is solid, but maybe there is a better 'ideal' to follow?
Maybe do what the N64 did. Have a very precise hall effect stick with octogonal gates, so you only really need a small radial deadzone and no angle snapping or vertical or horizontal deadzones.
@@elhomoflow I have no idea how or why the movement deadzone is set up the way it is. Despite the unscaled angles and deadzone/curve issue, the camera implementation could have been mirrored on the movement. But those are really it. Nightdive already offered great options in the Quake titles. If they just tweak the implementation they're there. The Quake source ports, Quakespasm and Yamagi, already have these options(radial deadzones, x/y sensitivity, curve slider) and good implementations. Yamagi also has an angular restriction option, and the restriction itself is properly scaled.
@@elhomoflow "use mouse and keyboard" "... Limitations of a gamepad" To paraphrase one Nightdive developer. Such amazing console ports that are held back by this one glaring, yet largely unnoticed issue.
@@daneelaart Not ideal, but unfortunately those are pretty common. Call of Duty and Battlefield have also adopted smoothing in hipfire and none in ADS, and both have different sens ratios between them. Hopefully they can be addressed at some point, but it's still worlds better than the older Counter Strike ports as a more direct comparison.
This is a great visual representation (and your other videos) of deadzones and controller improvements. Will be using to inform a pitch i have at my AAA game studio to improve/implement deadzones. You're awesome.
@@NightWanderer31415 Yes. The antideadzone script I used here can alter the curve using following a power function(x^a, where 'a' controls the intensity of the curve) using the degree value. In the script, degree values under 1(but above 0) will slow down the curve, and values over 1 will speed it up. These are the options this script uses and you can see how they degree option changes the curve.: www.desmos.com/calculator/hlbt068sjo
I received a notification asking how the I set up the circular deadzone and if it could be used in the N. Sane Trilogy. The user either removed the comment, or RU-vid vanished it. The circular deadzone was set up by using a TitanTwo device(~$140). It's pricey, but I set up a script(Deadzone, Antideadzone and Curve Manipulation) to set up a radial deadzone, and counter the in-game deadzone fully for the circular deadzone example. The N. Sane Trilogy's deadzone is a very strange clover-shaped deadzone, and my script can generally address its deadzone by using the 'diagonalScale Min/Max' values to counter the clover shape, and the antiAngular options to counter the restricted diagonal movement. N. Sane Trilogy changes the deadzones per level(wider levels, more 2D oriented and bonus levels), so it won't counter all of them unless each version is set up as a slot in the device to flip between.
I'm one of the guys that message over reddit, I didn't sea you posted this untill today. I want to thank you deeply :) Came for the dead zones and learned (my English isn't that good) a lot more than I expected 😅 I knew it, I knew it! So increasing your sensitivity also affect your ads modifier 😮?! I don't understand the point of radial being .01 higher than axial, do they have a reason to do this? Thanks man, really I appreciate your work.
It’s shark week! Thanks for doing this, even though it’s like tilting against windmills. Your videos already made change possible in some games because they always explain the issue in an easy way.
Thanks for helping me understand why this game felt all weird, and helping me understand what deadzones and axials are in general. This was a huge help that I'll keep in mind for all games I play on controller
This script and its simpler versions can be found on the online resources tab in the GtunerIV software for the TitanTwo. This version won't work on the Cronus devices. This uses decimal values which aren't supported on those and the TitanOne. I have made a simpler version of this script for the TitanOne, which I've heard worked on the CronusMAX. You can download it here to see if it works on the ZEN: Script: www.consoletuner.com/gpclib/?s=1671 Visual Tool: www.desmos.com/calculator/pksqyyy5qq It functions almost identically to the T2 version, but it only supporting integers can cause some loss to precision(numbers will round down), mainly if you want to use small deadzones and larger antideadzones, and slower curves with the midpoint values. It's not dramatic, but you might notice some jitter in those cases.
@@lucacappelletti2084 I've updated the TitanOne script(same above link) with the angular version of the restriction and anti angular restriction options. The axial/cross-like restriction is possible, but both the angular restriction and anti restriction could be implemented at the same time. You might try to see if that restriction works for you. I can add the axial if you have a specific need for low-end values to be more restricted, but otherwise it's similar enough to the angular version to be redundant for larger tilts.
