/diy/ - Do It Yourself

>>1652317
the 80s were great OP

>>1652317
ive been wanting to build an Atari Punk Console like this for ever.
https://www.youtube.com/watch?v=NLewGdYa1ts the guy has a link in the description for a schematic.

Trimming out a front door

>>1652573
Man, I'm getting bad vibes about that microwave.

>>1652317
I'm into analog video synths. That's VCO's adders, delay lines, etc like in audio synths but used to generate video patterns. I've been designing and building the modules for this bad boy for the last six months and now it's time to put it all together. I've actually got enough modules done to fill in another two huge cases. I designed it to produce a vaporwave aesthetic. It can do that wireframe mountain sort of thing.

>>1652642
what the shit

link?

>>1652727
I haven't done any documentation yet but I'm happy to answer questions.

This guy generates circles by solving the Pythagorean theorem using op-amps. When combined with other modules it can make visuals like crazy interdimensional wormholes or a moon reflecting on a turbulent sea.

>>1652737

>>1652737
Holy fuck please elaborate and show pics, this is one of the cooler things I saw on /diy/

>>1652642
This is absolutely baller, please show results

>>1652738
Noice

>>1652739
>>1652740
To the left is a wave folder. These are commonly used with audio and make a metalic sort of noise. When used with video you're sort-of going from a nice smooth analog pattern to wireframe.

To the left is a "number muncher". It's a bunch of comparators and diode-logic that renders a 7-segment digit onto non-euclidean space.

>>1652747
This is what you get when you combine them. The tiling effect comes from how the folder chops the space into a bunch of smaller rectangles.

>>1652748
>>1652747
That's awesome

All of your own design?

>>1652748
>>1652738
Can't get enough of this stuff, pls post moar

Also, would you happen to have some info/links to theory behind this or maybe something you used for thinking these things up?

>>1652748
This is cool as fuck my dude.

>>1652317
I'm making a modular synth in a suitcase. Waiting for the plywood to arrive, desu. I still don't know how many of what modules I wanna build.

>>1652865
> would you happen to have some info/links to theory behind this

Unfortunately it's an ancient dark magic I had to figure out for myself and there's basically no information online to make that easier. My end goal is to sell these modules as DIY kits with beginner friendly booklets. I've been using nothing but through-hole components so anyone can solder them.

This is the video encoder/decoder board. It takes in three voltages representing H,S,V and generates an NTSC signal. It also extracts this data from an external signal and can synchronize to the sync pulses and chroma-subcarier of it. This means it can add effects to an external video source as well as generating a new one from scratch.

>>1652737
you are a genius and i want to follow this project, you should make regular updates that thing is bad ass

>>1652893
different anon, but if you made a youtube channel where you go over some theory and show some of the results, I would most definitely subscribe and watch them religiously.
This shit is something I've always wanted to do, but never had the resources to do.

>>1652893
These are the function generator boards. I have some that generate horizontal patterns (frequencies that oscillate within a scanline) and some that generate vertical patterns (frequencies that oscillate over dozens of scanlines). The Pythagoras and number modules take in both these signals and generate an output that's some f(x,y).

They're based on the ICL8038 voltage controlled function generator chip with some extra bits to add voltage controlled amplitude. That chip isn't manufactured anymore so I've got to redesign everything. Having these modules modulate themselves and eachother can generate some very elaborate patterns with only two or three of them.

>>1652893
Ah I see. If and when you do, please make a thread for that, I know for a fact many anons would find that very interesting. I've been looking into some small projects for CRTs but this is the final redpill on that.

>>1652737
>This guy generates circles by solving the Pythagorean theorem using op-amps
what the actual fuck

explain this man, holy shit, can this used to make neverending melodies?

>>1652747
>wavefolder
>number muncher
what PCBs are you using for these?

>>1652898
>ICL8038
fucking aaaaa
being literally experimenting with that chip a month ago, but dropped it because found the sinewave too distorty for what i wanted.

some questions:
are you using bipolar +15/-15?
are you FM modulating it? via what pin? 7?
ever had trouble with purity of the sine wave?

are you putting these as kits soon? id buy it all lmao

>>1652893
>Unfortunately it's an ancient dark magic
Not that ancient, the old electronics magazines often had video synth projects, there was at least one book on the subject in the 70s or so and a few video synth kits were produced as well. Quite a few of the synth magazines showed how to use a modular audio synth to do video work with a few extra modules.

This is mostly a mixture of analog computing circuits and synth circuits.

Good work but stop trying to be special, you are not the sole source of this knowledge, it is just mostly forgotten. Video synthesis was a common practice of those in the DIY modular synth community in the 70s and 80s.

>>1653090
>>1652893
Forgot to say, well done, it is an impressive bit of work.

>>1653090
why does he need to solve the pythagoras? wtf. and that's more easily done with microcontrollers. and 'solving' with op amps sounds like a joke.

>>1653098
Listen anon its just a bunch of old fags fapping each other in a dirty rest room

>>1652893
But all of pattern generation can be done quite easily with a, gulp, arduino, right? Analog stuff is cool but you can code any kinds of fancy patterns easily.

>>1653098
>>1653106
>bro why do anything analog when you can just use a microcontroller
Fuckin' zoomers, man

>>1653106
>>1653098
Analog computing gives infinite resolution and adjustability along with no need for DAC to feed the analog display. It is a simple process to execute the basic mathematical functions of addition, subtraction, multiplication and division in the analog realm with quite simple circuits, addition and subtraction are exactly what the opamp was designed to do and their first uses back in the vacuum tube days was in analog computers.

Analog computers still have some advantages over digital and have never gone away, they are still used in many places in science.

>>1652317
I was thinking the other day that I'd like to make diy mastering modules so I can fine-tune the sound of my nicer bookshelf speakers. Maybe a compressor, eq, filters, etc. If anyone has suggestions for resources towards accomplishing that it would be appreciated

>>1653126
>so I can fine-tune the sound of my nicer bookshelf speakers

I guess actually listening to music is out of the question.

Spooky synth

https://www.youtube.com/watch?v=4TlchgzKW18

>>1653072
>>1653098
Here's how to solve the Pythagorean theorem with op-amps. Opposite and adjacent sides of the triangle are represented by analog voltages. You can then use a sequence of logarithmic and anti-logarithmic amplifiers to get an analog voltage for the hypotenuse.

>>1653074
I design my own circuits and the PCBs for them.

>>1653078
I'm powering the ICL8038 with Gnd/12v and FM modulating with pin 7. The sinwave is pure enough for my application.

>>1653090
The problem is almost every chip specifically indented for analog video stopped being manufactured a decade ago. You can't copy the circuits out of books from the 70s and 80s anymore.

>>1653106
>>1653098
Programming microcontrollers is my day job. I want to do something else as a hobby. Also with an Arduino the best you can hope for is Atari 2600 blocky slow moving graphics.

>>1653126
>Maybe a compressor, eq, filters,
top quality 80s and 90s racks are dirt cheap. you wont diy anything of worth besides some tube pre for phonos.

>>1653179
is that a tranner?

>>1653390
>just saw the hands
nvm.

>>1653388
>dirt cheap
How cheap are we talking here?

It's not really fair to say the opamps are solving anything other than the differential amplifying function they normally do, with gain controlled via feedback.

The diodes are used in the feedback loop to take advantage of the exponential in the diode current function to produce the opposite effect on the output voltage: log(x).

Anyone with basic level algebra should know you can pretty much compute everything once you have exp(x) and log(x) since multiplication and division, roots and powers become simple addition and subtraction.

So since functions like tan(x), sin(x) and so on are all actually symbols representing combinations of addition+subtraction and logarithms, there you have it.

Have you tried a wave multiplier ala serge? Using the non-linear transform via a diode clamp on each stage (positive inputs) in combination with difference from the input (negative inputs) you can get a function with tons of zero-crossings.

This produces a similar effect to a "wave folder" which is really just a single stage rather than being in series.

Common series combinations are 3, 4, 5 and 6, since noise becomes a major issue with greater stages.

