FXpansion D-CAM: Synth Squad for sale now!

[Angus_FX]

@ Krim - if you're in London, drop me a line - if I'm about in the office, you can pop by and see us.

[andy-cytomic]

So Shy, are these demos a little better to show the analog modeling?

I think that demo actually shows mostly how un-analog it sounds, compared to the analog demo and generally. I also think it's good that unlike most other developers you weren't afraid to post dry demos of the synth. Personally, except the smooth sound at high frequencies and with high overdrive (thanks to the offline oversampling), none of the demos really appealed to me and I think there's the usual digital harshness and lack of flexibility (literally, with fast attacks for example). But there's also something other digital synths lack, and it at least sounds like it's mostly thanks to how your overdrive is integrated with the basic sound.

Having automatic oversampling is really useful with the synths, especially with loads of drive. I thought that some of the growl of the sh-101 filter was still evident in the demo, definitely not as interesting sounding as real analog, but in essence the right direction. Thanks for the feedback, I really appreciate honest opinions.

Here are two more examples. One is of all oscs at full volume with square and with random LFO (filter at max and no resonance), and one with just self resonating filter and decay from slow to very fast. This is the kind of demos that would show any virtual analog synth's limitations even much more clearly than the high resonance filter demo.

Thanks for posting more examples I'll stick the synth squad on similar type stuff. I'm not trying to hide behind marketing here, I've done as much as possible with current cpu limits to get a great sound.

What I think your synths could be great for are things that are very hard and/or expensive or impossible to do with analog synths. To be able to do that with decent analog-inspired quality (by that I mean "not crap", which I think almost all other virtual analog synths are) is what would really be useful. I've still heared nothing that could replace a real analog synth, and you don't even claim that. So, all good. In short that's what I really think and it's nothing personal at all, so no offense is meant at all. I can definitely understand all the excitement about the synths. People have grown way too tired of big claims with nothing to back them up and these synths bring some fresh air free of bullshit.

Completely agree with you here, you're spot on. I'm very happy to be called "not crap, which I think almost all other virtual analog synths are", I take that as a huge compliment

The amount of cpu required to really be near indistinguishable from analog is beyond current computers, but I look forward to the next few years when I can really dedicate loads of cpu to the task.

Andrew Simper

[andy-cytomic]

truly fast, analogue sounding attacks are something that are normally lacking in software synths (as well as a few hardware analogue synths I could mention).

This is one thing that I have always felt a problem with softsynths - but have to say not many talk about it here - prob for fear of starting an argument...But to me its plain to hear synths like the Virus and Blofeld has stonger attacks than most softsynths particularly in percussive sounds. However I did notice in the Arp presets Strobe has some real clout. I find that to be really important to me. And yes not all hardware have strong attack. I sold my Studio Electronics ATC1 because of this. And also the Matrix 1000....

If Andy or anyone would like to chime in - I'd love to hear any reasons as to why this is - as at one point I wondered whether is was just not possible to implement agressive attack qualities in native software instruments...I guess its just in the nature of the coding?

Most analog envelopes are on only of the order of around 1ms at their fastest, which is around 88 samples at 44100. People typically can't tell the difference between a slight curve and a straight line at that speed, but just like oscillators, it's the discontinuities (jumps in the level or derivatives of the signal). When you have these fast transitions in modulator signals you need to band limit them just like you do with oscillators, to remove aliasing. This will be most evident when the modulator has a perceivable pitch, like running an lfo or looping envelope at near audio rate. This is why I made Cypher, which allows audio rate anti-aliased modulation between the oscillators. You can modulate the waveform, amplitude, and frequency of one osc from another. In "low" mode you can tune each osc to exact frequencies and even bmp sync them. I'll post some examples for you soon. If you want to do extreme lfo modulation, or ultra smooth envelope effects, Cypher is the one to turn do.

I hope that all made sense, I tried to make it clear, but I've been doing this stuff for a while so sometimes miss out on points to clarify understanding. If you have any questions fire away.

Andrew Simper

[pdxindy]

Most analog envelopes are on only of the order of around 1ms at their fastest, which is around 88 samples at 44100. People typically can't tell the difference between a slight curve and a straight line at that speed, but just like oscillators, it's the discontinuities (jumps in the level or derivatives of the signal). When you have these fast transitions in modulator signals you need to band limit them just like you do with oscillators, to remove aliasing. This will be most evident when the modulator has a perceivable pitch, like running an lfo or looping envelope at near audio rate. This is why I made Cypher, which allows audio rate anti-aliased modulation between the oscillators. You can modulate the waveform, amplitude, and frequency of one osc from another. In "low" mode you can tune each osc to exact frequencies and even bmp sync them. I'll post some examples for you soon. If you want to do extreme lfo modulation, or ultra smooth envelope effects, Cypher is the one to turn do.

