I think part of the trend of making hardware less complicated in some aspects is that it is generally assumed that any hardware will be combined with a computer at some point. There is less of a need for 128 voice polyphony workstations anymore if you don’t need/like those sounds. You can’t meaningfully interact with so many tracks at once, so because digital multi tracking is ubiquitous these days, the focus has shifted to fun and intuitive interfaces for generating interesting sounds and performances. Also most people’s music isn’t so complex that so many voices in a single instrument is necessary. In the case of the Digitone, the flexibility of the engine and the sequencer, plus meaningful performance functions for interacting with the tracks means that it puts out more complex music than instruments with 2-3 times the voice count
On the other side for me, as a guitar player I am a stickler for low latency, and certainly look forward to the day you don’t even have to think about it anymore, no matter what gear or amount of plugins or whatever…
I love creating music with old gear, new gear, homemade gear, I love technology best when you don’t notice it, that is to say when the interface between the man and the machine is as transparent as say using a pencil and paper - it doesn’t have to be simple either but for me it has to have some elegance and not feel disjointed in any way.
Ha, that’s cool you will probably like these guys too:
On the subject of limitations and in particular 8bit video game music, I’d encourage every Elektronaut to get a Gameboy and LSDJ and/or nanoloop and work within those ‘limitations’ for a while, I can’t recommend it highly enough, cheap to get into, portable and it will change your game (pun intended) 
You can get nanoloop as an iOS app now! Just learned that recently. Nanoloop on gameboy was my favorite thing for road trips or plane rides when I was younger
You’re right, I do dig them! When The Advantage came to town for a show, they were billed as a “video game metal” band, but Bit Brigade sounds more metal.
Yeah I mean my Sound Canvas 88 is 64voices and 32 part multitimbral… I think what it actually is going on is companies care way more about higher quality fx and filters going into these units which costs a lot of dsp or cpu power to run. I think the Roland System 8 ACB stuff is a pretty good example on how you don’t see the super high quality digital stuff showing up in workstations or when you run it on a cpu you see how much these modern synths eat up cpu. I guess I would be curious to see how much a digitakt or a digitone would do to a cpu… the filters are really nice on them so my guess is it would be a decent hunk of cpu. I’m pretty sure filters take an exponential amount of cpu the higher quality you go on them although if you stay out of the aliasing ranges maybe that doesn’t matter too much. So maybe there is an argument to be made for lesser quality and more polyphony at times.
Sure, there’s lots still around. My conversation had drifted here and there and by that reply I was talking about all gear including computers.
You can certainly create setups with crazy low latency with the right gear, but with other gear and plugins and whatnot it’s not to hard to get to a point where it’s not enjoyably playable for a live instrument like guitar…
Latency is worlds more noticeable playing guitar than playing my OT and Rytm for example. Once it starts getting over about 10ms you start loosing the connection with the groove, 20 and it’s almost killed. Ideally 1-5 is a nice range but you need to spend a bit on a thunderbolt interface, have a nice computer, and run high sample rate and low buffers to get down there, which limits other things you can do.
Just jamming on sequencers I could probably have ten times the latency or so and not sweat it and it would seem fine.
DX7 runs oscillators with a resolution of 30kHz, which is 3.2 times less than Digitone, iirc.
So actually bigger/cheaper DSP power seems to be used for hi quality sound generation.
And oscillators sampling rate have necessarily impacts on every piece of the chain of sound processing, so I guess the difference with 80’s treatments is really exponential when you add each brick of the chain.
The SY77 (1989) featured 16 voices of advanced 6 operator FM (three feedback loops, two filters and 45 algorithms) and 16 voices of AWM synthesis. The AWM could be routed to modulate the FM voices.
It also featured 4 multi FX units, 2 of which offered stereo chorus and flanger plus tremolo and symphonic, and 2 of which offered reverbs, delays and distortion. The FX units could be routed in series or parallel.
The sequencer had 16 linear tracks and one pattern track. The event list for each track could be fully edited so both micro-timing and parameter locks were possible, albeit tedious to apply.
Actually I still have the TG77, such a wonderful beast!!
That’s a valid point. it’s also been the same all the way through history for every musician though. I’m glad they didn’t stop in the 1800s and think “right this must be all there is to it”. We use tools now that no one could have ever guessed or even thought possible and there’s still more future out there. Still, interesting thought
Right, just to clarify, that’s why I added this below.