>>1653413
If you can pick them up yourself you'll find stuff for <$100. The reason for that is almost absolutely everything done in analog audio can be beaten once you're running at 4x rates such as 192k.

The only things that can't be beaten (non-linearity, distortions, etc) are honestly a trade-off... since high sampling rates = lower aliasing = higher processing cost = zero noise vs. very high noise levels (~-80 dB minimum if you're lucky) for analog processing.

As far as your typical mastering setup with shelf EQs, compressors and so on you'll have to decide for yourself. Nearly everything today though is mastered digitally. A lot of sources are still analog since it's very difficult to beat analog synthesis, effects and so on if any non-linearity is present. That's assuming chaos is desirable though, and where you need predictability and reproducibility (such as preset settings!) digital still wins.

>>1653270
>The problem is almost every chip specifically indented for analog video stopped being manufactured a decade ago. You can't copy the circuits out of books from the 70s and 80s anymore.
The chips that were designed specifically for it were largely of the 80s, the arrival of affordable video cassette recorders made them a viable product with a market large enough to justify IC fab. People have been hooking modular synths and analog computers to cathode ray tubes since at least the late 60s, the electronic magazines and journals are full of circuits designed around standard parts which even if are out of production are easily subbed, and they give a very good bit of information on theory and circuit design as well. If memory serves, PAiA had a simple video synth built around common cheap parts, there was at least one book on the subject that went well into theory. There is a great deal of information on this subject for anyone willing to do some digging.

>>1653419
It is perfectly fair, it may not be the complete story, but perfectly fair, you are just a pedant.

>>1653431
IC fabs have also, generally speaking, become more expensive to set up while being usually cheaper per component over time. They've become more and more strongly mass production oriented. You can still get space on a mask but it's not a common service and costs about as much as having a whole mask produced as 15-20 years ago. Some places more or less specialise in the kind of processes of the late 90s/early 00s though, that's why you have some suspiciously late entrants like some of the elbrus processors on nodes of that period, it's in some ways cheaper.

is it possible to mod a yamaha synth to have audio input so i can send the audio from my computer to the synth speakers ?

>>1652317
never heard of the Fucking 18 is it a famous brand?

>>1653431
>you are just a pedant.
Well, I am an engineer...

>>1653482
The question is why would you want to? It's likely better to use the output, mix elsewhere and send the mixed signal to monitor speakers or headphones.

Love these threads. I still haven't started this delving into since the last thread though, apart from playing with programmable synths like Pure Data, Csound SuperCollider and Nyquist.

I'll probably finally buy a soldering iron next month and get to work.

>>1653472
>it's in some ways cheaper.
go back to the 70s now, IC fab was fairly expensive back then, regardless of quantity, it was the 80s when it became cheap enough to justify smaller runs aimed at general consumption. Are you a pedantic engineer as well? It is an interesting time though, extreme low quantity IC fab is getting quite affordable!

>>1653566
>Well, I am an engineer...
Well then, to be fair no one actually said it was the opamps that solved it.

>>1653579
>Pure Data, Csound SuperCollider
I was really pushing those three hard in that thread, are you the one that I discussed building a synth around a small embedded system like a rpi and run one of the above as the back end?

>>1653568
I want the sound from a DAW to go to the speakers of the synth

>>1652608
based and banditpilled

>>1653111
yeah grandpa, you should solve the phythagorian with vacuum tubes for warmth. don't forget the NOS capacitors.


>>1653115
> infinite resolution
but the resolution of your tv is not infinite. so what's the point. and 24 bit DACs are dirt cheap, and if that's not enough you can upgrade to 32. or you can build your own DAC if you are bored.

> analog display
yikes, why? an infinite resolution on a low res analog display??

> in the vacuum tube days was in analog computers.
wow, so i just joked about vacuum tubes in my reply to the boomer above and you seem to be serious... wow. yeah let's go back to tubes and and analog tvs in 2019. you guys are trolling right?
>>1653270
> I want to do something else
ah finally the real answer. well that's at least a fair excuse. it is completely insane from the practical point of view but i admit i can see how that can be fun. you should really use tubes man.

>muh slow arduino
well that's just an example you can use stm32f7 at 200MHz or something even faster. you can use FPGA for extra speed.
so stm32f7 + 24bit DAC + digital TV will probably look a lot better than your current setup but i can see how that can be boring comparing to solving equations with opamps...

>>1653115
>they are still used in many places in science
like?

>>1653740
Cool your autism there, champ.

>>1653740
>the resolution of your tv is not infinite. so what's the point
I was giving an advantage of analog computers, nothing more.
>yikes, why? an infinite resolution on a low res analog display??
Because that is what OP used, don't ask me.
>wow. yeah let's go back to tubes and and analog tvs in 2019
I gave some history, I did not suggest using tubes for anything.

You really have poor reading comprehension.

>>1653744
Mostly as small purpose built computers, but the larger heavily programmable ones are coming back as well, mostly as hybrid analog/digital computers on a single chip. Far from production things though, mainly built in very very low numbers for single labs.
Google can help some, but their results seem to focus on the big variety and bickering about whether or not we should bother with them, but I did not dig much, no time to spoon feed you right now.
https://www.google.com/searchq=analog+computers+used+today

>>1653755
>no time to spoon feed you right now
You said they are still used today yet your source is "lol, just google it"?
Everything I see today in science ("scientific computing") is digital.

>>1653779
Not him, but to be fair it sounds like the grand sum of your lab experience is in undergrad lab courses.

>>1653656
>go back to the 70s now, IC fab was fairly expensive back then, regardless of quantity
Uh, sort of. 70s fabs were more like lab scale production, the feature size on the silicon is the same or not far off the feature size you find on real good homebrew boards now (approx 1 mil). Expensive for different reasons. Some people are starting to build 70s level of technology silicon stuff again in their garages so if you're interested in that it doesn't seem like an impossible task.

The 80s and 90s are known as the golden age of silicon for a reason, easier to design and build and simpler processes for interconnects as you can use alu no prob. Up until the late 00s you could even take a college class on ASIC design and you'd get one made at a fab it was cheap enough, but as things have gotten more complex and business harsher the landscape's changed.

>>1653740
>24 bit DACs are dirt cheap

Not video rate DACs. The cheap one's you're thinking of are meant for audio and have an output rate of 200kHz or less. An NTSC scanline has 52us of active video data. If you want to do 1000 pixels per line you need at least a 19.23Mhz DAC. (1/(52))*1000. I took a quick look at Digikey and there are no 24 bit DACs that fast. If you look for 16bit DACs fast enough for video the cheapest is $17.

>>1653884
Ah damn, I don't even know why I brought up a DAC probably because the poster I replied to mentioned it. Since I was talking about an all digital design as an alternative to analog, no DAC is required. I don't know how exactly thats done but probably there is some encoder that can do that or maybe you can write pixels directly in a parallel mode, like 16 bit or something.

>>1653884
>NTSC scanline has 52us of active video data. If you want to do 1000 pixels per line
wait NTSC is analog, how can you do 1000 pixels per line with NTSC?

>>1653904
by signaling 1000 times per line.
CRT computer screens used analog signalling too.

>>1653656
>are you the one that I discussed building a synth around a small embedded system like a rpi and run one of the above as the back end?
Yes! But honestly one can go so deep into synthesis theory and it's pretty cozy to just do it on my laptop. That Pure Data book is awesome. I think I'm going to wire up some hardware for the first time before the end of summer.. I still need to grab a soldering iron though.

I saw a really inspiring video of some guy using Machine Learning over an instrumental database (cellos, I think) and he had synthesized a very natural sounding cello, and would add random variables to decay the cello into madness, it was cool.

Here it is: https://youtu.be/wGkdb6YlLgg?t=681

>>1653914
what kind of CRT screens had 1000 pixels?
and computer CRT screens didn't use RF signals and NTSC is an RF signal, right? i am confused. if you have a CRT NTSC TV it accepts an NTSC RF signal which is like 300 scan lines, right. where is 1000 pixels coming from?

