digshadow has quit [Ping timeout: 240 seconds]
ZipCPU|Laptop has joined ##openfpga
azonenberg_work has joined ##openfpga
flaviusb has quit [Quit: Leaving.]
flaviusb has joined ##openfpga
digshadow has joined ##openfpga
azonenberg_work has quit [Ping timeout: 248 seconds]
digshadow has quit [Ping timeout: 240 seconds]
digshadow has joined ##openfpga
m_w has joined ##openfpga
ZipCPU|Laptop has quit [Ping timeout: 260 seconds]
<rqou> azonenberg: so my second bhm arrived, and it seems to be in better shape
<rqou> but it doesn't have original olympus eyepieces but some third-party one
<rqou> is it worth swapping?
<rqou> i'm still working on union-ing my two mediocre microscopes into one good microscope
<jn__> you could combine your two mediocre microscopes into one mediocre stereo microscope ;)
balrog has quit [Ping timeout: 248 seconds]
<azonenberg> Eyepieces are not super critical
<azonenberg> But if you can get olympus ones they are probably better than no-name ones
<azonenberg> that said i have not noticed a diff between olympus and amscope
<azonenberg> if you spend your money on one thing, get good objectives
balrog has joined ##openfpga
<rqou> hmm ok
<rqou> oh btw azonenberg
<rqou> microscope 2 that i obtained was definitely set up for DIC at one point
<rqou> it has an analyzer installed
<rqou> but somehow no polarizer?
<balrog> rqou: does it have the prisms and all?
<rqou> it has two prisms that are mismatched with the objectives
<balrog> :/
azonenberg_work has joined ##openfpga
digshadow has quit [Ping timeout: 240 seconds]
pie_ has quit [Ping timeout: 240 seconds]
nikitakit has joined ##openfpga
digshadow has joined ##openfpga
digshadow has quit [Ping timeout: 248 seconds]
nikitakit has quit [Ping timeout: 260 seconds]
Hootch has joined ##openfpga
_whitelogger has joined ##openfpga
digshadow has quit [Ping timeout: 248 seconds]
<rqou> ping azonenberg azonenberg_work
<azonenberg> o/
<rqou> two things
<rqou> 1) will you ever test the no-libudev PR?
<rqou> 2) if you ever see nikitakit log on again, poke him
<azonenberg> I've been incredibly busy lately
<azonenberg> :p
<rqou> i thought you were finally (fingers crossed) done with traveling for a while?
<azonenberg> Yes, i got back this weekend finally
<azonenberg> Now i'm catching up on all the stuff i didn't do over the last six weeks
<azonenberg> both around the house and things i've promised people i'd do :p
<rqou> ah
<rqou> anyways, thing 2 is more important
<azonenberg> ok
<rqou> nikitakit is someone i know IRL who said he would be interested in PAR algorithms
<azonenberg> And i still have that first aid class next week that i have to read a 300-page medical textbook for
<rqou> but explicitly no RE
<azonenberg> i'm on page 138 of the first pass so far...
<rqou> sounds fun
<rqou> i need to lern2rotations by thursday morning
<rqou> (will probably annoy oeuf about it)
<rqou> and i have two phone interviews coming up
<rqou> and i still need to actually send in job applications to more companies
<azonenberg> Fun fun
<azonenberg> Yeah, openfpga stuff is probably on hold for the next ~2 weeks for me due to more pressing obligations
<rqou> so far one company is "eh, i would take it" and the other company i definitely won't take and am just doing interviews for practice
<azonenberg> after that i'm going to see where things are, probably get back to the greenpak characterization
<rqou> damn you're super behind too :P
<rqou> and i thought the problem set life was bad :P :P
digshadow has joined ##openfpga
digshadow has quit [Ping timeout: 240 seconds]
digshadow has joined ##openfpga
sgstair_ has joined ##openfpga
sgstair has quit [Disconnected by services]
sgstair_ is now known as sgstair
pie_ has joined ##openfpga
pie_ has quit [Ping timeout: 240 seconds]
sgstair_ has joined ##openfpga
sgstair has quit [Disconnected by services]
sgstair_ is now known as sgstair
sn00n has quit [Read error: Connection reset by peer]
sn00n has joined ##openfpga
digshadow has quit [Ping timeout: 248 seconds]
m_t has joined ##openfpga
sgstair_ has joined ##openfpga
sgstair has quit [Disconnected by services]
sgstair_ is now known as sgstair
pie_ has joined ##openfpga
<sn00n> are the BHM still a good choice?