I hope that all made sense, I tried to make it clear, but I've been doing this stuff for a while so sometimes miss out on points to clarify understanding. If you have any questions fire away.

Andrew Simper

Thank you very much for taking the time to make such explanations, and to post the audio samples of what you are talking about. It is quite educational!

There are lots of discussions on kvr, but without audio examples it usually devolves into subjective opinion. I am not expert by any means, but I have a good ear. I look forward to the audio examples from Cypher and its audio rate modulation capabilities!!

[andy-cytomic]

Here are two more examples. One is of all oscs at full volume with square and with random LFO (filter at max and no resonance)

Well but that example I'd say you are somewhere in europe running your sh-101 off a cheap wall wart transformer. Your sideband frequencies from mains hum bleed through are at to your osc are at 50 Hz, at -48 dB down from the main haromic, ie big, hence the cheap wall wart. This is the kind of thing that makes analog sound interesting. I have actually added this to the synths optinally via the "analog noise" knob, but the adaptor I had was only at around -70 dB of bleed through.

Andrew Simper

[Shy]

Right , my "good" tranformer broke down, gotta get another. However, I doubt it has any noticeable effect on the synth's sound (except being an annoying background sometimes).
Thanks for taking the time to reply to everyone.

[pdxindy]

truly fast, analogue sounding attacks are something that are normally lacking in software synths (as well as a few hardware analogue synths I could mention).

This is one thing that I have always felt a problem with softsynths - but have to say not many talk about it here - prob for fear of starting an argument...But to me its plain to hear synths like the Virus and Blofeld has stonger attacks than most softsynths particularly in percussive sounds.

But isn't Virus basically a softsynth? Isn't the comparison between true analogue and digital representations, not between hardware digital synths and digital synths you run on your computer?

I think the Virus sounds really good, but it does not sound like a true analogue synth to me. I don't think it is even particularly trying to sound analogue.

To me, the conversations on kvr get confused really quickly when talking about analogue. As one example, in many cases a really big layered/unison sound is associated with 'analogue'. The term has become so vague.

Here, I see Andrew actually trying to model the behavior of a true analogue synth. This task necessitates having a simple synth architecture because of the CPU requirements of modeling a 'simple' component.

Synths Like Zebra, Tera, etc etc are complex digital synths that sound excellent and have a very wide range of timbres but they are not trying to be analogue. I'd put the Virus in that category.

Looking forward to more audio examples!!!

[andy-cytomic]

Right , my "good" tranformer broke down, gotta get another. However, I doubt it has any noticeable effect on the synth's sound (except being an annoying background sometimes).
Thanks for taking the time to reply to everyone.

It has a big difference on the sound. It permeates every voltage in the synth, and makes it buzz more. This isn't necesseraly a bad thing, I quite like the extra buzz, but it does explain why I wasn't getting some of the background fuzz in the resonant filter example. If you want ultra smooth (ie borting) voltage use batteries, which the sh-101 allows you to do.

I love the sound of synths just working when you reduce the voltage till they start distorting and generally fuzzing out, loads of fun

I'm happy to reply to everyone, I've had some great posts from everyone that have jogged my memory on many of the things I've spent time thinking about for the synths. Thanks to everyone for asking questions and generally taking the time to be intersted in the thread. I really appreciate all the help you are giving me towards genrating good tech doc and audio examples.

Andrew Simper

[SWAN808]

truly fast, analogue sounding attacks are something that are normally lacking in software synths (as well as a few hardware analogue synths I could mention).

This is one thing that I have always felt a problem with softsynths - but have to say not many talk about it here - prob for fear of starting an argument...But to me its plain to hear synths like the Virus and Blofeld has stonger attacks than most softsynths particularly in percussive sounds.

But isn't Virus basically a softsynth? Isn't the comparison between true analogue and digital representations, not between hardware digital synths and digital synths you run on your computer?

I think the Virus sounds really good, but it does not sound like a true analogue synth to me. I don't think it is even particularly trying to sound analogue.

To me, the conversations on kvr get confused really quickly when talking about analogue. As one example, in many cases a really big layered/unison sound is associated with 'analogue'. The term has become so vague.

Here, I see Andrew actually trying to model the behavior of a true analogue synth. This task necessitates having a simple synth architecture because of the CPU requirements of modeling a 'simple' component.

