Gambold wrote:>You can't help but assume that they are unnecessarily padding the file to make it harder for people with old boards to use? <
I don't know if it's unnecessary or not.
You said "I can't help but assume it's to move people like me into new keyboards." meaning you couldn't help but assume it was unnecessary (that is, only being done to screw users of older boards). That's the part that didn't make sense to me. Why
must you assume the worst (i.e. "can't help but")? And if you're going to assume evil motivations, why would you necessarily believe anything else they said anyway?
Gambold wrote:The new technology evidently doesn't mean anything for XL. Which seems a little odd - if larger bitrates or longer attack portions are being used, why haven't the XL sample sizes doubled too?
That's an interesting point. First, as an aside, I don't think bit rate is a good theory here to begin with, because in the context of instrument playback, unlike longer attack portions, higher bitrate is unlikely to make a noticeable audible difference (and may even require different playback electronics). But you're right, longer attack portions would seem to have to affect all sample sizes, so that's probably not what they're doing either. (The reason I say "probably" rather than "definitely" is that there are other possibilities... like they already were using longer attack portions on the XL but hadn't been saying so, or they felt that longer attack portions were not as important if they weren't stretching... but I think these are less likely answers.) Same with the idea of including more velocity layers.
So if I had to guess, I'd say it was probably that the S/M/L are using less sample stretching than they did in the old days. That would be something that would tend to make them sound better, but would not increase the size of the XL. And they could choose to do such a thing because an increasing percentage of their users have the space to spare and would probably prefer the better sounding piano in a still pretty manageable size (it's not like there aren't already plenty of smaller pianos available for those who need them), OR they could have determined, because of the sonic character of the particular piano being sampled, that they could not get satisfactory results if they did as much stretching as they did on some of the earlier sample sets.
If someone is so inclined, one could load up the XL and the S of the newest piano, and by comparing note by note, determine how many notes are truly sampled vs. stretched; then load up an XL and an S of an older piano and do the same thing. Then you could determine whether the new S samples do or do not have less stretching than the old.
Again, I think it is somewhat academic, but it would be interesting to know.