150 ms is hardly intolerable lag. I've played so many online games I can't even begin to guess how many hours of my life I wasted, and if I'm pinging 150 or less I am happy. Of course less than 100 ms is ideal, but I can game with the best of them at 150 ms. And good luck finding a dial up connection (of any kind) that will give you a ping of anywhere under 200 ms...
Wiimote latency - my final question answered
- Thread starter Colmino
- Start date
I myself would love to see a video that accurately shows the lag between the wii action and the actual movement.Then the lag between the wiimote and the console is acceptable.
I doubt this is caused by lag in the wireless connection at all as you have said yourself, the other buttons work fine. Rather, the delay is likely there purposefully. In the case of the system menus or other pointing functions, it is likely there to reduce unnecessary jittering of the cursor. A mouse can be set still, but a pointer is still being held by a human hand and not perfectly still. In the case of gesture recognition like in Wii Sports, it needs to take the time to let you finish your gesture, or at least complete more of it.
- Thread Starter
- Thread starter
- #23
Greetings again. It's gratifying to see that most people are willing to discuss this matter in a progressive fashion.
The wireless mouse I was referring to was, as I said, a 3d variety. Not a standard mouse with no wire, but instead a mouse that you can move around in the air in front of you. I am not sure what technology is used, but it does not require the situation of any devices near the TV. It was not an expensive item at the time I purchased it, several years ago.
The comments about human reflex time are about as poignant as those suggesting that humans cannot discern more than a few hundred colors, or twenty frames per second. If the 150ms latency were to disappear overnight (or even halve), most people would notice the difference and be happy with it. Fact.
It is true that there do not appear to be any reports about the Wiimote's latency. That's the main reason why I've made this thread. A similar fiasco transpired with the X-Arcade multiplatform joystick, which exhibits similar (100 to 150ms) lag. I was among the first to question this phenomenon. It took a dozen emails to the manufacturers for them to admit that it was a problem. Now it is common knowledge for anyone who cares to investigate it. Eventually there will be enough "twitch" gamers who own a Wii for the Wiimote's conspicuous latency issues to be well-established. It cannot be regarded as advisable to suggest that the current lack of such reports means that it is not a problem, or that I invented it or am imagining it.
150 milliseconds sounds like very little, but in fact, one can, for example, swish the Wiimote back and forth about four times a second without trying, and this is fast enough motion for the onscreen hand to be out of sync so that it appears to be moving in the exact opposite direction of the Wiimote's movements. A spring effect. Not desirable.
I'll address this insightful response directly:
The wireless mouse I was referring to was, as I said, a 3d variety. Not a standard mouse with no wire, but instead a mouse that you can move around in the air in front of you. I am not sure what technology is used, but it does not require the situation of any devices near the TV. It was not an expensive item at the time I purchased it, several years ago.
The comments about human reflex time are about as poignant as those suggesting that humans cannot discern more than a few hundred colors, or twenty frames per second. If the 150ms latency were to disappear overnight (or even halve), most people would notice the difference and be happy with it. Fact.
It is true that there do not appear to be any reports about the Wiimote's latency. That's the main reason why I've made this thread. A similar fiasco transpired with the X-Arcade multiplatform joystick, which exhibits similar (100 to 150ms) lag. I was among the first to question this phenomenon. It took a dozen emails to the manufacturers for them to admit that it was a problem. Now it is common knowledge for anyone who cares to investigate it. Eventually there will be enough "twitch" gamers who own a Wii for the Wiimote's conspicuous latency issues to be well-established. It cannot be regarded as advisable to suggest that the current lack of such reports means that it is not a problem, or that I invented it or am imagining it.
150 milliseconds sounds like very little, but in fact, one can, for example, swish the Wiimote back and forth about four times a second without trying, and this is fast enough motion for the onscreen hand to be out of sync so that it appears to be moving in the exact opposite direction of the Wiimote's movements. A spring effect. Not desirable.
I'll address this insightful response directly:
This is a good guess, and I am reserving final judgment until I have either laid eyes on a game which can prove to me that the lag isn't inherently unavoidable, or I've seen about a half dozen games up close and personal. A telling fact, however, is that the jitteriness of the Wiimote pointer is quite dramatic even while the latency is everpresent. Try moving the pointer up/down, as this vector showcases the most dramatic inaccuracies. Consistently, I watch the pointer shuffle left/right as it travels up or down, at a rate which defies the likelihood of any sort of lag-engendering, jitter-softening filter.
The latency has nothing (read very little) to do with the communications between the wii-mote and the wii. It has everything to do with the speed at which the wii-mote can discern its location in relation to the sensor bar; this is why pointer based functions have the lag and the "classic" functions do not.
