|
syscall girl posted:also put a camera up there just because put a motion detector and speaker on it; when it detects motion blare "YOU ARE OUT OF COMPLIANCE, CITIZEN" from the speaker
|
# ? Jul 8, 2014 01:24 |
|
|
# ? Apr 24, 2024 05:07 |
|
Panty Saluter posted:i never understood felt turntable pads, seems like it would hold static electricity and such This thing literally came, all original, in box. . . It's really a pretty thing to have, outside of the crockpot comments. . . I'm waiting on a new audio technica cartridge to come in, along with a phono pre-amp, and a new slipmat. The felt mat don't make sense to me, but considering that fucker was made the same year I was born, I have a piece of 31 year old felt that is still in pristine condition. That's kind wild.
|
# ? Jul 8, 2014 01:33 |
|
SRQ posted:#1, which is now complete and boring: Build the ultimate computer as of 1999 for video games. I play quake and old dos games on it occasionally, it's fun to have. I wish your project was not posting!! lol
|
# ? Jul 8, 2014 02:31 |
|
enotnert posted:full fatty tube driven jerky post some other things that can only be sold to goons
|
# ? Jul 8, 2014 03:02 |
|
Panty Saluter posted:i never understood felt turntable pads, seems like it would hold static electricity and such its dwarfed by the amount of static electricity built up via a spinning vinyl record this is why the separate ground wire used to be super important
|
# ? Jul 8, 2014 07:26 |
|
N.Z.'s Champion posted:Trying to implement Google Material's ripple buttons in a different way. It's rooly ugly right now but it's just a proof of concept ok now it's slightly less ugly but still pretty ugly
|
# ? Jul 8, 2014 07:38 |
|
is there an SPI programmable sound generator that would be comparable to something like an AY-3-8910 or other simple square wave beep blooper?
|
# ? Jul 8, 2014 08:23 |
|
Luigi Thirty posted:is there an SPI programmable sound generator that would be comparable to something like an AY-3-8910 or other simple square wave beep blooper? there's the new AYY-L-MAO, have you tried that one?
|
# ? Jul 8, 2014 08:29 |
|
Trig Discipline posted:there's the new AYY-L-MAO, have you tried that one? that sounds like some kind of designer drug
|
# ? Jul 8, 2014 08:38 |
|
if you want that kind of sound you're talking about someting a lot more complex than just square waves, so try the yamaha ym2612 https://www.youtube.com/watch?v=DCHuoQAqrUU i know nothing about programming them but they make pretty music (it was the fm synthesizer in the sega genesis) the comedy option would be a mos technology SID chip but gat drat those things are hard to find these days https://www.youtube.com/watch?v=MOhzkWZHBfM Sagebrush fucked around with this message at 08:40 on Jul 8, 2014 |
# ? Jul 8, 2014 08:38 |
|
oh you wanted spi. no, probably not, not that i know of
|
# ? Jul 8, 2014 08:44 |
|
two or three pwm outputs on any mcu is basically equivalent to the 8253 timer chip used for sound in an original ibm pc that's pretty primitive compared to a SID, but its a start. write some software envelopes/lfos to modulate the whole mess and you're off to the races. if you really want to get fancy use another timer to clock a switched cap filter dammit now i sort of want to make a synth out of an avr
|
# ? Jul 8, 2014 08:50 |
|
and you'd even be able to have an spi input
|
# ? Jul 8, 2014 08:50 |
|
Base Emitter posted:two or three pwm outputs on any mcu is basically equivalent to the 8253 timer chip used for sound in an original ibm pc somewhere in my closet is a radio shack project kit that I turned into a "synth" and by that I mean a grating oscillator hooked up to a 10k potentiometer to control the frequency it can't be that hard
|
# ? Jul 8, 2014 09:40 |
|
so im trying to work out what kind of led driver i need for an led but the datasheet only gives me the wattage and the voltage of the led. is that enough info to get forward current from?
