Quantcast
Why could some digital hardware synths sound better than software? - Gearspace.com
The No.1 Website for Pro Audio
Why could some digital hardware synths sound better than software?
Old 10th September 2012
  #1
Lives for gear
 
StarfishMusic's Avatar
 
🎧 10 years
Why could some digital hardware synths sound better than software?

Obviously features, FX, presets, and the like are different but how come some digital hardware synths sound better (to me) than similar software synths? It really does go against my philosophy of not to buy items which are redundant with the computer because they are indeed computers.

I love analog gear and synths and have plenty, but a virtual analog hardware vs a soft synth for instance like Sylenth or Massive makes no sense to me. Could it actually sound better? Why would it even exist? That one's not in my direct comparisons because I've played viruses but never owned one to AB. Truthfully they've never struck me as awesome but I haven't played the newset ones.

What I have AB'd are the few digital synths I've owned which are all renowned for their tone I believe. They are the Yamaha Fs1r, Roland Jd990, Yamaha An1x, and also my old Casio Cz101 I used to have. All sound distinctly better to me than similar style synths in the computer. Not the sample size, not the craziness of their sounds, but just the sheer clarity, warmth, sheen, whatever you want to call it.

What sparked me to write this thread Is I recently spent a ton of time playing Native Instruments FM8. Though the FM8 does some crazy stuff the Fs1r doesn't, just the clarity, sparkle and integrity of the Fs1r seems far superior. Both are FM synths, shouldn't the FM8 at least be in the same leauge? It has that same dirty sound character alot of software synths seem to have (not samplers/romplers) I realize the fs1r is a 16 operator synth and the FM8 6 op but still I'm not talking about the complexity of the sounds. Just simple sounds too on the Fs1r sound superior. This makes no sense when the Fs1r converts out of digital from itself and gets reconverted into my software making it second generation. The FM8 stays in the computer and only comes out of my DA. My converters and clock for my DAW have to be better than what can be in a 14 year old synth! Im using an SSL Alphalink and antelope Ocx. They do an excellent job of recording that tight clean FS1r sparkle so it has to be from the softsynths. Anyone have similar experiences?
Old 10th September 2012
  #2
Lives for gear
 
Eric J's Avatar
 
🎧 10 years
Dedicated DSP for one. A hardware synth can utilize 100% of the power of it's CPU purely for the purposes of sounds. Your computer CPU must handle a multitude of tasks which means it has less available power to dedicate to synthesizing sounds.

Look at DIVA, for example. Sure it sounds great, but it can easily being even a relatively modern computer to it's knees.
Old 10th September 2012
  #3
msl
Lives for gear
 
msl's Avatar
Different convertors, the sound of the specific dac.



.
Old 10th September 2012
  #4
Gear Maniac
 
FullCount's Avatar
 
🎧 5 years
Because in a hardware solution, the synth is making it's decisions and building the sound by passing electricity and audio through components, from one section to another; each leaving its own imprint on the sound. In software, a program is making decisions on what the waveform should be based on all the synths settings in a mathematical way that doesn't involve components outside the CPU, or better yet, in software, the CPU spits out the audio waveform result of all the vst's parameters, which then finally gets sent through the DAC. In software, there never is an actual audio signal until the very end after computation.
Old 10th September 2012 | Show parent
  #5
Lives for gear
 
StarfishMusic's Avatar
 
🎧 10 years
Quote:
Originally Posted by Eric J ➑️
Dedicated DSP for one. A hardware synth can utilize 100% of the power of it's CPU purely for the purposes of sounds. Your computer CPU must handle a multitude of tasks which means it has less available power to dedicate to synthesizing sounds.
So your telling me the DSP dedicated to generating sounds in a 1998 module is more than my 2009 core 2 quad dedicating a portion? Not saying you're wrong, It's just hard to swallow.

Quote:
Originally Posted by msl ➑️
Different convertors, the sound of the specific dac.
So current moderately high end $2500 SSL alphalink convertors and an antelope clock just give a less pleasing sound than a 14 year old module? I know the clock issue is a raging debate here on GS, but I use it too when I record the Fs1r so that can't be making the "dirty" sound can it?

