What's new

2C Audio B2 or Perfect Storm 3.0

B2 has the silkiest highs and the sexiest dampening filters for high freqs. They truly shine at 4x oversample, but that's where it has one of the most insane CPU loads I've experienced in a reverb.

Luckily, it has the option to have a global 1x in Realtime, and 4x for Offline, so you can monitor and mix with 1x (do try and check how 4x sounds), and then it will switch to 4x for Offline Rendering. Wonderful.

yes, 4x OS is literally processing 4x the number of samples, so it is literally 4x the amount of work PLUS the work to do the oversampling (sample rate conversion) process at extreme quality (two long filters, zero stuffing, decimation etc). Worse, it ends up using a ton of memory/cache which is really the main limit to performance.

But it does indeed sound sublime.

Filters most certainly gain some benefit from oversampling. Moving from 1x to 2x gets you 90% of the benefit, and moving from 2x to 4x gets only the last 10% or so. 2x is a reasonable compromise. If your host rate is 44.1, and you switch between 1x and 2x you can notice the curve changes some. This is accurate do to the way digital filters work (BLT wraps analog freq response from 0 to inf, to digital freq response of 0 to Pi (half sample rate), so digital hiCut/loPass filters always have exactly 0.0 magnitude at half sample rate even if you have a high cut (low pass) filter set to 20,000hz. the analog version would be quite different. Oversampling helps this a lot as well as other things.)

Note if you are already running at a host rate of 96K, you might not need 2x, and you certainly don't need 4x. And personally IMHO you should NEVER run your host higher than 96k. 192 etc host sample rates and audio files is complete marketing stuff. Some dsp processes certainly benefit from 192 and even much higher, but running the whole host that way and keeping things on disk at such rates is crazy. Oversample only what is needed and if the base SR is 96k, this is already enough wiggle room to maintain perfection even with extreme dsp happening.

btw, regarding "sexiest dampening filters" do you have any particular favorites in terms of the filter options in B2. I've been making some more in the past week actually...
 
Last edited:
yes, 4x OS is literally processing 4x the number of samples, so it is literally 4x the amount of work PLUS the work to do the oversampling (sample rate conversion) process at extreme quality (two long filters, zero stuffing, decimation etc). Worse, it ends up using a ton of memory/cache which is really the main limit to performance.

But it does indeed sound sublime.

Filters most certainly gain some benefit from oversampling. Moving from 1x to 2x gets you 90% of the benefit, and moving from 2x to 4x gets only the last 10% or so. 2x is a reasonable compromise. If your host rate is 44.1, and you switch between 1x and 2x you can notice the curve changes some. This is accurate do to the way digital filters work (BLT wraps analog freq response from 0 to inf, to digital freq response of 0 to Pi (half sample rate), so digital hiCut/loPass filters always have exactly 0.0 magnitude at half sample rate even if you have a high cut (low pass) filter set to 20,000hz. the analog version would be quite different. Oversampling helps this a lot as well as other things.)

Note if you are already running at a host rate of 96K, you might not need 2x, and you certainly don't need 4x. And personally IMHO you should NEVER run your host higher than 96k. 192 etc host sample rates and audio files is complete marketing stuff. Some dsp processes certainly benefit from 192 and even much higher, but running the whole host that way and keeping things on disk at such rates is crazy. Oversample only what is needed and if the base SR is 96k, this is already enough wiggle room to maintain perfection even with extreme dsp happening.

btw, regarding "sexiest dampening filters" do you have any particular favorites in terms of the filter options in B2. I've been making some more in the past week actually...

I run at 48k. I'll only go to 96k if I'm specifically asked. True, 2x at 96k is the most anyone would ever need. But that's why there's the offline setting which is just great.

I absolutely love Air II A and Air II B.
 
This thread encouraged me to try it again (Breeze 2, i think I had tried a pre release version).
Whilst I find the GUI superb, it once again reminds me how personal sound is.
I have only had two reverb developers who work I really love... exponential audio and relab.... why? for me, I love a reverb that sounds like it is part of the sound, that it blends in vs just being on top of the sound. Almost like the difference between the sugar used to bake a cake vs, the sugar used for the icing... Alas Breeze 2 hasn't changed that opinion that for me there are only two developers who have the reverb being part of the sound right...

