Wikipedia talk:Featured sound candidates/James Scott - Frog Legs Rag

Appears in Frog Legs Rag and James Scott (musician). I'm VERY pleased at how the collaboration with Juju came out. Because of the nature of this, it's tweakable if anyone spots something (and we agree, of course), though try not to rethink everything we've done =) Adam Cuerden (talk) 07:39, 23 February 2011 (UTC)


 * Nominate and support. Adam Cuerden (talk) 07:32, 23 February 2011 (UTC)
 * Support as co-nominator.  Jujutacular  talk 13:13, 23 February 2011 (UTC)
 * Fundamental query—was this performed live, or were the notes placed in position electronically? My problem is that the performance is completely robotic. There appears to be not a single deviation from a prefabricated tempo, pulse, and rhythm. As such, it is devoid of musical interpretation. Sterile. Even a piano roll of the period (binary code, too) would contain the results of musical interpretation. Tony   (talk)  09:30, 23 February 2011 (UTC)
 * Is that a question or a statement? ;-) Yes the notes were placed digitally, but the fact that you're asking tells me you know that what you're saying is a little off: there are variations in tempo and pulse. If you'd like to propose changes, we're listening. This is ragtime however, which generally is performed quite mechanically.  Jujutacular  talk 13:13, 23 February 2011 (UTC)
 * There are also hundreds of subtle changes to rhythm and phrasing. There's a big difference between legato, staccato, and everything in between, all of which can be (and is) used to add an interpretation. Adam Cuerden (talk) 23:11, 23 February 2011 (UTC)
 * Question Why are we using the synthetic piano instead of feeding the MIDI into Juju's piano like your last collaboration. This isn't me saying that the decision is good or bad, only that it's different from how you did your other ragrime collaboration.  S ven M anguard   Wha?  04:57, 24 February 2011 (UTC)
 * We used the identical process as last time (Adam' source file, my piano sampler).  Jujutacular  talk 05:06, 24 February 2011 (UTC)
 * Okay now I'm confused. Piano sampler? So neither of the rags were played by actual pianos? Adam's MIDIs were fed into your computer software?  S ven M anguard   Wha?  06:44, 24 February 2011 (UTC)
 * Here's how a piano sampler works: the people that produced it recorded an actual piano playing every note at various volumes, then created software that would play those individual notes when a MIDI file is input. So in a sense, you're listening to an actual piano. This is the exact mechanism you get from a high quality digital piano.  Jujutacular  talk 11:26, 24 February 2011 (UTC)
 *  Strong oppose.  Removing my oppose because I don't want to appear to be obstructive, and anyway there's enough support here to gain promotion. I would be pleased to see discussion / careful thought before any more of this genre are nominated. Tony   (talk)  13:30, 4 March 2011 (UTC)Plain unmusical unless it's significantly worked on.  Tony   (talk)  12:21, 28 February 2011 (UTC) I want a live performance or a MIDI construction that sounds like it. "hundreds of subtle changes to rhythm and phrasing"—Could have fooled me. "This is ragtime however, which generally is performed quite mechanically."—Not that mechanically. You really notice it at the cadences; e.g., 0:19, 0:37, and 0:57. 1:31 is disappointing. There's a bit of tweaking at 1:12, yes.  Tony   (talk)  10:23, 24 February 2011 (UTC)
 * Which is, of course, why you only opposed after asking if it was based on a live performance. Because you were so easily able to tell. Adam Cuerden (talk) 18:24, 24 February 2011 (UTC)
 * No, I opposed because it's unmusical—it wouldn't have mattered whether it was a live recording or a MIDI construction. I quite like the timbre of the synthesiser, and I have no issue whatsoever with the concept of carefully arranging each not in a MIDI file: on the contrary, I'd love to do this myself. But I find the arrangement of the notes not nearly nuanced enough, even for a motor perpetuum genre like a rag (the thing is, the mechanical nature of the accompanimental motif demands some give and take at the structural joins, or it becomes unrelenting. To me, it sounds as though it's a Finale performance from the notation alone—that is, with utterly no variation from the notation. Listen hard, as I did after your comment, and you can only just make out a bit of fiddling. Not nearly enough, though. Tony   (talk)  07:54, 25 February 2011 (UTC)
 * I sought expert opinion of a piano specialist, Mr David Kinney, who is employed by a conservatoire. He says it is totally unsatisfactory on a musical level. Ragtime is not "robotic", he says. He turned it off after three phrases. Tony   (talk)  09:23, 25 February 2011 (UTC)
 * Wow, so a few seconds of the actrual file in question are all he needs to agree with you when you ask him to agree with you.
 * I'm also impressed that, while complaining about how rigid the beat is, and how you can tell in the first few seconds, neither of you noticed that it starts with a gradual accellerando over the first four seconds. I thought I'd give you a little time to notice. And yet, while complaining about how the notation, you didn't notice.
 * There's lots of fiddling, every bar has four or five changes in note length. Adam Cuerden (talk) 11:57, 25 February 2011 (UTC)
 * After reading the comments you attribute to this expert, I'm quite sure he's a quack. No remotely competent expert in any field would render judgment on something after only viewing a fraction of it, period. You've indicated that he's done so twice. I'm sorry to be so blunt, but if this Mr. Kinney really did what you said he did, I wouldn't trust him to even tell me what instrument he's hearing, because it's clear that he lacks the will to put in even a basic amount of effort in assessments and lacks the expertise to know that musical pieces change over the length of their performances. Also, to the video, I have never known a competition musician to use "cheap" instruments. Also, if someone's going to offer their opinions, the need to have their own account and they need to say those opinions themselves. Opinion by proxy is not as credible as matching opinion to a single account.  S ven M anguard   Wha?  21:28, 26 February 2011 (UTC)
 * Quack or not, these kind of comments are also inappropriate because they invite Essjay-style review. Announcing that you have this expert friend is unnecessary, and sets bad precedence. Wikipedia is at risk to this type of problem, and in my opinion Featured Sounds is especially so. If you want to discuss nominations with friends and discuss your findings here, that's fine with me, but they should be given the same treatment as any other user's opinion. I respect your opinion and your friend's, and I actually agree that this file could use a bit more humanization, but announcing a level of expertise has no place here.  Jujutacular  talk 23:11, 26 February 2011 (UTC)