Personally, I have not done alot of testing with the IR independent wii-mote motion capture capabilities (as in, the stuff that makes all of the sports games work, not the menu cursor). If you don't know what I mean here (understandable, I'm not explaining it well) cover the red part of the front of the wii-mote, whatever still works is what I am talking about. If there is indeed an equal "lag" with these functions, its entirely possible that its a simple design oversight and they have the updates tied to the same loop. If thats the case, this can be fixed in a firmware release.
Nintendo has stated that they worked diligently to try and find a solution that would let the system work up to par without the sensor bar, but were incapable. I don't know enough about the alternative technologies to make a guess as to why, but if the O.P. can give us a product/brand name for the motion capture mouse he is using, that would go a long way towards helping.
For those interested, the functions involved most likely look something like this.
Television/Sensor Bar
---No data is transfered by either of those, the sensor bar is simply there to act as an area of reference for the wii-motes IR camera. It transfers no data to the Wii, the only reason it plugs in is to draw a very small amount of power; you could just as easily cut the wire and splice it onto a battery pack and have no connection with the Wii at all.
Wii-Mote IR Camera
---the IR Camera records where the Wii-Mote is pointing in reference to your screen; this is most likely where the vast majority of the lag comes from.
Wii-Mote Motion Capture
---the two motion sensors in the wii-mote possibly add a small amount of latency, but alone I highly, highly doubt anyone would notice it.
Wii-Mote Classic Buttons
---the normal buttons should add no real latency at all.
---very fast, latency free wireless connection with the Wii. There should be very, very little lag created by the wireless connection.
Wii!
---normal buttons most likely go straight through to whatever function code they are bound to, just like every other console.
Then the Wii gets to decide what your remote is doing, and where it is (these are two separate questions - turn around, hang out upside down on your roof, walk into another room, the "what your wii-mote is doing" motion functions should still work fine").
"Where it is" is definitely the more difficult, and time consuming question to answer. The question we have to ask is, is it simply because of the slow/low quality IR camera in the front of the Wii-Mote, or is there something more to it? If its simply the IR camera/sensor bar setup I would be really surprised if a third party company did not eventually release a much speedier setup.
________________
EDIT::.
------------------
I just wanted to add that I really doubt this lag was/is intentional to allow for say, full gestures, or jitter reduction in the pointer. Jitter reduction could/should/most likely is handled separately from the communications all together, if its even there. The scale of movement is actually pretty large, so small hand jitters really wouldn't be captured anyway (at least, not highly noticeable, if you really watch though the hand in the Wii Channels Menu definitely has a bit of jitter to it). Some games, though, without a doubt have it coded themselves. Rayman for instance, has a much - much smoother pointer for its menus, and definitely has some sort of post capture jitter reduction.
As for full gestures, the Wii doesn't work in a way that would really account for that. Its not like:
Wii-Mote preforms gesture, reports to Wii.
In which yes, that would make sense. Instead, the Wii-Mote constantly updates its position to the Wii and the Wii can wait for a complete gesture on its own. To get a real feel for this, play the golf game. When your practice swinging, make him swing - and then at the last second accelerate ALOT. The "power" bar will change accordingly to reflect the complete swing.
Similarly, when your putting, alot of people seem to have trouble making the Wii actually understand their put. This isn't because its insensitive, and a common way to "fix" it is to give it a real hard swing until it recognizes. The problem is that you never really finish the gesture, and end up slowly decelerating at the end; if you instead swing very slowly like you want, but stop the wii-mote abruptly at the end, or even flick it backwards a little, the swing will be much better recorded because the Wii understands you wanted to stop the gesture at that point. This is not a hardware thing though, gesture handling is done entirely through the software - so another golf game could come out tomorrow with much, much more responsive controls.
Personally, I have not done alot of testing with the IR independent wii-mote motion capture capabilities (as in, the stuff that makes all of the sports games work, not the menu cursor). If you don't know what I mean here (understandable, I'm not explaining it well) cover the red part of the front of the wii-mote, whatever still works is what I am talking about. If there is indeed an equal "lag" with these functions, its entirely possible that its a simple design oversight and they have the updates tied to the same loop. If thats the case, this can be fixed in a firmware release.
Nintendo has stated that they worked diligently to try and find a solution that would let the system work up to par without the sensor bar, but were incapable. I don't know enough about the alternative technologies to make a guess as to why, but if the O.P. can give us a product/brand name for the motion capture mouse he is using, that would go a long way towards helping.
For those interested, the functions involved most likely look something like this.
