nice
true1 power output connectors so safe i cant even get standard fluke probe in contact with terminals
neutrik ftw
Posted by renesis at 19:36 | permalink | 0 comments
lordpil's blog
ive reproduced every track except the pink noise one
that one i gotta get better at digital filters
like hes got some s domain transfer function and i have to make sure i can just z domain that shit
seems like it works, get same fft as white paper
but then when i put the soft clip i get some HF ring i know isnt in the real spec pink noise
Posted by renesis at 18:31 | permalink | 0 comments
so im like, fuck this and fuck you, im just gonna freq_delta**(1/sweep_cycles) to get the multiplier and do this shit continuous with this given sweep times
and then, for his pink noise soft clip shit, he mentions you have to scale the shit appropiately but doesnt say how much like he wants me to mr wizard this shit
wooooooo 10min track of 0.5s exp sweeps half done!
no timecop i was
and i ate and never worried about bills and rent and went to shows like every fucking week
but yeah the doc describes bunch of test procedures so i might as well automate those
Posted by renesis at 18:24 | permalink | 0 comments
holy shit this thing is generating in realtime
which is really fucking slow
so i used this algorithm to generate this exp sweep
well, this fucking math equation in the design white paper for the spec
and its got diff algorithms for the diff sweet rates
and there is a phase offset variable
and yeah when you splice the shit together, and pass the last phase into the init phase of the next generator, the shit dont line up
like, he went in and manually offset the phase?
which seems likely, because he mentions for the symmetric downsweet, he just inverted and offset a half cycle
*downsweep
Posted by renesis at 18:18 | permalink | 0 comments
anyway yeah it should prob be us
that just solves
Posted by renesis at 12:57 | permalink | 0 comments
yeah i dont get it
at 5khz your shortest time is 200us?
okay look with a 1ms youre going to have a count of 5
right at 10 ms itll be 50, so almost half midi res
youre scaling tho so i guess youre skipping vals
Posted by renesis at 12:52 | permalink | 0 comments
anyway, how fast does it scan?
like thats a full scan?
yeah thats pretty quick
throw some RC on it
other way than what?
ha
what the isolated switch timing like?
key press is prob like 2-5 ms?
ns?
jezus fuck ns?
whats a slow press?
Posted by renesis at 12:47 | permalink | 0 comments
i have an idea for kbd
i need money
college is stupid
what is your lamecurve
how was this doing velocity again?
and youre trying to do that with scans instead of ints?
or they did do that?
interrupts
some gates or something with a lot of pin change ints
Posted by renesis at 12:42 | permalink | 0 comments
ima find a bum and steal his bike trailer
i think its okay because bums keep stealing my bike lights
Posted by renesis at 09:43 | permalink | 0 comments
'All in a matter of a second'
ive heard this a lot lately and find it completely unimpressive
like 10ms, neat, 0.1s, usable, in one second massives amounts of very bad things can happen
Posted by renesis at 07:52 | permalink | 0 comments
neat, i wonder how they plan to do all that
still think supply chain management gonna go to blockchains big
Posted by renesis at 06:39 | permalink | 0 comments
Top | Add to Technorati Favorites
© 2007 lordpil. XHTML 1.0! CSS! Site design by GNAA Blog Engine by pbx | MULTI2 | ian hanschen | lolwat