|
# ? Jul 8, 2014 10:06 |
|
also so leds need a certain voltage to turn on. if the voltage is sufficent can you restrict the current to make them dimmer or is it easier to pwm them
|
# ? Jul 8, 2014 10:07 |
|
echinopsis posted:so im trying to work out what kind of led driver i need for an led but the datasheet only gives me the wattage and the voltage of the led. is that enough info to get forward current from? yea it is. power = voltage * current. if the led is 3.6v and 5 watts then it's (5/3.6) amps = 1.38A white leds are almost always around 3~3.6v so you can estimate the current draw p. easily from the power figure alone echinopsis posted:also so leds need a certain voltage to turn on. if the voltage is sufficent can you restrict the current to make them dimmer or is it easier to pwm them you can restrict the current to decrease the brightness yeah, but restricting the current means burning some of it up in heat and wasting the energy. leds can be strobed at 1000s of hz no problem so it's a lot more efficient to pwm at maximum intensity than to try and screw with the current note: if the led says 5 watts, 3.6 volts you are perfectly capable of running it at 1 watt by only allowing 350ma through. as long as the voltage is right the led will light up, just very dimly if there isn't enough current. you can also overdrive leds, put 2a into that 5w led and now you've got a 7.2w emitter but this will seriously kill the lifetime of the chip (even more so if you don't keep it cool). exponentially too, like if driving a chip at 30ma gives you 50,000 hours, then driving it at 45ma might give oyu 1000 hours and at 60ma maybe you get 10 hr before it starts to get dim and lovely looking Sagebrush fucked around with this message at 10:28 on Jul 8, 2014 |
# ? Jul 8, 2014 10:25 |
|
Sagebrush posted:yea it is. power = voltage * current. if the led is 3.6v and 5 watts then it's (5/3.6) amps = 1.38A quote:you can restrict the current to decrease the brightness yeah, but restricting the current means burning some of it up in heat and wasting the energy. leds can be strobed at 1000s of hz no problem so it's a lot more efficient to pwm at maximum intensity than to try and screw with the current ok cool. im thinking about these leds Input Voltage: DC 12-14V Power: 10W http://www.ebay.com/itm/10W-COB-LED-Lamp-Light-Bulb-12V-24V-1000LM-Warm-Pure-White-for-DIY-120x36MM-/261308193618 so more ampage. but yeah i epxerienced that first hand running 12v into a 12v led with no current limiting and they got dull very fast. in less than a minute thanks though
|
# ? Jul 8, 2014 11:11 |
|
fyi if you go for that cob flat emitter thing you're gonna have to mount it on a big rear end aluminum plate for heat sinking. get some thermal epoxy and glue it down. excess heat is what kills the led, and even a new efficient led at least 60% of the input energy is turning into heat. and all of it is coming right out the back of the semiconductor junction, the worst possible place for it. fun fact this is part of why it's taken so long to get led headlights in cars. the light output is temperature sensitive and you have to avoid both the situation where the emitters get hot in arizona and only put out half the light they're supposed to, and the opposite where the guy in canada starts his car up in the middle of winter and for the first ten minutes his headlights are twice as bright as is legal
|
# ? Jul 9, 2014 00:19 |
|
Sagebrush posted:fyi if you go for that cob flat emitter thing you're gonna have to mount it on a big rear end aluminum plate for heat sinking. get some thermal epoxy and glue it down. excess heat is what kills the led, and even a new efficient led at least 60% of the input energy is turning into heat. and all of it is coming right out the back of the semiconductor junction, the worst possible place for it. interesting. does this mean that possibly watercooling LED arrays in, say, grow room setups might become a profitable effort
|
# ? Jul 9, 2014 02:36 |
|
Sagebrush posted:fyi if you go for that cob flat emitter thing you're gonna have to mount it on a big rear end aluminum plate for heat sinking. get some thermal epoxy and glue it down. excess heat is what kills the led, and even a new efficient led at least 60% of the input energy is turning into heat. and all of it is coming right out the back of the semiconductor junction, the worst possible place for it. thermal epoxy... the poo poo you put on your CPU? ive got some of that. or is it more like a glue? quote:fun fact this is part of why it's taken so long to get led headlights in cars. the light output is temperature sensitive and you have to avoid both the situation where the emitters get hot in arizona and only put out half the light they're supposed to, and the opposite where the guy in canada starts his car up in the middle of winter and for the first ten minutes his headlights are twice as bright as is legal yeah just last night i was reading about using RGB LEDs for poo poo like stage lighting but due to this effect the colour would change and it needs a feedback system to actively control the power to the individual LEDs to maintain a steady colour. cool
|
# ? Jul 9, 2014 03:13 |
|