enriq has joined ##openfpga
enriq has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
pie_ has quit [Ping timeout: 240 seconds]
sgstair_ has joined ##openfpga
sgstair has quit [Disconnected by services]
sgstair_ is now known as sgstair
xdeller has quit [Quit: Leaving]
enriq has joined ##openfpga
Hootch has quit [Ping timeout: 264 seconds]
Hootch has joined ##openfpga
defparam has quit [Quit: Oh no... i'm going down...]
digshadow has joined ##openfpga
pie_ has joined ##openfpga
xdeller has joined ##openfpga
test123456 has joined ##openfpga
eduardo_ has joined ##openfpga
eduardo__ has quit [Ping timeout: 264 seconds]
Hootch has quit [Ping timeout: 248 seconds]
sgstair_ has joined ##openfpga
sgstair has quit [Disconnected by services]
sgstair_ is now known as sgstair
<rqou> azonenberg_work: apparently you were right and hardware store acetone really does leave residue if you try to use it to clean optics
<rqou> ordering reagent grade now
<qu1j0t3> ugh!
<qu1j0t3> at least the new acetone can clean the old acetone off
<azonenberg> rqou: well the other thing is, even if the solvent itself is pretty pure
<azonenberg> it will get residue from the surface clean
<azonenberg> surface being cleaned*
<azonenberg> So what you need to do is flood the surface with enough clean solvent to remove the dirty solvent
<azonenberg> A common practice for silicon wafer cleaning is acetone followed by methanol or IPA, then an optional DI water rinse
<azonenberg> then blow dry with N2
<shapr> google says IPA = isopropylalcohol, turns out it's not a beer
<azonenberg> shapr: lol
<azonenberg> Yes, in the lab context when someone says IPA they mean 2-propanol
<azonenberg> Not beer :p
<shapr> good to know :-)
<azonenberg> rqou: so baiscally you need pure solvents but you also have to use enough that the solvent evaporating from the surface isnt contaminated
<enriq> has more alcohol % indeed :)
<enriq> good afternoon
<rqou> azonenberg: what reagents do you forsee me needing other than acetone, methanol, and isopropanol?
<azonenberg> I dont actually like methanol for toxicity reasons
<rqou> meh, i got a respirator already :P
<rqou> apparently the recommended optics cleaning solution is acetone+methanol
<rqou> this evaporates slower than pure acetone
<azonenberg> I prefer reagent alcohol
<azonenberg> It's 90% ethanol / 5% methanol / 5% IPA, you can get ACS grade that is pretty free of solid impurities etc
<azonenberg> has enough methanol to be considered dantured and is nondrinkable / not taxed as a beverage
<rqou> hmm ok
<azonenberg> but is a low enough concentration that it's not quite as nasty as pure methanol
<rqou> wait wait wait
<rqou> potassium nitrate is used in food curing?
<rqou> that makes things so much easier :P
<azonenberg> Yeah, you didnt know?
<azonenberg> and why
<rqou> (for obtaining fuming nitric acid)
<azonenberg> lol
m_w has quit [Ping timeout: 240 seconds]
<azonenberg> if you dont need live decap
<azonenberg> i say just use 70%
<azonenberg> thats easy enough to find
<rqou> fuming nitric acid is annoyingly difficult to obtain, but i can just distill drain cleaner + potassium nitrate
m_w has joined ##openfpga
* azonenberg would absolutely not want to distill HNO3
<azonenberg> but thats just me...
<azonenberg> and dig has no problem getting FNA
<rqou> but "meat curing compound" and "drain cleaner" are probably even easier :P
<azonenberg> lol
<rqou> btw nitric acid distillation temps are way lower than sulfuric
<azonenberg> i'm not concerned about temps
<azonenberg> i'm concerned about a) strong oxidizer @ elevated temperature
<azonenberg> and b) lots of NO2 vapor
<azonenberg> if something leaks
<rqou> btw, phenolphthalein or digital ph probes?
