Accuracy Comparison of Motion Controls between Joy Con and Pro Controllers

[Squirty]

RESULTS EDIT:

Hey everyone! Sorry I'm late giving final results, but I wanted to give a quick update:
The simple summary is: I'm having a very hard time finding any substantial differences between Pro Controller, separate Joy Cons, or Joy Cons in Grip. I've done the test and have about 45 min of video footage I'm trying to analyze. It seems that the only notable difference is whether you have a good calibration of motion controls on your controller! I will eventually get the full analysis completed with video data, but at this point I think it's pretty safe to say that the differences in accuracy are very subtle, if there are any at all.

Here's a video comparing Splatoon 1 and Splatoon 2 motion controls:

Summary of differences between Switch and Wii U:

• Sensitivity of +5 in Splatoon 2 is equal to -0.5 in Splatoon 1 (Unchanged from Test Fire)
• Latency of Switch Pro with Dock in Splatoon 2 is ~3 frames (50ms) slower than Wii U Splatoon (Unchanged from Test Fire)
• Slow movement differences (Unchanged from Test Fire):
  • Gyro drift correction is more aggressive for Joy Cons and Wii U than for the Pro controller. (This means Joy Cons and Wii U are worse at picking up very slow movements than Switch Pro.) Update: I had this happen where Pro Controller was worse than Joy Con, but it was fixed by re-calibrating. It seems that calibration is the only thing that matters.
  • Splatoon 2 controls do not drift as much/long as Splatoon 1
    • This is hard to explain so bear with me. Video will help eventually.
      
    • When you rotate the controller very slowly, the game doesn't pick up this rotation and instead calibrates itself to think that this rotation is actually the controller's rest state. When you stop rotating the controller very slowly, the game compensates to think that you suddenly are rotating, causing a rotation in the opposite direction in the game. The duration and amount of this opposite rotation after you stop rotating the controller is shorter and smaller than in Splatoon 1.

Test Fire video:
ORIGINAL POST:

Hi there!

On day 1 of Splatoon 2 I want to know which controller configuration is the best for motion control accuracy. It is possible that there is only one inertial measurement unit (IMU) within the Switch Pro controller (likely) and it is effectively known that there is one IMU per Joy Con controller. With two IMUs in the dual Joy Con + grip configuration, it is possible that sensor fusion could allow the Joy Con controllers to be more accurate than the Pro controller.

To determine if there is a difference, my current plan is to simply strap both controllers together to form a single rigid body and see if one takes longer to drift than the other or has any other inconsistencies. This approach may introduce bias because one controller is in front of the rotational center and the other is behind the rotation center -- This bias may be negated by simply redoing the test with the controller positions swapped and seeing if it affects the results.

Some background on me: I'm the one who did the latency comparison for Smash 4: https://smashboards.com/threads/fin...ency-comparison-gc-vs-gamepad-pro-etc.372697/ -- One thing I learned from that test was that, while the test was very precise, it took me a year to complete. I want to have the results for the Splatoon 2 test within the first few days, so I am looking to make a much less complex test that still gives fairly accurate results.

Let me know if you have thoughts on this. I'd love to ideas of how to ensure that the test is precise and can be repeated by different people with the same results.

Testfire Update
During the testfire I performed some initial tests:
(direct link)

See this reply for a quick summary.

First Splatfest Update
See this reply.

DAY 1 TEST PLAN
Here's what I plan to test on day 1 using a 60fps video capture of three Switch. One Switch will use Pro controller, one will use Joy Con + Grip, and one will use a separated Joy Con:

• Slow movement responsiveness and behaviour (slow movement followed by sudden stop, performed in both directions)
• Fast movement out with medium speed movement back to center (performed in both directions to a few different target rotations)
• Flick movements (fast movement with rapid change in rotational direction, performed in both directions to a few different target rotations)
• Swap the Switch that each controller is synced to (to verify results are only affected by controller type) and redo test
  
• Remount controllers at 30' angle and redo test

I am also considering performing a comparison of these behaviours between the Wii U and the Switch Pro controller.

- Squirty

 

[NotAPerso]

I have a feeling that when joycons are paired as one they will only use one of the gyros instead of both.
Probably an easy test if you just play a game with them both detached to see which gyro it will use. If it only uses one then it's equal to the pro controller.

 

[Squirty]

Yeah, I'm expecting that as well -- which I will also expect means that they will need to software correct for the IMU being off-center, while they might not need to for the pro controller...

 

[Inyo]

I like the term "sensor fusion" and will use it frequently from now on.

In terms of controllers, all reports seem to indicate that the Pro Controller is better on ergonomic grounds. Obviously nobody was able to test latency but it'll be interesting to see the results.

The swapped position of the right stick is going to take getting used to.

 

[Squirty]

The greatest part about the term "sensor fusion" is that it's actually the correct technical term haha! :D https://en.wikipedia.org/wiki/Sensor_fusion

 

[khord22]

I like the term "sensor fusion" and will use it frequently from now on.

In terms of controllers, all reports seem to indicate that the Pro Controller is better on ergonomic grounds. Obviously nobody was able to test latency but it'll be interesting to see the results.

The swapped position of the right stick is going to take getting used to.

Yeah I feel that will be the hardest thing for me to get used to on the switch, hopefully I get used to it quickly. I also feel you might want to check gry when attached to the console in portable mode and also possibly joy cons w/o the grip?

 

[Inyo]

Yeah I feel that will be the hardest thing for me to get used to on the switch, hopefully I get used to it quickly. I also feel you might want to check gry when attached to the console in portable mode and also possibly joy cons w/o the grip?

Joy-Con without the grip would be interesting to test. They both have gyro in them, so would you have to move both in the same way or would it only care about one?

 

[Lyn]

I think the pro controller will likely end up being the most viable one, but it will be interesting to see how all of the schemes play out.

 

[Squirty]

Hmmm... I'm thinking it will be too tricky to judge joycon attached to switch to be worthwhile to test, to be honest. Testing joycon without grip sounds like it shouldn't be any harder than with grip.

Right now my thought is to get a lazy-susan type thing and mount a board on it, so it comes out perpendicular. Then strap pro controller and joycon+grip to this board. This way I can spin both controllers the exact same and judge their drift and responsiveness.

This strategy depends on some things that are currently unknown. Specifically, I don't know if there will be any split screen local multiplayer on the switch. If there isn't, I might need two switches in a local 1v1 turf lobby to do the test instead.

 

[Squirty]

I am doing a comparison of Splatoon 1 vs. Splatoon 2 motion controls during the test fire right now:

www.twitch.tv/squirty33/

During this stream I will be comparing things like how easy it is to make controllers drift for Wii U vs Switch!

 

[RespawningJesus]

Hmmm... I'm thinking it will be too tricky to judge joycon attached to switch to be worthwhile to test, to be honest. Testing joycon without grip sounds like it shouldn't be any harder than with grip.

Right now my thought is to get a lazy-susan type thing and mount a board on it, so it comes out perpendicular. Then strap pro controller and joycon+grip to this board. This way I can spin both controllers the exact same and judge their drift and responsiveness.

This strategy depends on some things that are currently unknown. Specifically, I don't know if there will be any split screen local multiplayer on the switch. If there isn't, I might need two switches in a local 1v1 turf lobby to do the test instead.

There is not going to be split screen multiplayer for Splatoon 2. If it can handle it, then I will be impressed, but I am 99% positive it won't feature it.

 

[Squirty]

I am doing a comparison of Splatoon 1 vs. Splatoon 2 motion controls during the test fire right now:

www.twitch.tv/squirty33/

During this stream I will be comparing things like how easy it is to make controllers drift for Wii U vs Switch!

Hmm, yeah, maybe that will be the case. I assumed it would be split screen, but maybe I'll just need to use 2 Switches for that test. :/

 

[Spraylan]

Had a little trouble with gyro aiming on the pro controller, didn't try the joycons. I used gyro controls exclusively in Splatoon 1, and it seemed to me like the shape of gamepad gave a better sense how to use them. I really wish I had a way to test both controllers in a single game.

 

[dimefox]

Joy-Con without the grip would be interesting to test. They both have gyro in them, so would you have to move both in the same way or would it only care about one?

As it works in Zelda, it only uses one joy con gyro when they are disconnected and I assume that it only uses one also when using grips or attached to the console.

 

[RespawningJesus]

As it works in Zelda, it only uses one joy con gyro when they are disconnected and I assume that it only uses one also when using grips or attached to the console.

Can confirm, only the right joycon works for the gyro.

 

[knubie]

Great work on those comparison tests squirty! The pro controller looks pretty solid. Do you have a summary of your findings?

 

[dogjoke]

For anyone reading who is unsure of how the test was conducted, Squirty built a multi-tiered lazy susan and secured the Wii U Gamepad, the Switch Pro Controller, and the Joycon Grip to each tier. This configuration allowed him to move controllers for each console with the same degree of rotation and at the same speed and velocity while playing the Splatoon 2 test fire and Splatoon 1 live simultaneously. He changed the sensitivity of each game between to see how the settings correlated.

I'm sure he'll want to be more thorough in reviewing footage before publishing his results (he did spend a year testing Smash, after all), but some of his preliminary findings were as follows (please bear in mind that I'm summarizing in my own words what he shared on our team discord server and am taking some liberties with inferences based on what he said):
-There is nothing inherently wrong with how the Switch handles Gyro. There is no evidence to suggest that the Splatoon 2 demo attempts to take any shortcuts or overcompensate in any way that is more aggressive than Splatoon 1.
-Sensitivity scaling (or perhaps just the default point of reference) is different between the two games. Anecdotally, the closest correlation he could find immediately following the test was between +5.0 sensitivity on the Switch and -0.5 on the Wii U. Take this information with a grain of a salt, because there are probably some factors that can't be accounted for, like the placement of the accelerometers in each controller and how their distances from the point of rotation might affect results or how off-center placement of the accelerometers (which is almost certainly the case in the Joycon Grip) might similarly affect results. The takeaway therefore should not be that -0.5 = 5.0 from Wii U to Switch, but just that there is a noticeable difference.
-Adjustments will need to be made when learning to play on the Switch. We're not going to get a 1:1 ratio that allows us to transfer our muscle memory informed by hundreds to thousands of hours of gameplay seamlessly from one game to the other. We basically already knew this given the change in size, weight, and shape between controllers, but testing suggests that there will be other factors that add to the learning curve.
-The Switch will handle subtleties in motion control better than the Wii U. Squirty described the Switch's response as "less confused" than the Wii U's when doing things like slowly rotating and coming to a stop.
-Drift correction is least aggressive on the Switch Pro Controller among the three configurations tested. This is a good thing for any Switch owners who might have been worried they'd have to bench their $70 USD investment.

tl;dr version: Switch has a lot of upside, but a lot of people will experience growing pains while adjusting to Splatoon 2.

I look forward to Squirty politely letting me know I'm an idiot and that I've completely misrepresented what he said. :P

 

[Squirty]

Hey everyone!!

Please see this test that was done with the test fire:

(direct link)

Summary of some key things are:

• Sensitivity of +5 in Splatoon 2 is equal to -0.5 in Splatoon 1
• Gyro drift correction is more aggressive for Joy Cons and Wii U than for the Pro controller. (This means Joy Cons and Wii U are worse at picking up very slow movements than Switch Pro.)
• Splatoon 2 controls do not drift as much/long as Splatoon 1
  • This is hard to explain so bear with me. Video will help eventually.
    
  • When you rotate the controller very slowly, the game doesn't pick up this rotation and instead calibrates itself to think that this rotation is actually the controller's rest state. When you stop rotating the controller very slowly, the game compensates to think that you suddenly are rotating, causing a rotation in the opposite direction in the game. The duration and amount of this opposite rotation after you stop rotating the controller is shorter and smaller than in Splatoon 1.

Sorry, I've had a busy few days after the test fire and was slow in getting my results posted. I will revisit with more details between Pro and Joy Con when Splatoon 2 is released. Hopefully I will be back before then with a better summary video of these initial results as well.

 

[knubie]

Excellent analysis as always Squirty.

Have you noticed any difference in latency between the two versions? I recently created a thread expressing my concerns about the added latency in the switch's gyro controls, and it looks like some other high level players are chiming in as well with the same sentiment.

https://squidboards.com/threads/switch-gyro-has-considerably-more-latency-than-wii-u.29011/

 

[Squirty]

Excellent analysis as always Squirty.

Have you noticed any difference in latency between the two versions? I recently created a thread expressing my concerns about the added latency in the switch's gyro controls, and it looks like some other high level players are chiming in as well with the same sentiment.

https://squidboards.com/threads/switch-gyro-has-considerably-more-latency-than-wii-u.29011/

I did not notice any difference in latency during my test, but I wasn't testing for it (so it might be there). I can totally test for this once Splatoon 2 has come out using my setup. I would use a low-input lag PC monitor (Dell professional) for the Wii U and the built-in Switch display recorded at 60fps and both controllers having motion input from my setup I used for the initial comparison test. I will plan to do this when Splatoon 2 comes out.