I know some vintage digital sounds cool for it's weird crunchiness like an emulator II or S950, or others have analog filters envelopes like my Emu EIII, but compare straight more modern old HI FI digital like an Fs1r or JD990 I still have trouble believeing their DACs are really what could sound better. I mean I'm no electronics or software engineer so the real answer could be beyond my scope. Do some dedicated hardware DSP just sound better than computer? I remember doing extensive AB testing about 6 years ago of Pro Tools TDM plugins vs the exact same plugin run natively. There was a tiny difference, but hardly the difference between a Fs1r and FM8
Old 10th September 2012 | Show parent
  #6
Lives for gear
 
Eric J's Avatar
 
🎧 10 years
Quote:
Originally Posted by StarfishMusic ➑️
So your telling me the DSP dedicated to generating sounds in a 1998 module is more than my 2009 core 2 quad dedicating a portion? Not saying you're wrong, It's just hard to swallow.
No. I'm saying a DSP chip in a digital synth can dedicate 100% of it's power to making sound while your computer CPU cannot (since it has to eat up cycles to perform other tasks).
Old 10th September 2012 | Show parent
  #7
Lives for gear
 
StarfishMusic's Avatar
 
🎧 10 years
Quote:
Originally Posted by Eric J ➑️
No. I'm saying a DSP chip in a digital synth can dedicate 100% of it's power to making sound while your computer CPU cannot (since it has to eat up cycles to perform other tasks).
I guess the question is: is the full DSP power from 14 years ago still more than the dedication of say one 50th of a modern computer that's a thousand times stronger than a computer from that era? i'm not saying it's not, it's just my phone's CPU is a few times stronger than any computer from that era. Why don't modern synths have a knob that lets you scale how much computer power they take then if that's the source of sound quality?

Quote:
Originally Posted by FullCount ➑️
Because in a hardware solution, the synth is making it's decisions and building the sound by passing electricity and audio through components, from one section to another; each leaving its own imprint on the sound. In software, a program is making decisions on what the waveform should be based on all the synths settings in a mathematical way that doesn't involve components outside the CPU, or better yet, in software, the CPU spits out the audio waveform result of all the vst's parameters, which then finally gets sent through the DAC. In software, there never is an actual audio signal until the very end after computation.
Not to be rude but you just described an analog vs digital synth, I've been talking about a digital hardware synth vs a digital software synth. Both are software synths technically. No electricity is passing through individual components other than the CPU and ROM/RAM until the DAC unless your talking digital synths with analog components like an Emulator III which I am not. Does a Jd990 have analog filters? I don't think it does.
Old 10th September 2012 | Show parent
  #8
Gear Maniac
 
FullCount's Avatar
 
🎧 5 years
Quote:
Originally Posted by StarfishMusic ➑️
Not to be rude but you just described an analog vs digital synth, I've been talking about a digital hardware synth vs a digital software synth. Both are software synths technically. No electricity is passing through individual components other than the CPU and ROM/RAM until the DAC unless your talking digital synths with analog components like an Emulator III which I am not. Does a Jd990 have analog filters? I don't think it does.
Yes, but in software, a single program is making the decsions as to what the resulting waveform should be, all contained within the CPU. In hardware, even though it has chips as well, there are components outside of those chips involved.
Old 10th September 2012
  #9
Lives for gear
 
🎧 5 years
I think the dedicated DSP is a red herring for anything that hasn't been made in the last few years or so. The Virus TI or Creamware stuff being examples of that.

With older digital synths I think it's mostly about the DAC and output amplification section.

Though I wouldn't be surprised if there's also some interesting things being done in the old code to make up for limited technology. Programming which doesn't make it into software, because it's assumed there's no need for it, but that actually has an effect on the vintage sound.

Last edited by FubarSnafu; 10th September 2012 at 02:11 AM.. Reason: Grammar
Old 10th September 2012 | Show parent
  #10
msl
Lives for gear
 
msl's Avatar
Quote:
Originally Posted by StarfishMusic ➑️
So current moderately high end $2500 SSL alphalink convertors and an antelope clock just give a less pleasing sound than a 14 year old module?
Yes.

