level 8 awaits

Comments and questions about Absolute Pitch Painter
aruffo
Site Admin
Posts: 1696
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

level 8 awaits

Postby aruffo » Fri May 16, 2014 5:14 pm

So I'm actually partway into the Large games of Level 8. Nothing new to report just yet...

...although I have been noticing that some of my judgments must be based on scale degree rather than proper chroma. Not always, and not consistently, but sometimes. I notice this most evidently in the orange tone. Sometimes it will actually feel like a seventh-- that is, I can recognize it because it feels like it's "almost" orange/blue and wants to resolve upwards. Other times it will have a dissonant feel that I am reasonably sure is the result of a scalar ratio rather than an absolute quality. And if I can tell that that's happening with the orange, it's probably happening with the others in ways that I can't tell.

That was bothering me until I remembered that most absolute listeners were trained in fixed-do systems. Although, of course, they mightn't have initially understood the relative sense of each tone, I'm willing to accept the possibility that in training, here, it'll be more important to establish a solid base of "absolute" perception and gradually (but consistently) undermine the relative elements of that basis by expanding octaves, introducing new instruments, and using APA to perceive chroma more clearly.

aruffo
Site Admin
Posts: 1696
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Postby aruffo » Mon May 19, 2014 1:55 am

Egad. For a while, as I began the Large games, I thought that this would be no problem at all, because I was identifying the tones without too much difficulty.

But having a greater number of categories means that there are many more border tones lurking in wait. They're driving me crazy! I mean, it's probably a good thing, because you can bet your boots I'm paying close attention through the whole game-- although this last game I let out an agonized groan as my final score topped out at 1,999-- but those border tones are what stand between me and the Jumbo games (and, presumably, level 9). It's obvious from the graph:

Image

I'm still prone to some height judgments in the green-to-blue/orange area, and yellow is still my worst category, but what's really killing my game are the borders between grey/blue and grey and between brown and red. My mental borders are simply in the wrong place. As I'm playing, "don't guess" is practically a mantra-- but I will hear what I clearly perceive to be a grey tone, and it's grey/blue, or I'll hear what is obviously a brown tone and it's lower red. When I make a bad height judgment I feel dumb because, when I listen to it again, it's obvious that I just wasn't paying attention... but for those high grey/blue and low red tones, I simply can't tell unless I leave them aside, find a tone that represents its category more clearly, and compare, and I don't do that because when I hear them I think I know them!

Even so, because these border tones are the trouble, then it seems likely that I'll become pretty darn sensitive to those borders. (I hope!)

aruffo
Site Admin
Posts: 1696
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Postby aruffo » Mon May 19, 2014 11:21 am

Only one more 2000+ score before I can move past the Large games.

It is obvious-- but it's frustrating in practice-- that if you leave aside all the ambiguous tones and drop only the obvious ones, then when you get toward the end of the game, all you've got left are ambiguous tones. In my last game, I had the scoring multiplier at 12x for a cheeringly long time... but then all of a sudden POW I had trouble even keeping it at 2x. (Final score: 2006).

aruffo
Site Admin
Posts: 1696
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Postby aruffo » Thu May 22, 2014 12:07 am

As I've been playing Jumbo games on level 8, I've been awfully impatient with myself. Whenever the multiplier dipped too low, I'd restart, even if there was still plenty of opportunity to recover. After a while, though, I wondered why I was having such awful trouble keeping the multiplier up-- why couldn't I keep myself from making mistakes?-- and the identifying stats told the tale.

Image

Yellow was the culprit. I knew I didn't know yellow, and so it appears I had (without really realizing it) essentially given up trying and been using height-based judgments instead.

Once I realized this, I started listening again for the "chroma" (I continue to put that in quotes because I am fairly certain that the quality incorporates scale degree) and, when I encountered tones which had to be yellows, I'd play them and listen and feel... that blank, empty space in my brain, like the minor sixth when I'm identifying intervals. I wondered how I could somehow mentally attach some kind of identifier to this tone; I again considered melody triggers but again rejected the strategy as indirect. I wasn't sure what could work.

But I considered... I know by now pretty much what every tone sounds like except yellow. The border tones are still difficult, of course, but the problem there isn't that I don't recognize them-- it's that I misjudge which of its mixed qualities is the definitive one. So what would happen, I thought, if I trusted myself to judge that, whenever I felt that blank space in my mind, and couldn't recognize the tone as either green or red, I was hearing a yellow?