>>1653926
>what kind of CRT screens had 1000 pixels
All the later computer CRTs could do at least 1024x768 and decent ones could do higher than that.

I think the other poster used 1000 pixels per line as a hypothetical example. For NTSC the horizontal resolution was closer to 720 dots per line.

>>1653904
>>1653926
>>1653939
The NTSC standard has 525 lines but the lines are not broken up into pixels. On an analog NTSC CRT the brightness of the phosphor represents the voltage at that exact moment in time. Any horizontal resolution you may have heard of comes from whatever is providing the video source not the TV. If you were to provide an analog sawtooth wave rising across a scanline the TV would basically show infinite shades of gray; it's not broken up into discrete pieces unless you want to get pedantic and talk about individual electrons and phosphor atoms.

Here's a video explaining more.
https://www.youtube.com/watch?v=Ea6tw-gulnQ

>>1653786
>Not him...
okay, maybe you can link me a single paper that is not from the 60s

>>1653779
You are kind of dense, pic related is an analog computer currently in production, that is about as far as I am going for you.
>>1653812
And to finish things off, chips like video synthesizers would have been to expensive to produce for how small the market was in the 70s given the shape of the industry, by the 80s the market was strong enough as well as the industry. My point was clear and correct, I did not feel a need to give a history of IC fabrication to make it.
>>1653923
Have you figured out where you are going with the hardware yet? Or just playing around with the software for now? PureData is good fun, especially with the Csound object, really reduces some of the tedious patching and increases the power of PureData greatly. Which pd book are you talking about? Miller Puckette's?

The video is interesting, but I do not think I can agree with him that such additive synthesis is the future, it will be part of it, but not dominant, it is ancient and adding in machine learning is far from new and does not really change things that much, just removes some of the tedium at the loss of some control but makes live playing more realistic. It is odd, he spent all that time making that complex patch which could be done in pd by pasting that data into a table, or in Csound with an ftable in a fraction of the time and with more control, or just skipped copying and run the sample through ~sigmund in pd or the gen routine in Csound for extracting fft, forget which it is. I like what he did with the cello sound though.

>>1654038
Christ, just search edn or ieee
https://ieeexplore.ieee.org/document/1493879
https://www.edn.com/electronics-news/4036696/Analog-computer-trumps-Turing-model

First two results I got from each. There is not a great deal written on analog computers these days because they do not change much, they are just another tool in the lab like a beaker or meter.

>>1653926
Black and white CRTs do not have pixels, you have to work out the resolution with how fine you can focus the beam and position it/turn it on and off while scanning. Generally speaking you can get insane resolution but you end up with a low framerate because of scanning speed limiting everything.

>>1654046
>And to finish things off, chips like video synthesizers would have been to expensive to produce for how small the market was in the 70s given the shape of the industry, by the 80s the market was strong enough as well as the industry. My point was clear and correct, I did not feel a need to give a history of IC fabrication to make it.
We're talking across each other. If you're planning to sell kits, I'm saying you might be able to produce short run/small run ICs if you really wanted, but that the market is smaller in general. Food for thought.

>>1653656
Any good ways of getting into supercollider? Ive been meaning to learn it for a while now but never got around to it.

>>1654114
My fault, I misread your first post.

>>1654143
I would just get The SuperCollider Book. There are some decent tutorials on the web, but they all tend to either focus on those things that SuperCollider has become popular for and ignores everything else it can do, or written by a college professor for use in their class and are full of odd little gaps which the verbal part of class is supposed to fill in. It is a good book t hat has been in print for nearly a decade now so can be found used for cheap and it contains some important information that is at best poorly documented on the web.

>>1654113
>>1653939
>>1653955
OK, so what would it take, in theory, to build a digital video synth that could feed a digital signal into a modern LCD TV? Can that be done with a micro like STM32F7? Or you'd need some specialized chips? Maybe a micro driving something like TFP410? THe datasheet says it supports "pixel rates up to 165Mhz" but datasheets always lie as you need to read between the lines and derate etc. What kind of speed is required for, say, 480p at 30fps? Also I don't know if the color data are RGB pixels or compressed data or what?

>>1654546
Hmmm answering my own question, is my thinking correct?
640x480x24x60/1000/1000/16 = 27mhz
(16 bit parallel interface for 24 bit color at 60Hz). 27Mhz is not too bad. The max speed of the APB bus is about 100Mhz so it can support full HD it seems?
1920x1080x16x60/1000/1000/24
assuming 24 bit parallel interface with 16 bit colors or 30fps instead of 60 which requires about 60 to 80 Mhz clock speed

>>1654546
You would need an HDMI serializer chip like a DS90UB949 and a powerful enough microcontroller to generate video data. Don't underestimate how processor heavy that can be; think of the DOS games that ran without graphics cards. A rough estimate: 640x 480 pixels * 60fps * 20 CPU instructions per pixel = 368.64 MHz.

>>1654554
>DS90UB949
What is the difference vs TFP410?
I don't understand the idea. Send the entire pixel stream over a serial interface? That sounds awful. I was thinking about a parallel interface like TFP410 which receives pixel data on a 24 bit bus and outputs HDMI /DVI signal. I think?
https://www.ti.com/lit/ds/symlink/tfp410-ep.pdf

>>1654554
Ah sounds like it serializes data to be transmitted over coax so basically an RF signal like an off-air antenna signal? I am talking about sending direct HDMI signal to the TV HDMI input.

>>1654561
Never mind, I just read up on TDMS.
So basically TFP410 converts parallel data from micro to TDMS

>>1654552
>27Mhz is not too bad
For a video synth project you'd want to define each pixel with some fairly elaborate equations taking dozens or hundreds of CPU instructions to execute. The processor needs to be orders of magnitude faster that the output bitrate.

>>1654600
A typical MPU clock speed is 150 to 200Mhz and most have an FPU.
>elaborate equations
well depends on which equations. the pythagorian? stuff like sin/cos can be done with a look-up table i guess. fractals can be hard. 2bh I'd just do that on a PC and connect it to a TV. Or just watch on PC. Same end result lol. I don't see a point really. I mean if I wanted to do stuff on a micro I'd connect a 3" LCD to it and that would make more sense than connecting a micro to a TV. But even with a tiny LCD, I'd rather use a raspberry instead of a micro.

>>1653390
greetings from /tttt/
put me down as Yes for $5 please (are normal people's trans-dar really this bad that there is even a question?)

I was posting in /ohm/ a few weeks back about making a shitty analog sampler/"synthesizer" from a broken cassette player

I managed to put the pwm circuit on a semi permanent solder board so that's covered now.

Now I need to figure out how to do Midi to CV, and also make an infinite cassette tape.

I tried doing the latter last week and fucked up the magnetic tape real bad (did you know scotch tape will just pull the magnetic compound straight off? I sure didn't)
It's also extremely finicky getting the circumference of the tape just right so it actually has enough friction for the cogs to turn the tape. Even after scuffing the plastic cogs with sandpaper it didn't do jack shit. I wound up buying some rubber bands. If my next iteration's circumference is still a bit too loose for it to grip I'll just wrap the rubber bands around the cogs for friction.

>>1654822
>enough friction for the cogs to turn the tape
That is more the fault it being a shitty walkmen or dictophone with cheap mechanism. Rip the guts out of any decent deck from the 80s/90s and the pinch rollers will pull the tap past the heads. Find one with 3 heads, erase, record and play. This will allow you to use it as a delay as well as a sampler or to layer samples and as long as the pinch rollers are in good shape your looped tapes will last much much longer.

To step it up to the next level of complexity, get an old floppy drive, gut everything but the disk transport and motor, stick your tape heads and circuitry in there and use floppies instead of tape, no need to make loops anymore and they almost never die from mechanical failure like tapes do since there is no tension on the disks, they just spin. Now you can sample, delay, layer and even easily save and load samples on the fly.

You can also do the mellotron mechinism for analog sampling fun, it has it uses but personally I like tape and floppies.

>>1654649
normal people dont see tranners every day.