Jonny 290 posted:interesting. does this mean that possibly watercooling LED arrays in, say, grow room setups might become a profitable effort i imagine "profitable" would be more dependent on the weed market than the physics of the led, but cooling is definitely a good idea if you want max output. it's like any other semiconductor -- works better when cold echinopsis posted:thermal epoxy... the poo poo you put on your CPU? ive got some of that. or is it more like a glue? thermal epoxy is silver bearing has the same effect as the poo poo you put on your cpu, but it also hardens and holds stuff in place. if you're gonna fasten the led to the aluminum plate some other way then regular thermal goo is fine. depends on if you want the easy way or the removable way (thermal epoxy is not coming off) comedy option: jb weld is a metal bearing epoxy, give that a shot
|
# ? Jul 9, 2014 04:18 |
|
hey bloody i used diptrace for a thing and my needs were extremely suited, thanks
|
# ? Jul 9, 2014 20:44 |
|
Sagebrush posted:if you want that kind of sound you're talking about someting a lot more complex than just square waves, so try the yamaha ym2612 huh? no, i don't want to figure out how to program a yamaha FM synthesizer from a datasheet. i'm talking about oooold sound (this game uses 5 of them plus an analog drum machine). it's a 3-channel tone generator plus noise chip that was used in all kinds of early 80s toys and games. apparently the 8910+arduino combo is common anyway since they're dead simple to program and you can run them with a 2MHz crystal, a shift register, and a speaker e: and if I wanted to program a SID i have a C64 already Luigi Thirty fucked around with this message at 22:47 on Jul 9, 2014 |
# ? Jul 9, 2014 22:34 |
|
Sagebrush posted:the comedy option would be a mos technology SID chip but gat drat those things are hard to find these days half the c64s on ebay have the SID removed, because you can sell the SID to a sperg for more than a complete c64 is worth
|
# ? Jul 9, 2014 22:44 |
|
heh so i coded a led flashing light thing on my ardiuno! hahahahacode:
del=del+iterate; if (del>100 OR del<0) { del=del*-1; } this isn't working. i saw something once in a program and it was used to ping pong an iterator. the alternative is using 2 for loops like i did early, one going up and one going down, but that seems inefficient
|
# ? Jul 9, 2014 22:50 |
|
echinopsis posted:del=del+iterate; what are you trying to do
|
# ? Jul 9, 2014 23:22 |
|
i'm confused by what that loop is trying to do? sound chip chat: the SN76489 is another option since it comes in a 16-pin DIP package and is very similar. the 8910 is a 40-pin dip since it has two 8-bit parallel I/O ports in addition to the sound hardware for some reason.
|
# ? Jul 9, 2014 23:28 |
|
oh I think you want "iterate=iterate*-1" instead of del though drop the multiplication and do "iterate = -iterate" just in case the compiler doesn't do it for you
|
# ? Jul 9, 2014 23:42 |
|
oh yeah what a fuckin moron Luigi Thirty posted:i'm confused by what that loop is trying to do? so the idea is that del is iterated up by 1 up to 100 and then down by 1 back to 0 and ad infinitum. when iterate hits the boundary it changes sign anyway the idea is that the flashing goes fast and slows down and then ping pongs back its a left to right and back flashing set of LEDs like in nightrider. onmly i try to make it so that the not active pins have a small amount of brightness relative to their distance from the active LED
|
# ? Jul 10, 2014 00:18 |
|
I was redoing my circuit and I accidentally connected the ground wire coming from my board to AREF on the arduino for a few minutes. would this kill anything? i'm trying to rebuild this two-board system on one board from my schematics (not the PCB I ordered, a different one) and i can't get it to work right. comedy answer: the schematics are wrong e: comedy answer was right, the timer circuit was wrong. Luigi Thirty fucked around with this message at 03:41 on Jul 10, 2014 |
# ? Jul 10, 2014 01:19 |
|
got my first OSHpark fab in today, just some small breakout boards and whatever to see how good they really are. dag, yo. they good.
|
# ? Jul 10, 2014 02:11 |
|
echinopsis posted:oh yeah dno't use delays in ur code unless you actually want the chip to pause execution and ignore all i/os for whatever length of the delay. better way is to use a repeating timer. code:
Luigi Thirty posted:sound chip chat: the SN76489 is another option since it comes in a 16-pin DIP package and is very similar. the 8910 is a 40-pin dip since it has two 8-bit parallel I/O ports in addition to the sound hardware for some reason. this is neat stuff and i'd be interested in finding out what you do w/ it. cheap, low-overhead sound is something i'm always looking for on microcontrollers and those chips seem like neat ways of doing it
|
# ? Jul 10, 2014 04:36 |
|
Sagebrush posted:dno't use delays in ur code unless you actually want the chip to pause execution and ignore all i/os for whatever length of the delay. better way is to use a repeating timer. i know why delay is bad and in other cases i would avoid it but in this case doing what you said seems inefficient coding because ALL this is doing is flashing lights and I have 0 intention of doing anything else with it and its so small and simple anyway if I ever DID want to do more with it it'd be so simple to rewrite. or am I still wrong?