Synths Like Zebra, Tera, etc etc are complex digital synths that sound excellent and have a very wide range of timbres but they are not trying to be analogue. I'd put the Virus in that category.

Looking forward to more audio examples!!!

In the example I was refering to 2 things really - firstly the speed of analogue envelopes but also what seems to be the speed of some of the hardware VA envelopes when compared to software. Thus yes that is the reason why I think it is strange given that hardware VA and software VA are similar. However there is a difference in that hardware VA runs off its own CPU as opposed to a jack of all trades computer CPU - thus I was interested as to whether there are coding limitations of some sort natively that arent there when on a discrete hardware DSP platform...it seems strange how a lot of bespoke DSP solutions have in my opinion higher quality plugins than most native given that they are essentially the same. SCOPE synths, UAD, Duende plugs, Virus TI etc...anyways thats another argument! Dont want to open that can of worms (whoops I just did).

[Angus_FX]

@ SWAN808 - it's more because their inherent pirate-resistance and the overall greater investment involved allows for a greater spend on R+D than most native outfits (or, in the case of Duende, safe porting of high end algos which already have a large R+D investment behind them). Access, for instance, probably has more total man-years of R+D behind them than any VSTi developer (with the exception of maybe NI) and have essentially dedicated all that effort to (repeated repackaging + enhancing of) a single product. Similarly, most of the best Pulsar/Scope synths were developed over a period of years by John Bowen, who was literally designing and building synths before most VSTi developers were born. Universal Audio have a DSP software team few VSTi firms could hope to match.

Also, perhaps, hardware prohibits or makes difficult some of the more bells-and-whistles features that are easy on software VSTis, maybe there are less distractions.

And, as those cards are accelerators, people tolerate a greater amount of load per plug-in than they often do with native. The fully modelled compressors on the original UAD-1 chewed a huge amount of power.

D-CAM: Synth Squad has been a long, long time in the works, and that's given us time to investigate things in more depth than many, most obviously on the raw DSP side but also in relation to the synth layout, workflow, and parameter ranging and choice. We're taking a bit of a gamble that people will a) appreciate it, and b) do the right thing (not the w***z thing), but so far the response is very encouraging!

All the best,
Angus.

[SWAN808]

D-CAM: Synth Squad has been a long, long time in the works, and that's given us time to investigate things in more depth than many, most obviously on the raw DSP side but also in relation to the synth layout, workflow, and parameter ranging and choice. We're taking a bit of a gamble that people will a) appreciate it, and b) do the right thing (not the w***z thing), but so far the response is very encouraging!

All the best,
Angus.

Thanks for the reply Angus - that makes sense and its good to hear a developer discuss this as I had only been speculating in my mind...interesting how native DSP development is in such a way controlled by the problem of piracy. As a UAD owner myself it is interesting to see companies like Manley - classically a devout 'hardware' company- stepping up to work with them. Obviously the piracy issue is a real strongpoint for UAD (along with their coding skills, userbase and general awesomeness)...

One question I had was I take it you considered Syncrosoft as a protection for SYnthsquad? I dont know too much about it but it seems the only real safe option atm for anti-piracy...I only ask because Im interested in the way this problem is dealt with by developers.

[Angus_FX]

I detest dongles and dongle vendors. It will be a sad day for us and our customers, and a victory for the pirates, if Syncrosoft or PACE ever sees a cent from us. Most (though not all) of the things we've made have not been pirated too badly despite minimal copyprotection; bear in mind that the more apps that use iLok or Syncrosoft, the greater the prize for the hacker that breaks it (and the longer they've been at work on doing so).

To repeat what was said earlier in the thread - the system we're going with is basically the same as BFD2 has, which is our own challenge/response system with "online" (automatic) and "offline" (via key files and a second, Internet-connected machine) activation, and a fairly generous policy on hardware changes and machine authorizations. (insert drummer joke relating to ease of use here).

[MitchK1989]

Angus, as a laptop user who pays for all of their software, I am very glad to hear you say that.

[pdxindy]

I detest dongles and dongle vendors. It will be a sad day for us and our customers, and a victory for the pirates, if Syncrosoft or PACE ever sees a cent from us.

That is one significant reason you get my money!

Love GURU and will be getting Synth Squad as soon as it is released.

I REFUSE to get stuff with dongles!

[zvenx]

Each method has its advantages and disadvantages...... I guess to each his own...to me that stance makes you miss out on some wonderful stuff out there that require dongles...I am sure most of Angus customers however appreciate their method of protection.
rsp