>>1655090
I didn't even know they made ones with 3 heads, that's what I really wanted all along.
Part of the original plan was to have the cassette constantly being written to, and then read from, but my cassette is most definitely a cheap, shitty chinese cassette player.
I also want to be able to control it via midi, so I can play one of the cassette tapes like a monophonic organ
>layer samples
Do explain, we might be thinking of the same thing

I didn't even consider using a floppy disk. Do you mind elucidating on that as well? How does that work for audio out though? For the tapes, the audio is literally stored on the tape, no encoding, you just play it through the cassette player and it takes care of it. With the floppy drive it sort of sounds like trying to play a flash drive which doesn't make sense (the way I'm thinking of it now)

this is just a prototype, once I know what I actually want and get more familiar with this concept, I'll look into getting a higher end cassette player to modify

>>1655218
>Do explain, we might be thinking of the same thing
When you have separate record and erase heads you can turn off the erase head so it just overdubs the new material on top of the old.

>elucidating
A floppy is the same thing as a tape, one is just round and one is long and skinny. Just mount the tape heads under the floppy disk, and use as normal, they will write the analog data onto the floppy just as easily as they can write it to tape. With the 3 1/4 floppies you have to cut away sections of the bottom side of the disk so the heads can get close enough to the disk itself or increase the drive current on the heads so they can write through the plastic. With the true floppy old 5 1/2 and 12" disks you can write through the case easy enough generally, might need to boost the drive a bit.

>>1654046
>Have you figured out where you are going with the hardware yet?
No - not at all. Well, I know I'm going to be building some DIY modules down the road, but I don't know much of anything about electronics, circuits or electricity whatsoever so it's a slow climb. Like most of us I also have a lot of other things going on, so I'm quite content with fiddling around with SuperCollider and friends while I keep the hardware idea on the back burner. Any recommendations to speed the hardware project up are absolutely welcome though. I downloaded Art of Electronics and few other electronics books, one of which uses an Raspberry Pi for noodling, but ideally I could find a resource that has no intermediate steps and just walks a person through creating a simple, functional module.. And Art of Electronics is a BIG book, so it's kind of daunting/less directly actionable.

And yeah, Miller Puckette's. I will probably move on over to the SuperCollider book afterword. Then Csound.. Then hopefully I'll remember to pick Nyquist back up :-) The author is potentially releasing a free online class for it as early as this fall.

>It is odd, he spent all that time making that complex patch which could be done in pd by pasting that data into a table, or in Csound with an ftable in a fraction of the time and with more control
When I watched it I kinda felt the same, like he was using the wrong tools. But yeah the end result was really cool imo.


But yeah I would really appreciate some recommendations for the hardware side of things, I could use some more motivation here as it's been shelfed with no progress for months now. One tutorial I saved and will hopefully complete soon is here:
>https://hackaday.com/2015/02/04/logic-noise-sweet-sweet-oscillator-sounds/

But I'd rather something that takes me from nothing to building a real module, I want to just get my hands dirty and move the project forward.

>>1655759
>Any recommendations to speed the hardware project up
I would not worry about speeding it up, use the software to try out ideas and experiment. Start working out what you require out of a hardware synth and figure out different ways to meet those requirements, try them out in software and see how they work and how well they fulfill your needs, when you have something sorted out that seems to fit the bill, start building.

>Miller Puckette
His book really is a synthesis theory book that uses pd, not a pd book, it goes quite deep. Both the Csound and SuperCollider books spend more time on the basic mechanics and techniques of the software and when they get into synthesis theory they are using it to show how the software works. Miller's book should leave you with a very good base for Csound and SuperCollider and make their books a quick study and will fill in the gaps that miller does not explore. The Csound book has a fair amount of physical modeling, granular and algorithmic and the SuperCollider Book goes for Machines learning/listening and performance, both human interface and live coding. The three books work quite well together.

>But I'd rather something that takes me from nothing to building a real module,
If you are unsure of what you actually want to make the logic noise series is a good place to start, skip the bread board and build on stripboard, pad per hole or the like, cram as many of those circuits as you can into a small box. It is good cheap fun and you do not find yourself throwing a pile of money into things that do not meet your needs and you never end up using.

>>1655891
Thanks for the feedback. I'm eager to move toward hardware because I'm already on the computer so many hours daily, so I try to minimize any additional computer time. Also, it's just a stupid personal goal to stop being an /ohm/ noob. That being said, I'll push that Logic Noise series up in priority, and thanks for the tips.

But, I'm left wondering if there's a module or any component of the system I can start building now without significant drawbacks (e.g, a piece of hardware that remains roughly the same, regardless of overall system design / goals). It would be nice to have a more tangible and direct project to work toward, i.e "build x as the first component of my dope eurorack" as opposed to "noodle around some tutorials / read some books that I believe may indirectly get me closer to building the first component of my dope eurorack". I hope you catch my drift. It's just that I only have so much time, so getting tangible progress would be a huge milestone.

>his book really is a synthesis theory book that uses pd, not a pd book
Yeah, that's what I liked about it over the other books that I briefly skimmed over. It

>the Csound book has a fair amount of physical modeling, granular and algorithmic
sounds comfy

>>1652317
This is my half finished pitch-bent arcadiator. not so much a synth, but I couple it with my Yamaha PSS-30 and hopefully my Commodore 128 Cynthcart.

>>1656085
If you want to start building then just sit down and plan out what you want to build then start building. My point was if you build a pile of stuff without actually knowing what you want you can end up with a pile of gear that is nearly worthless, hard to sell home made gear. So if building just because you want but do not actually have a clear idea of what you want, just go cheap and have fun with it until you figure out what it exactly is that you want.

>>1653740
Welcome to the middle, sonny boy. Right now you think the future is cool. You look back on older tech as "old crap". In a few years you will progress to hipster faggotry and you will pay obscene prices for "cool old tech".

TL;DR: you're a faggot in the past, present, and future.

>>1656526
>projecting this hard
what the fuck are you even trying to say, retard?

>>1656439
>>1656439
>just go cheap and have fun with it until you figure out what it exactly is that you want
This is the idea, yeah, and I would continue refining what I truly want through software and tinkering. I just don't know enough about synthesis, modular or electronics to even know a reasonable first step. I'm considering just following this for a cheap intro:
>https://www.youtube.com/watch?v=5DKU1m5_b_8

Does that look good? He lays out the price and it doesn't seem too bad. I will not be doing this in a week though, I would be lucky to have it done by end of year.

>>1652317
Hi I'm from >>1657019
long story short, i have a working cheapy hammond organ that i was going to cut off the base off to make it easier to store (all circuits are in the keyboard half anyways), BUT i guess there are midi mods and shit you can do to these instead? i dont have a schematic but i can get one for about $20 bucks

lmao what a total waste of time this thread was. thanks for nothin' retards

>>1657307
k

>>1657307
wow, rude.
we are trying our best anon.

Bump

>>1656678
>>https://www.youtube.com/watch?v=5DKU1m5_b_8
someone tell me if this guide is retarded before i ignorantly plunge into this and follow this video series religiously, im sick of waiting

>>1658614
It looks like a good guide and the circuit is legit. What I like to do instead is recycle the ATX power supply from an old PC. Anything is better than spending over a hundred on a purpose built synthesizer power supply like a rich idiot.

>>1658645
>Anything is better than spending over a hundred on a purpose built synthesizer power supply
does he really do that? I was already thinking about using recycling a PSU, but wasn't sure if it'd work at all, so thanks for the tip

>>1658793
I don't think the guy in your video does, but some people do.

>>1653740
>> infinite resolution
>but the resolution of your tv is not infinite. so what's the point. and 24 bit DACs are dirt cheap
Sooner or later you'll run into some weird edge case where digital just seems to get glitchy and harsh, and where analog might handle things better. Depends on whether analog noise or digital artifacts are more acceptable to the eye/ear, I suppose.