|
# ? Jul 10, 2014 04:41 |
|
Sagebrush posted:this is neat stuff and i'd be interested in finding out what you do w/ it. cheap, low-overhead sound is something i'm always looking for on microcontrollers and those chips seem like neat ways of doing it well if they ever get here and if they actually work i'll do something with them. someone in china either has a warehouse full of them or is making bootleg ones but it's not like a sound IC from 1980 is hard to build
|
# ? Jul 10, 2014 05:00 |
|
i understand that a lot of the sids you can get on ebay/aliexpress/w.e are fakes , but what does that mean? do they still make sounds? like i don't care if they model every filter quirk and rc wobble perfectly, and will make the most perfectly -authentic- sound reproduction of your 1982 video shame sound track, as long as they can still make the different waveforms and envelopes well enough to sound like c64 music.echinopsis posted:i know why delay is bad and in other cases i would avoid it but in this case doing what you said seems inefficient coding because ALL this is doing is flashing lights and I have 0 intention of doing anything else with it and its so small and simple anyway if I ever DID want to do more with it it'd be so simple to rewrite. or am I still wrong? if literally all it's doing is blinking led then nope go nuts, there's no problem w/ using delay. usually the first thing that gets screwed up is if you want to read a button though. your code is stuck in a delay (or god forbid multiple delays within a loop) and you press the button but it can't read it. maybe u can use an interrupt but it's better to just put it on a timer. buttons are pretty common next step once you have leds going (eg turn em off, cycle pattern). anything more advanced you definitely want to use timers to give tehchip execution time and ensure correct timing when other instructions need to run also i wouldn't say it's "inefficient" though. whe you use delay() im pretty sure it just does NOP NOP NOP NOP NOP NOP 16 million times for every 1s of delay. not exactly "efficient" :P Sagebrush fucked around with this message at 05:26 on Jul 10, 2014 |
# ? Jul 10, 2014 05:24 |
|
Sagebrush posted:if literally all it's doing is blinking led then nope go nuts, there's no problem w/ using delay. usually the first thing that gets screwed up is if you want to read a button though. your code is stuck in a delay (or god forbid multiple delays within a loop) and you press the button but it can't read it. maybe u can use an interrupt but it's better to just put it on a timer. yes. an upcoming project is to make a clock for my kids i want to use that 7 segment. pretty straight forward and red LED is in fashion and the neopixel strip (actually 2 of them, on either side of clock) to do a sunrise light thingy like this https://www.youtube.com/watch?v=N9mvZABAoOI there are opther sunrise clocks but not just one light getting brighter but the correct colour (or similar) as a sunrise like in the video but also on the strip of neopixels starting at the bottom and working its way up to the top and then put it behind dark glass so you can only see the lights and not all the other stuff and because i dont want it connected to pc it'll have to have function buttons to set the clock and stuff. (i realise this might make it quite dim.. im not necessarily trying to light the room. id just a single RGB led at the top to do that if necessary and i WONT be able to use delay then at all reminds me i should order a real time clock module and i guess some buttons quote:i wouldn't say it's "inefficient" though. whe you use delay() im pretty sure it just does NOP NOP NOP NOP NOP NOP 16 million times for every 1s of delay. not exactly "efficient" :P echinopsis fucked around with this message at 05:33 on Jul 10, 2014 |
# ? Jul 10, 2014 05:31 |
|
Sagebrush posted:i understand that a lot of the sids you can get on ebay/aliexpress/w.e are fakes , but what does that mean? do they still make sounds? like i don't care if they model every filter quirk and rc wobble perfectly, and will make the most perfectly -authentic- sound reproduction of your 1982 video shame sound track, as long as they can still make the different waveforms and envelopes well enough to sound like c64 music. never bought one but according to a google search apparently the common problem is non-working filters. i would have assumed they'd be any random DIP with the numbers labelled off, i'm not sure how they're coming up with a SID-like chip without a filter. maybe its an unsuccessful cloning project (the filters the only real analog portion) fyi re your other post in A/T i don't think SID does FM synth, iirc it's just a straightforward oscillator-filter-VCA chain. there are fm synth chips on soundblasters and so on, though, i don't remember which chip that is
|
# ? Jul 10, 2014 05:53 |
|
the yamaha ym2612 sounds like an interesting one, possibly still manufactured, and it does fm synth apparently. then again might suffer the same probs as the knockoff sids. seems like a neat project to mess w/ though. today's spare time project: i fixed my speakers finally. i have had these nice little yamaha m10s since i was about 12 years old (my dad got em for me) and for the last, oh, five years they've been occasionally popping and buzzing and then getting super quiet and making a 60hz hum when you turned the volume up or when you pushed the input wires around. never really cared that much cause you just shake the right speaker and it glitches back in. so today i got sick of it and decided to open them up, figuring that my electronics knowledge had advanced enough to probably diagnose the problem, and it was: a cracked solder joint on the input plug. durr. i could have fixed that with my radio shack soldering iron when i was 8. re-soldered that one and all the other plugs too for good measure and it's rock solid now Sagebrush fucked around with this message at 06:18 on Jul 10, 2014 |
# ? Jul 10, 2014 06:13 |
|
|
# ? Apr 24, 2024 05:07 |
|
yr clock is pretty racist imho
|
# ? Jul 10, 2014 06:16 |