Wii sensor bar - Grand Slam Tennis

GreatWaves · 14 Mar 2010 at 00:18

Over the past week I've been playing GST for the first time. I'm wondering what people have thought is better in terms of the position of the sensor bar. I have it BELOW the set currently. Seems like it's a better position for ground strokes. I've stopped using the motion plus though. Can't get used to the M+. Seems like it makes it more erratic, but maybe I can use it when I get better. I don't use the chuck.

I've gotten a lot better and enjoying it, good exercise, but will take a while before I start fairing better against the experienced players online.

Where do you keep the sensor bar for GST? Above or below?

cheers

Kreeeee · 14 Mar 2010 at 00:42

I thought tennis games didn't use the sensor bar?

GreatWaves · 14 Mar 2010 at 00:56

As far as I can tell it does. It can't detect the full motion, but it detects which side you're hitting on and helps determine speed of strokes. Does it not?

Kreeeee · 14 Mar 2010 at 05:50

The sensor bar is literally just a pair of LED's (well, 2 clusters) and it doesn't do anything clever.

It's only ever used when the Wiimote is pointing at the screen and producing a cursor.

GreatWaves · 14 Mar 2010 at 08:12

Kreeeee said:
The sensor bar is literally just a pair of LED's (well, 2 clusters) and it doesn't do anything clever.

It's only ever used when the Wiimote is pointing at the screen and producing a cursor.

Right. Moving the cursor around on the screen (and how it does it) is not something super clever, but you can easily (and creatively I suppose) do more with that than just moving a pointer on a menu. Would seem a bit odd to have this prominent sensor bar but only use it for literal pointing-on-the-screen tasks.

Demon · 14 Mar 2010 at 09:13

GreatWaves said:
Right. Moving the cursor around on the screen (and how it does it) is not something super clever, but you can easily (and creatively I suppose) do more with that than just moving a pointer on a menu. Would seem a bit odd to have this prominent sensor bar but only use it for literal pointing-on-the-screen tasks.

I agree with Kreeee, the sensor bar is just for the cursor/pointing, it's such a simple system. that's all it's really capable of, you do get an angle of rotation in there, but I dont think there is a much you can do with it reliably.

However, turn it around, wii mote by the tv, and sensor bar on your head, and voila,

[/QUOTE]

GreatWaves · 14 Mar 2010 at 09:34

Demon said:
I agree with Kreeee, the sensor bar is just for the cursor/pointing, it's such a simple system. that's all it's really capable of, you do get an angle of rotation in there, but I dont think there is a much you can do with it reliably.

I'm not saying it has to be magical, complex or clever tech to accomplish what I suggest. Within a certain "cone" in front of you, "it can see" the remote. So it can see part of the "pull back the racket" motion. It can see part of the "hit the ball" motion. Those are absolute points of reference that compliment the relative input of the accelerometer.

Anyway, so far I've noticed differences in trying different sensor positions when experimenting on the practice court with predictable alternating flat shots (to take out variability from "how the opponent hits it") while trying with large samples of hitting it back consistently myself

fini · 14 Mar 2010 at 09:52

An easy way to tell if it's using the sensor bar at all during the match is to get someone to cover it up or turn it to a weird angle and see what happens.

GreatWaves · 14 Mar 2010 at 09:59

fini said:
An easy way to tell if it's using the sensor bar at all during the match is to get someone to cover it up or turn it to a weird angle and see what happens.

My thoughts exactly, but I'm not home right now. Apart from the different positions I've tried I have not tried something "that should not work". It just seemed like common sense to me that the game would utilize the motion data within that "cone" in front of you. It's absolute coordinate data (with some under-sensitive or over-sensitive variability)... why would it not be used.

But I'll try later.

Demon · 14 Mar 2010 at 10:06

GreatWaves said:
My thoughts exactly, but I'm not home right now. Apart from the different positions I've tried I have not tried something "that should not work". It just seemed like common sense to me that the game would utilize the motion data within that "cone" in front of you. It's absolute coordinate data (with some under-sensitive or over-sensitive variability)... why would it not be used.

But I'll try later.

Just had a thought, it could be that when the sensor bar is detected (when the wiimote is pointed at it), it could allow re-calibration of where the wii-mote is orientation wise..

This would be plausible, because the wiimote uses accelerometers or gyro's to determine position etc, and these do 'drift' over time, so having a periodic absolute reference every now and again may be helpful.

If that's true, that I would suspect you are correct in that having the sensor bar top or bottom would make a difference as it would alter the reference point slightly when calibration occurs.

Probably a load of rubbish, but I do know Sony's Move uses a similar dynamic calibration to keep accuracy.

GreatWaves · 14 Mar 2010 at 10:20

Right and it has more than one point to use as reference within the "cone of motion". It also senses larger extents of motions better than small slow motions (like when trying to point to a menu selection... can be jittery or under responsive).

EDIT: I just got around to watching that youtube video further up. I think it just demonstrates very well that the sensorbar/wiimote link is accurate enough to give very good motion reference points for complimenting the accelerometer. I haven't had a chance to do the test with GST yet, but it wouldn't make sense to me why they wouldn't use such accurate sensing to compliment the relative motion data of the accelerometer. I know that the guy is using PC software there, but obviously the hardware is capable.

GreatWaves · 14 Mar 2010 at 12:24

I haven't had a chance to play tennis yet, but got a chance to just do a pointer test at the main menu. 1)With the sensor bar at an awkward angle (or with one side covered), the accelerometer manages to compensate IF you move the pointer more slowly. If you twitch it back and forth even a little, it loses the pointer. 2)With the sensor bar in a proper position, it does not lose the pointer when you similarly twitch it rapidly.

