KVR Audio

thelizard · Post by **thelizard** » Wed Jul 05, 2017 6:29 pm

Soarer wrote:
Soarer wrote:Looks good but are there no controls for a single modules effect?
I do see that it still allows for a lot of effects with the Geometry effect etc.
Or is "Amount" actually an effect control i.e. changing the spectral content, rather than just a dry/wet control?

Yep, Amount is an effect intensity, not a dry/wet amount. Each Amount control has been heavily tweaked to feel great and respond well to modulation. They run in serial from top-to-bottom, so you can create very complex spectral effects just with the three operators.

The DSP flow is Analysis->Speed->Geometry->Ops 1-3->Spectral Compander->Analog-style LP->Mix.

EDIT: To follow up, in the teaser image, you can see this in action. The first region, Expand, takes the frequency bins and spreads them out (check out the visualizer with the dark, empty spaces in the first region). The Amount control determines how expanded they become. This actually acts like a pitch shifter! The Smear controls lets each bin bleed into its neighbors, with Amount determining how much bleed occurs. In the visualizer, the blurry section is actually the Smear effect.

Soarer · Post by **Soarer** » Wed Jul 05, 2017 6:45 pm

Now that's what I like to hear!!!

Sounds freakin fabulous!

tl · Post by tl » Wed Jul 05, 2017 8:13 pm

thelizard wrote:SpecOps is a massive spectral effect with a performance-focused interface.

... and I can't wait to lay my hands on it.

I'm sooo excited about this plugin being released.

Cheers,
tl.

acousticglue · Post by **acousticglue** » Wed Jul 05, 2017 9:39 pm

how does it work in real time? Latency?

thelizard · Post by **thelizard** » Wed Jul 05, 2017 9:48 pm

acousticglue wrote:how does it work in real time? Latency?

Latency depends on the size of the FFT (really, it's an overlapping STFT, but FFT is the more well-known term, which is why it's on the interface) along with your sampling rate. So, an FFT size of 512 will provide 512 samples of latency. At a sampling rate of 96kHz, that's about 5-6 ms of latency.

EnderWillSaveUs · Post by **EnderWillSaveUs** » Wed Jul 05, 2017 10:29 pm

thelizard wrote:
acousticglue wrote:how does it work in real time? Latency?
Latency depends on the size of the FFT (really, it's an overlapping STFT, but FFT is the more well-known term, which is why it's on the interface) along with your sampling rate. So, an FFT size of 512 will provide 512 samples of latency. At a sampling rate of 96kHz, that's about 5-6 ms of latency.

ETA on release? Super excited!

incubus · Post by **incubus** » Thu Jul 06, 2017 5:35 am

It'll be ready when it's ready.

Christ, doesn't anyone understand..................

Soarer · Post by **Soarer** » Thu Jul 06, 2017 8:33 am

One other important thing: How about stereo width? Can you get this wide lush stereo width that is characteristic of many FFT processors? I don't see any pan or width control yet. Anyway, that is a feature request - would be great!

Check my soundscape track here that is made with a Max for Live spectral freezer - it has stereo processing:
https://soundcloud.com/soarer/thawing

thelizard · Post by **thelizard** » Fri Jul 07, 2017 5:03 am

After nearly a year-and-a-half of work, we finally submitted the SpecOps alpha builds today.

Soarer, there's not a dedicated Stereo Width effect, but after the final interface cleanup, there's room for it. We'll test it this weekend. That being said, a lot of the effects can already create stereo interest. Nice track! Which freezer is that?

Soarer · Post by **Soarer** » Fri Jul 07, 2017 8:44 am

But it does process in stereo right (i.e. separate processing for each channel)?

I used a Max for Live effect called Freeze which was processing in mono and I then added separate processing for each channel. Also added an LFO which does the on/off switching.

http://www.maxforlive.com/library/device/346/freeze

mt3 · Post by **mt3** » Fri Jul 07, 2017 10:40 am

thelizard wrote:
cron wrote:Great to get a proper look at this thing. Excellent stuff! I look forward to no longer bugging you about SpecOps and bugging you about the next thing instead.
It's a creative compressor. Have at it

What if I lack actual creativity? What would happen to me in that case? Through-zero creative reduction?

acousticglue · Post by **acousticglue** » Fri Jul 07, 2017 11:01 am

Your nads would turn to dust and be a guest star on Big Lebowski. Two.

thelizard · Post by **thelizard** » Fri Jul 07, 2017 4:38 pm

Soarer wrote:But it does process in stereo right (i.e. separate processing for each channel)?

I used a Max for Live effect called Freeze which was processing in mono and I then added separate processing for each channel. Also added an LFO which does the on/off switching.

http://www.maxforlive.com/library/device/346/freeze

Yes, it has true stereo processing. No mono averaging here.

CPU usage is about even with Fault. Ableton is reporting about 7-10% on my 2014 MacBook Pro. The most expensive effect, Smear, raises it to about 13%.

Soarer · Post by **Soarer** » Fri Jul 07, 2017 5:19 pm

Oh thank god!

- thank you I mean

Then artificial widening doesn't actually matter that much but it can sound great of course. I guess that is achieved by adding some difference/randomness to the spectral bands on each channel...

Thank you for listening to our suggestions.

Soarer · Post by **Soarer** » Fri Jul 07, 2017 7:53 pm

I was always awed by the noisy spectral beats of Yasume from the golden era of IDM in the 2000s.
Now I think spectral processing is mostly used in the Star Wars SFX department

Listen here to 2112 Cresent Hights by Yasume for those cool crispy FFT beats (and lovely tunes):

New things from Unfiltered Audio