>>1652642
Posting progress

>>1653270
I know jack shit about electronics, so I'm feeling like I am staring a ancient magic book. How the hell do you represent formulas in circuitry is beyond my knowledge but amazes me.
Hopefully you project comes to good end, anon.

oh hey, there's an actual DIY synth thread... noice
I'm juzt getting into synths and decided to build my own (analog) modular synth in a hipster breifcase... because why not?

I need a keyboard (because every CV input device I can find is $100+ and digital), so I decided to make my own

this is my (current) diagram, each of the top group of resistors would be coupled (via a keyboard key) to the CV out line, acting as a voltage divider, and the lower group acts to raise the low-end voltage of the divider (with a handy tuning pot)

it's mabe to output C3-C5 (the orange 4.25v input line is C5, the blue 2.25v line is C3, and the green 3.25v line is C4)

my question is this: how bad is this design and how can I do better?

>>1659386
>everycircuit
such a comfy program

>>1659427
expensive AF though, but I think I might buy it anyway, it's pretty great

>>1659386
>how bad is this design and how can I do better?
A lot of the earliest synthesizers used a design like that. However hitting two keys at once made it react like a key in the middle had been hit and musicians hated that. If you connect the keys to the voltage divider points through a diode the control voltage output will be that of whichever voltage is highest. Musicians liked that a lot more. Later polyphonic synths used sample-and hold circuits to get two or more control voltages out of the keyboards from multiple keys.

>>1659480
this is exactly what I needed to know, thank you!
the idea was to model it after the korg ms-10 keyboard (I never did find a shematic or diagram but it's just like this)

I'm planning to get a box made to slot into this case (the box insert will be missing its bottom so I can just pull it out and work on it, it won't be screwed in), and have the keyboard I'm building plugged in via a jack on the top face

>>1659505
>I never did find a shematic or diagram
I can not believe you even tried, they are widely available and have been for decades.
www.synthfool.com/docs/Korg/MS_series/KorgMS10-ServiceManual.pdf
>but it's just like this
Not quite.

>>1659537
yeah, I actually have that PDF. the diagram I'm looking for is the keyboard itself (you know, the part with the keys?) that pdf basically just shows where the keyboard plugs in to, not how the keyboard works

>>1659540
Yes, I know what a keyboard is.

The electrical part is top left on the schematic, right below where it says 'klm-130,' which is the designation for the keyboard circuit and PCB. Your circuit is an expo design, theirs is a linear.

The mechanical is just a buss bar with a piece of music wire/spring steel sticking off it for each key, the plastic key pushes this down to short the divider to the buss bar.

>>1659547
so is this basically the setup in the ms-10 (minus a few keys) then?
I've been messing with this keyboard wiring for like 3 days now, I just need something that works (and only passes the highest key's voltage to CV_OUT)

I'm not going to buy any compunents until I know my keyboard will work, and so far I jave zero confidence in my ability to make it work

I know I can make a synth work, it's just the perfect voltage division that's killing me

>>1659156
Moarrr

Show effects

okay, I think I'm doing this right, I just want confirmation.

>>1659480
is this what you meant by added diodes? it seems to work just like you said, so I'm quite hopeful.

I was getting some weird capacitance effects, probably because my diode chain just terminates without leading into anything, so I turned junction capacitance down to 0 and it went away (it'll be fed into a vco anyway, so it's not like the charge is just gonna sit there)

obviously red is lower voltage, followed by orange and green, where blue is the CV out line

I switched (red, orange, green) in that order, twice (cycling them on, and then off, in sequence) and got the (blue) output on top, so I'm going to say it's a success.

if no-one tells me otherwise in the next 12 or so hours, I'll move forward with ordering all the components I need for the keyboard and VCO(s? I might use 2, idk) and get on my way building my first synth.

>>1659547
also sorry, I didn't realize that was the keyboard in tbe corner, I've just been frustrated at this project for a few days now (which means it's a good challenging project I can learn from) and was a little too agressive towards someone trying to offer honest help.

and obviously, yes, I will expand this to more than 3 keys.

>>1659637
Diodes could work, but you are going to need a good input voltage to deal with the voltage drop, 4.25volts will only get you 7 keys before you run out of voltage and it does not follow a suitable curve to produce our standard 12 tone scale.

Here is a good run down on an exponential keyboard controller with gate and triggers.
http://musicfromouterspace.com/analogsynth_new/THE_CAVE/KBD%20Controller%20Old%20Design%20I/Analog%20Synth%20Keyboard%20Schematic.htm

>>1659662
the vco I'm using has a built-in linear-to-exponential voltage converter and my keys are all 0.83v apart (1 volt/octave) (I just sampled 3 keys to make sure I could get the outputs to not interfere with each other)
what voltage drop are you talking about? (honestly curious because I'm new to this)

>>1659670
>>1659662
oh, diode voltage drop... well shit, let me see if I can fix my fixed fix

>>1659670
Ahh, I thought you were using the diodes in the divider instead of resistors. The diodes are not needed, just follow the setup used in the link I gave and you will have low note priority, so when multiple notes are pressed the lowest note sounds and you save yourself 30 odd diodes and all that soldering.

In the most basic form you just need the current source (IC2-A and Q1), the resistor chain and a buffer (IC1-C), but to make it actually useful you also need gate so the note knows how long to sound, a CV memory, so when you release the key the pitch remains constant until the envelops release cycle finishes and sometimes a trigger to fire the envelop, depending on your envelop. If you are using standard momentary switches instead of a keyboard you can use DPST switches and use it as two SPST switches, one for CV and one for gate, this will mean your gate circuitry is just buffer and a one shot to get the trigger, no need to extract the gate.

>>1659637
This is what I meant when I said add diodes. The core concept here is that when you hit two keys at once the diode blocks whichever one connects to a lower voltage from flowing to the output. Setup the resistor ladder to account for how there's a 0.7v drop through the diodes.

The problem with how you laid things out is for some keys the signal from the voltage divider needs to flow through several diodes in series to reach the output. It drops by 0.7v multiple times instead of just once.

>>1659679
Doing it like this also works just as well.

>>1659679
Did any classic analogs use high note priority? I can not think of one that did not just save the expense of all those diodes and labor and just called it good with low note priority?

High note priority gives no real advantage over low note, I do not see the purpose of soldering all those diodes.

>>1659685
Low note priority would use just as many diodes; the same circuit with the diodes pointing in the other direction.

Without the diodes the circuit isn't low note priority or high note priority. Say one key connects to 4v and another to 2v. Hitting both those keys together would put 3v at the output. When the first synth keyboards worked like that musicians had to be super careful never to hit two keys at once since it would produce some average voltage between them that didn't correspond to a music note.

>>1659685
Answering my own question, went through a few dozen schematics of classic mono synths, they all seem to be low note, can't find one that is high note, it seems to be something of the virtual analog era and analog revival and just exploiting the fact that it was free to implement since they generally run the keyboard through a micro so switching priority is free.

>>1659689
>Low note priority would use just as many diodes

No, every classical analog has low note, no diodes, look at the link I posted for an example, it functions the same as every classic analog, just uses modern components.

>>1659694
The keyboard in your schematic gives low note priority by driving the resistor ladder with a current source circuit. That takes fewer components but the design isn't as straightforward as putting a diode on every key.

>>1659698
Yes. The current source can be as simple as a single transistor current source, there is no reason to bother with all those diodes and all but anons last schematic show a current source of 83,2uA, which suggests he planned on using one, so no diodes needed.

>>1659593
I call this one " it burst into flames because I'm drunk and it's not read yet".

>>1659703
honestly I had no fucking clue what I was doing with the keyboard input (way over my head) so I said fuckit and decided to settle for a midi keyboard and a midi/cv converter

also I spent $250 on parts for 2 VCOs, a good psu, and a delay filter today

hopefully the guys at work can cut and form my enclosure (although knowing the forming department they'll find some way to fuck it up)

Okay, I just ordered the midimuso CV12 ( http://midimuso.co.uk/index.php/cv-12/ ) and am unsure in whether or not I messed up
I want to hook it up in a monophonic synth, but now that I stop and think about it, I don't know how it handles multiple key-presses
will it only play a note if it's the first one you pressed (or only if channel 1 is free) (worst possible option)
will it play the last not you pressed (like the KORG Prologue in mono mode (this is ideal))
will it play the first note you press, and ignore other keys until you release the first, immediately jumping into the next key you're holding(i can work with that)

i wouldn't care so much, except for the fact I just paid $38 for the thing

will it work the way I want it to, /diysynth/?