"Pleasing" is relative, I'd just say different, my ASR10 kicks ass, and is super "warm" almost "creamy", my Nord Lead is "cold" and "clinical", neither are transparent, which can be "dull" sometimes



.
Old 10th September 2012 | Show parent
  #11
Lives for gear
 
StarfishMusic's Avatar
 
🎧 10 years
Quote:
Originally Posted by msl ➑️
Yes.

"Pleasing" is relative, I'd just say different, my ASR10 kicks ass, and is super "warm" almost "creamy".
I can't say I don't agree with you. I was a programmer on EMU E4s for awhile. Worked for Sonic Reality. I recently used some drums and Strings from an E4xt Ultra in a song. Have to admit they sound very different from the same exact multisamples in a software sampler. Why do you guys have to make me miss my E4? The good news is if they really do sound better or at least different in some ways then we better shut down this thread quick and keep E4s and real wavestations dirt cheap. BTW I owned a wavestation AD, thought it sounded horrible and I think the korg legacy cell software one sounds better.
Old 10th September 2012
  #12
Lives for gear
 
🎧 10 years
Haha, what about the quality of the programmers? Especially with FM there used to be way more money in the products and perhaps the programmers that were better compensated and probably teams to evaluate the sound quality.

Now with fewer people and less money, options are less diverse. It's not like just anyone can do that kind of stuff. You gotta have a big brain to program it and good ears to make sure it actually sounds good.

Summary of Gibson v. Crowe
Old 10th September 2012 | Show parent
  #13
165099
Guest
Quote:
Originally Posted by FullCount ➑️
...In software, there never is an actual audio signal until the very end after computation.
Quote:
Originally Posted by StarfishMusic ➑️
...Not to be rude but you just described an analog vs digital synth, I've been talking about a digital hardware synth vs a digital software synth. Both are software synths technically. No electricity is passing through individual components other than the CPU and ROM/RAM until the DAC unless your talking digital synths with analog components like an Emulator III which I am not. Does a Jd990 have analog filters? I don't think it does.
Quote:
Originally Posted by FullCount ➑️
Yes, but in software, a single program is making the decsions as to what the resulting waveform should be, all contained within the CPU. In hardware, even though it has chips as well, there are components outside of those chips involved.
I'm going to jump on this as well. Whether the synth uses, FM, modeling, wavetables or whatever, it is all zeros and ones and isn't an audio signal until it hits the DAC which is exactly the same as on a computer and the only real difference between the two is the quality of said DAC's, which on a modern audio interface, are (generally) going to be much better than those found on a digital synth from the 80's or 90's.
This is the reason why the VC1 D-50 card for the V-Synth has a setting called "D-50 Mode" which degrades the V-Synth's DAC to approximate that of the original D-50. DAC's aside, the other factor that often plays into the distinction between hardware and computer-based digital synths is the onboard effects.

anyway, on a modern synth like if say Roland were to release their V-Synth as a plug-in instrument there is no reason for it to sound any different whether in the box or outside.
Old 10th September 2012
  #14
Gear Addict
 
🎧 10 years
I like older digital synths, so , it may be because of the "lower bit" effects/DA converters. Wavestation sounds "colder" than Wavestation VST.
Old 10th September 2012 | Show parent
  #15
Gear Guru
 
Yoozer's Avatar
 
1 Review written
🎧 10 years
Quote:
Originally Posted by StarfishMusic ➑️
I guess the question is: is the full DSP power from 14 years ago still more than the dedication of say one 50th of a modern computer that's a thousand times stronger than a computer from that era?
Nope.

Even then, in the end, it's all a matter of MIPS.

A simplified comparison: Jack and John have to work in a factory loading boxes into a truck. John can only dedicate 6 hours per day to work but gets the job done in 2 hours, Jack has the full day but needs 20 hours to load in all the boxes because he's not as strong. The end result in both cases is that the truck is filled with boxes within a single day, so being able to spend more or less time on it is not the issue.