What has happened is that this strategy actually seems to work. I still haven't gotten another over-4000-score game, and the stats aren't changed enough to show a difference, but when I hit that blank, I drop it in yellow, and it is usually right. In fact, amusingly enough, the blank is starting to be replaced by "oh, yes, that's that one I don't know."

abminor
Posts: 87
Joined: Wed May 23, 2007 7:55 am

Postby abminor » Thu May 22, 2014 6:51 am

I had a similar problem with green in level 4. I didn't know it. My mind was going blank every time I heard it. I think it's because I stopped playing APP for a while right after I completed the medium game, so I forgot what I learnt.

My strategy was a bit different than yours though. I tried to concentrate on the green eggs and ignore the others for a while and it seemed to help me relearn this color.

However I feel that what really helped me the most was trying to pay attention to scale degrees as I was playing. Although there was time when the color was in contradiction with the scale degrees, because the tuning has shifted to much. So for this level I think I can hear the color but I use the scale degree to check myself. I 'm wondering if I'm on the right track though because learning scale degrees is not my purpose. Is it a strategy that has to become prevalent as one progress through the levels ?

I feel the times I use the more what I believe is chroma is when playing with only two categories in the small and medium games. would it works to learn to identify all the pitches this way or is it a necessity to have to choose between all categories ? I don't know.

aruffo
Site Admin
Posts: 1696
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Postby aruffo » Thu May 22, 2014 8:36 am

I mentioned in another post that I think scale-degree listening must be inevitable-- and that, between APA, the planned expansion of APP's range, and the fact that most AP listeners learned music with fixed do, I'm not [yet] concerned about it.

A key assumption that I make about learning (whether ETC or any other context) is that our brains and bodies will attempt to fulfill a task in the most efficient, least effortful way possible. Therefore, the task should be designed so that the best way to accomplish it is by using the target learning-- and it should be expected that, if there is a better and easier way to achieve the result, people will do it (often called "cheating").

So in this case, there was a choice either to frustrate height judgments (a la Shepard tones or multiple octaves) and scale-degree judgments (by some kind of multikey-confusion element) or to allow them. As I mentioned in that other post, I think it makes more sense to allow them but gradually undermine their effectiveness.

aruffo
Site Admin
Posts: 1696
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Postby aruffo » Fri May 23, 2014 12:10 pm

Add another two successful Jumbo games.

It's really quite strange that I can identify yellow by recognizing it as "that one I don't recognize." Sometimes that's a positive identification, where I recognize it as the unrecognizable one-- contradictory as that seems-- and other times I listen and confirm that empty blank in my head before dropping it (correctly) on yellow.

I have three major pitfalls that trip me up:

One, of course, is yellow, because I don't recognize it, and if I'm not careful then my mind realizes it doesn't know yellow and subtly switches to a height strategy-- which my mind believes it does know, because judgments are rendered more quickly (even if they're not exactly correct).

The second is that border between brown and red. The problem there is exactly the same as before-- that the lowest red has a certain extra quality that is shared by the highest brown. If I am not paying careful attention, I'll recognize it as a brown, and make a mistake. If I am paying attention, I'll recognize it as a border tone, and wait until I find a red to compare it to, and get it right.

The third is a confusion between gray and gray/blue. When I'm chugging along, and a gray/blue pops up, I will frequently mistake it for gray. If I wait a little while, and compare it to either a blue or a gray, then its gray/blue character is blatantly obvious, and I wonder how I could ever have made such a mistake-- but until I do, it still sounds perfectly gray.

This last one is actually sort of encouraging, because it's been shown and known for some years now that the worst-recognized tone is A-flat. In this game, gray/blue is A-flat and gray is A. If I'm at the point where an A-flat sounds like an unmistakeable A (when I'm not paying attention) then that seems like a kind of progress.

aruffo
Site Admin
Posts: 1696
Joined: Tue Dec 14, 2004 12:09 pm
Location: Evanston, IL

Postby aruffo » Mon May 26, 2014 2:05 pm

Well, there's the end of level 8, and it looks like this:

Image

and these stats are coincident with my experience. I didn't find it difficult to recognize most of the tones, but I kept stumbling on border tones, and yellow was, obviously, by far the worst category. I managed to pull myself out of height judgments largely by remembering that yellow was the one I didn't know, but I still slid back into height now and again.

The border between brown and red is still the worst border, which should prove interesting when Level 9 splits the red category in two. The border between green and orange is also weak, which should be just as interesting whenever the green category gets split.

The discrimination scores look like this:

Image

If I discount blue as an end effect, then the "good" categories here are grey/blue (grue?), red, green, orange, and blue/orange (blorange?); the "not-so-good" categories are grey, brown, and yellow. Which is interesting, because in terms of identification, grey and brown are usually my best (as you can see in the previous chart).


Return to “Absolute Pitch Painter”

Who is online

Users browsing this forum: No registered users and 1 guest