Why do Rock Band Drumsets need to be recalibrated so often?

Solution 1:

There's a brilliant tutorial on completely manual calibration here by a pro-level RB player:

I can't express enough how useful this video is. Even if you think you know everything there is to know about this topic, it's worth a watch.

How to calibrate, per this video:

1. Completely turn off audio (set volume to 0), so you can focus on the video and hitbox.
2. Go into quickplay and hit purposely early and late to figure out where the hitbox is for you. How early can you hit? How late can you hit? That's the size and position of your hitbox!
3. Based on your measurement of the hitbox, move the hitbox using the positive video bias (towards you) and negative video bias (away from you) to get a centered hitbox that allows you to hit equally early and equally late
4. Now it's time to focus on the audio.
5. Ideally this takes two people and a simple song with a steady beat; one person closes their eyes and plays to the beat, the other person observes to see if you are hitting early or late. I know, this sounds freaky, but per the author, this is very very difficult bordering on impossible to do by eyeballing it solo.
6. If you are hitting early, set negative audio bias to compensate. If you are hitting late, set positive audio bias to compensate.

Warning: different types of instruments may have different lag -- wireless/wired, guitar vs drums vs keyboard, etc. There's only one global audio/video lag setting, no way to set it per instrument. Thus, you should think of the hitbox as the "all controller" compensation area you use to deal with this.

Solution 2:

OK, so you are saying that you can sit at your drums, calibrate and get value x, then reboot your PS and power cycle your TV, then when you calibrate again you get value y. And x and y are different enough to cause problems playing. Is that right?

Just so we're on the same page: The game engine will produce an event (say, a snare hit). This event will cause the video to appear in a certain way (a dot in the snare lane crossing the line) and the music to sound a certain way (a snare sound). This event takes place at a single instant in time.

However, TVs and stereos have latency for signal processing, buffering, mixing, DRM, etc. So the game engine cannot know when a human will see or hear what it outputs. So the game engine actually splits that event into three events, and allows you to tune when it outputs the sound and when it outputs the video relative to the abstract event within the game.

So the only variables that come to mind in this situation are:

1. the game's data for the calibration settings
   • you said that this does not change during a single session, but you need to change it between power cycles for the gameplay to be accurate
2. the game's ability to observe those calibration settings accurately and consistently
   • it's hard to believe that this would fail or drift, it's an important part of the game and likely very well tested. I would expect other parts of the game (eg, video quality) to be intentionally compromised before this is allowed to degrade.
3. the amount of time it takes for the video and audio to reach your eyes and ears
   • it's probably safe to say the speed of light and sound are not changing :) But if you are playing in a huge room and moving the drums, that may have an impact. Sound travels about 1 foot in 1 ms, so moving 15 feet will cause the variation in sound that you noted.
4. the latency within your TV and/or stereo
   • I don't know much about this technology, but it might be reasonable to assume that the manufacturers do not consider a +/-20ms drift to be critical. I doubt a normal human would consider that drift to be noticeable for movies or other types of games. It may be that the CPU within your display locks to a specific time slice for syncing it's DSP/HDMI/etc functions, and chooses one arbitrarily on each boot. For example, it may be that your display picks the next multiple of 50ms and uses that as the starting point for the interval of its overall sync clock frequency.
5. the latency within your interconnects
   • I don't know a lot about HDMI, but it is not a simple protocol and is probably not guaranteed to be realtime or driftless. There should be no latency with the analog interconnects that your friend has.

I have never heard of any of the 6 or so people I know having calibration issues like you describe with this kind of system. It is possible that you have played on two setups that are particularly ill-suited to time-sensitive application such as Rock Band. It is also possible that you yourself are particularly sensitive to small drifts that other people would not notice. I personally do not think that a 16ms difference would be perceived by me while playing a game like this.

One thing you could do is power-cycle your TV during the calibration process. When the display comes back on, you may be able to notice that the new video is off from the old video (assuming you keep good time :). This would tell you that your TV's latency is variable between boots. You could do the same thing for your stereo. I can't think of anything you could do to your PS3 to test if there is a problem with the game itself, but as I mentioned above, I think this is the least likely to be the actual problem. If you have an iPhone, there are free metronome apps that may be able to help you identify the problem. I would guess there are also free metronome apps for Android.

In the end, I doubt there is much you can do aside from figuring out what the variable is and replacing that equipment. Some stereos allow you to tune the latency, but since your problem is variability over time, this would just be another way to "calibrate" your drums. Or you can live with frequent recalibrations and/or adjust your playing style to accommodate the variation.

HTH.

Are you saying that you would bring your drums to your friends house? Or that (s?)he would bring his PS3 to your house? Calibration does not happen in the instrument, AFAIK. It merely changes the way the sound and video are "synced" prior to output from the console so as to match your perception and your environment. So if you calibrate the system for one environment/console/display setup in one place, you would have to recalibrate for another environment/console/display in another place.

If the above is not the case, and the drums do not move between sessions and no settings get changed, it may just be perceptual on your part. How long is the duration between sessions? Do you play with different individuals?

In my experience, after a long duration between sessions, it does take some time to get back into the groove of the audio/video sync. It will never be perfect.

As well, different co-players play differently based on their own reaction timing and physiological quirks, and it is human nature for us to sync to these other humans in the room. I've played with one person who was consistently late (but within the range of "correct" in the game) and it can be very disconcerting, especially if you are playing at different levels.

I don't know the science, but I would be very surprised if ~40ms would be significant. I feel like there are times when I am up to 100ms off from the video, but in sync with my co-players and/or the sound, and that I can mentally shift the video to compensate.

I have had good luck with using the video as just a rough guide, and relying mainly on keeping up with the music as well as the others that I am playing with. I have seen people approach these games as precise video challenges and not give enough weight to the music/co-players, and they never do very well.

Or are you saying that a calibration you made simply does not survive the occasional console reset? That sounds like a plain old bug... Are you shutting down the console "properly"? Or just killing the power to it? I have Rock Band Beatles and Guitar Hero World Tour for the PS3, and one calibration per game is all I've ever needed.

Solution 3:

Are you always playing with the receiver in the same mode/setup?

The amount of sound processing that the receiver needs to do for some special effects should have an effect on lag.

I would make sure that you are playing with the least amount of receiver processing possible and consistently using that mode.