Quote:
Why don't modern synths have a knob that lets you scale how much computer power they take then if that's the source of sound quality?
They do! It's called oversampling and algorithm quality, though not all synths offer it. Even then there's no way to dial it up smoothly, because what usually happens is that one component of the synth runs a less CPU-intensive version of the algorithm.

Quote:
Does a Jd990 have analog filters? I don't think it does.
Nope. What does play a role, apparently (and it's usually lumped in with "converters") are the filters and opamps at the output stage.

Quote:
Originally Posted by foodeater ➑️
Haha, what about the quality of the programmers? Especially with FM there used to be way more money in the products and perhaps the programmers that were better compensated and probably teams to evaluate the sound quality.
The theory of FM synthesis is known. The patents on FM synthesis have expired; what Yamaha had back then was a fast and working implementation of it.

If a DX7 were a Super Nintendo with equally devoted fans, we would've had bit-accurate emulation by now. Building a software emulation of the chips in the DX7 should not be an insurmountable problem, but it means hooking up several probes to the processor and carefully studying the output stage. FM synthesis just doesn't get that much love, mostly because nobody's gotten the interface right yet.
Old 10th September 2012
  #16
Deleted User
Guest
No doubt, hardware digital synths (and reverbs) sounds to me broader, deeper and more focused, regardless if it is a cheap crappy box or an high end machine. That's a main reason I still have my XV-5080 in my slimed down studio setup. It would be easy to just sample and rebuild the favourite sounds in a software sampler as Kontakt and play through my nice sound card.

But the XV-5080 (or almost any other favourite sound module or keyboard of ones choise) still sound broader, deeper and more focused and with more edge. Just like with the DX7 vs FM8 that is discussed in another thread.

One reason is of course the D/A. I think that the old D/A (some good one that were popular at that time were Delta Sigma and another one was Lambda if I remember right) are more dirty and have some faults, comparable to older analog synths and unstable VCOs, overdrived filters and other "faults" etc that gave that special sound (no further comparison with the analog domain).

I was studio trained the old school way with 2" 24 track recorders, analog mixer consoles etc and learned to let the sound source pass as less electronics as possible, because all electronics gives color to the sound (in that case unwanted). In a digital synth is a lot of electronics that the signal pass through. I guess it colors the sound a lot.

I don't buy the argue with 100% processor power. During the heydays of digital synths in the 90s the manufacturers and resellers talked of megaflops (I have a small clue of it but far away from an expert of it) etc regarding to processor speed. Today a regular computer can handle at least comparable Gigaflops or Teraflops. 1% of my MB Pro can easy exceed all "megaflops" of my XV-5080 (or any other digital synth of that time).

I can also bit crunch the sound and get it more dirty, but it still does not make the trick...
Old 10th September 2012
  #17
Lives for gear
 
Persemone's Avatar
 
1 Review written
🎧 5 years
Don't confuse 'better' with 'different'. Rather than examining absolute numbers, processor power, Moore etc, isn't it possible you just prefer the individual sound character of your digital hardware?

We were doing that 40 years ago with Moogs and Arps, without resorting to capacitor roll calls..
Old 10th September 2012
  #18
Lives for gear
 
🎧 15 years
Converters aren't the issue, there's an inherent weakness, blurryness, lack of texture and cohesion with most of the plugins regardless of converters.

DSP power also is an unlikely explanation.

It comes down to the algorithms and implimentation. Most DSPs use integer math, which produces artifacts different from the rounding errors the floating point processing in your DAW makes. Also, on the hardware you had to take shortcuts and use trickery because of the limited resources, producing more artifacts.

Up-/downsampling in modern plugins also is far from transparent and many ways more destructive and detrimental to good sound.
Old 10th September 2012 | Show parent
  #19
Lives for gear
 
🎧 10 years
Quote:
Originally Posted by living sounds ➑️
It comes down to the algorithms and implimentation.
This.

Quote:
Originally Posted by FullCount ➑️
In hardware, even though it has chips as well, there are components outside of those chips involved.
In hardware digital/DSP synth, a single program is making the decsions as to what the resulting waveform should be. Beside DAC/analog path, that's it, as in "that is whole story".