* azonenberg currently uses universlal indicator paper
<azonenberg> but i dont do titrations much/ever
enriq has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<rqou> hmm do i really need titrations or do i just assume "fumes --> good enough" for nitric acid? :P
<azonenberg> again, if you are not doing live decaps
<azonenberg> sulfuric drain cleaner is adequate with no reaction/purification
<rqou> but i do want to try a live decap
<azonenberg> You can do live with sulfuric although results w/ copper bond wires are mixed
<azonenberg> Also, conjecture: HNO3 can be dehydrated at room temp with some kind of zeolite/molecular sieve
<rqou> nope
<azonenberg> I just dont know what material, if any, absorbs/adsorbs a lot of water without being eaten by HNO3
<rqou> according to "those canadian youtube chemists," acids destroy zeolites
<azonenberg> I didnt literally mean actual zeolite
<azonenberg> i meant, something with a similar structure of hydrophilicity
* shapr googles that word
<azonenberg> i suspect something has similar properties but is stable at low pH
<shapr> oh, it's just hydrophilic as an adjective, duh
<shapr> er, adverb
<azonenberg> It just isnt practical do to do at scale, so they use distillation for mass prod
<azonenberg> But at lab scale it might be doable
<rqou> but why would you use a slow adsorption process if you can just buy drain cleaner and meat curing compound?
<azonenberg> Because distilling
<rqou> why do you hate distilling?
<azonenberg> I dislike distilling corrosive, toxic materials when there are safer alternatives available
<azonenberg> Like just buying WFNA
<rqou> hmm, where?
<azonenberg> Havent looked, i know of plenty of places that sell 70%
<rqou> that's much easier
<azonenberg> so if you're distilling anything start with that
<rqou> that doesn't work
<azonenberg> but still, just buy it
<azonenberg> sigma etc probably sells it
<rqou> it's azeotropic
<rqou> i've been told dealing with sigma is a massive pain
<azonenberg> thats why you mix it with a dehydrating agent w/ low vapor pressure
<azonenberg> Like sulfuric
<azonenberg> if you distill sufluric + KNO3 you still get an azeotrope, no?
<azonenberg> also you can probably get WFNA from digshadow :p
<rqou> sulfuric+KNO3 doesn't get an azeotrope
<rqou> where would the water come from?
<azonenberg> ah so just solid KNO3, not aqueous
<rqou> yeah
<azonenberg> Also wow ebay is cracking down i guess
<azonenberg> you used to be able to get at least 70% and possibly stronger there
m_w has quit [Ping timeout: 248 seconds]
sgstair_ has joined ##openfpga
sgstair has quit [Disconnected by services]
sgstair_ is now known as sgstair
<rqou> oh i just discovered the siliconpr0n decap page
promach has quit [Ping timeout: 246 seconds]
<rqou> azonenberg: i like the "disadvantages:" "Multiple healthy/safety hazards"
<azonenberg> lol
<rqou> not just one, but multiple :P
enriq has joined ##openfpga
m_w has joined ##openfpga
<rqou> anyways, i really need to work on how to clean up without constantly swap thrashing
<rqou> lab maintenance is hard
<azonenberg> rqou: lol, yes it is
<azonenberg> i'm having this problem here
<rqou> swap thrashing?
<azonenberg> also trying to figure out how to rearrange my desk to make more efficient use of space
<rqou> lol i'm still working on that
<azonenberg> well i want to move back to the L shape i had in the past
<azonenberg> But i might have to move some furniture around to make that practical
<rqou> i can't have an L at all
<rqou> instead now i have three desks in a row
<azonenberg> it'd stick out into the middle of the room
<azonenberg> but i think i can do it
<rqou> i definitely can't because the bed needs to fit in this room too :P
<shapr> join a nearby hackerspace?
<azonenberg> rqou: easy solution
<azonenberg> nix the bed
<rqou> lol
<rqou> where would i sleep?
<azonenberg> Foam pad on the floor
<azonenberg> :p
<azonenberg> Only half kidding
<azonenberg> Couch?
<rqou> but those take up more space than the loft bed i have now
<rqou> (╯°□°)╯︵ ┻━┻ there is a protest again today (╯°□°)╯︵ ┻━┻
<rqou> f*ck you nazis
<jn__> who's protesting for/against what?
<rqou> alt-right is trying to arrange a protest on UCB campus again
<rqou> azonenberg: what's your favorite "victim" chip to get started with decap?
<azonenberg> pic12f683
<azonenberg> somewhat well studied, known semi-invasive UV attack (although new chips have copper bond wires that take a bit more work to preserve)
<azonenberg> older ones were gold
<azonenberg> large optically readable 350nm process
<rqou> do i need a pickit or whatever?
<azonenberg> To program it? yes
<azonenberg> pickit2 or 3 will both work
<rqou> no legacy 12v mclr programmer?
<rqou> or does the programmer still do that?
