Cytomic 'The Glue' Compressor

[andy-cytomic]

I've not updated the internal algorithms at all, but I have updated the linear phase oversampling filters to reduce the latency down to 100 samples maximum. Unless you are using oversampling then there shouldn't be any difference at all, and even with oversampling enabled I wouldn't expect there to be any audible difference.

[meloco_go]

I've not updated the internal algorithms at all, but I have updated the linear phase oversampling filters to reduce the latency down to 100 samples maximum. Unless you are using oversampling then there shouldn't be any difference at all, and even with oversampling enabled I wouldn't expect there to be any audible difference.

I use oversampling, and there is a difference)) But I like that difference)

[andy-cytomic]

You should always judge the quality of a plugin by listening to it in context, but if you suspect something is amiss it can be useful to check out what a plugin is doing on some basic test signals and looking at the time domain wavefeorm, or the spectrum.

In doing an accurate analog model of the envelope follower section of a compressor there is a completely smooth and gradual transition to the diode charging up the envelope follower capacitor. A good way to test this transition is with a 993 hz sin wave test signal, a fast attack, a slow release, high ratio, and a small amount of total compression depth, and then look at the resultant spectrum with a spectrum analyser. Here are some plots of exactly these tests using identical settings on three compressors which are modelling almost exactly the same circuit, the red spectrum content is undesirable aliasing caused by a non-smooth envelope follower transitions:

[andy-cytomic]

I've not updated the internal algorithms at all, but I have updated the linear phase oversampling filters to reduce the latency down to 100 samples maximum. Unless you are using oversampling then there shouldn't be any difference at all, and even with oversampling enabled I wouldn't expect there to be any audible difference.

I use oversampling, and there is a difference)) But I like that difference)

If you use oversampling and a 44.1 kHz or 48 kHz base rate then the new filter is flatter for longer up high, perhaps this is what you are hearing? This may lead to a slightly brighter sound, but the difference is only above 19 kHz, so you will need very good ears and speakers to hear the difference - you must have both

[meloco_go]

If you use oversampling and a 44.1 kHz or 48 kHz base rate then the new filter is flatter for longer up high, perhaps this is what you are hearing? This may lead to a slightly brighter sound, but the difference is only above 19 kHz, so you will need very good ears and speakers to hear the difference - you must have both

Well, I checked myself on WinABX software and got 7/10 out of my first trial. That is not 10/10 of course but the probability of guessing which that software also calculates was 17.2%. Given that I was distracted during the test and was in rather noisy environment I'd say that's pretty good. And when I listened I wasn't "aiming in the dark" there were subtle but significant differences which I was listening for.
And the funny thing, this wasn't top end. The older version had sort of "grainy" feel in it, it was as the mid frequencies were denser and masked the bottom end, on the newer version the mids are cleaner and they do not get in the way of the bottom.
Wonder what could that be, maybe there is less noise accumulation on the shorter kernel? Or there is miniscule amplitude response ripple in the older version?
Anyway, that sort of differences are also what separetes good and bad converters in my experience.
Food for thought, huh?))

[andy-cytomic]

It is indeed food for thought and something I would love to be able to analyse in a scientific manner. I have a couple ideas what the difference could be so I'll get in contact via email and find out exactly what is going on.

[penguinfromdeep]

I just downloaded Duende demo for the kicks and compared (by ear) the bus compressor to Glue. The Duende bus compressor distorts really easily with fast attack times with material that has lot of low frequency content (dance type beats), Glue stays really smooth even without any oversampling or sidechain enabled. Damn, I was really looking forward to Duende but I was really disappointed. Does the hardware really distort nastily like this? To be honest it doesn't sound like good analog distortion. Glue sounds better to me

Duende processed

Cytomic processed

Listen how the kick is the distorted in the Duende examples... Not in a good way imho.
Here's another, more extreme example with more stuff and lots of compression:

Duende Processed

Cytomic Processed

[xxmonty78xx]

http://photo.rukes.com/premsg/slides/premsg%20090.html


BUSTED! STEVE ANGELLO MASTERCHAIN exposed!

Please can some people explain this to me!!!!

[dusted william]

http://photo.rukes.com/premsg/slides/premsg%20090.html


BUSTED! STEVE ANGELLO MASTERCHAIN exposed!

Please can some people explain this to me!!!!

where?

dw

[xxmonty78xx]

in the photo link above my previous post you can see his chain to the left of the youtube video..

[dusted william]

the Glue
Ozone 4
TRacks something.something
I can't read that last one.

[xxmonty78xx]

yeah i know what it is.

Logic Gain Plug
the Glue
Ozone 4
T-Racks3
PAZ Analyser

I really need an explaination as to what kind of settings they have etc.
why these plugins etc.

[ObsoleteAcc99]

[DELETED]

[izonin]

These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".

[jam92189]

mm i like my compressors but i have been thinking of getting the glue, I might just have to next time I have 100 bucks(I just paid all the bills) thanks for the examples