Quote:
Originally Posted by Eric J ➑️
No. I'm saying a DSP chip in a digital synth can dedicate 100% of it's power to making sound while your computer CPU ...
... can dedicate 90-95% of it's resources to do number crunching. On the other hand, modern Intel CPU runs circles around average modern DSP. There is some top tier stuff that can compete (and costs quite a bit), but if you want real edge in hardware you would have to go down long and winding FPGA route.

edit: BTW, Korg Kronos runs on Intel Atom. That's crappy Atom, not some highend (strike that: even average) CPU. It's algo and details of implementation, full stop.
Old 10th September 2012 | Show parent
  #20
Lives for gear
 
🎧 5 years
Quote:
Originally Posted by Eric J ➑️
Dedicated DSP for one. A hardware synth can utilize 100% of the power of it's CPU purely for the purposes of sounds. Your computer CPU must handle a multitude of tasks which means it has less available power to dedicate to synthesizing sounds.
As a modern CPU is by a factor between 10 and several thousands faster than the usual Synth DSP, this isn't the reason alone - the problem is: The computer has to emulate the analog parts of a Synth (that don't cost a Synth anything in performance) too, which is harder to achieve and does consume a lot of it's power.
Old 10th September 2012
  #21
Gear Guru
 
SWAN808's Avatar
 
1 Review written
🎧 10 years
One reason is that developers can put more research and development and money into coding for a hardware device - because there is less financial risk of it being pirated. Money that they spend designing it to sound good will not get stolen....Also they may have more money at their disposal due to being part of a large company eg Roland. So in short - they could have spent more time working on the code of the synth. For example - the reason why Access Virus still sounds so good compared to the average soft synth is because of the development they have put into it over the years.

Then there is the potential coloration of digital to analogue converters...another thing worth mentioning is basically the process of taking a digital signal and sending it thru the process of DA->AD can add a certain element because its taken the digital signal out of the samplerate stream and turned it into full res analogue signal - which then gets resampled into the computer as new and independent data...its not unusual for people to notice a difference even running a signal out of their converters back in thru a loop.

Not all digital hardware sounds better than plugin software but those are 2 potential reasons why...
Old 10th September 2012 | Show parent
  #22
Lives for gear
 
iksrazal's Avatar
 
🎧 10 years
Quote:
Originally Posted by Yoozer ➑️
What does play a role, apparently (and it's usually lumped in with "converters") are the filters and opamps at the output stage.
That makes the most sense to me. Just about anything sounds better out of 1/4 jacks, lots of people have claimed as much out of the Virus. My Oasys is just Linux with laptop hardware, and its TRS out sounds miles better than anything I've ever gotten out of my Linux desktop.

Admittedly you can get more real-time tweaking and less resource sharing done with Linux but I still think the output stages, and the typical lack of the equivalent hw in computers, are the difference.
Old 10th September 2012
  #23
Gear Addict
 
🎧 10 years
I think the biggest reason is R&D. A lot of R&D time and money goes into some of those high-end digital synths that sell or used to sell for thousands, compareed to a $99 plug-in.
Old 10th September 2012
  #24
Lives for gear
 
Michael E's Avatar
 
🎧 10 years
By the way, how do you think - why we don't have any DSP based virtual synths in small package?
In another words like Virus but without controls & converters. Just USB.

Like hardware video converters which do the work about 5 times faster than just software.

Old 10th September 2012 | Show parent
  #25
Deleted User
Guest
Quote:
Originally Posted by Michael E ➑️
By the way, how do you think - why we don't have any DSP based virtual synths in small package?
In another words like Virus but without controls & converters. Just USB
Yes, that would be something!

Another reflection: Software synths from the heydays of many digital synts had very long latency at the computers at that time, but the latency in these hardware synths are about nothing, at least those I played. That's even true of digital synths and samplers that were manufactured much earlier as well.
Old 10th September 2012 | Show parent
  #26
Gear Guru
 
Yoozer's Avatar
 
1 Review written
🎧 10 years
Quote:
Originally Posted by Michael E ➑️
Like hardware video converters which do the work about 5 times faster than just software.
There are some significant differences; the conversion here does not have to be done in realtime, but you can throw as much or as little data at it as you want. With synths, you want continuous, uninterrupted audio.

