Not to be that guy, but Massive does FM via the modulation oscillator, though it's only 2 operator FM. Also to continue being that guy, it doesn't actually do FM at all, but Phase Modulation, which is what the DX7 (and later emulations like FM8) also use.

Every major DAW already does this- 32 bit float is standard for internal processing, and quite a few of them support import/export of 32 bit files as well. Some even work at 64 bit precision internally (Studio One and Reaper both come to mind).

Sadly, I think that mine (FL Studio) doesn't, or at least it doesn't between channels when mixing into channels. It definitely goes above clip, but horrible things are be to expected when mixing it with other channels.

I assumed it was the norm as FL Studio is gaining immense amounts of traction with the more recent updates to advanced competence. I might have been wrong in that assertion.

Either way, my DAW does not represent it in a way that makes sense to a software engineer.

I would certainly love to actually work in HDR, as opposed to it being a ghost in the machine.

They've got fairly aggressive automated systems for identifying and flagging material that's registered with rights-management agencies. It works for both individual tracks and material featured in mixes. I've even had my own music taken down preemptively before since the label had already registered the forthcoming track. I was able to fill out a form with some personal info and they put it back up pretty quickly, but it was still annoying (though I completely understand why they have it set up that way).

Not sure how other people are confounding the detection algorithms. They're even resilient to changes in pitch. As a result, a lot of DJs I know can no longer post mixes to soundcloud and have to use other means.

Since the later 8.x versions, Live has had a "reduced latency when monitoring" option which allows you to do just that.

I must have missed that - thanks, I'll look into it.

This is anecdotal, and I haven't worked there for close to 5 years, but back when I was at a mall Apple store, our store was clocking more revenue per square foot than any other store in the (upscale) mall. If I remember correctly, it was by a factor of 2 or more. Granted, it was only a 30 foot store, and the items are pretty big ticket anyway, rent there is more expensive etc etc, but all that being said, Apple stores are REALLY good at selling Apple products.

I once heard that the rent was $6,000 a month for a cart in one of the prime areas at our upscale mall. Usually these sell bamboo plants or cell phone accessories.

I wonder if your store was average for the AAPL universe.

I think you're referring to this: http://en.wikipedia.org/wiki/Missing_fundamental

Which, since it's a psychoacoustic phenomenon, wouldn't hold true when the partials involved are above the audible spectrum.

Sort of puts arguments about "leaky abstractions" in the appropriate perspective. Even assembly has a lot of turtles holding it up...

In fact on an x86 machine the byte code that programs are written in gets translated into a specialized micro code which is then executed.

So even assembly isn't the lowest language.

I don't like this kind of mocking because usually, I don't care what methods are called internally, I care what the outputs and side effects are. There's almost always a better approach than expectations.

Wow, I just checked my apps, and sure enough, Bing was authorized. I, too, have never used Bing. That kind of scares the hell out of me.

This is fascinating. I think a really interesting project for someone (other than me) would be to collect each image, and have a whole bunch of people rate each one on a subjective scale from, let's say, 1 to 100. Then, draw a scatter plot of the ratings against time. I wonder what curves might show up? I also wonder whether there are any other datasets that might be used to determine if there's any common thread between learning processes like this.