Good luck on the sale.
rsp
 
Last edited:
This thread encouraged me to try it again (Breeze 2, i think I had tried a pre release version).
Whilst I find the GUI superb, it once again reminds me how personal sound is.
I have only had two reverb developers who work I really love... exponential audio and relab.... why? for me, I love a reverb that sounds like it is part of the sound, that it blends in vs just being on top of the sound. Almost like the difference between the sugar used to bake a cake vs, the cake used for the icing... Alas Breeze 2 hasn't changed that opinion that for me there are only two developers who have the reverb being part of the sound right...

Good luck on the sale.
rsp

with Exponential i would say Phoenix would be the closest match to Breeze 2.

for me, Breeze 2 wins out.

now, R4 is a different beast altogether - possibly closest to B2, not sure - B2 having a "dual engine" seems unique in reverb offerings.

CPU usage may be key between R4 and B2.
 
I a/b;ed it with R4 actually because Phoenix/Nimbus are invisible.... I figure it would have a better chance vs R4.
I actually use Nimbus much more than R4........ Also I meant software developer in my post two up.
But let me stop talking about EA in the midst of a 2C thread..

rsp
 
I a/b;ed it with R4 actually because Phoenix/Nimbus are invisible.... I figure it would have a better chance vs R4.
I actually use Nimbus much more than R4........ Also I meant software developer in my post two up.
But let me stop talking about EA in the midst of a 2C thread..

rsp

you know what "they" say about being too thin,

or having too many reverbs.
 
I a/b;ed it with R4 actually because Phoenix/Nimbus are invisible.... I figure it would have a better chance vs R4.
I actually use Nimbus much more than R4........ Also I meant software developer in my post two up.
But let me stop talking about EA in the midst of a 2C thread..

rsp

First, EA and Relab and others make cool stuff. No prob to enjoy their efforts of course. It's great to have so many great options these days!

Second, if you are willing, do you have a favorite preset or few from Phoenix, and do you have some example of the type of material you are working with? I'd be curious to explore the topic to see if I can hear what you are hearing. Just for my personal experience/experimentation.

Also, check the Balance/Mix switch feature in Breeze 2. This is a bit of a semi-hidden easter egg. It can definitely glue the source with the tail more and recess the source back into the sound-field if that is the goal. Sometimes this works wonderfully, and sometimes its actually better to have the dry signal sound quite dry and upfront and center.

The kind of perception you mention can be controlled by the amount of pre-delay and the balance between dry, early energy and late/diffuse energy. For example if you put an instrument and a mic 2 meters apart from each other and put them in the center of the Taj Mahal, the tail may be quite extreme in length by normal standards, but there will be a large pre-delay before the onset of the tail, which will make the result sound still fairly close perceptually. If you want the source to sound farther back there needs to be some kind of energy with little to no delay compared to the dry signal.

In Breeze 1 there is always some degree of natural pre-delay which is dependent on the Size parameter. Physically this means the model is something similar to both the instrument and the virtual mic/listener being close the center of the room. This works great on things such a sample libraries that were already recorded with ERs from the natural room they were recorded in. For very dry libraries some form of early spatialization can be useful. Aether can accomplish this with its ER engine. B2 can accomplish it by using one engine to function as ERs and the other to function as the tail.

In Breeze 2, the Classic Modes still have this degree of natural pre-delay, to maintain preset compatibility with 1.0. But the Hall, Chamber, Hyper-Plater, and Colored modes can have very little to zero pre-delay which will have the effect of moving the sound perceptually back into the space more.

Additionally the "Balance" mode will apply spatialization to the dry signal if desired to further glue it into the space. This works quite magically with the Chamber modes in particular. But again, it should not be an "always on" scenario. Sometimes we would indeed like the sound to be sitting front and center in front of the diffuse tail. It depends on the needs of the mix.

