While I'm indifferent to the whole 2000/1000 thing, I think there needs to be a requirement for very short spinners that there's a recovery period after the spinner. It should be at least a 1/1 at typical BPM, and only be required if Auto gets 3000 or less? Maybe 4000?

something like...

"At 130 or higher BPM, spinners of 3000 or shorter should have a 1/1 recovery period for circles/sliders, and minimum lenght for spinners should be 1000."

?

I agree with this, though we should define a different limit for the easiest diff of the set. Both on recovery time and minimum lenght.

Also, under no conditions should objects AFTER the spinner be visible BEFORE the spinner begins.

instead of measuring it with beats since they differ across different BPMs how about you measure it in miliseconds or full seconds (if applies) personally i think at least 1 seconds after a 2k spinner should be enough recovery time, and if you DO use the minimum, you cant put a stream right after, it's pretty hard to click a stream after recovering from a spinner

Yeah, I'd agree on this one here. As for when the next object appears, it goes without saying that the score explosion graphic upon spinner's end should never cover up gameplay elements

Also true, sometimes the note is centered and timed in just a way that it's approach circle can mix up with the spinner's approach circle, but i guess that would be for another rule, although if this one stands that there should be a minimum time length until next note, the statement that i just said would never be a problem

There should never be notes less than 1/2 after spinners anyway. Even 1/2 should be extremely rare. Spinners aren't supposed to replace sliders. If something follows right after, it's better to use a slider.

I agree to force a recovery time under a certain auto bonus, but it should be at least 3/2 for BPM over 180 imo.

If the spinner itself is long enough, no recovery time (or no more than a 1/2) is necessary. I think these spinners work fine: http://osu.ppy.sh/s/25365

Human reflexes arent measured dinamically by tempo, human reflexes are measured in a flat ammount of miliseconds i still think measuring the minimum recovery time with beats doesnt make any sense, or are you saying that if a song is fast (like 200 bpm) they should have less recovery time?, and if not, then why are you gonna have different ammount of recovery per BPM? that would turn it more into a guideline because it would become case by case scenario.

If you use time (like 1 second for example) it will be the same ammount of recovery regardless of bpm, if the 1/2 tick didnt make it past 1 second? too bad, just start on the next tick, faster BPMs would obviously need more ticks to cover 1 second, etc.

Note: I'm using 1 second as an example, but how long exactly could be discussed here, i still stand by my idea of using time instead of beats.

I think we can all agree that milliseconds are the best metric to follow.

We can't use auto to make a full judgement on this issue since auto is computer controlled, not human. The other factors that can screw with the spinner length issue is the Overall Difficulty level settings. I would still stick with no lower than 2000 unless there is a true exception to it I've never heard of. And about the recovery time thing..... I seriously agree about counting in seconds than in beats since as it was said, the BPM varies, and if I did try to count by the beats, what about maps with doubled or halved BPM?

Overall difficulty is factored into Auto's score bonus, so that's irrelevant.

Time between the end of the spinner to the next note is a poor metric to use, because it also depends on where the next note is. A time period after an Auto+1000 spinner as short as 350ms is very easy to hit as long as the next note is near the center. Any note in a corner after a spinner is far more difficult to hit because spinners restrict your cursor movement and force it towards the middle.

Auto+1000 spinners are very easy to hit, and only require a speed of around 300 spm.

We ALL know Auto can easily spin +xxxx amount of bonus points, but if you can actually pay attention, you can also see how much effort auto still has to put out to spin before those bonuses. And I did say that Auto play shouldn't be a true measure on making judgement calls for "good" and "bad" spinners. If you forget to apply the factor of human error then the logic wouldn't make sense.

Yes Auto can be a factor of how hard a spinner is to hit, because it always spins at the exact same rate every single time. If that's not a usable benchmark, then clearly nothing is. Spinners that have Auto getting 1000 bonus points are 300'able by humans. Period.

With the right settings and no harsh "trade off" moves like a short spinner with a note just a few 100 ms away. Also, Auto provides a *constant* spin compared to a human player which is another reason to not just say "okay, Auto can do it".