DSP austerity in 2018

Could you imagine the cost of a box like that if Elektron made one…Hahah >.<
Would be Apple Mac prices.

No, they’re not. They’re general purpose microcontrollers.

1 Like

Regardless of what the hell is powering these digital boxes, the question is why are they not much more powerful?
Computers today for instance are insanely powerful compared to computers 10-15 years ago, yet these new digital hardware synths seem even less powerful than they were way back then? They are going backwards?
Technology is refining and is becoming smarter, more powerful and efficient by the hour yet the modern digital hardware synth seems to be doing the opposite?
Elektron are not the only company providing under-spec’d digital hardware synths today, it seems like the trend throughout the industry unfortunately.
It doesn’t make sense to me at all :thinking:
Thank god for the mighty power of the modern DAW and computer, incredible plugins and effects we have available, making these digital hardware synths unnecessary today

3 Likes

4 monophonic voices ? Not to discard your point, but most string players can play chords. That makes quite a difference.

I would say that the mighty power of the modern computer is unnecessary for a lots of musicians. Not only unnecessary but overwhelming as well. And then there’s the frustration of the interface.

Are you sure they really go the opposite ? 8 voices FM synth with onboard fx, filters, sequencing and bargain price*. I don’t think we ever had this.

(*) yes, bargain… DX7 in my country was 14000F when it came out in 1983… minimum salary was 3700F at that time. Now DN is 779€ while minimum salary is 1500€.

6 Likes

What about the SY77?

people want more features, 8 track sequencer , parameter lock ( 70 per pattern thats insane ! ) soundlocks , external inputs, 2filters : that is not classic DX features, and yes the price is sweet compared to the other boxes ( the mono was 6 voices and no dynamic allocation that meant 6 mono or 1voice with 6 poly ) but no, people are whining !

1 Like

Can someone post some kind of example where crazy extreme amounts of processing power are used to make music, this power that is being talked about not found in hardware.

I want to hear what it can do if actually used…

1 Like

A few months back I was getting really deep into physical modeling synthesis. That’s something that takes a ton of CPU power, especially when it’s homemade in Reaktor and not optimized in the least. I could only get 6-8 voices going reliably with my newish iMac. Reaktor is the only soft synth I use, but I imagine if you were softsynth focused you would experience processing bottlenecks with higher sample rates and complex arrangements. Most of the time I use the computer as a glorified tape recorder/editor, but if I’m working in Reaktor I’ll max my CPU out every time :clown_face:

1 Like

True, but I’ve maxed my cpu doing far less interning things than I could do with hardware…
I’m interested in hearing actual real audio unattainable by a hardware setup that consists of multiple devices…
Just to see what it sounds like…

1 Like

Here’s one that was really maxing out the CPU. Kinda a mix between baroque guitar and Post rock noiseiness All sequencing is generative, which is a lot easier in software. 8 voices of physmod synth (the only hardware physmod synth I know of are the yamahas from the 90’s that are one or two voice). The mixing and fx processing is also all happening in realtime. One giant Reaktor patch all my own designs.

8 Likes

That sound really good just out of my IPad speaker, I’ll have to grab my phones and do it some justice later…

1 Like

What is ‘powerful’ in this context? Is there a suitable UI solution to harness this ‘power’ in hardware form factor vs computer/software? I agree that it’s probably not that big a deal to add more voices to something as this is just scaling, but if power equates to features, then UI is just as critical in the conversation. Computers have a large screen which can be used to draw whatever interface is needed to access and control a huge number of parameters, but a hardware interface is far more limited.

2 Likes

I’m quite sure people could leverage more power, but I wonder if the general vibe is more gassy and just thinking it should be there, without having great uses for it. I’m going beyond talking just more tracks on the Digitone, but the state of hardware and software power in general.

I have not reached the limits of my gear + 2012 computer, not at all, and I’ve heard some super crazy way far out stuff with so much processing it’s insane, but I don’t normally use that and generally back off quite a bit…

I could put a nuclear reactor in my car, but I don’t think I’ll use it… YMMV :joy:

I admit, there’s still room left for optimizing for super low latency and doing more on one machine, and faster more powerfull dsp will be better, it’s just to me doesn’t feel like that much of a limitation, more like we are pretty unlimited right now and the future will get crazy stupid unlimited…

4 Likes

Physical modelling is pretty interesting from a technical point of view, however (and no disrespect intended to @knobgoblin here) most of the time it sounds like something that could be achieved with simpler means such as electro-acoustic techniques and various less intensive synthesis methods, and a combination of these. In a way it is like when digital synths first came along and you could create semi realistic sounding more traditional instruments on a synth, although I get that there is more to it than that.

Not to say that it is less valid than any other instrument or technique, but since we are talking about the percieved need of more DSP power it really just depends on what outcome is desired and which methods are available/used, for an example if I wanted to create some interesting plucked string sounds or membranes being struck or whatever I’d probably make some rudimentary homebrew string and membrane devices, grab some transducers and mics and record them into say the Octatrack add some fx, pitch them to extremes, filter and whatnot to get similar results as could be done with physical modelling, it would be fun and wouldn’t require heavy DSP.

Of course it all depends on what you want to do, I bet that it is fun to create complex reaktor patches or whatever, in just the same way it is fun using any other technique or tool that you enjoy, I guess the point I’m trying to make is if it is meaningful to the artist then its great, if it is just specs alone then it is up to the artist to decide if it is a requirement or not, specs do not necessarily enable or disable inventiveness, enjoyment or capabilities, that falls upon the artist to provide.

I believe that someone who is talented on one thing could be just as talented on another thing and specs have not as much to do with it as many people seem to believe.

6 Likes

Just to clarify, I have nothing wrong with technological advances, more powerful dsp, new future music gear, or anything, and that stuff is interesting and exciting for me. But at the same time I think it’s absolutely not needed to currently make great music.

For me it’s quite possible to both not need something at all while fully welcoming it at the same time… :joy:
Just thought I’d get that out there…

1 Like

Even though processors for general use continue to become more powerful, it doesn’t seem to be in ways that are useful in production for me. I can run a synth from my phone if I want, but that’s a horrible interface. I can run a demanding (and very musical) softsynth like Kaivo, but even on a laptop with reasonable specs it isn’t viable as a live instrument because it taxes the CPU so heavily. (Unless a buy a really really top-range machine, for which price I could also buy very sophisticated specialist hardware.) I have absolutely no ideological objections to using a computer as an instrument, but I enjoy specialist machines more despite their limited processing power. Interface and design is more important that processing specs.

I should get off of this thread, but how about this one, haha:

Every bit of music that anybody has ever heard including their top 10 favorite tracks were all recorded on gear that exists before now…

When I think about my favorite music, it’s not max processor music… Are people actually hearing things in there heads more complex than anything they’ve ever heard before and just waiting until tech catches up? :thinking:

Even if so it’s seems like a small nitch super mega process music genre than some sort of new norm requirement…

3 Likes

That’s a good point too. Seems computers and smart phones are much more powerful but with the current way they handle music programs way less efficient, and that hardware seems to run the same sort of stuff with much less processing power on an intentionally designed processing scheme…

Not much to add to this thread, other than to say that I started listening to The Advantage’s album Elf-Titled again for the first time in years. It’s an album of video game cover tunes - most, if not all, from NES games. The NES had a 5-channel sound generator, with the 5th being a low-grade sample player.

Thus I thought of this thread. The Advantage has/had a 4-piece lineup - each member corresponding to one of 4 channels on the NES sound engine, with samples added to the mix here and there. :wink:

3 Likes