This is an area I have spent more time on in the past few months as well....
 
Last edited:
Hi Andrew.
Mostly I use reverb on 1) Voices 2) Acoustic Guitars/ukele 3) some string and percussion libraries in a supportive role just to help glue the stuff together. 4) Drum Libraries like Superior Drummer 3 etc. 5) Sometimes melodica, but by far I use them mostly on vocals. I tend not to like to hear reverb more feel if it it makes sense. Since Nimbus and R4 are out, I only end up 'using' Phoenix and R2 if they are already on older projects and I need to do a quick edit to send to a client.

Hard to tell you my favourites.... There were as an update from EA that basically killed all the presets you had in your Favourites and I haven't had to rebuild them, but for Nimbus I usually start for Neutral Hall and on R4 Med Hall 2....For voice overs anyway those are the ones I usually start with.
I of course tried to match them in Breeze 2, unsuccessfully....using pre-delays etc.... but in fairness I didnt' touch the Balance/Mix parameters. I always use Reverbs as Sends (old habits die hard :) ).....

thanks for responding.
rsp
 
is it possible to turn off early reflections actually? ive read the manual a bit but couldnt find out how yet sadly ;)

probably i just didnt read enough yet.

one last side note... the "depth" in breeze 2 is fantastic :)
also the cpu usage is worth a comment. you can really use a different reverb (as insert) for every instrument in your template probably.
 
is it possible to turn off early reflections actually? ive read the manual a bit but couldnt find out how yet sadly ;)

probably i just didnt read enough yet.

one last side note... the "depth" in breeze 2 is fantastic :)
also the cpu usage is worth a comment. you can really use a different reverb (as insert) for every instrument in your template probably.


from the manual:

Tip: Since Breeze 2 is ultra efficient on modern CPUs, we actually recommend considering using it as an insert directly on tracks. Breeze 2 also has a special Mix Mode called “Balance” mode, where the “dry signal” will be spatialized as well as the wet signal, and in this mode the numerical mix value will control the balance between this early spatialization and the late energy of the tail. This mode special mode works best when Breeze 2 is used as an insert, and is explained in more detail later in the manua

-

took me some time to get used to using Breeze 2 as an insert rather than a effects send.

bought the expansion packs - yow!


2C Audio B2 or Perfect Storm 3.0
 
from the manual:

Tip: Since Breeze 2 is ultra efficient on modern CPUs, we actually recommend considering using it as an insert directly on tracks. Breeze 2 also has a special Mix Mode called “Balance” mode, where the “dry signal” will be spatialized as well as the wet signal, and in this mode the numerical mix value will control the balance between this early spatialization and the late energy of the tail. This mode special mode works best when Breeze 2 is used as an insert, and is explained in more detail later in the manua

-

took me some time to get used to using Breeze 2 as an insert rather than a effects send.

bought the expansion packs - yow!


2C Audio B2 or Perfect Storm 3.0

yeah i found that one. its simply impressing and i have no clue how they did it, but you can run way more instances than any other reverb plugin i know off.

i just wonder if i also can only add a reverb tail without any ER somehow.
 
yeah i found that one. its simply impressing and i have no clue how they did it, but you can run way more instances than any other reverb plugin i know off.

i just wonder if i also can only add a reverb tail without any ER somehow.

Balance: the special feature setting where a spatialization process is applied to the dry signal in a subtle manner to give it an instantaneous spatial impression and sense of audio source width. The Mix/Balance control then mixes between the instantaneous spatial impression (ambience or AMB for short) and the normal reverb engine (labeled LATE here).
 
Balance: the special feature setting where a spatialization process is applied to the dry signal in a subtle manner to give it an instantaneous spatial impression and sense of audio source width. The Mix/Balance control then mixes between the instantaneous spatial impression (ambience or AMB for short) and the normal reverb engine (labeled LATE here).

hmmm... i guess i quite dont get it. so the balance up to max will just send a tail? i want to disable early reflections completely for some stuff.
 
Top Bottom