Hardware vs. Unison Blind Test

UniversalAudio · Jun 15, 2017

Hi all,

The Jacquire King/Jamie Lidell hardware vs. Unison shootout has been going really well and has generated a lot of great discussion so thanks for your participation!

Along the way, I've been asked by several people to post some unlabeled files for blind testing, so here you go!

There are 5 A/B pairs to check out. Either one can be hardware or Unison/UAD.

Your job (should you accept it) is to answer the question - "Can you pick out the hardware?" and select A or B.

Hopefully, I set it up correctly - thanks to Matt for all the help!

Enjoy and I'll reveal the sources and results when it looks like it's slowed down.

-GK
Gannon Kashiwa

PLEASE NOTE: you must use the desktop/web version of the forum in order to cast your vote!

Drums:
View attachment Drums A.wav
View attachment Drums B.wav

Full Mix:
View attachment Full Mix A.wav
View attachment Full MIx B.wav

Bass:
View attachment Bass A.wav
View attachment Bass B.wav

Lead Vox:
View attachment Lead Vox A.wav
View attachment Lead Vox B.wav

Guitar OD:
View attachment Guitar OD A.wav
View attachment Guitar OD B.wav

Oli P · Jun 15, 2017

You can only vote one thing in this poll...I suppose you should be able to choose for each example?

UniversalAudio · Jun 15, 2017

Oli P said:
You can only vote one thing in this poll...I suppose you should be able to choose for each example?

Hi Oli,

Yes, that's the idea. I'm not sure how to get around it...

Please stand by while we figure it out...

-GK
Gannon Kashiwa

simonjewel · Jun 15, 2017

UniversalAudio said:
Hi all,

The Jacquire King/Jamie Lidell hardware vs. Unison shootout has been going really well and has generated a lot of great discussion so thanks for your participation!

Along the way, I've been asked by several people to post some unlabeled files for blind testing, so here you go!

There are 5 A/B pairs to check out. Either one can be hardware or Unison/UAD.

Your job (should you accept it) is to answer the question - "Can you pick out the hardware?" and select A or B.

Hopefully, I set it up correctly - thanks to Matt for all the help!

Enjoy and I'll reveal the sources and results when it looks like it's slowed down.

-GK
Gannon Kashiwa

Question is Gannon will they self destruct in 35 seconds?

UniversalAudio · Jun 15, 2017

simonjewel said:
Question is Gannon will they self destruct in 35 seconds?

Yes - the tape can only be played back once!

BTW, Matt is fixing the poll so you can vote properly.

Cheers!

-GK
Gannon Kashiwa

Matt Hepworth · Jun 15, 2017

Hang loose, guys. I gave Gannon bad info. Give me just a couple minutes to fix it.

Matt Hepworth · Jun 15, 2017

Fixed! Sorry!

UniversalAudio · Jun 15, 2017

Matt Hepworth said:
Fixed! Sorry!

Thanks Matt!

-GK

Eric Dahlberg · Jun 15, 2017

Files aren't level matched, quite far off. Bass A is louder than Bass B. Drums B is louder than Drums A. I didn't check the remaining files.

Pro · Jun 16, 2017

Software has really come a long way. A couple of these like the lead vox I prefer the version that I think is actually the software. There are small level differences but it's partially because of timbre more than mismatch.

UniversalAudio · Jun 16, 2017

Eric Dahlberg said:
Files aren't level matched, quite far off. Bass A is louder than Bass B. Drums B is louder than Drums A. I didn't check the remaining files.

Hi Eric,

Thanks for checking the files! I used NuGen LMCorrect2 and VisLM2 to analyze them and they all came out within a couple of tenths of a dB in Peak, Leq(A) and LUFS.

I did notice that the peaks on the bass examples were different in Pro Tools, but it seemed like the overall level was matched pretty perfectly to my ears and the overall loudness metering.

What tools did you use to measure?

-GK
Gannon Kashiwa

Screen Shot 2017-06-16 at 9.31.28 AM.png

Eric Dahlberg · Jun 16, 2017

I added all the files to a Studio One Mastering Project and used the analyzer there. The peaks are similar but the average RMS's are as much as 1dB apart (there's a stray peak at the end of Bass A but I'm not counting that). I could tell just from looking at the audio waveforms, that's what prompted me to analyze them.

Oli P · Jun 16, 2017

There is definitely something funky with how this poll calculates the percentages

Beechwood · Jun 16, 2017

Very much looking forward to calimike getting all these right!

Personally I don't care about the differences. Seems like the main sonic benefit of the hardware is the variation between units. Not the only benefit perhaps, but I'm certainly not confident I've got these answers right.

Unit to unit variation pretty much explains the differences I hear. I actually do kind of wish there was a way to choose between a couple of different units with the plugin versions.

kedavis · Jun 16, 2017

Regardless of which is which, both sound good and can make a great sounding record.

BlueFang · Jun 17, 2017

The biggest difference I hear is in the Full Mix test. Mix A sounds more organic. Mix B sounds tighter - more precise. The A/B test is fun, and I am sure it is something your engineers do all the time. However, I am not interested in figuring out which one is which because the HW/SW option is not a choice I feel I will ever be faced with.

You are preaching to the choir with your plugins. We all love your plugins. We all spend thousands on plugins. But sometimes, it is the simple things we want. Like midi / OSC / HUI control of the Console app. I still don't understand how you guys are smart enough to reverse engineer some of the most complicated non-linear circuits in the world, and yet, can't create a simple assignable midi application.

Sorry to derail this thread. But as someone who has invested heavily in the UA platform, I am tired of using my mouse to adjust knobs.

Ashford · Jun 17, 2017

The previous example gave the listener the opportunity to judge how close the software emulation came to matching the real hardware. The real hardware was the reference and the only question was how close did Apollo sound to the reference. For me there was no aspect of whether the sound was "better" or "worse".

In many cases the Apollo may handle aspects of the signal chain more elegantly than the hardware and as result may be perceived to sound "better". I do not consider a blind test such this really meaningful as there is no absolute reference and the results will be biased by personal preference. It is however fun!

RalphK · Jun 17, 2017

i could clearly hear the difference between the two and yes i perceived a level mismatch... but didnt give it much importance, the difference between the two is clarity and warmth... i beleive i'd pick either based on the sound i am looking for... Daft Punk already did that

SuperUAFan · Jun 19, 2017

Cool test! There are differences, but I couldn't really say which is the hardware. Maybe the warmer one, but sometimes hardware can be brighter and have more transient detail than we think. Results pretty much say that so far most people are like me.

Cass Anawaty · Jun 19, 2017

Level matching has to be done by ear.

Hardware vs. Unison Blind Test

Can you pick which one is hardware?

Drums A

Drums B

Full Mix A

Full Mix B

Bass A

Bass B

Lead Vox A

Lead Vox B

Guitar OD A

Guitar OD B

Official UA Representative

Active Member

Official UA Representative

Established Member

Official UA Representative

Master of the UADiverse

Master of the UADiverse

Official UA Representative

Purveyor of musical dreams fullfilled.

Active Member

Official UA Representative

Purveyor of musical dreams fullfilled.

Active Member

Venerated Member

Member

Member

Venerated Member

Active Member

Member

Shareholder