Television/Sensor Bar
---No data is transfered by either of those, the sensor bar is simply there to act as an area of reference for the wii-motes IR camera. It transfers no data to the Wii, the only reason it plugs in is to draw a very small amount of power; you could just as easily cut the wire and splice it onto a battery pack and have no connection with the Wii at all.
Wii-Mote IR Camera
---the IR Camera records where the Wii-Mote is pointing in reference to your screen; this is most likely where the vast majority of the lag comes from.
Wii-Mote Motion Capture
---the two motion sensors in the wii-mote possibly add a small amount of latency, but alone I highly, highly doubt anyone would notice it.
Wii-Mote Classic Buttons
---the normal buttons should add no real latency at all.
---very fast, latency free wireless connection with the Wii. There should be very, very little lag created by the wireless connection.
Wii!
---normal buttons most likely go straight through to whatever function code they are bound to, just like every other console.
Then the Wii gets to decide what your remote is doing, and where it is (these are two separate questions - turn around, hang out upside down on your roof, walk into another room, the "what your wii-mote is doing" motion functions should still work fine").
"Where it is" is definitely the more difficult, and time consuming question to answer. The question we have to ask is, is it simply because of the slow/low quality IR camera in the front of the Wii-Mote, or is there something more to it? If its simply the IR camera/sensor bar setup I would be really surprised if a third party company did not eventually release a much speedier setup.
________________
EDIT::.
------------------
I just wanted to add that I really doubt this lag was/is intentional to allow for say, full gestures, or jitter reduction in the pointer. Jitter reduction could/should/most likely is handled separately from the communications all together, if its even there. The scale of movement is actually pretty large, so small hand jitters really wouldn't be captured anyway (at least, not highly noticeable, if you really watch though the hand in the Wii Channels Menu definitely has a bit of jitter to it). Some games, though, without a doubt have it coded themselves. Rayman for instance, has a much - much smoother pointer for its menus, and definitely has some sort of post capture jitter reduction.
As for full gestures, the Wii doesn't work in a way that would really account for that. Its not like:
Wii-Mote preforms gesture, reports to Wii.
In which yes, that would make sense. Instead, the Wii-Mote constantly updates its position to the Wii and the Wii can wait for a complete gesture on its own. To get a real feel for this, play the golf game. When your practice swinging, make him swing - and then at the last second accelerate ALOT. The "power" bar will change accordingly to reflect the complete swing.
Similarly, when your putting, alot of people seem to have trouble making the Wii actually understand their put. This isn't because its insensitive, and a common way to "fix" it is to give it a real hard swing until it recognizes. The problem is that you never really finish the gesture, and end up slowly decelerating at the end; if you instead swing very slowly like you want, but stop the wii-mote abruptly at the end, or even flick it backwards a little, the swing will be much better recorded because the Wii understands you wanted to stop the gesture at that point. This is not a hardware thing though, gesture handling is done entirely through the software - so another golf game could come out tomorrow with much, much more responsive controls.
Last edited:
vagrant
₪ ۩ ₪
I have noticed 0 latency as far as the wiimote pointing is concerned. In FPS or otherwise.
Perhaps you should lower your daily valuum dosage?
Perhaps you should lower your daily valuum dosage?
Squall7
A li'l bit different
Who said that humans can't discern 25 frames per second? I'm refering to the camera speed of the footage used to take the analysis. At least, I assume you're using footage and not trying to discern the level of lag in real life.Colmino said:
How is a lag that is nigh on unnoticable for most people, a problem? I really don't see this as a problem, as it's not going to render games unplayable. Neither does it (based on everyone elses reports) take away from the enjoyment. I'm not saying you've fabricated the issue, I'm just questioning whether this is significant enough to get peoples' backs up.
And what are you refering to as "twitch" gamers?
You have to forgive many people on here, after accusations of the Wii "damaging their TV screens", causing "sores" and generally breaking things about. The Wii is getting some negative press merely because people don't know how to use it (especially apparent on websites like Joystiq). Damaged TV screen? Should've held onto the controller better. Causing sores? You should've taken Nintendo's advice, and took a break of 15 minutes for every hour. Breaking things? Either shift them out of the way or don't use so many extravagant gestures. Each time there has been a load of people blaming Nintendo for their own stupidity.
That's not to say you don't have a case. Maybe you do. It's just that I have a very critical eye when it comes to things.
I'm sorry, but I don't think you can "swish your arm back and forth 4 times" in the space of 150 milliseconds. Well, maybe the Flash could do it, but not the average gamer (or even a good gamer for that matter). Maybe once. Twice if you really went at the task.
