http://www.vintagesynth.com/forum/viewt ... =1&t=67917
As Stab Frenzy presented a way of defining analog vs digital that I wasn't familiar with. I'd like to let the conversation resolve itself and leave the poor little Evolver thread alone. Let's NOT make this about which is better...I just want to find out why a DCO would be considered digital rather than analog since I'd always been under the impression that DCOs are analog with a digitally controlled clock.
Apart from defining DCO vs VCO I think it would be interesting to discuss what scientific properties make VCOs sound the way they do in comparison with DCOs--again not subjectively in terms of preference it rather objectively in terms of measurable information.
Where the conversation left off:
Stab Frenzy wrote:That doesn't make sense and wouldn't work.samuraipizzacat29 wrote:firstly, my bad on the z3000. I thought it fed a frequency counter that then fed the information back around to the pitch input.
It's so you can see what frequency your oscillator is oscillating at, a very handy thing for people who use them to make music. This is how the tuning on the Andromeda, Little Phatty and AFAIK Monotribe works as well.samuraipizzacat29 wrote:it doesn't. it's just a vco with a frequency counter on the back that feeds an led. I don't see the point now that I know that, but what do I know.....?
I have been this whole thread, but you don't really seem to be listening to what I'm saying.samuraipizzacat29 wrote:I wish someone of more authority than me would weigh in.
I'm in awe of your determination and certainty even after you admit you don't know what you're talking about or have a basic understanding of the topic.samuraipizzacat29 wrote:I don't know whether a CMOS ic "counts" as an analog output or not. All I know is they're made with transistors, the exact same transistors that make up everything that we consider an analog oscillator. like dis:
http://www.electronics-tutorials.ws/wav ... table.html
-same mosfet transistors. a clock signal (in this case the charge/discharge cycle of the capacitor) turns the MOSFET on/off creating the hard edge of the pulse wave. The only difference between that and a DCO is that the clock signal is created by a microprocessor, then fed to a frequency counter, that then feeds the MOSFET. Or in the case of the poly800 (per the datsheet) the computer feeds a counter which then feeds a cmos "tone generator".
and the output doesn't need to be deciphered by a D/A converter, it goes straight to the filter, vca, and out to the jack. that makes it analog IMHO. It doesn't make it sound good , but it makes it technically analog, and technically a DCO. to my knowledge anyway.
Guess what computers are made from? Transistors! There's transistors in a lot of things, they don't make things either analogue or digital. I already explained what the difference between an analogue and a digital circuit was in my last post, I'm not going to repeat myself. Go read it again.
Did you know the earliest digital circuits were made with valves? They used stupid amounts of power, put out craploads of heat and broke all the time, but they were still digital. You think 'what could be more analogue than a valve?' but they can still be used for digital purposes as well. We use transistors in digital circuits now because they can be made really small and not use much heat or put out much power.
Digital circuits aren't some magic thing with little digitals flowing through them, they're made up of normal analogue components just arranged in a way that the only piece of information about the signal that matters is whether it's over the threshold to be logical high or not. You can still measure the analogue voltage at a point, it doesn't need to go through a DA convertor unless you're doing more complicated multibit reconstruction. Theoretically you can even reproduce CD (or better) quality audio from a 1-bit digital signal using only a LPF with a high enough frequency signal. That's how DSD works, and there's a few recorders out there that use that method as well, Korg has a whole bunch of them.
To demonstrate the point, I can program an Arduino (which is a completely digital microcontroller) to use one of its output pins to make a square wave. I change the amount of time the digital output is high or low and I get different frequencies, I change the duty cycle and I get PWM; all without using a DA convertor, an analogue waveshaper, anything. The output is digital but it can be used as the input for an analogue circuit. It's a digital oscillator, not a DCO.
I'm not saying any of this to put you down by the way, I'm just trying to correct some misconceptions. Hopefully now you and anyone else who reads this thread has a better understanding about the differences between digital and analogue circuits.
He's said his reasons quite a few times; he's an engineer and he thinks that the DCOs he uses are the best tool for the job. I think that maybe your opinion of what feels and sounds right are clouded by your prejudices against DCOs if your appraisal of the Z3000 is anything to go by. There are faults that you can find with the Evolvers but the oscs aren't the biggest thing that needs to be improved IMO, and all the people buying up Mophos, Tetras and P08s seem to agree.samuraipizzacat29 wrote:I don't get why Dave thinks he needs DCOs in his synths. That's the only reason why I don't own any, because I don't like how it sounds.... (that's why I continued posting here, because I do think it's essential to DSI evolver). And that's coming from a guy who had some breed of evolver as an essential part of his setup since his early days of synthing. I mean, "oscillator slop" is great and all (it really does add the same effect as oscillator drift, just inject random numbers into the clock output of the microprocessor), but it still just doesn't "feel" and "sound" right. I know i'm grasping at straws with that one, but I'd rather have digital oscillators (wavetables or samples) or something more drifty and smooth as opposed to clicky (I never liked the filter stepping either, that and the vca are digitally controlled as well, ewwwwwwww). I'd rather have more flexible dsp than a sort of halfway "it says analog on the box" implementation. I'll run my ipad through a monotron that was semi-jest. how about a moog mf-101? like give me an awesome digital oscillator, or give me a unique analog oscillator, but this halfway "we're going to try to tame everything so that you can unleash it later" ends up being kinda lame imo.
It would seem to me that the whole thing needs a redesign considering arturia's (edit:proposed) takeover of the cheap, full-featured monosynth market. I put them next to each other (as an average consumer) and I don't understand what I'm getting for the extra $700ish. the minibrute sounds dern good (even if I've only heard it from my laptop ) so it would seem that Dave knows very few are going to be GAS-ing for a MEK for a while and there's plenty of used market so why bother for a few years? I just hope he puts something out that I really want to buy. Like I suggested, better wavetable implementation and real vco's. I can dream.....
Comparing a Minibrute to a MEK on price is incredibly unfair; MEK has four times the oscs, 1.5 times the keys, four times the LFOs, four times the filters and three delay taps the Minibrute hasn't got. Comparing them on sound is subjective so people can make up their own minds on that, but the Evolver can certainly make a whole lot of sounds the Minibrute can't get close to.
Please this time let's stay on topic OK?