Wednesday, July 29, 2015

i can't make sense of this. i'm starting to understand the tonal differences a little better, but it really makes no less sense to me as to how i ended up where i am.

the guitars null flat. there's some bleed, but it's from the bass part, as it was transferred from tape. tapes bleed. it's not an incomplete null, it's bleed. that's settled.

i can get the drums to very nearly null. i need to cut the eq on them by about 2 db at 50 hz, then boost the master eq by about 2 db at 80 hz. there's an izotope plugin on the master, and it comes in before the eq in the signal path, so that 80 hz boost has to be where it is and that 50 hz cut has to be where it is. i can't say to myself "maybe i split the difference and forgot". they're not interchangeable. in order for it to null, i have to cut the bass into izotope, and increase it out of izotope.

but, of course, anything i do to the master will affect the guitars. then, they no longer null. nor can i say "maybe i cut the guitars before they got to izotope". it doesn't work like that when you've got a plugin in the way.

this suggests two distinct signal paths in the earlier mix. but, i don't mix like that. i've never mixed like that. i'd never do that.

see, the funny thing about this is that it's an entirely rational mixing decision. you might decide the bass is coming in to the mix a little too heavy, take it down and then increase it on the way out. i don't remember doing that here. but i've done this countless times. it's a pretty normal mixing strategy. it doesn't seem like a random error. it seems like there's a fucker at a mixing desk.

i'm still convinced this is crazy. it's the kind of thing i'll never actually believe. i'll disconnect the internet. i'll rant about it all day. but, it's insane.

i'm the kind of person that would think i hallucinated it if god herself came down and slapped me upside the head. i could have a bruise. i'd think i fell somewhere. there's certain things that no amount of evidence can be convincing of. it's a sort of problem with being a cold rationalist. even if it's true, i can't accept it.

yet, i've determined that the old mix somehow sent the guitars and drums through different izotope instantiations, because there was a slight bass boost on one of them and no bass boost on the other - and the probability that this could have cancelled out any other way is so negligible it can't be seriously entertained.

my brain is unable to interpret that as a broken clock, because i can null with what i could call "elementary mixing operations". once a mathematician, always a mathematician. a 2 db boost at 80 hz is a common operation. a 2 db cut at 50 hz is a common operation. a broken clock would create the messy and random complications i was hearing, not uniform cuts at widely used frequencies - that in some sense cancel each other out.

i think, at this point, there's nothing of further value to be gained by trying to null these files. i'm going to shift the focus to determining if the device is currently consistent and flat. if i can prove this, i'll then go back to mixing it from scratch.

let be me clear on the following point: i cannot use mixes that i cannot recreate. whatever the cause, these mixes are useless to me.
see, what's weird is that, on closer analysis, the guitars are actually nulling. which indicates that the file that's outputted is not actually a clean render. i don't know how to make sense of that. did i render a mix without saving it? i have no recollection of this, and it's very uncharacteristic of me. but it's just about the only thing that makes any sense.

well...i suppose that the corruption in the device may have been bus related. but i think that's kind of stretching it.

dropping this external mixing hypothesis is proving difficult. if i can create a mix that nulls without modifying the master volume or eq, i'm left with a head-scratching paradox. but, that seems to be exactly what i'm in the process of doing.

even this doesn't make sense. the file save date on the project is june 29. the render is dated at basically the same time. there's no space for shenanigans. and these aren't things you can easily fake.

but, if this corruption on the bus is a real thing, what i may be doing is manually reconstructing it. i could have output the file through these corrupt buses, which could be the cause of the warped output.

i suppose that would reduce the initial issue to a cubase problem. it seems remote, but i can't argue with the obvious.

more testing is required.

i can't prove there was actually a driver problem until i wiped the registry. there was some other unknown problem before that.

rap news 34

mostly spot on. the way this works is like this:

0) people like donald trump write trade treaties like nafta.
1) nafta allows a us factory to move from arizona to mexico, where it can pay workers a third of the price.
2) the workers at that factory lose their job.
3) the factory blows up the local economy in that region of mexico, forcing the locals into poverty.
4) that sets off a chain reaction of migration that pushes people out of mexico and into the united states, partially because
5) sub minimum wage pay in arizona is still more lucrative than work in mexico.
6) this benefits producers that create products for export but it reduces the pool of possible labour opportunities in arizona, creating structural unemployment levels.
7) what low education white arizonans are able to physically see and intuitively understand is that illegal mexican immigrants have jobs and they don't.
8) donald trump (carefully avoiding neatly shaved mustaches) stands up and blames it all on zee mexicans, in an attempt to generate a political base and distract from his own guilt and responsibility. mass unemployment creates social unrest that needs to be controlled.
9) the media agrees, in an attempt to construct a subscription viewer/reader base and further control the social unrest.
10) low education white arizonans vote for donald trump, but they do not expect him to reverse the policies that set off the chain of reactions.

it's a bit of a stretch to suggest it was planned from the start. but power has a tendency to perpetuate itself by profiting from it's own crises.