<azonenberg> hmmm i forget if the 350nm parts need it
<azonenberg> If they do use it, it's just an enable to say "go into programming mode"
<azonenberg> its not actually used internally for flash, they have an on die charge pump
<rqou> ugh pic ICSP suucks
<azonenberg> Why?
<azonenberg> it's a legacy protocol from ages ago, and is simple as a result
<azonenberg> greenpak4 programming is no different, except it's a new protocol and is simple for costs savings
<rqou> needing the 12V is already annoying enough
<azonenberg> pickit provides it
<rqou> and it's IME really slow too
<azonenberg> if it's used at all
<azonenberg> i'd have to read the spec
<azonenberg> pic32 ICSP is very similar to SWD
<azonenberg> it's JTAG muxed onto a bidir data + clock
<rqou> that's much better
<azonenberg> pic32 also has full jtag
<azonenberg> (which i htink you have to enable over the 2-wire bus before you can use it)
<azonenberg> but they're 180nm
<azonenberg> actually it might be on by default?
<rqou> what about AVR?
<azonenberg> either way its too big and complex to do as a first RE project
<azonenberg> old avr is 350nm too, but i dont know anything about the ISP protocol
<azonenberg> re pic12 isp
<rqou> is debugging closed?
<azonenberg> you know, i havent looked
<azonenberg> i havent really touched the 8 bit pic stuff in ages
<azonenberg> pic32 debug is MIPS EJTAG which is well documented afaik
promach has joined ##openfpga
m_w has quit [Ping timeout: 260 seconds]
enriq has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
pie_ has quit [Ping timeout: 255 seconds]
m_w has joined ##openfpga
m_w has quit [Ping timeout: 240 seconds]
<cr1901_modern> azonenberg: "if it's used at all" Very likely it is... actually PIC programming annoys me for a similar reason. It can't be done by "bitbanging from a 3.3v GPIO chip like FTxxx" without extra hardware
enriq has joined ##openfpga
pakesson has joined ##openfpga
<azonenberg> So a lot of the newer XLP pics have low voltage programming options
<azonenberg> the 350nm ones do not
<azonenberg> on that note i plan to make a pic programming io card for starshipraider at some point
<rqou> azonenberg: did you hear about my troll-y idea for a "red magic probe" "blue magic probe" and "white magic probe?"
<rqou> there's an open hardware project called "black magic probe" that is a debugger tool for cortex-m
<rqou> it handles all of the debug logic itself, so you just "target remote /dev/tty*" in gdb to debug
test123456 has quit [Quit: Leaving]
<rqou> my troll-y idea was to make a "blue magic probe" to do the same thing for AVR
<rqou> and then a "red magic probe" for PICs
<rqou> and a "white magic probe" to be a unified build
<azonenberg> i would just make it starshipraider plugins
<azonenberg> then you do target remote starshipraider:4141
<rqou> but these are much cheaper :P
<azonenberg> In other news
<azonenberg> i found an xc6slx25 and a magjack in the garage sitting around from abandoned projects
<rqou> oh goddammit
<rqou> this is why you're always distracted :P
<azonenberg> Meaning, my BOM for this board is gonna be an LDO if i cant find a 1.2V one handy
<azonenberg> and a few resistors
<azonenberg> This is gonna be awesomely fun and cheap and simple to design
<rqou> but it takes up a bunch of time
<azonenberg> And an excuse to get rid of my last spartan6 (actually i think i might have a lx16 sitting around somewhere...)