(the documentation I can find, but I honestly don't inderstand it [or MIDI]well enough to be able to answer my own question): http://midimuso.co.uk/wp-content/uploads/2017/08/CV_12_ORAC_Manual.pdf

>>1660263
I did not see anything saying how it functions in mono mode, most likely it is either low note priority or last note pressed.

>>1660268
welp, I guess we're gonna find out once I get my synth built up enough to output some waves
I will report back with my findings in... probably like 2 weeks

okay, I need a dual VCA setup, and I found this diagram (super simple really, most of the bits there seem to be for power supply noise reduction), but I'm not sure where the 'Level' and 'Envelope' points on the left go to.
I did find this on the right, which tells me all 4 are 100k pots (and not just resistors), but I still don't know where to connect them to.
( http://synth.glitched.org/ built this one, he has a few pics of the board but I can't fill in the pieces from there)

any help you can give would be greatly appreciated

>>1661384
Envelope and Level are just CV inputs, solder them to jacks and plug any outputs you want into them, The pots would go between the resistors and the jacks, they are just depth controls for the CV inputs.

>>1659784
where do those mini CRTs come from?

>>1659156
>anxious 1950s operator noises

>>1661445
They're old mini CRT TVs from my local recycling depot. I cut new aluminum faceplates, fixed some bad caps, and spliced in a wired connection to the antenna only ones.

>>1661470
How straightforward was it? Been wanting to get into that sort of thing after I got myself an analog oscilloscope.

>>1661472
It's usually very straightforward. I documented how to do it on a previous project:

https://hackaday.io/project/20373-zappotron-super-sequencer/log/55032-crt-television

Notsomuch /diysynth/ but i has a tangentially related queshun and this seems a better place to ask than /sqt/. I've got an Octave the CAT that doesn't fire up. Anybody know a decent synth repair guy in Indiana? Am circuittarded.

>>1653656
>Pure Data, Csound, SuperCollider
What other software tools do you use? I tried some popular DAWs but the interface was a headache, imo. While I haven't used them much, I'm attracted to the style of Ecasound and Sox, which I think I picked up here from the last thread. I would love a complete workflow with an interface I enjoy, I do not enjoy turning knobs with a mouse, or hopping through menus.

Just trying to get an idea of what a complete workflow might look like. I play piano and have been meaning to do more composition on the computer, but finding interfaces/workflows I agree with has been non-trivial.

>>1661666
Those 5 programs are pretty much all I use, couple times a year I open up audacity for something or futz about with alsa modular synth, but pretty much those are all I use.

I have been playing about with Nama, a front end for Ecasond, it gives a different syntax and some easier ways to view things, not sure if I prefer it over straight Ecasound yet.

https://freeshell.de/~bolangi/cgi1/nama.cgi/00home.html

>but finding interfaces/workflows I agree with has been non-trivial.
Sometimes you just have to accept what you have and work within its limits.

>>1661429
this makes so much more sense now, thank you
it looks to me like they're just 2 ports for 1 input, I assume I can get rid of one, right? (getting rid of its potentiometer too of course)

>>1661958
Sure, or you can skip the pots and have two full strength inputs, or one with a pot and one without. Or one can have a pot and no jack, to act as a level control, or you can use a switching back and have it act as a level control until something is plugged in and then it is a depth control for what is plugged into it. Lots of way to deal with it.

>>1661635
>doesn't fire up
Could just be a dead fuse or old capacitors, easy enough to fix. At least take it apart and see if anything looks suspicious.

>>1652893

>>1653744
For some reason electron microscopes still use them a lot while mass specs ime don't, they go straight to the ADC stage.

>>1661695
>Sometimes you just have to accept what you have and work within its limits.
I agree - and to be clear, this is why I've sought out those 5 programs. I had tried Renoise, Reaper and some others (not for more than 10 hours each) but just couldn't do it. And while I haven't spent a ton of time with the aforementioned 5 programs, I do find them immensely more ergonomic, I just wanted to know if these were sufficient for the general set of production needs before really diving in, and I'm happy to hear that this is the case.

Do you have anything available online that I can listen to, just for fun?

>>1661695
Also, have you played with C++'s STK library? Or anything similar. Sounds like it might be nice very far down the road. >https://ccrma.stanford.edu/software/stk/information.html

does anyone know where I can get larger bits of stripboard? the pack I got off amazon was the biggest they had in a 10-pack and they're barely big enough to fit a decent VCO on

>>1662718
etch your own pcbs, its good fun

>>1662730
I plan to at some point but don't have the equipment or space for it right now (hell, finding a spot to solder here is a pain)

>>1662733
I know how you feel about space, I only have a 2mx3m bedroom and a living area that's mostly a kitchen and 3mx3m in total. Thankfully I have a local maker space I can use to solder and nab passives from.

>>1662736
there's a maker space I could go to (fucking awesome place, they even have tesla coils, a car lift, and a radio tower for fpv drone flights) but they charge a ridiculous membership fee, and it's about 30 miles away

>>1662739
actually nevermind they closed

>>1662739
>>1662740
Huh, perhaps they weren't getting the membership needed to stay afloat. A lot of the ones I know of are subsidised somehow, and don't have nearly that much fancy kit.

Mine is free for university students only, so I guess they do charge a ridiculous membership fee.

>>1662741
they had a capacitor bank that could store enough energy to power the city they were in for a good 30 seconds, I think they just overspent (it was a private nonprofit)
I mean seriously I went on a tour there and they vaporised an aluminum plate for fun, these guys weren't cheap

>all this time and effort to make wavy purple squares and learning zero appreciable transferable skills

This is fucking hilarious

>>1662751
stage audio techs get paid pretty well, and hipsters will pay out the ass for good quality handmade 'hi-fi' (analog) equipment (especially in a well-made wood enclosure) to play their records through

>>1662751
still better than playing video games, the tele or reading books like a gay

I wish these components would hurry up and get here, I'm waiting on a capacitor and the main chip for my VCO (the CEM3340), plus have 8 more jumpers to solder in

also waiting on my front panel to get here before I start adding the potentiometers

does anyone have any tips on how to mount boards behind a panel? (I have 4" of depth to work with and plenty of 2D space behind, but the only thing I have to mount to is my sheet metal (plain steel, will be painted, probably white or grey) front panel insert (I'm using a nice(ish) decorative MDF box as a case [see above], and I don't want to cut/drill into it)

>>1662548
Ecasound can do most anything that a graphical DAW can, it's weak point is midi, but it is not bad and not an issue if you sequence from Csound/SuperCollider/PD themselves, they all can sequence each other or sync to each other. I am not sure if Ecasound ever got hardware latency compensation, so if you are pipping in sound from hardware sources you may have to manually correct, it handles software latency correction just fine though. There is a work around for hardware latency correction with Nama, it is documented on the Ecasound mailing list and can possibly be applied directly too Ecasound as well.

I played with STK for awhile, but I quickly realized it offers little advantage since you have to go through a great deal of extra work to do things like file and audio i/o, all those things the others handle for you. It was fun but once I came to a situation where I needed both jack and midi, I saw how much work it would take to implement those into the existing project I called it quits. If you want the STK tools you can compile Csound against the library and get the entire toolkit as Csound ugens.

I have nothing online, I may be able to dig up a Csound or SuperCollider project and paste bin it though. I lost most of my work from the past 2 years after a moment of stupidity during a computer backup a few months ago, so not sure if I have anything worth while to share. I will give a look later this week.

>>1664253
also don't mind my resistor clusters, for some I didn't have the right value resistor so I put a few in series that were the right combined value
it really detracts fron the look of it, but it should work the same when it's done

>>1664258
Cool! And yeah I'm curious to hear what you have, softsynth or not. A bit off topic here, but do you have any recommended resources for composition? I can make little melodies for a few bars but I haven't yet been able to develop anything into a full song. The same problem persist on the laptop.

