A few observations (maybe some misunderstandings)


#1

I’ve only been testing out Circle2 for a couple of hours, but a few things have stuck out to me. Not sure where this type of feedback/observations should go in these forums.

  • There doesn’t appear to be a way to adjust oscillator phase other than hard sync. I’m guessing the oscillators are free-running otherwise? Would be nice to be able to adjust phase, particularly when using unison
  • Filter frequency only goes to about 10Khz. At first I thought the analog saw sounded incredibly dull, then I realised I needed to disable the filter to hear the full frequency spectrum. Seems odd for an INIT preset to be substantially cutting frequencies
  • Cannot target master tune with modulation. This one’s more of a nice-to-have
  • Cannot adjust unison spread/width. The unison sounds very narrow and I need to use chorus to widen it
  • There’s a strange lag when MIDI notes are playing and you adjust glide. It can take a few MIDI note-ons for the adjusted glide amount to “catch up”
  • There’s no way to bypass individual modulations other than bypassing the modulator itself
  • Is the maximum modulation amount very low? For example, an envelope with max sustain assigned to an oscillator’s coarse pitch with maximum modulation amount only raises the pitch 9 semitones. Do the same in a lot of other synths and only dogs will hear the sound that’s produced (exaggeration, but you get it)

#2

Thanks for the feedback. We’ve always been careful of implementing niche features that get in the way of experimentation and usability, but you raise several great points here which I’ve logged as feature requests.

We’re aiming to begin developing Circle3 over this year so they’ll be considered for that if they don’t find their way into Circle2.


#3

Thanks for the response Hugh. I don’t think you are implying that all my comments pertain to niche features, but I appreciate what you mean, and I understand that experimentation is one of Circle2’s guiding principles and greatest strengths.

Adjustable parameters such as oscillator phase and unison spread are features I’ve come to expect from modern synthesizer plug-ins, since they are both useful and seem to show up frequently. I was actually surprised and a little disappointed to see that they weren’t a feature of Circle2.

As cool and fun to use as Circle2 is, especially with regard to experimentation, I am somewhat hesitant to make the synthesizer a staple in my plug-in library because I know that some of the features listed in my original post are things I want to make use of regularly, and they are present in other plug-ins I own. That’s not to say Circle2 has nothing to offer me, it’s just that I’d rather not have that “if only” feeling when using your product.

Great to hear that you will consider the points I raised for future implementation.

Thanks.


#4

Hey @VimDoozy

First, thanks for the great feedback, we really appreciate you taking the time to put together some well thought out feedback for us.

On the unison, we do actually have a spread feature, it is controlled via the “global detune” knob in the settings panel. I’ll have to confirm in the code, but as far as I remember it also sets a phase amount so that the oscillator phase won’t be the same across the voices.

On the filter frequency, what we did was to implement the filter as a model of the Sequential Pro One. With this and other analog synths, the cutoff control rises to 10Khz with out any modulation. If you then attached a modulation, say the ADSR control, it exponentially opens the filter cutoff as it goes through the phases. If you start at 10Khz (max filter freq without modulation) and then apply the adsr to the cut-off, it will rise all the way up past 20kHz and beyond.

The actual calculation for the cutoff:

cutoff = knobCutoffpow(1.059,modulation92.0) with modulation going from 0 to 1.

I notice that Native Instruments all do controls that run the full range, it was a decision on my behalf to go the old analog route as it was what I was used to previously. Definitely something we can switch up in a future version, say circle3 :wink:

Bypass modulation can be achieved by setting the amount to zero. I’ve seen bypass in other synths, but generally they make it harder to do the connection in the first place so bypass is something that you need as the effort to reconnect the modulation is a pain. Does that make sense? I’d be interested in seeing some comparisions where bypass is implemented so as to have a check and see why they do it.

Modulation should be +/-24 or +/-12 semitones from the base value of the knob.

Gavin.


#5

Hey @Gavin_FAW

Thanks for the thorough explanations.

I’ll have to test a few of the things you mentioned and get back to you. Perhaps I overlooked a couple of things.

I did twiddle about with the global detune parameter and I just wasn’t getting the width I was expecting - which is why I started looking for a width/spread knob. It just seemed to thicken, but not widen. Typically detune and width/spread are different parameters. I’ll take another look.