About your last points, I also suggest games to add a Orientation setting for joysticks: this would be a setting that allows you to change the orientation of the X/Y axis so you can tilt it from 0 to +45/-45 degrees. From My experience this would make a better sense of the direction our thumbs are place in the joystick. For example: I was good at a Racing game, I found a "way" to grab the controller so I could drive and react with precision with the Left joystick, however the way the controller is designed (Xbox One controller) made me move my thumb in a way that if I wanted to move my thumb completely horizontal from its anatomic point of view, then it didn't match the horizontal axis of the joystick, so I had to constantly re-orientate my thumb for that which is very annoying. This can also apply for the Right joystick under certain escenarios as well, so yeah, and Orientation setting would help a lot.
OMG you just mentioned every single problem I noticed on Halo joystick settings with such clarity, you also mentioned the aim assist option to disable it, I completely agree with that! If the game allows you to fine control your joystick sensitivities, accelerations and speed curves then you don't need aim assist anymore (in fact aim assist turns out to affect and mess up with your accuracy once you perfected your joystick settings). Thank you for the video, in the past I wanted to do something like this to show every single problem that Infinite (and other Halo titles) have, but you just made it and its well done and explained.
Hi where can I get the software? I'm using a Flydigi Apex 4 but I can't be precise in the micro movements because they are too sensitive and I would like to calibrate it
The only controller software used in this video is a script for the TitanTwo to clamp the maximum range for 0% circularity error. That's a pricey device($140) if there are other options. If you're playing on PC, you might try running games with Steam's Input customization and edit an output curve that can help with the micromovements.
@EternalDahaka How did you manage to modify the curve from 2:50 ? Does it replace the default Destiny 2 one or (like DS4 Windows does) does it superimposed on destiny 2's aiming curve ?
That is just an overlay for the video. I used a function close to Destiny's shape to demonstrate how it would change. I don't use it here, but my antideadzone script for the TitanTwo(and any similar device) functions like DS4Windows and applies the curve on top of the game's curve.
WTF I thought I was the first person to do something like this. You beat me by 7 years! Is there some way I can contact you about an issue I am having with Steam Input? I'm a long time PC gamer and Quake enjoyer who recently embarked on a joystick journey after I found out I can have universal joystick input in every game using the Joystick Mouse option in Steam Input. But there is something very strange happening with the response curve affecting the diagonals. A really wide (exponential) response curve slows down the diagonals considerably (like 5x slower than cardinals). A Square-shaped deadzone fixes the issue because it speeds up the diagonals massively, but it comes with an axial deadzone built in. Reducing the deadzone so much that I get drift doesn't get rid of the axial deadzone, only the center one. I want to hassle Valve about it, but I don't understand enough about joysticks to even begin. Can the axial deadzone be removed from the Square-shaped deadzone, or is that just how square ones are, with an axial built in? And why do the diagonals get lost with a wide response curve? I compared Valve's Joystick Mouse to Titanfall 2's in-game joystick and noticed that even when I match them perfectly (same sens, vertical, response curve) Valve's response curve ramps up way later than Titanfall's. Valve's starts ramping up at like 70% joystick deflection, while Titanfall's is slow for less of the curve and starts ramping up at like 50% (very rough estimates, I don't know how to actually measure this). Is this why the diagonals don't get lost in Titanfall, but do when using Steam Input's Joystick Mouse? It's hard to communicate this in a comment. Hope you are open to a chat or something because I can't find any info online about how all this stuff works. You look like the only person that knows everything about this stuff. Like literally know more about joysticks than the developers making games for them.
Oh and btw when I said square-shaped deadzone, I didn't mean the inner deadzone is shaped like a square instead of a circle. It's the one that Valve describes as "mapping the output from a circle into a square, making output in the diagonals approach their maximum value more quickly."
hi brother, sorry for mi english mistakes, I speak Spanish. how I can take one copy and test this scripts in my Pc games ? I'm wanna test deadzone , antideadzone in Fc24 for example, thanks
If you have a TitanOne, you can just search this script in the Online Resources tab in Gtuner. You can also download the script from here: www.consoletuner.com/gpclib/?s=1671 Otherwise you can always use gamepad-tester.com to gauge the deadzones, and Steam's Input customization to edit the antideadzone.
Sorry for the late reply. I'm finishing up the tests, but a video will be a while. The acceleration changes with sensitivity and I haven't figured out how it changes. I'll try to send the graphs when I'm done. As a summary, the deadzone options are accurate, and circular. The curve options are exactly identical to Call of Duty: Warzone's. CoD curves: i.imgur.com/yNNDITZ.png
Hey bro, what program did you use to have less % on the circularity ? Im having a problem with my new xbox series x controller. When I want to move the stick in an absolute straight line along the x or y axis, it tilts slightly up or down, which is a problem when playing Rocket League because I can't make perfectly linear side turns. It's really sad because in my country it was very hard for me to afford this original joystick, and I bought it to be able to play Rocket League. Is there a program that can help force the axes to be more linear?
