Switch gyro has considerably more latency than Wii U

[knubie]

As I was playing Breath of the Wild on my Switch, I noticed right away the first time I got a bow that the motion input felt a bit laggy (similar to how Wii felt back in the day). I've put a lot of hours into Splatoon on my Wii U (which is very responsive) so it was pretty noticeable to me right away.

One of the first things that blew me away about the Wii U gamepad was when I booted up Nintendoland and noticed just how accurate and responsive the motion controls were in that game. I hadn't experienced anything like it, and I believe it's one of the reasons Splatoon felt so good.

I did some research into the Wii U gamepad's tech back then and found the key element it had that was missing in every other controller that features motion inputs was something called a "magnetometer" which offers so-called "nine-axis" controls. From this [article](http://www.nintendolife.com/news/2012/11/wii_u_gamepad_offers_nine_axis_controls):

> Becky Oh, chief executive of PNI Sensor Corporation, the company behind the GamePad’s geomagnetic sensor, has spoken in an interview with GamesBeat about how the technology improves the accuracy and reliability of the GamePad.


> "[The gyro and accelerometer] are good at tracking relativistic change. But it doesn’t tell you absolutely where you’re pointing and where the pointer is. What the magnetic sensor does is use the Earth’s magnetic field as a reference. It can always guide [the GamePad] back to what the absolute position is."


> Oh believes the new geomagnetic technology will allow the failings of six-axis devices — such as the Wii Remote — to be eradicated, saying she thinks several genres of games can now be played much more responsively.


> "I think games such as first-person shooters, driving games, or some type of flying game would be a good candidate for this type of technology. Sony’s Sharpshooter [Move peripheral] did something like this, but when we played with it we saw it was not accurately tracking. There was both latency and inaccuracy. In that case, hardcore gamers would go back to using a joystick or game controllers, but if you had a very accurate way — with no latency or very little latency — to use the gun to point what you’re shooting. I think that does change the way the game is played."

I did some research and I could not find any information regarding whether the switch has a geomagnetic sensor like in the Wii U gamepad, and after comparing bow aiming in BOTW with Splatoon and Nintendoland on Wii U, the difference is absolutely noticeable.

People who don't like gyro aiming won't care either way, but I'm afraid the exclusion of such a feature will completely ruin games like Splatoon. I could have lived with this if it was limited to the Joycons, but it doesn't even look like the pro control has this feature either. I'm hoping this is just some odd way the developers have implemented gyro aiming in BOTW and it will be more responsive in Splatoon, but I don't think that's the case.

Does anyone have a copy of BOTW for both Switch and Wii U that can compare?

 

[dylan_not_bob]

I do sometimes have issues aiming the bow in BotW, but usually have no problems with aiming the magnet or statis power. I can't remember what else I was doing the other day, but I remember thinking that the gyro felt pretty alright with whatever it was. For the record, I am using a pro controller as well.

It will be interesting to see how it feels during the demo next week!

 

[Maave]

I think this is just the devs intentionally nerfing the motion for people who don't normally use motion controls. The motion controlled puzzles are much more sensitive than the bow. It seems like the motion for the bow and camera and whatnot were "smoothed" to avoid jitter. I also noticed there is a minimum threshold for movement or else the camera won't move. If you try to move really slowly it doesn't work. So I think it's a nerf for people who don't use motion normally. None of the people playing the Splat 2 demo (like ThatSRB2Dude) noticed laggy controls, they said it feels just like Splat 1

 

[RespawningJesus]

I have had no problems with the bow in BotW. I aim with the right analog stick, and then make adjustments with the gyro, which is what I imagine is how it will work in Splatoon 2.

 

[Ikaheishi]

I played the Splatoon 2 demo version about a week and a half ago at a special event Nintendo put on and I didn't notice anything wrong with the gyro controls. It felt just like playing Splatoon itself on the Wii U… Actually a little smoother than Splatoon. The only thing I had a problem with motion-controls wise was that the sensitivity was set lower than I'm used to.

I'd also like to note that I have the Wii U version of Breath of the Wild and I've noticed (and have died a few times because of) some latency with motion controls in the game. The motion controls in BotW just aren't as smooth or precise as Splatoon, which is a bit disappointing.

 

[spoxbox]

tbh this thread scared me a bit with the first post but it seems i can rest easy

 

[Award]

The gyro and accelerometer are actually higher resolution than that in the gamepad, so it should actually be capable of GREATER accuracy and smaller movement. The bow is definitely set to very low resolution by design, however controlling it with separate joycons and using the right one alone for motion aiming is terrific. The motion puzzles (roll the ball into the hole) in shrines however show off how VERY finely accurate the motion controls can be. Now that I think about it, the aiming feels similar to Twilight Princess HD bow aiming on WiiU. I'm guessing Zelda always uses the same aiming code since Ocarina 3D.

 

[Cyan]

Not sure if BOTW is a fair game to compare to Splatoon, because I'm pretty sure BOTW runs at 900p30 when docked (and 720p30 undocked/Wii U). The fact that the framerate is half that of Splatoon (which definitely runs at 60FPS) would alone be enough to mess with the fluidity and precise aiming you get with motion. I played the Splatoon 2 Demo about 2 weeks ago in London, and the game felt just as good as Splatoon 1 for aiming, I would even say it felt more responsive and more sensitive than 1 ever did (I always use -1 in Splatoon though, never really messed with the sensitivity). There shouldn't be anything to worry about, and the devs wouldn't nerf motion controls to make stick more viable, because deliberately making poorer controls for a game would end in a huge outcry from fans and make the game bad when it shouldn't be. You do have the Testfire in 9 days' time, so you have 6 hours worth of sessions to mess with the game and compare to Splatoon 1 but I felt no real difference or degrade in quality from 1 with motion (except for the Joycon Grip being awkward with the shoulder buttons).

 

[knubie]

I think this is just the devs intentionally nerfing the motion for people who don't normally use motion controls. The motion controlled puzzles are much more sensitive than the bow. It seems like the motion for the bow and camera and whatnot were "smoothed" to avoid jitter. I also noticed there is a minimum threshold for movement or else the camera won't move. If you try to move really slowly it doesn't work. So I think it's a nerf for people who don't use motion normally. None of the people playing the Splat 2 demo (like ThatSRB2Dude) noticed laggy controls, they said it feels just like Splat 1

I actually find the motion puzzles even less responsive than the bow aim, which leads me to think that it must have been intentional. ThatSRB2Dude did actually mention that there was "a bit of input lag" but that he was still able to aim accurately. Whether or not the input lag he experienced is as noticeable as it is in BOTW remains to be seen. I hope it's not.

The gyro and accelerometer are actually higher resolution than that in the gamepad, so it should actually be capable of GREATER accuracy and smaller movement. The bow is definitely set to very low resolution by design, however controlling it with separate joycons and using the right one alone for motion aiming is terrific. The motion puzzles (roll the ball into the hole) in shrines however show off how VERY finely accurate the motion controls can be. Now that I think about it, the aiming feels similar to Twilight Princess HD bow aiming on WiiU. I'm guessing Zelda always uses the same aiming code since Ocarina 3D.

Do you have a source you can link to about the gyro/accelermeter being higher resolution? I'd be interested to look at the details on the chip that's acutally in use (couldn't find any details on ifixit). The main issue I've experienced, however, is not lack of granularity, but a noticeable ~100ms input lag.

Not sure if BOTW is a fair game to compare to Splatoon, because I'm pretty sure BOTW runs at 900p30 when docked (and 720p30 undocked/Wii U). The fact that the framerate is half that of Splatoon (which definitely runs at 60FPS) would alone be enough to mess with the fluidity and precise aiming you get with motion. I played the Splatoon 2 Demo about 2 weeks ago in London, and the game felt just as good as Splatoon 1 for aiming, I would even say it felt more responsive and more sensitive than 1 ever did (I always use -1 in Splatoon though, never really messed with the sensitivity). There shouldn't be anything to worry about, and the devs wouldn't nerf motion controls to make stick more viable, because deliberately making poorer controls for a game would end in a huge outcry from fans and make the game bad when it shouldn't be. You do have the Testfire in 9 days' time, so you have 6 hours worth of sessions to mess with the game and compare to Splatoon 1 but I felt no real difference or degrade in quality from 1 with motion (except for the Joycon Grip being awkward with the shoulder buttons).

The 30fps would only add, at most, an additional 16ms of latency which would be hardly noticeable I think. The latency that I've noticed is closer to 100ms or more.

And to be clear, I'm not saying I've experienced a lack of accuracy with the motion controls, but there is a noticeable 100-200ms latency. Which is a HUGE amount of input lag for a competitive game like Splatoon.

 

[knubie]

Looks like the joycons (and presumably the pro controller) use a STMicroelectronics SH641 according to this article at techinsights, which "has a full-scale acceleration range of ±2/±4/±8/±16 g and an angular rate range of ±125/±245/±500/±1000/±2000 dps."

Compare that to the Invensense MPU-6500 on the HTC Vive controllers and the accuracy looks identical (however the vive has the benefit of external sensors).

I still don't know what this means for the Switch in terms of latency of motion control input.

 

[blu]

I have heard that the latency might be due to wifi issues in the Switch, so I would check your wifi connectivity to see if its strong enough, or if its disconnecting during gameplay. If its a single player game try airplane mode to see if the gyro controls are smoother.

 

[knubie]

So Tomorrow Corp just released a few games for the Switch, and some of them use the gyro in the joycon as a cursor. Someone recorded a demo of the cursor in action here:

If you download the video and run it frame by frame you can see that there is about a 3 frame delay between when he moves the joycon, and when the cursor moves. Since the video is running at ~30 frames per second, that translates to a ~100ms delay, which is exactly what I predicted the latency to be in BOTW.

Now this latency may be attributed to "smoothing" where the software samples the position across 2-3 frames in order to make the cursor appear less jittery. Hopefully this is the case, and the smoothing is removed in Splatoon, otherwise it does not bode well for the game.

 

[Astral]

1ms monitor meta

 

[knubie]

So it seems I'm not the only person concerned about this issue. ThatSRB2Dude has recently released a video where he has expressed his dissatisfaction in the latency of the switch gyro controls.

My only hope is that this is some kind of a software issue, and not a limitation of the hardward.

 

[Squirty]

I performed a comparison test during the testfire, but I did not specifically test for latency differences between Wii U and Switch. I plan to do that when Splatoon 2 has been released using my setup:

https://squidboards.com/threads/acc...ols-between-joy-con-and-pro-controllers.28620

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.

 

[Inyo]

I didn't notice any delay using motion controls. Obviously a pro like Dude will sense it more than a middle-tier player like myself would, but I did not have any issues and the game felt the same to me, maybe a bit MORE sensitive since the Joycon in the grip don't weigh near as much as the Gamepad does

 

[knubie]

After doing some testing with the high speed camera on my iPhone 6s, I've come to the conclusion that input latency on the switch is *very high overall* at about 130ms. This is true whether hooked up to my lcd tv, or in handheld/tabletop mode. I ran several tests in BOTW, including jumping, closing the inventory, and tilting with the gyro. All tests averaged around 32-33 frames of input latency. And at 240fps that comes to around 130-140ms.

Doing the same experiments, connected to the same TV, using Splatoon 1 as the test subject, the Wii U had an average input latency of only 16-17 frames (70ms).

Now, granted, these are two different games, so it could just be that Zelda in particular has a very high input latency for whatever reason, but this would help explain why others have described Splatoon 2 as being less 'snappy' than the original.

I've uploaded a couple of comparison videos for those that want to take a look frame-by-frame as I did.

https://www.dropbox.com/sh/3sskoj14ihvyzhu/AAB-pO1hf95TsjEr1uuQIbDca?dl=0

 

[Maave]

Thanks for the detailed tests. I did notice it in the test fire once I tried the charger. It was very slight but enough to feel it when making snaps.

 

[Ulk]

In BotW the motion controls actually worked quite well for me. I'm pretty good at performing precise shots.Not any less than I was in Splatoon 1. Splatoon 2 did feel a little different I must admit, but I was hardly able to tell, since I use extremely low motion control settings (-4 to -2).

 

[Dessgeega]

Thought: Unless the pro controller is proven to be consistently better in this regard, I think it's kind of a moot point. Everyone has to use the same controllers, so the playing field will be level regardless of any latency or lag.