There's always a difference. Even two pieces of the same brand gear can sound different.
IMO this kind of difference is not worth a couple of GB of disk space per plugin and big CPU load (which leads to not being able to put a lot of AA plugins across all channels and mix them simultaneously). Besides most of AA plugins' GUIs are very clumsy and not intuitive.
I've been verbally attacked by avid AA plugins users on another forum for saying something like that

 

I like the idea of this test and appreciate your effort!

Since you mention that you had to level match the processed files, I suspect that the gain reduction between the Algo - Nebula pairs was not really identical?

Nebula plugins are known for unreliable metering (their metering is a raw approximation, not an actual measurement as far as I understand).

If you would find a way to ensure that the GR between the Algo and Nebula is indeed the same, this test would be much stronger.

Not to mention that you had to randomly choose one out of many available algo emulation plugin, hence the results can not be generalized to all algo plugins.

I suggest you rather take one neutral workhose algo compressor which can be set to do all kinds of compression action (think Klanghelm DC8C) as the representative of algo capabilities. Then you pair it with the Nebula ones with closely matched parameter settings (double checked with tools that can measure / visualize compression action, not solely rely on Nebulas displayed parameter values, since they are often off).

Waves is not a good choice, since they are known for bad modelling and lack of sophisticated code that would ensure anti-aliasing and reduced interharmonic distortion. They are not up to the state of the art. IKM mostly the same, except for their Tape, which suffers from another issue: lack of CPU efficiency.

Cheers!

 

Last edited: Jan 19, 2021

DMGAudio TrackComp 2 is also a good allround comp with emulations of most of the famous analog comps!

 

After some extensive "hofa blind test" tests with TimP lbus against a vast amount of algos and my g SSL clone, I have to say nebula is better, at least on buses.
I picked out lbus and the hardware every single time against the algos correct on my 2 and drum bus and preferred them. Hard to describe but the transients are somehow more present and it has more depth. The positioning, the space of an element of a track is different too. It is somehow clearer. Which I prefer.
Night and day? No. Better? Yes.

 

Last edited: Jan 19, 2021

Cool idea. I can do a comparison on a full mix of bass, drums and guitar. Any requests on which plugs to use? I could do a full SLL channel emulation from AlexB vs PA SSL for example.

 

I think you should mix algos for example the satson pre/buss + SSL native channel + dmg buss comp

 

Are you able to use a plugin analyzer like plugin doctor to carefully match gain reduction and attack/release times for compressor? You can skip EQs because there aren't sound quality differences to find there. EQs like Equilibrium are perfectly capable to sound the same as Nebula ones.
If anyone wants more proof of this (I made a comparison on the previous page) you can link me an EQ setting from a Nebula/Acqua library that is available as trial (so I can download it and use it) and provide me a dry loop sample and I will match the AA EQ setting with DMG Equilibrium (for example) till they sound the same (of course I'm not talking about added harmonic distortion). To do a preset for Nebula programs you can load them in a vst container like DDMF Metaplugin and save the preset there (right click inside the window > "save graph")or send a Reaper/Ableton project.

 

Of course EQ's (albeit flat) will not be skipped. These are part of the channel strips that will be compared. And I'll get the comp settings and gain matched as close as possible with PD.

 

After years of use, there are sounds PERFECT in Nebula :
OwnHammer cabs (too bad he stopped ...)
Mechanical reverbs (springs, oil cans, plates ...)
EVERYTHING related to low-mid distortion : transformers sound, tube saturation ... INSIDE other devices too (like neutral EQ or compressor)
Consoles
Mastering EQ
Tapes
...
But heavy COMPRESSORS still s*cks
And heavy distortion too (guitar amps and more)

Nebula can't handle HEAVY signal treatment.
Everything else is almost perfect sounding like the real thing (good or bad)

I now tend to use only reverbs, cabs on tracks.
And everything else on master only. To lower resources.
It is more than enough to experience the magic.

I will still use Waves API comp or Dynamite for heavy drums
And still put a Nebula comp AFTER on bus
Nebula is limited yes, but not that much.

 

In my opinion it would be better to compare each section separately to better understand what is doing what especially when there will be harmonic distortion imparted from the saturation stage mixed with that from compression, otherwise it would be a comparison of channel strips (sampled/modeled from different units and therefore slightly different from the beginning) and not if Nebula is better (technically and sonically)than algos. Same can be said with single processes but at least we can reduce and better understand the variables going on at each stage. Another vote for DMG Trackcomp 2 for compression and one for Softube SSL 4k or UAD SSL 4k mk2 channelstrips.

 

It would have been better to analyze them in Plugin Doctor and share screenshots with us. Single audio file comparison is tricky

 

Nice offer. I had difficulties to reproduce this setting with EQuick. Can you match this sound-wise with Equilibrium?

 

Not better, just an addition because sadly, PD doesn't tell the whole story either.

 

Here screenshots from plugindoctor matching AA Ultramarine 4 EQ with DMG Equilibrium. Didn't care to Frequency/phase match from below 17 hz and phase from above 10k hz because they already sounded the same (AKG K701 and KRK V8 mk2)

 

Ears (with good monitoring) are the ultimate judges. It happens that at least for EQs the sound is not distinguishable. Already done other comparisons years ago prior to the one I posted here and a handful of respected mixing engineers that listened in their studios agreed that they were undistinguishable.

Sure. It would be more accurate if you could share a preset via metaplugin or ableton/reaper project but I can try matching by GUI.

 

Yep.

It's not about +/- 0,1dB. What I couldn't match was the sound. Feel free to not stick exactly to my pic, no prob. Just try if you can match the sound.

 

Ok, downloading right now, acquarius server is slow as f**k but I should complete the task before dinner. Is it ok to you if I reuse the same Linndrum loop that I used previously or do you prefer to send me an audio sample of your choice to process?

 

No need to rush.

Sure, why not?

 

Here we are. Audio samples and Equilibrium preset in the zip folder. A Freq response differences no more than 0.1 dB from 13 to 20khz, negligible phase differences. I could match them pretty much exactly but I didn't bother because there weren't already audible differences.
EDIT: reuploaded the zip folder with Equilibrium preset (I deleted unused bands that were off on the first upload)

 

Last edited: Jan 19, 2021

Yep, looks and sounds convincing.