- Thread Starter
- Thread starter
- #27
The immediate way in which it is a problem is in how the latency, in conjunction with general accuracy issues which in turn are compounded by said latency, stifles the options of the developers. I liken it to the public perception vs. the reality of the Virtual Console - a disparity that Nintendo seems content to leave it to the consumers to self-disabuse - with the perception being that "all" the games from the old Nintendo consoles can be played on the Wii, and the reality approaching perhaps 0.1% with a theoretical max of maybe 10%. The developers theoretically have a stick that they can turn into a gun, a pointer, or a steering wheel... but at the end of the day, the gun is no quickdraw (fear the day they finally port a NES light gun game over - strange that they avoided such an obvious choice, yes?), the pointer is on a spring, and the racing game itself is designed more like a rollercoaster than anything requiring quick reflexes, seeing as how you can't add 150ms to somebody's reflexes and still call them "quick".
I will clarify. This refers to the bracket of gamers for whom reflexes are a recognized important facet of their gameplaying experience. To stereotype, lump all avid multiplayer FPSers into this bracket. Such people, as I suggest, will tend to be capable of identifying minute (<50ms) and not so minute (100ms+) response time latencies with ease, and likely with disapproval.
You may want to give what I originally said a closer read. To reiterate, I suggest that one can perform a flicking motion four times per second without risking exhaustion.
I think in all honesty the latency in Wii sports is either to make it slightly less likely for you to whip the remote across the room, because in zelda it's pretty much bang on. Even Wii Tennis is close. I think 150 ms is higher than actuality. It's accellerometer is what senses the movement, the sensor is only for pointer based things, such as navigating the menu and aiming things like arrows and slingshots in zelda. the latency i've experienced has been no moreso than on a wired controller. theoretically it should be less, because there is no resistance from a wire. Well, not nearly as much as if the signal had 8-10 feet of controller wire to get through. I would say it's equivalent to your average midi keyboard with a short cord. Noticeable if you (like you and I) are in the business/hobby/whatnot of looking for some things, not noticeable really otherwise.
The games compensate for your lag -- Zelda included. I could tell this right on.
There's an inherent delay to making a full motion as opposed to just pressing a button, anyway.
Z-targeting (when available) also helps get around it a LOT.
And... if you're playing online, it's a non-issue. Why? Everyone else will have the same lag. It's fair.
There's an inherent delay to making a full motion as opposed to just pressing a button, anyway.
Z-targeting (when available) also helps get around it a LOT.
And... if you're playing online, it's a non-issue. Why? Everyone else will have the same lag. It's fair.
Squall7
A li'l bit different
Erm, how does it stifle options for developers? Considering that the lag itself includes the time it takes to compute and translate movement onto the program itself. Perhaps a predictve system for movements would do better, but so far, experience from everyone (Including people in this thread) deem the lag as not a problem.Colmino said:
Nowhere have I seen it stated that "Every" game featured on a Nintendo console will be playable on the Wii. Besides, it's one of the things with essentially "legal" emulation - the permissions of companies that still own the rights. Sure, you can download games onto an xbox and emulate them, but in the end, if it's not sold to you with the user agreement, it's not legitimate and not legal. Considering I have (and I assume a lot of people) have 0.15 seconds to start and immediately stop a stopwatch (which I know when I'm going to start, and therefore have an advantage), I don't think it would make that much of a difference. Besides, one thing about the technology that the Wii uses - it's relatively old - to cut down on expenses. They could develop technologies that would reduce down lag time to 0.05 seconds, but in the end, if the consumer has to pay an extra £50 for it, it's not going to be cost-effective. It's the beauty of the Wii, old technolgies repackageed into new things - but the Wii is more than the sum of it's parts.
I've personally played many a game in which reaction times were a major factor, and I can say that with a hand on my heart, I can tolerate it, if I even notice it. Most of the time, I would be able to accomodate for lag time, no problem. Look at Goldeneye for the N64 - quite long lag times relatively speaking, yet it was still an amazingly playable game for both hardcore FPS veterans and casual gamers alike.
Sorry, it's just that the use of the word "Swish" usually connotes an extravagant movement. Now, flicking several times is a little more understandable. Yet, somehow I still doubt if someone would be able to go that fast. Even so, you may not really notice that the computer is that far behind you unless you do it for a relatively long time (a few seonds). Even so, what the problem you're inferring is more to do with relexes rather than anything else, and whilst the computer may recognise the difference in time of 0.15 seconds, a mere human will not. If it were compounded to 0.25 seconds or 0.3 seconds, I think it would start to get noticed, but still workable. 0.5 seconds would start to get annoying - but it's much shorter than that.