So it "makes do" with accelerometer data when necessary without tracking or without good tracking from the sensor, but with more rapid movements, it gets lost without good sensor bar input. That's what I would expect.

But the question still remains, does Grand Slam Tennis "ignore the sensor bar" as is being said here. I don't have time now for GST, but will check that later.

Vonhelmet · 14 Mar 2010 at 12:32

GreatWaves said:
I haven't had a chance to play tennis yet, but got a chance to just do a pointer test at the main menu. 1)With the sensor bar at an awkward angle (or with one side covered), the accelerometer manages to compensate IF you move the pointer more slowly. If you twitch it back and forth even a little, it loses the pointer. 2)With the sensor bar in a proper position, it does not lose the pointer when you similarly twitch it rapidly.

So it "makes do" with accelerometer data when necessary without tracking or without good tracking from the sensor, but with more rapid movements, it gets lost without good sensor bar input. That's what I would expect.

But the question still remains, does Grand Slam Tennis "ignore the sensor bar" as is being said here. I don't have time now for GST, but will check that later.

It doesn't use the accelerometer for pointing. If you cover one led, it makes do with one.

GreatWaves · 14 Mar 2010 at 12:38

vonhelmet said:
It doesn't use the accelerometer for pointing. If you cover one led, it makes do with one.

The only thing that doesn't sound right is that then why would rapid movement as such make a difference? I can see it being less accurate with one side covered, but if it's not the accelerometer, then why does rapid movement make it get lost?

If I don't reply, it's just because I'm just off to something else atm.

GreatWaves · 14 Mar 2010 at 15:32

Well. I turned on GST. Flipped the sensor bar to face the other way. And I found that all I have to do is sit on the couch and flick the Wiimote where ever to hit forehands, backhands, etc. Arg.

I've been playing GST for less than a week, so I haven't yet gotten so attached to it, but now I wonder what the heck is the point of this game? And it makes me realize the "really good players" are doing just what I described: sitting on their back sides on the couch flicking the wiimote like a small fishing rod. That obviously makes it a lot easier to "anticipate forehand or backhand" when you don't have to do it physically.

I guess using the Motion+ forces you to actually do something (not sure altogether what), but I can't get the hang of the M+. I find that it causes the player to hit a backhand when I'm hitting a forehand, etc. Just erratic stuff like that.

Well, I pretty much lost interest in that activity entirely. I wanted GST for some aerobic exercise, not for flicking my wrist.

Demon · 14 Mar 2010 at 15:38

GreatWaves said:
Well. I turned on GST. Flipped the sensor bar to face the other way. And I found that all I have to do is sit on the couch and flick the Wiimote where ever to hit forehands, backhands, etc. Arg.

I've been playing GST for less than a week, so I haven't yet gotten so attached to it, but now I wonder what the heck is the point of this game? And it makes me realize the "really good players" are doing just what I described: sitting on their back sides on the couch flicking the wiimote like a small fishing rod. That obviously makes it a lot easier to "anticipate forehand or backhand" when you don't have to do it physically.

I guess using the Motion+ forces you to actually do something (not sure altogether what), but I can't get the hang of the M+. I find that it causes the player to hit a backhand when I'm hitting a forehand, etc. Just erratic stuff like that.

Well, I pretty much lost interest in that activity entirely. I wanted GST for some aerobic exercise, not for flicking my wrist.

I know what you mean, My 4 year old can beat the AI easily at boxing by sitting down and just waggling..

I think Sony's Move, and MS's Natal stand a chance of taking things further, Natal for sure will pretty much require full body movement in games that use 1:1 tracking of the body, sitting down won't be an option, but perhaps developers won't want to 'force' people to interact this way, after all most people tend to be a little lazy.

Streeteh · 14 Mar 2010 at 15:46

This is the problem with using accelerometers to detect movement, they detect acceleration, not distance travelled or position (though the motion+ does help detect orientation etc a little, but it's still poor imo).

Have a look at Sonys new product, move, which is the same sort of concept as the Wii-mote but uses a video camera to more accurately map your own movements.

GreatWaves · 14 Mar 2010 at 15:59

Thanks for the suggestions. I'll sit on the fence for now and read reviews and then decide when something is worth getting.

Feels a bit silly that inside a week I didn't start questioning what it's really doing. I was really feeling better from the engaging exercise, so didn't think about it.

Makes you wonder also what in the heck the engineers were thinking. They have a working system for at the very least triangulating the left to right position of the remote. Why not develop it to for use in the games??? They could have made two satellite led sources further apart. I would think that could work quite well within games especially when cross referenced with an accelerometer.

Anyway whatever...

JaFFa · 14 Mar 2010 at 16:10

Even though Kree has already pointed it out I will hop on as you really are reading way too much into the sensor bar. It is simply a light, it doesn't 'sense' anything. No data is sent back to the wii, the cable is merely for power. The remote does all the work through a set of accelerometers. Position of the bar will make no difference at all. It's just a reference for pointing so the IR cam in the front of the remote can produce a cursor. That's why its a relative pointing and not direct.

Kreeeee · 14 Mar 2010 at 16:15

Argh, if I wasn't on my iPod I'd post a big response about it. I made a light gun game for the wii using 2 sensor bars to allow the player to duck behind and lean around objects as a big project in my final year.

It really is that simple, it isn't even used to measure the vertical angle of the wiimote when pointing at the screen. I even used 2 candles instead of the sensor bar at one point.