Cytomic 'The Glue' Compressor

[Funkybot's Evil Twin]

Andy, question: why do I seem to be getting more gain reduction/compression at the 2:1 ratio than I am at 10:1? It seems most noticeable with the fastest attack times.

Try this:

1. Load up a live drum track and insert The Glue on the buss
2. Set attack to .01, release to .1, ratio at 10:1, range around -40
3. Set the threshold so the the standard peak meter is showing about -5db of reduction when the kick hits
4. Now cycle between the 10:1 and 2:1 ratios

Result: At 10:1 it sounds like I'm getting less compression than 2:1. The new peak meter confirms this. In my case, the peak meter shows gain reduction up to about but never exceeding -9.5db of reduction at 10:1. At 2:1 however, the gain reduction will go as high as about -12db. Also, in cases where I'm doing some light leveling, I've seen the meter not move at all at 10:1 while bouncing around a bit at 2:1.

Expected Result: I'd expect the deeper compression (more reduction) with the 10:1 ratio than with the 2:1 ratio.

FYI - I notice the same kind of thing with the BFD2 Bus Compressor.

[meloco_go]

That must be because at 2 it is actually a soft-knee compression, so compression starts quite a bit below threshold. The higher the ratio, the smaller the knee.

[andy-cytomic]

Andy, if it is possible to do without disclosing your know-how's, can you please tell what did you do to improve attack behavior of the Glue?
I remember you said, you've used advanced equation solver, was that the key?
I'm asking because there's a lot of talk about how to get close to analog sound you have to focus on nonlinearities and stuff, but from what I gathered that's opposite of what you do -- getting the virtual circle do what it was planned to do without imperfections.

I can't go into much detail on how I improved the attack sound specifically, but I can let you know the approach I'm using. The sidechain section of The Glue is highly non-linear, and the circuit equations are solved each and every sample for the non-linear voltages and currents of all the critical circuit components. I use a modified implicit trapezoidal non-linear solver to do this, which has been optimized for real time use. This is the same core circuit mathematics that circuit simulation packages use. I keep the main signal path completely linear so all the character is coming from the gain control coming from the sidechain, which means that when the compressor is not in action you get a completely clean (no drive or any distortion at all) output signal. There are plenty of imperfections modeled in The Glue, this is needed to get a sound like the analog version, but I keep all the imperfections and non-linearities in the sidechain section only, since I believe this is the most useful thing for a mastering compressor.

[andy-cytomic]

Andy, question: why do I seem to be getting more gain reduction/compression at the 2:1 ratio than I am at 10:1? It seems most noticeable with the fastest attack times.

Try this:

1. Load up a live drum track and insert The Glue on the buss
2. Set attack to .01, release to .1, ratio at 10:1, range around -40
3. Set the threshold so the the standard peak meter is showing about -5db of reduction when the kick hits
4. Now cycle between the 10:1 and 2:1 ratios

Result: At 10:1 it sounds like I'm getting less compression than 2:1. The new peak meter confirms this. In my case, the peak meter shows gain reduction up to about but never exceeding -9.5db of reduction at 10:1. At 2:1 however, the gain reduction will go as high as about -12db. Also, in cases where I'm doing some light leveling, I've seen the meter not move at all at 10:1 while bouncing around a bit at 2:1.

Expected Result: I'd expect the deeper compression (more reduction) with the 10:1 ratio than with the 2:1 ratio.

FYI - I notice the same kind of thing with the BFD2 Bus Compressor.

Each ratio has a different threshold offset. This is done so that when compressing around -5 to -10 dB you can switch ratios and more easily compare the sound. This is modeled directly from the analog circuit of the original.

[andy-cytomic]

Being a novice/hobbyist bought glue a few weeks. Installed the 64bit today and been playing with it in a few tracks. Messing around using it in channel strips and masterout. It really works for me and I'm stickinng with it. I know there's a lot of threads on the net comparing it to waves and such. For my money it really does the job written on the label. Wish I discovered it earlier. Great product, great price and nice to see a solid developer delivering top notch goods. You deserve success, sterling result and thank you.

Thanks! Please help spread the word about The Glue to your friends as this really helps

[IIRs]

Please contact Sonic Charge with your Microtonic serial number and they will find your discount code for you and possibly update your email address at the same time

Ok, got the discount code, grabbed the demo, am now very tempted! Is there a time limit to the offer?

Quick question while I'm here: the numeric readouts under each knob don't update in realtime: I have to move the mouse cursor away from the knob then back again to see the new value. Testing in Reaper and Tracktion. Anyone else get that behaviour, and is it the same in the registered versions?

[andy-cytomic]

Please contact Sonic Charge with your Microtonic serial number and they will find your discount code for you and possibly update your email address at the same time

Ok, got the discount code, grabbed the demo, am now very tempted! Is there a time limit to the offer?

Quick question while I'm here: the numeric readouts under each knob don't update in realtime: I have to move the mouse cursor away from the knob then back again to see the new value. Testing in Reaper and Tracktion. Anyone else get that behaviour, and is it the same in the registered versions?

The deal is limited in time. Can you please email me a few more details about the system you have? The email address and list of useful info is here: www.cytomic.com/contact - I can then forward you the full email set out by sonic charge which contains all the details of the discount.

[adl]

1.016 works fine in Live 8.1.3 on my Win XP SP3 desktop PC btw.

Really like the RMS needle, great update!!

[Funkybot's Evil Twin]

Each ratio has a different threshold offset. This is done so that when compressing around -5 to -10 dB you can switch ratios and more easily compare the sound. This is modeled directly from the analog circuit of the original.

Thanks Andy. The new beta sounds and works great (in Sonar, using XP) BTW.

[dub3000]

sounding great here!

[highkoo]

Thanks! Please help spread the word about The Glue to your friends as this really helps

phew!
a friend 'spread the word' to me a few nights ago, by slamming one of my (our) drum tracks through your plugin, and now i finally see why this thread is here.
i am an idiot, but i dont think ive heard drums get pushed quite like that. instantly impressed.
im guessing its a bitter pill for Waves owners to buy this.
i am not really shopping.. but i think this will be the next compressor i would buy.

[skank]

Yikes!

148 pages, when did it get this big ?

Anyway this is fast moving up my must have list.

[LarsErik]

I have no idea how the real british big console buss compressor sounds or behaves but I do know that I think The Glue is a great sounding compressor.

Keep up the good work Mr andy_cytomic.

Sincerely
LarsErik

[djanthonyw]

Latest update sounds amazing!

[djanthonyw]

Latest update sounds amazing!