<azonenberg> Yes, but i need to stay sane
<azonenberg> (ish)
<azonenberg> I'm REing obfuscated java @ $DAYJOB
<azonenberg> gotta unwind somehow
<rqou> oh you probably know about this already, but i recall a person who wrote a tool to help with this
<rqou> he took a dictionary of verbs and nouns
<rqou> and auto-renamed every class with a noun and every function with a verb
<rqou> un-overloading all functions
<azonenberg> Lol
<rqou> apparently this helps a lot
<azonenberg> This obfuscator just has like a.b.c.a
<rqou> so instead of a.a.a.a.a.a(b,c,d)
<azonenberg> nothing fancy/long
<azonenberg> i'm just using eclipse auto refactoring
<azonenberg> they didn't obfuscate strings
<azonenberg> or any of the external libraries they used
<azonenberg> so interface names and interface-implemented function names are cleartext, and strings
<azonenberg> its mostly just tedous renaming
<rqou> apparently it helps the human brain when you change that into e.g. cat.dog.apple.pear.eat(leaf,hotdog)
<rqou> at least comparing this is less error prone than single letters
<azonenberg> yeah, i've mostly been naming methods by what their contents do
<azonenberg> or at least trying to name the classes
<rqou> of course
<rqou> this tool i read about was intended as a first pass
<azonenberg> like, if i see a bunch of JsonObject stuff I'll call it DoJsonStuff(0
<azonenberg> ()*
<azonenberg> then figure out what it does with the json later
xdeller has quit [Quit: Leaving]
<azonenberg> Or SubsystemXClass1, SubsystemXClass2, SubsystemYClass1, etc
<azonenberg> I do the same thing with functions when REing assembly
<azonenberg> except disassembly tends to not be full of fluff :p
<rqou> so one thing i hate about REing java is that convincing javac to allow me to invoke a method for testing is ridiculously hard
<azonenberg> i generally go full static
<rqou> last time i tried this i f*cked around with classpath for hours and eventually gave up
<rqou> in general classpath and/or the java compiler model just doesn't make sense
<azonenberg> java generally seems like a huge pain in the neck
<azonenberg> i understand the C model :p
<azonenberg> and HDLs
<rqou> fun challenge: try to make java, scala, and clojure work with a circular dependency
<rqou> afaik the only way to do this is to start with them all be java, and then rely on the old .class files to replace the scala/clojure parts
<azonenberg> yes and then how do you do a clean rebuild?
<azonenberg> My response: "don't do that"
<azonenberg> :p
<rqou> you're screwed :P
<azonenberg> If you can't build from a clean slate, your architecture is fscked
<azonenberg> Period
<rqou> make them separate .jars and call it a day :P
<rqou> each build involves grabbing the last good build of the other projects from the build server
<azonenberg> ...
<rqou> but i can actually see this happening
<rqou> imagine each jar is a whole project, originally done by three separate companies
<rqou> and then a huge M&A happens :P
* azonenberg hears "enterprise" and runs the other way
<rqou> now what if they hire you to do a security audit? :P
<azonenberg> Then i do static analysis
<azonenberg> Also, it's pretty rare i do java
<azonenberg> this is actually an embedded linux gig and they just needed help on one section
<azonenberg> and i wasnt booked
<azonenberg> i havent touched java in months before this, and i don't miss it one bit :p
<rqou> i take it you wouldn't enjoy working on a minecraft mod? :P
<azonenberg> Um, no :p
<rqou> but then you can touch java, classloaders, opengl, and drama all at once :P :P
teepee has quit [Ping timeout: 240 seconds]
* azonenberg hugs plush CPLD and quivers in fear
<rqou> wait do you actually have one?
<azonenberg> Lol i wish
<azonenberg> i was actually talking with ally a while ago about making a full PCB bed set
<azonenberg> The pillows would look like crystal oscillators with pink 'ceramic' bottoms, four gold pads on the underside, and an Abracon logo + frequency marking on top
<azonenberg> then some plush ICs
<rqou> you mean some dakimakuras? :P
<azonenberg> the sheet would be green with traces and gold pads applique'd on
<rqou> you should bring a CPLD dakimakura to a weeb con
<rqou> just for lulz :P
<azonenberg> and well, idk of any chip that looks like that
<azonenberg> actually...
<azonenberg> i know of a chip that would work
<azonenberg> but it wouldn't match the scale i'm going for
<rqou> giant 68k? :P
<rqou> long DIP is long
<azonenberg> i actually calculated a consistent scale for say a 5x7 mm oscillator to match a standard bed pillow
<azonenberg> But a TXB0108 in WLCSP, while the right size/shape
<azonenberg> is about 10x too small to match
teepee has joined ##openfpga
pie_ has joined ##openfpga
uovo has quit [Quit: moo.]
oeuf has quit [Read error: Connection reset by peer]
oeuf has joined ##openfpga
oeuf is now known as egg|zzz|egg
<rqou> wtf UCB is actually showing off those cat ear headphones as a success
<rqou> apparently they went through one of the university's startup accelerators
<azonenberg> loool
<rqou> i mean, they did eventually ship it, which is already better than a lot of hardware startups
<azonenberg> lol
sgstair has quit [Quit: .•«UPP»•.]
sgstair has joined ##openfpga
enriq has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<qu1j0t3> NVIDIA releases an open-source Deep Learning Inference
<qu1j0t3> chip design (based on Xavier), with full verilog source:
ZipCPU|Laptop has joined ##openfpga