How do you think about making music anyway? Do you just have music in your head that you get out, or you poke around like it's a blank canvas and develop ideas like that? Just wondering, and thanks. I've been struggling with this. At any rate, I'm gonna spend more time learning theory to help in composition.

>>1664590
Sit down and figure out some songs your like, program them into the synth of choice, do some songs you dislike as well, and generally get into the habit of dissecting the music you hear everywhere you go, figure out what exactly it is that you like and dislike like about a piece. Eventually it will become automatic and you will hear the underlying structure of every song you hear. Little if any music is not built upon music that came before it, even the avant garde build upon what came before them and borrow from everything, so start paying attention, look at how others do what you want to do and experiment. Music theory can be worth while, but I would get a firm base in experimentation first, theory tends to be towards a style and it takes a good amount of time and study to move it beyond that, but worth the time, I just find it worthwhile to develop ones own style from ignorance first unless your interest lies completely within a well established style.

I generally just find a little noodle that I like and build upon it, I decide where it should go in a song, I figure out how I want it to start and end, different qualities I want the piece to have and then I figure out how to travel between all those different places, experiment, borrow from things that I hear. For most people writing music is more a process of revision, going over an over, reworking and rewriting. I rarely actually finish a piece, it may sound complete, but it is still being tweaked.

I will try and dig something out from my past this week, but not sure if I will have the time, supposedly my parents are coming to town so do I do not know if I will have much free time.

>>1664702
Thanks, probably some of the best advice for this I've heard. I like that your method removes any waiting around for creativity or inspiration to strike, and instead gets into reverse engineering sounds. I was also sort of worried about getting into music theory prematurely and being too robotic (I know myself), so I'll continue to expirment for now and just pick it up as I go and maybe spend some nights/weeks here and there diving into it and applying new stuff. At my skill level, it'd probably be mostly a waste of time anyway as I couldn't apply anything beyond the most basic theory.

I can make little noodles but I have trouble expanding them, I think I just don't think about it enough, really. I'm having a lot of fun learning PD right now though, this thread bumped up it's priority so I'm trying to spend a lot more time learning synthesis. It would be nice to have a basic drum machine / patch programmed this month, I could get a lot of utility out of that.

Anyway, thanks for taking the time. Still interested in those uploads if you can find them!

the CV-12 kit I bought is ready, now all I need to do is add the power connector and remote MIDI port, and wait for my keyboard to get here
we should hopefully have first sounds within a week or two

(yes, I know the chips are missing, this is intentional, I'm waiting until I solder the last bits in to put them in the sockets)

>>1652642
based and redpilled

>>1652563
i cant be bothered open it in youtube is that the 'tari ST yamaha chip?

I should have bought that DX-7 at the pawn shop when I had the chance even though it’s digital. Synthesizers are the shit.

VCO done, minus the adjust pots and patch jacks. hopefully we'll have sound soon (chips will be put in their sockets last)

also yes, I know those patch wires are just beautiful, no need to compliment me (I don't have proper wires for this, so I'm using what essentially amounts to speaker wire for this first board, until I can get proper wire)

>>1667252
also resistor chains because it needs resistor values I don't have... it'll work but it ain't pretty

>>1667252
>I know those patch wires are just beautiful,
You could have put them on the solder side, out of sight.

>>1667354
and solder them surface-mount style? now that's just barbaric.
(it's single-sided stripboard, I would rather it be ugly than solder and place components (even wires) on the conductive side)

>>1652748
my guy this is truly radical

Anyone played with a korg monotron? Think I might bust it open to run a sequencer into it

>>1652748
how did you get the numbers and shit to appear? what module is that?

>>1654649
dilate

bump

wow, a synth thread.
sweet!

don't have time to go through it, have to go to work.

bump in the mean while with my humble setup.

>>1659593
>Ominous droning hum and crackling sounds.

i don't have any real synths but i did get a cheap nintendo ds just for korg ds-10 if anyone wants to learn how to use a basic synth for cheap its a good alternative

>>1670228
Funny I just pirated it the other day and it was a lot slicker than I was expecting. Makes me kinda wish there were more ds synth games
>tfw ywn have a rack of nintendo DSes with different synths loaded on each one.
How comfy would that be?

>>1670228
this is where I started, now I'm building my own synth from scratch

Also update: the power supply works and can supply USB power for my keyboard (or charge your phone if you really want it to)

>>1670218
is there any way to capture the crt input and convert it to a digital format? because I'd love some hypnotic screensavers :P
(I'm also wondering if it's vector- or raster-based or some combination)

>>1670635
It's a standard NTSC signal (raster) so I could get a capture card. I should probably get around to buying one of those soon.

I officially have one CV pitch and gate out (it's going to be mono), along with five 0-5v CV outs (4 knobs (half-decent rotary encoders) and the mod slider pad) and another 0-5v CV out that snaps to 2.5v when not actively interacted with (the pitch bend slider pad)
it may be cheap(ish) but this keyboard is perfect for my uses (when paired with the midi reciever (Midimuso 12CV) I bought (which I really need to redo before showing off, it looks terrible right now))

the sticker just came with my oscillator chips, I figured it went well with the rest of the project (as well as any of it goes together, it's semi-intentional patchwork at this point)

>>1670668
I got one from Ali for next to nothing, works fine and it's nice to watch old VCRs on a PC

http://musicfromouterspace.com/
The guy who made one of the better resources for synth diy on the internet died a few years ago :(

does anyone have a schematic for a good ±12volt-powered VCO (that tracks 1volt/octave)? the VCO I built appears to have fried itself. All I need is a triangle and square wave, but ideally it would be able to do PWM square, triangle, and sawtooth (and not require a vco chip that costs $15 apiece [looking at you, CEM3340])

any help would be appreciated

>>1674199
What makes you think you fried it? If all you did was do something stupid and kill the 3340, a new 3340 will be comparable to starting over, at most a few $ more expensive. Fried modules are generally not difficult to repair.

>>1674283
well I built it to spec, plugged it in (sans panel or jacks) and got intermittent oscillations before it just stopped working

I was working off this deaign from Look Mum No Computer with really no clue what I was doing, and I'm still bot too confident with it yet

there is a chance it still works and I had a loose power connecrion or something, but I'm not quite to a troubleshooting-ready step yet (I'm basically working on an end table in a tiny apartment bedroom, it's hard to work on comfortably for more than 5 minutes)

>>1674291
It is bad solder somewhere, almost certainly, it will take less time to learn to troubleshoot than it will to start over. Think /ohm/ has some circuit troubleshooting links in th OP.

>>1674320
I'll check all the whole thing for continuity. knowing the connectors I'm using a plug probably pushed a socket pin up off the contact side... we'll see when I get home
in the meantime I'm braiding a keychain... because why not?

>>1652346
>disco
>tv evangelists
>reagan
No.

>>1653388
rack fx units aren't exactly known for producing great compression...especially the kind you'd want strapped across a final stereo mix...

>>1654822
can confirm that making infinite cassette loops is fucking maddening for all the reasons you mentioned.

>>1662751
have you heard of this thing called "creative expression" ??

>>1670218

>>1675481
It is more a matter of learning how to compress, much of the issue people have is they look at what their studio idols do, set their compressor the same and then blame the compressor when it sounds like shit. You can not set a Alesis 3360 the same as an SSL 4000 and expect it to sound good, they have very different responses, if you take the time to understand compression and your compressors you can get good sounds out of most anything,

Good write up on the tecnical side of dynamics processing by Rane, once you understand this the rest is experimenting, training your ears, and dissecting what has come before.
https://www.rane.com/note155.html

Personally I would say that if you can hear the sound of the compressor, you have too much compression, do a retake and pay more attention to your playing dynamics so you do not need so much compression. If you want that compressed sound break it up into multiple stages, instrument, mic, channel, mixbus, four small amounts of compression adding up to the same restricted dynamic range without that over compressed sound, assuming you do not use the same compressor with the same setting for each stage.