I was only getting +9 semitones max modulation on the Osc pitch from the base value, but I’ll try it again. Maybe I didn’t have it set quite right. Even if it is +12, that’s far too low to add a nice snap to the attack (which is what I had in mind). But there’s probably other ways to achieve this.

Thanks!


#6

I think you might be looking for stereo unison? That is where the voices are spread out across the stereo field. Its hard to achieve as we would need stereo voices, thus doubling the processor usage, or another way is having a voice mixer that takes the mono voice output and turns each into stereo before sending to the effects. That way you could pan some voices left, right etc…


#7

I think you might be looking for stereo unison?

Yes, that’s what I meant. Thanks for confirming.


#8

@Gavin_FAW

2.1.1 has introduced a very troubling issue for me. The coloured modulator circles catch and insert themselves upon slots which are passed over by the cursor before the mouse button is released. It’s essentially brushing modulators across the slots.

Check out the modulator slot issue here

On a seemingly related note, the modulation amount slider is bugged too. It disappears the instant you release the mouse button after having utilised it, and it will persist on screen if you summon it and subsequently click anywhere else on the interface other than the modulation slot it relates to.

Watch the modulation slider bug here

I also have a feeling that the frame rate of the coloured modulator circles when you drag them about has worsened, but I could be wrong about that.

On less concerning note, I had the sense that your 20 min demo period was not actually running for the full 20 mins. I confirmed it today with a timer. 9 min and 30s remained on the clock when the noise and detune kicked in.

I also tested your advice with regard to the filter cutoff raising way up past 20KHz and beyond when modulated with an envelope. My tests show that the cutoff is both visually and audibly raised only a marginal amount. I honestly think there is something wrong with your modulation, which is why I can also only get an envelope to raise the pitch 9 semitones.

I apogolise, I have not yet set up my mac to be able to capture audio and screen together, but I’ve used an EQ display and tuner to demonstrate the following:

Filter only being raised marginally by env

Watch the EQ display. With filter off there’s a huge jump in amplitude of frequencies around 20KHz. When the filter is switched back on and an envelope is applied to the cutoff with max amount, it registers only a small bump to high frequency amplitude compared with no modulation.

Watch the env raise pitch +9 semitones

Look at the tuner display here. Playing a C with env applied to coarse pitch raises the pitch to an A. I can’t see how I’m making a mistake here. The modulation amount is as high as I can go and the env sustain is at maximum too.

Finally, regarding your comment about bypassing modulators, I think that your recommendation to set modulation amount to zero is nothing more than a workaround - and an inconvenient one at that. This method has its shortcomings. It demands that you summon the amount slider and then reset the value manually 0 - and you’re further inconvenienced by the fact that this slider is one of the few interface elements which cannot be reset to 0 with a double-click, or other shortcut.

A more convenient workaround, in some cases, is to drag the modulator out of the slot momentarily. This provides instant bypass, but it prevents you from moving on to something else, since you’re still holding the modulator circle. True bypass provides fast access disabling, and fast access re-enabling when you see fit to undo bypass. A simple Right-click > Bypass would suffice, but I’m guessing you may be avoiding this because your units in the synth aren’t currently uniquely labelled.

Thanks Gavin.


#9

@VimDoozy, just testing here and I’m seeing the same thing with the coloured circles catching. That is totally new and not sure where it has come from. I’m looking into it right now.

Gavin.


#10

@VimDoozy, I’ve fixed the bug you highlighted with the drag and drop modulation, great find, I’ve also added a column in our test check.

Here is the latest OSX build, RC2
http://www.futureaudioworkshop.com/downloads/Circle-2.1.1.RC2-setup.dmg
http://www.futureaudioworkshop.com/downloads/Circle-2.1.1.RC2-Splice-setup.dmg

If you can check it out when you get a chance and again thanks for the invaluable feedback and help with testing.


#11

@Gavin_FAW

Confirming that the drag and drop modulation bug is no longer present in the OSX RC2 build.


#12

Thanks @VimDoozy for checking!

I’m going to check everything else you have posted about, was focused on the mod circle bug the past day.

Gavin.