 * Oh look, how dare you insult my friend of more than 25 years. You have a hide, both of you. He is a fine musician and an excellent judge. Your problem is that you can tell exactly what the problem is from listening to no more than a few phrases. It's plain boring and unmusical . You should be ashamed of yourselves. And Sven, you don't even have the guts to use your real name on WP; yet you want to take his real name and insult him? (He was fine with my using his real name, but I guess he didn't count on this kind of nastiness.) Tony   (talk)  12:19, 28 February 2011 (UTC)
 * I can't speak for Sven, but what do I have to be ashamed of? I told you I respect yours and his opinion, but that announcing a level of expertise has no place here. I could tell you about mine and my friend's musical backgrounds, but you would have to take all that with a grain of salt. If we can't learn from an experience like Essjay, what would it take? I cringe at the thought.  Jujutacular  talk 13:18, 28 February 2011 (UTC)
 * My not using my real name is irrelevant, as is your friend's using of his. What's reliant is that his method of assessment is totally wrong, the equivalent of looking at the top corner of the Mona Lisa and saying "oh yes, there's some yellows and greens and such but I really don't see the point here" without ever looking at the piece as a whole. If Roger Ebert or any other music critic wrote "the first 7 minutes of the movie were not up to snuff so I walked out of the theater, I give it zero stars" no one would listen to his advice. It's bad practice and everyone knows it's bad practice.  S ven M anguard   Wha?  21:31, 28 February 2011 (UTC)


 * Support on one suggestion - could the right hand accents be just a hair softer? I like how you're highlighting the bass in spots - don't change that. But the right hand accents (you know, on the syncopated chords) get a tad repetitive after a while. Otherwise, lovely rag - better than late-night coffee. Or...early morning coffee. Gosh, I need better study habits... &mdash;  La Pianista  ♫ ♪ 10:51, 28 February 2011 (UTC)
 * Sure, I'll go through and do a revision. Adam Cuerden (talk) 10:56, 28 February 2011 (UTC)


 * Ah, I hadn't thought of suggesting changes to it. In that case, may I suggest that a few tweaks could overcome the unrelenting robotic feel. Here are some suggestions for "marking" the periodic structural points, allowing it to "breathe" a little; I wonder how hard it is to implement them in the file (i.e., can you make one note/chord sound a little later and have all of the subsequent events move later to make it work?).
 * 1) 11 s—Just a tiny bit more time before the RH melody comes in after the LH tonic? Either that or a tiny tiny delay in the sounding of the tonic? Or both? I'd like to be able to feel it perceptibly. Same at 29.9 s etc. That phrase is repeated so many times.
 * 2) 18.5 s—the trilly motif in the RH: can you do something with it? Again, what happens if it comes in a teensy bit later. Same at 38 s. I'd be inclined to make those descending octave-doubled notes a bit slower. They are the entree into the next structural unit. Maybe not at 46.5 again, though, since that's ongoing within the structure. But yes at 54.5, please, since that is a major periodic join; but not at 1:06, as before.
 * 3) 26 s—that highest note in the RH ... it is in tune, isn't it?
 * 1:13 s—The isolated RH chord sounds trivial, but it's a nice little idea. It is flicked over without a thought.
 * 1) The transition around 1:37 is so elaborate note-wise that it comes off well metronomically, as you have played it.
 * 2:13 s—Desperately needs pulling back in tempo, momentarily, just enough to mark the change.
 * 2:31 s—Mark it.
 * 1) Then last three chords: choked off, falls off a cliff. Some kind of greater emphasis (a slight rall.) is essential.