Sure, you can put a heavy softsynth on a chip that's optimized for it so that offline rendering is done faster, but if you're doing that anyway, you might as well build a keyboard case around it or a module so that you can still access it.
Old 10th September 2012 | Show parent
  #27
Lives for gear
 
Persemone's Avatar
 
1 Review written
🎧 5 years
Quote:
Originally Posted by Michael E ➑️
By the way, how do you think - why we don't have any DSP based virtual synths in small package?
In another words like Virus but without controls & converters. Just USB.

Like hardware video converters which do the work about 5 times faster than just software.
We do, don't we? Powercore, UAD? Virus or V-station?
Old 10th September 2012
  #28
Moderator
 
Don Solaris's Avatar
 
4 Reviews written
🎧 10 years
Another important aspect of the sound is data companding. A lot of the "16 bit" ROM-plers are actually 8 bit (sample ROM) devices. i.e. Roland JV-1080, JD-990, Korg Triton, X-5 etc.

Korg and Roland are kings in that department. Their algorithms are stellar!
Old 10th September 2012 | Show parent
  #29
Gear Maniac
 
🎧 10 years
Quote:
Originally Posted by living sounds ➑️
Converters aren't the issue, there's an inherent weakness, blurryness, lack of texture and cohesion with most of the plugins regardless of converters.

DSP power also is an unlikely explanation.

It comes down to the algorithms and implimentation. Most DSPs use integer math, which produces artifacts different from the rounding errors the floating point processing in your DAW makes. Also, on the hardware you had to take shortcuts and use trickery because of the limited resources, producing more artifacts.

Up-/downsampling in modern plugins also is far from transparent and many ways more destructive and detrimental to good sound.
Yeah, I'd agree with most of this.

The same problem seems to appear in all the plugin synths I've tried.

Sure they have the general feel or frequency response to the hardware synth they are trying to emulate (or are influenced by), but there seems to be some kind of resolution factor missing from the plugins which make them soft sounding and cloudy in the mid-range.

It's almost like the coders are emulating the frequency response only and forgetting the dynamics.

It's in the dynamic of the attack that a lot of the magic lies with hardware synths and can make the difference between a tight punchy bass and a round mushy non-entity.

Not on an ADSR scale but something of a much higher resolution.

So I guess even on the digital hardware there is something happening in the output stage that adds some extra magic that coders are either missing, or that chews up so much cpu power that they don't bother with it.
Old 10th September 2012
  #30
Gear Nut
 
HHaynes's Avatar
 
🎧 10 years
I have to agree on DSP optimization on dedicated platforms influencing the sound. There are compromises that are made in plug-ins (with the tacit understanding that it's powered from a shared resource pool) that you don't have to consider when 'on the metal' and owning 100% of the resources.

Things like oversampling and free-running oscillators/modifiers can make a huge difference in the quality if a sound. I have all of the Arturia plugins and the Origin keyboard. Even though I like the idea of blending and melding parameter in the software interface I'd say I use the hardware 90% of the time. And I can't blame it on the placebo effect of having the front panel hardware, as I've mapped in most of the controls on the Origin to correspond to the 'like' controls on the VSTIs.

The FX on the Origin is no small part of it. I think all hardware (especially workstation) synths have been like that since their inception. It's like a guitarist with a few choice foot pedals between his/her guitar and speaker. Where is the dividing line between performance and processing when the musician makes no distinction during the act? The FX are part of 'the instrument' and therefore is an inextricable part of the sound. Ask any DSP coder (after a drink or three) and they will tell you all about 'the gives and gets' of generalized versus dedicated platforms.

I have more to say/type about this but will switch to my desktop after my morning coffee.
πŸ“ Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 627 views: 104769
Avatar for damoog
damoog 27th April 2019
replies: 835 views: 177322
Avatar for jlgrimes11
jlgrimes11 30th August 2020
replies: 518 views: 41619
Avatar for audslu
audslu 28th December 2015
replies: 498 views: 60861
Avatar for GlenGood
GlenGood 4 weeks ago
Topic:
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump