New Wiimote in action

http://www.youtube.com/watch?v=acND4sO3pJs

I'm not sure if the lag is caused by the TV upscaling 480p video to HD res, or if it's caused by the Wiimote itself taking its sweet time transferring the data, but that's a pretty serious issue. Also the 1:1 tracking isn't as 1:1 as they'd like you to believe, but if the one second lag weren't there, it'd be acceptable.

I really can't see them letting it 'lag' like that. Must be something going on outside of the remote that's introducing that lag.

Anyways, aside from portion where he was smacking the dummy around with the sword, couldn't all those other demos be done to the same extent using just the accelerometer and motion sensor?

I've been wanting a spiritual successor to 'Die by the Sword' since day one of learning the Wii remote has the potential to do what that one game allowed you to emulate with mouse movement. Should be even easier for someone to program a proper 1:1 sword system with the M+ attachment.

I see lag, not much, but enough for me to loose interest for game...its nice but not exactly 1:1 tracking...I guess further optimizations are required..

perhaps this lag is due to an out of sync bloke in the lower left corner and the main screen

his screen was too small for me to see any lag but im willing to take an out of sync dual feed into the equation

I didn't see any lag, the thing on the screen moved at the same moments he moved the remote... maybe they updated the movie.

I think this is cool and I really look forward to playing some RPGs with it + other games. I see a tiny lag in the video, it's not too bad, but I'd prefer none.

Foppy:
No, you're just blind. :-)

Who typed that! Who typed that!

Ok, maybe I see 0.1 second of lag. But that's as far as I'll go! ;)

I don't see any lag to worry about either. Sounds like people are just being overly picky. "serious issue"? I don't think so.

I didn't see any lag in the Wii Sports Resort sword fighting demo at E3, so yea, definitely not something to worry about.

Depends how big the screen is they're showing it on...bigger screen introduces more lag.

This makes no sense to me. Maybe more perceived lag due to more distance to travel on larger screen but not more ACTUAL lag.

Since some folks seem to be doubting thomases, I took the liberty of converting the video to avi so I could examine the frames of the video more closely and check the timing.

Here are some screenshots of one of the sword swings:


This swing took half a second. The middle shot is halfway through the swing. And as you can see, in that shot the sword on screen still points to the right, even though his arm is all the way over on the left.

That means there is 500ms of lag, on top of your own reaction time of around 250ms to deal with.

For comparison, the lag in an online first person shooter might be 50-100ms on average, and in a worst case 250ms. With 500ms of lag such a game becomes unplayable.

Now obviously he can throw a ball with the wiimote, because lag doesn't matter there. But if things were flying towards him and he were exepected to bat them away, unless they were moving very slowly, he'd have a hard time of it. I guarantee you the baseballs in Wii Sports 2 won't be rocketing past you at 60mph, and sword fighting as we saw in the Nintendo E3 presentation will be more like a slapfight.


I think in the video he mentions "early detection" of motions. I think what that means is that their toolset lets a developer look at the first tenth of a second of a motion and decide how the user's going to follow through based on that and then start playing the predicted animation early so as to reduce the percieved lag.

Unfortunately, even if that is the case, he's demonstrating their system here. One would assume that said prediction would be part of this demo if that's a part of their toolset and he's talking about it. And if that's the case then that means the actual lag is even worse than we're seeing here, and it only looks as good as it does because of the prediction, and even with that, it's still less than perfect.


I know everyone really wants the Wii controller to be fixed, but I'm afraid you're just going to set yourself up for dissapointment again if you buy into this 1:1 thing. I haven't seen a single demo yet which has convinced me there isn't significant lag in response time.

haha, I smell anal rententiveness. But thanks for the scientific confirmation. But surely if the middle shot is HALFWAY thought a half second (500ms) swing, the lag is 250Ms. (which is still bad. Human reaction time is around 180ms)

You must be the one exuding that odor, Mister 250ms. :-)

Anyway, you're right, but regardless, that's still a lot, it'll be quite noticeable when playing, and it will make it difficult to hit fast moving targets.

And it's not my fault people here are blind and made me resort to drastic measures to prove them wrong. :-)

Flat screens have a lag of 200MS alone. Did you read Mick West's last two articles on game responsiveness programming (Gamasutra). When you're swinging your whole arm and not just pressing a button the lag will not seem as great either. Anyway even games with exceptional response times have 0.3 - 0.4 seconds of lag and not 0.25 according to his article and medium range games had about 0.7 seconds lag if I remember correctly (at 30FPS). That demo looked like a 30FPS demo too which would increase the lag.

LOL, absolutely right. Although the very phrase "anal rententiveness" implies that no odours would in fact be released doesn't it? Thus the verb (smell) in my initial phrase was inaccurate, I should have said "spy" but then that implies that I look at peoples rears, or "sense/detect" but then how do I do that? With my fingers (yuk!) or some kind of 6th sense or perhaps a scanning device of some kind. OK I'll stop now.

Jeremy Alessi: Thanks for the article suggestion they are interesting. On my PC lag is virtually non-existent (I have a CRT) as my games have a 200HZ logic loop and immediately process the input and output it next VSync (and that occurs 60 times a second normally, so 16.67ms) although I don't know what delays DX, the video card and the monitor add in, but it's surely not much at all or it would be obvious.

Also you can run sound tests by playing a simple noise when the mouse button is clicked and then when you click it, you can listen out for a difference (or even record it with a microphone and compare the start point of the physical mouse click and the sound played as a result).

Having worked with the Wii Remote for over a year now and Live Move for around 6 months, there's always this stupid lag when it tries to "guess" what the user is trying to do, by comparing to pre-recorded data.
It sucks.
With the Motion Plus, they just diminished the lag a bit. Without it, it's just... painful. On top of this, the more move you want to try to recognize, the longer it takes. So, for his little perfect example, he only had 5 moves loaded, probably with 2-3 recordings of sampling data for each (done by him so he always does his motion exactly the same each time).
LiveMove are the only one right now who have access to the new Motion Plus. Even tho we are registered developers with over 100 Wii Dev/Test kits, we cannot have access to them. Nintendo is keeping radio silence to us.

Nintendo dev tools have always been really poor.

I hope the Wii dies.