A boring performance can be turned into something with life if just a few adjustments are made. Otherwise, I still cannot support this nomination. Tony  (talk)  12:13, 28 February 2011 (UTC)
 * Initial changes File:Frog Legs Rag 3.mid - please provide any and all critique. Adam Cuerden (talk) 13:16, 28 February 2011 (UTC)


 * Comment As is, it needs more reverb, and could definitely use more staccato throughout on the up beats. The common ragtime eighth-note pattern is usually long-short, and I think that's what this piece is going for. I also noticed a lot of notes that were taken too far off-beat to where they sounded almost like mistakes--but that could just be my MIDI player. If these two things are cleaned up, then my vote would be to Support. — trlkly 21:26, 28 February 2011 (UTC)
 * Staccato on the upbeats sounds off in a lot of sections, as it de-emphasizes the syncopation. However, about a third to half of the notes are already modified in some respect, either stacatto, legato, or detache. Also, can you point to the time of one (or more) of the notes taken too far off-beat? I'll have a look and see what's going on at that point. Adam Cuerden (talk) 05:04, 1 March 2011 (UTC)


 * Half that reverb would be nice (halfway between it and the previous version). I agree that there wasn't enough.
 * Where I suggested making notes come in a bit later, you've adjusted the placement of only a single note. Thus, the RH melodic phrase begins as though with a 16th note jammed up against the next note. It produces an unintended extreme syncopated effect. What I was suggesting was that the whole texture slow up commensurately—in other words, that every subsequent note in both hands be just that much later to accommodate the slight delay in the entry of the RH melody. This was my initial concern: that having programmed the MIDI file, it might be a huge job to push every single note that much later. Change the temporal location of one note alone won't work. I did suggest this above, when I said "can you make one note/chord sound a little later and have all of the subsequent events move later to make it work?" Later on, the same phenomenon has appeared to produce an actual jarring of notes together. Perhaps my fears are confirmed, that it is not easy to introduce rubato, and what is referred to here as "interpretation" involves only accent, staccato, and volume (none of which have the "domino" effect required when the whole texture is bent slightly slower, as it were, for just an instant. Concerning comments above about sampling part of a performance and commenting on the file, my view is diametrically opposed. The first sip of wine tells me it has been watered down. The first page of a novel can tell me the writing is generally substandard. Finding an editing glitch or two in the first 30 s of a film tells me there are likely to be problems throughout. Examining a tiny corner of the Mona Lisa under a microscope might tell me that it's a fake. You have publicly defamed this man, using his real name. It is shameful. Tony   (talk)  12:08, 1 March 2011 (UTC)
 * It's fairy easy to introduce as much slowing on a note or rest as you want - have a listen, for example, to 1:11 (bar 35), where you get exactly the effect you seem to think is impossible to do with MIDI, in a very pronounced way. Since you seem unable to hear tempo changes of which there are about 60 (that's a bit out of date, but gives some idea),  I'm not sure what I'm supposed to do. One can't extend rubato indefinitely and still have the piece sound like a ragtime one. At the moment, to give an idea of what's happening, the first six bars start at 106, drop to 104, accellerando to 108, and then ralletando to 106. At bar 12, we go from 106 at the start to 110. then ralletando back to 106. At bar 18 - the join between the first theme and the second - the first note is held a bit longer (90 crotchets per minute), then we speed back up. I don't see how one can do much more than that sort of moderate, but fairly constant tempo tweaks without ending up with something like the delisted Entertainer FS, where too much rubato made it have to be delisted as not being played with the relative metrical regularity of a ragtime piece. And I can't very well put the righthand permanently out of joint with the left hand.
 * You're asking for a lot of things, but when I do them, you seem unable to hear them, and I can only presume that either you want something that I would find hugely excessive, you're listening to the wrong file, or I'm completely misunderstanding you.
 * I'm sorry, Tony, but I need to ask you to be a lot more clear about what you want, because I'm, in the situation where I do what you ask, then you seem to act as if nothing's changed. I am doing rather a lot of work to try and keep this sounding like a coherent whole - Ragtime simply cannot stop dead for more than a split second, while fulfilling your requests for pronounced tempo and phrasing changes at the joins. Adam Cuerden (talk) 14:15, 1 March 2011 (UTC)
 * ETA: Apparently, he was listening to the wrong version. Making some final changes. I'll throw this into "suspended" for the moment. Adam Cuerden (talk) 14:47, 2 March 2011 (UTC)


 * Withdraw for the moment. So much has changed that this should probably be revoted on. New file in 2-3 days. Adam Cuerden (talk) 13:35, 4 March 2011 (UTC)