I used a script for the TitanTwo adapter. I just clamped the stick values past 100% to get the circularity to 0%. There is no benefit in doing that, since the "real" error is still there, just hidden from gamepad-tester. Higher circularity was supposedly better in Rocket League anyway, at least before they added sensitivity options. I haven't tested Rocket League, but another user made a great tool showing the deadzone and dodge thresholds for it. The deadzone there is axial, so if you raise the deadzone size it will increase the ranges that you move perfectly left/right/up/down. It works with controllers so increase the deadzone enough so you can make perfectly horizontal movements easily. halfwaydead.gitlab.io/rl-deadzone/
I just tried something on ps5 version... please test this! Flip the sticks with the "south paw" controller setting in the game. Then go to the ps5 accessibility menu and flip the sticks again. Go back and test the game, are the dead zone values now switched? I feel like the aiming is better now?? Unless its in my head. Also seems to now make John not have to hold a or x to run now?? what did i just do?? it feels better to play...TEST!
Great video thabk you. So what do you thibk is better to run. Only radial or a combonation of both with axial being 1 less then radial. Do they add up together?
I'm not a fan of angular restriction, so I never use axial deadzones when possible. It's subjective though, and plenty much better players than me like using some level of axial deadzones. The deadzones roughly add up together, but this will become less accurate as the axial deadzone gets larger. The equation for the accurate total size would be: axial + radial(1 - axial) = total.
my 3rd party controller has 5% deadzone on both sticks, i could lower it to zero. Should I keep the 5% , and zero everything in game , or zero the controller, and increase in game based on drift ? Im having trouble finding a sweetspot , everything feels off in some way tho i have tried for 2 months now.
In-game options should be more convenient, but there shouldn't actually be a difference between the game options and controller options if the controller deadzone is implemented properly.
What on earth I thought my controller was broken till I saw this setting. It makes it 1000x harder to aim with this dead zone nonsense. I can’t possibly see an argument for having that be a thing.
I recently picked up the PS5 version and have been struggle to tune the camera speed to counter the extreme sluggish aiming. I was so frustative that I almost threw my ps5 controller out of window. Thanks for publishing this finding. I played RE2 and RE3 remake on PC with a Xbox controller with no such issues.
can you do a test on forza horizon 5? in my tests with emulated dualshock into x360 gamepad, I get an innate %25 deadzone on steering axis. the game even has a setting where you can reduce steering axis to %0 but it doesn't seem to apply for emulated x360 pad. for some weird reason though, values above %25 work (setting it to %30 gives you a %30 deadzone) would appreciate a lot if you can test that game out with your real x360 gamepad
Sorry, I don't own FH5. Are you using DS4Windows for the 360 emulation? If you haven't done it already, try looking at the options and setting the deadzone and antideadzone options to 0.00 if they're not set to that. DS4W seems to set both to 0.20 by default, which would apply a deadzone, but keep values after 20% accurate which would match what you're experiencing. Otherwise you might try using Steam Input's emulation to see if its emulation is different. Sorry I can't help more. Good luck.
@@EternalDahaka yes I use ds4windows and i have a dedicated profile with %0 dz ant %0 anti dz which I try to use as much as possible steam input is out of question as this is the game pass version of the game. if u ever sub to that i hope u remember to test this one out, else, no problem thanks for even considering and replying back !
@@cyberjunk7258 Have you tried using the antideadzone to address it? If the game forces a 25% steering deadzone and the slider is accurate above that, you can set the in-game deadzone to 25+% and DS4W's antideadzone to the same to counter it and use DSW4's deadzone for the one you want. If you want to use tiny deadzones that might mess with menu navigation depending on the menu deadzones, but in-game that should counter the steering deadzone accurately.
@@EternalDahaka yes, I already used that too. what I want is to get to the bottom of it and learn why it happens. it is really interesting that the slider literally doesnt work. it is why I wondered if it is a thing on other brands of gamepads that use xbox inputs because my friend have reported that such a thing is not happening on this 9th gen xbox gamepad
Rage 2 was the worst launch to my knowledge, i think ID software did a mumber on it at the time 🤫🫢🤭. No deadzone calibration for 4 months solid and it wasnt paired to the vehicle camera when they implented it. Plus the response curve was too aggressive from the central spring seat resting position. Thats like the inversion of aim acceleration the absolute opposite of what you would want 🫣. Deadspot 🎯 calibration & linear response curve options please you c***ing devs 🫠🤤👍🏼. (Industrial espionage from rage 🤷🏼♂️😬🤐)