The 80s and 90s multifx can do things just fine, as the 90s show, lots of good sounds came out of those rack units, there were studios filled with them and many albums recorded with not much more than an old board and a handful of multifx as the only outboard gear, especially when it comes to electronic music.

>>1675517
hey, was just hoping to see you here
do you do audio visualizers? techmoan is looking for someone to build one. or a whole bunch.
you could probably make some good money with this.

https://youtu.be/N1cWXamGZmM
you can skip to 10:20 for details

>>1675547
You definitely understand the concepts but honestly it sounds like you have not a lot of practical experience with pro audio equipment. Yes you can get usable compression out of fx boxes, but it won't be anywhere near the "weight"\"glue"\"mojo" of a dedicated compressor, the type of thing that likely has a big juicy transformer in it, or even tubes. I don't know if you were listening to garage releases in the 90s, but believe me on things like Nevermind, Siamese Dream, etc., the engineer wasn't using an fx rack for the drum or mix compression.

I have many things here in my space, including Lexicon and Eventide units that are mind blowing for many applications (way above what even the best modern boutique stompboxes or VSTs do), but they can't touch my dedicated compressors.

>>1675759
I never said they were great, nor did I say there could replace dedicated compressors, but the fact is if you can not make something work with them, you are likely a hack who just throws around buzz words and pretends to know what he is talking about. The main difference between low end and high is the same here as with instruments, the high end is far more likely to gracefully handle any idea you come up with where the lower end is going to impose more barriers, you have to work within its bounds and figure out where it sounds good and how you can exploit that. If you knew anything about the 90s beyond top 40 you would know there was far less rigidity in the industry, the indie movement was making its mark and you had electronic where the musicians were happily recording everything themselves with setups just as I mentioned, hell more than a few were recorded with nothing but a VS1680 and its internal FX.

You may know your shit, but all you did was demonstrate your arrogance and ego, you gave no actual information but you made certain to drop hints about your setup. Sure sign of a hack, lists gear instead of explaining anything.

>>1675714
> do you do audio visualizers
Yup. I stuck the video on /wsg/ since it needs sound and gets flagged by youtube for copyrighted music. Thanks for suggesting I contact that guy.

http://i.4cdn.org/wsg/1567450953329.webm

>>1670218
This is too awesome

>>1675517
Are you mixing a camera feed in? How do you do that?

>>1675928
no problem, keep us (probably just me) posted if he gets back to you, I'd love to see your work featured on his channel.

>>1667585
not him but I guess it's a tape or some analog signal from a pc

>>1675929
>Are you mixing a camera feed in? How do you do that?

I designed my video encoder board to be able to synchronize to an external video source. When two video signals have their sync pulses and chroma subcarier synchronized you can mix them with just a potentiometer without getting a glitchy mess. I also designed it to pull the hue, luminance, an saturation out of the external video as separate analog voltages. These can be fed into other video effect circuits. Eg the numbers warping from the camera luminance channel.

>>1675940
>>1667585
I built a module just for that. It takes in two sweep voltages which represent x,y space. A set of comparators break that up into the segments of a 7-segment display. A BCD to 7-segment chip that's meant for LEDs defines which segments are actually activated and a bunch of diode logic ties it all together.

If these x,y sweep voltages match the position of the CRT raster beam you just get a big boring number, but if modulating the signals you get the weird wavy effects.

>>1675950
...wow, that's really awesome
why don't you make a website or a channel for this?
It's a waste to have it only here

>>1675950

>>1675954
I have some videos on youtube but they only get a few hundred views. Not really worth the effort. I also post my stuff to a video circuits facebook group.

>>1675883
>lists gear instead of explaining anything.

What else is there to know other than working hands on with the hardware itself? Oh well m8.

>>1676031
Gain staging, impedance matching, the dynamics and frequency response of the entire signal chain and the source, the harmonic content and the dynamics of that harmonic content, how the signal chain affects that harmonic content, music theory, the list is long, that is bare minimum for being anything but the bottom. Eventually a musician or producer is going to say something like "that diminished chord is sounding major," and if you do not know all of the above you will not know how to deal with it. There is a deadline, if you have to waste time playing guess and check you are going to have very unhappy clients. Any successful engineer is going to be hired to record albums in unfamiliar studios full of unfamiliar gear with little time to get to know the gear, you need to be able to look at the manual and tech info for that gear, listen to the bands prior work and the demos of what they are going to record and have a damn good idea of what and how you are going to do your job before you even set foot in that studio if you ever want to move beyond recording bar bands and high schoolers.

Go read any interview with Butch Vig, the producer of both of those albums you mentioned, where he goes into technical details of a recording session, if you do fully understand everything he says, you do not have a chance. A good producer like Vig can make up for the failings of a less than ideal engineer, it is their job to speak both musician and engineer, but to be successful an engineer needs to be able to communicate well with both the band and the producer, it is probably the most demanding position in the recording process, the engineer has to make both the band and the producer happy, the producer only has to make the band happy.

Your attitude is fine for the bedroom studio, but it will not get your anywhere in the professional world. A studio full of the best gear will not make up for an inept engineer, a good engineer will make it work regardless of the gear.

Anyone have any good resources a complete idiot that knows nothing could use to build himself a string synth? I want a decent analog string synth but there's just nothing new and decent on the market. I definitely haven't got the money to buy a vintage model either.

No microcontroller has enough analog inputs to make a really good synthesizer. So I'm designing a multiplexer to combine 64 analog knobs and sliders into a single analog input!

https://www.youtube.com/watch?v=XvaW19IhjiI

>>1676921
this is why I'm using MIDI
it's gonna be a digital output from the controller anyway, so it's not like the inputs being analog really matters
that said, mine is going to be an analog
kinda-modular mono synth with digital control, different synths have different needs

also wow noise generators are easy (pic related, a super-simple white noise generator I threw together, will be useful for testing the filter I intend to make soon, and will go into the final synth eventually)

>>1676637
>decent analog string synth
What's your opinion on the Streichfett? I can't justify the cost, but it seems like a good goal for a DIY project

>>1677665
I do actually own the Streichfett. Bought it used for a much better price. I'd say it's ok. You can get some decent sounds out of it but it feels a bit sterile to me. The reverb and phaser effects that are built in aren't that great. It's powered by USB, which annoys me a bit. There's also no off-button. As soon as you plug the USB in it's turned on. I think the overall concept is kinda cool but the execution is nothing to write home about. That's actually why I've come to think about attempting to build my own string synthesizer.

>>1676637
If I really had to do an analog string synth I would just rip off the Korg PS-3100, dump the switching circuitry/jack panel, add a micro to trigger the fucking thing, and add a chorus parallel to the resonantor. Might strip the voices down a little, could simplify things a bit and cut out anything that is not stringy.

There is a reason string synths fell out of popularity with the rise of the poly synth. All an analog string synth is, is just a very stripped down synth to make up for the increased number of voices required. Most chose more power and dumped their string synths with the arrival of the cheapish 6 and 8 voice polys. In the 90s almost everyone had a string synth but rarely used them, they were just so damn cheap and fun at times, but offered little to nothing over 16 voice synths that were plentiful at the time. I do miss my old RS-09.

>>1676921
Why not use rotary encoders and use no analog inputs?

>>1677319
They can get more complex, pic related is a good one. It needs clocked noise and random pulse/dust added in though, which would triple the circuitry.

>>1677665
They are fun boxes but I really do not see the point in them, for the same price as a new Streichfett I can get a used Blofeld desk top, which will blow it out of the water when it comes to strings. Being able to separate you strings into 16 individual voices is just so damn handy, even if you keep all 16 voices the same, you can apply trem/vib to just one or a few voices/notes or have a few out of tune notes, slide just one voice and on and on. It's biggest drawback is that 16 voices is not always enough for strings, but overdubbing solves the issue in the studio and no one will notice the voice stealing in a live situation, its just the end of the release cycles being cut off anyways.

>>1675959