I don’t want to tell people how much dsp to use or how to make music and with what, but it’s still an interesting conversation… Whatever I’ve said on the thread are just some thoughts I’m having at the time… 
Things to ponder…
I like how this industry went from wooden end cheeks on analogue sounds warmer to mine has more DSP than yours in only a few years. 
Yeah I was just thinking out loud too. I’d just never really considered every sound anyone could be chasing is probably something they’ve already heard. Well, kind of. There’s obviously a lot of venturing into the unknown until you find something nice, I think that’s where the desire for more powerful gear comes in. I reckon it’s more about new turf than more power. More powerful just often means new stuff to explore. There aren’t many uncharted waters left but there’s a hell of a lot of unexplored depth in the ocean
Right, I don’t go chasing sounds I’ve heard, it’s just that based on the crazy vast amount of sounds I’ve already heard, it doesn’t seem like I need new things to still come up with crazy amounts of new sounds…
If that makes any sense…
And I think that comment was more like, everyone’s favorite music has been made on stuff that exists, so it seems they could make music they enjoy with stuff that exists…
Like, your favorite artist isn’t missing anything, what are you missing?
And soon enough it will swing back again. Flares and drainpipes.
Sounds like a home-made rocket launcher right there. [/wayofftopic]
IMO the best way to look at this whole argument, is which one is easier (and maybe more fun) to sit down at and start making music? It’s very subjective. The answer won’t be the same for everyone. Which tool someone chooses won’t be the same for everyone, and it absolutely doesn’t need to be. How much DSP available in the Digitone is completely irrelevant. The device has limitations. For me, I like to work within limitations. It makes me creative. (I suppose that comes from a background in SID/MOD type music, early synths, etc.) When I sit down with the Digitone, I’m up and running in seconds. I’m immediately thinking creatively about how I’m going to get the most out of what’s there.
For me, there’s nothing worse than sitting down with something so vast, that I spend most of my time thinking about where to start. This is why when I build modulars, they tend to be smaller systems focused on a particular workflow that I have in mind.
I like to have several very focused tools at my disposal rather than one that does absolutely everything. Since I started building instruments, (maybe 2004ish) I haven’t really written any music. I picked up the DT+DN+Blofeld combo, and now I’ve been making music again. I think that says something to what these little machines can do, and how their limitations actually serve them. They’re immediate. You sit down, make a few sounds, and in minutes, these sounds are in patterns and can be manipulated on the fly.
By the time I’m done with my day job, get home, get the kids situated, work on my synth builds for people, hang out with my wife, I have maybe a couple of hours before I go to sleep that I can do something like make music in. The above-mentioned combo has allowed me to make several little tunes in the last couple of weeks.
This workflow isn’t for everyone I imagine, and maybe some people might need more or less of this or that to achieve their musical goals. However, I don’t think adding an extra DSP to any of the above would make them necessarily better instruments. They can easily be supplemented with other synths. Exactly how many FM sounds at once would make the Digitone any better? Personally, I think four tracks of that, a track or two of analog, a few tracks of samples, maybe some wavetable, and maybe a hint of modular is more interesting that 16 tracks of just FM would be. Then there’s the “challenge yourself” aspect. Just how much CAN I squeeze out of four tracks of FM on the Digitone. I think you’d be surprised what you can get out of that. Throw in the Digitakt to relieve the Digitone of drum duties, and it opens it up that much more.
Adding more DSP is nice in some cases. However, in these compact and immediate devices, I really don’t think it would add too much to the experience. Just my point of view though.
That’s how I’ve come to look at gear as well. When I had more disposable income - thanks to a combination of cheap rent (no mortgage or other homeowner responsibilities) and decent income - I bought gear based on power rather than shortest path from powering on to making music… After buying a house, and thus taking on a mortgage, paying for repairs and all that stuff, I got a lot more discriminating with where I throw my money for gear.
In the case of the Digitone, despite having more capable FM synths in the form of the Nord G2 and softsynths, I find it increasingly appealing because people seem to be immediately productive with it, and the music sounds pretty good. I just don’t get on with desktop softsynths because I dislike mousing about to tweak synth params - don’t care how good they sound, just can’t get past that mouse-based user experience. The G2 has the same problem even though it can run on standalone hardware. Not enough of a FM-tweaking nerd to invest in the Montage, and besides it would take up too much space in my small abode.
I already have oodles of polyphony in the form of two 128-voice poly romplers. Such things are still out there in today’s market if max polyphony is a requirement.