Blind Test: Digital - Tape - Tape sim plug-ins

[akisD28]

Hi all,

For my final project for my MA in Audio Technology I recorded a couple of tracks simultaneously to ProTools (HD, 96K/24-bit) and an Otari MTR-90 (GP9, 30IPS). The main purpose was to see if it's beneficial to use a tape machine for recording compared to a digital system equipped with tape simulation plug-ins.

I have uploaded 3 versions of the same mix of Hole's Celebrity Skin. The procedure was the following:

1) I mixed the track using the straight-to-digital files, having total freedom as to what processing I could use, including tape simulation plug-ins as I would them in my standard mixing practice (that is not trying to simulate the particular tape recordings - see below).

2) I replaced the audio files with those recorded to tape and exported a mixdown. Any tape simulation plug-ins where of course removed (see below).

3) I processed the digitally recorded files individually with various tape sim plugs in order to approximate the sound of the tape machine as closely as possible and exported another mixdown. Each file was processed with 10-12 different tape sims (VST/RTAS/TDM) and the best version (best approximation of the given tape machine's sound) of each file was chosen for the mixdown. Of course, any other tape sims used during mixing were removed as well.

A few notes:

- During my original mix, I only used 1 tape sim on the drum bus and I used it differently and way more subtly than I had to when I tried to approximate the sound of the tape machine (where I actually processed each track separately). The difference it made was extremely subtle, but I thought I should better remove it when I used the tape/emulated files.

- The acoustic guitar in the bridge section was recorded to tape in all versions.

- The average RMS levels of the audio files between the 3 versions were matched with a precision of 0.01dB.

- The mix was done in the box in Nuendo 3.2.1 @ 96/24. For further details, please ask.

- The order of the files (A, B, C) is random.

Please reply to the following questions:

1) Which version do you prefer overall?

2) Which one do you think is done with the digital files, which one with the tape files and which one was processed with the plug-ins?

3) What do you like/dislike in each version? i.e. Do you prefer the vocals in one and the bass in another version? etc.

As a thank you, I can send you some impulse responses of the MTR-90 if you're interested - just reply to this topic and include your e-mail address and I'll send them to you. 8)

Thanks in advance,

Akis

PS: If you prefer .wav versions @ 44.1KHz/16-bit (all together as a .rar file):

http://akis.files-upload.com/9314/4416.rar.html

No need to register to download!

EDIT: If you want to hear newer versions, scroll down to post #21./EDIT

DISCLAIMER: I'm nowhere near nor I claim to be nor I expected to become an expert on anything before/while/after doing this. It was just an experiment that some of you may find useful. I certainly did. It doesn't prove anything.

[beatbed]

I enjoy A the most, it feels as if C has better bass/low-mid. Over all the panned guitars are to loud for me (my taste that is) so I didn't put the dial on 11 and rocked out.

All sound much the same, no gigantic differences in my headphones (Koss portapro).

...and that's my two cents!

[akisD28]

Thanks for replying!

[smokapot]

"A" Sounds better to me.

"B" and "C" sound almost the same as each other, and thats weaker than "A"

Nice Job

and thanks for posting them.

Now which one's the tape?

[akisD28]

Thanks for replying!

Send me your e-mail address; I'll send you the impulses and the results.

[wwittman]

but wait... you're not mixing from tape, right?

you're CONVERTING the analogue to digital before you mix?

so whatever the digital loses, is STILL being lost in all versions?

[akisD28]

Yes, I had to convert to digital before mixing. The difference is the analogue tape losses/distortion/non-linearities etc. that are only on the files recorded to tape first.

My aim was not to see if there are any benefits from using tape entirely as the recording/playback/storage medium, but rather if there is any benefit from recording to tape before dumping to digital. Sorry if I wasn't clear.

[akisD28]

links to.wav versions added.

[theblue1]

Originally Posted by akisD28

Hi all,

For my final project for my MA in Audio Technology I recorded a couple of tracks simultaneously to ProTools (HD, 96K/24-bit) and an Otari MTR-90 (GP9, 30IPS). The main purpose was to see if it's beneficial to use a tape machine for recording compared to a digital system equipped with tape simulation plug-ins.

I have uploaded 3 versions of the same mix of Hole's Celebrity Skin. ...[/size]

Um... I'm hoping you won't take this too personally but it always amazes me when someone is trying to set up a blindfold listening test to evaluate the relative fidelity/quality of some device or software and they use object material that makes it difficult to evaluate subtle differences in said fidelity/quality. (I make the distinction here, since I presume we are talking in this case not about fidelity but about the unique sonic qualities imparted by typical tape distortion.)

Anyhow... a range of test materials that would both challenge the processes/devices being evaluated AND provide material where differences in fidelity/quality are easily discerned by the subjects MIGHT be more useful, at least to guys like me. (Now, don't get me wrong. I WAS a punk rocker. But it was not exactly concerns about fidelity and sonic accuracy that drew me to punk in those halcyon days of the mid-late 70s.)


[I'm particualrly reminded of a "blind listening test" I've seen here and elsewhere which was elaborately done but which used as its test bed material a grungey, distorted emo mix in order to evaluate presumably subtle differences in hardware/softare summing...]

[akisD28]

The reason I chose this material is because it's exactly the type of material where I wanted to see if tape makes a difference. That's the type of music I make (well, sort of...) so it's not really in my interest how tape affects, for example, classical recordings. I should have probably clarified this in my first post, too.

[theblue1]

Well, I was kinda thinkin' that... Them classical guys are usually not going for the heavy tape satch...

cheers

[akisD28]

Originally Posted by theblue1

Well, I was kinda thinkin' that... Them classical guys are usually not going for the heavy tape satch...

cheers

Any comments on the tracks?

[bcgood]

I honestly don't like any of the recordings. They sound really harsh to me and sonically are very displeasing to listen to. I'm not trying to be a dick but all three don't sound as good as many digital or analouge recordings that I like so I'm afraid saying one is better than another only prooves which one is the best of your three specific examples and you wouldn't be able to extend that to other recordings. Sorry to be so negative...

bcgood

[akisD28]

I have absolutely no problem if you don't like my mix. It is my opinion that the differences between the 3 versions are still obvious, though, and...

Originally Posted by bcgood

I'm afraid saying one is better than another only prooves which one is the best of your three specific examples

...that is the point. ;o)

[tuRnitUpsuM]

I'll go on a limb for ya... i'm not scared... if im wrong... i'm wrong... but i still make a mean pasta.


1) 1073
2) 1176
3) distressor stuck on nuke with brit mode engaged... and freakin laser beams.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
... i'm lightening up the mood... ear training if u will... smiles make ears work better... is fact.

k gettin down to biznas


1) enjoyed A the most
2) A = tape mix , B = digital files, C = digital with tape sim processing?
3) instead of mix by mix... i'll state why i liked A the most...and why i think its the tape file mix... A is more forgiving... i can hear digital over accuracy in B as well as C but not as much as B. By this i mean... digital to me sounds un-organic... pronunciation is overly clear...to the point of being sterile... and even if touched up is still eminate. Vocals can be a tell tale sign ...not always... as nothing is ever written in stone musically speaking of course.


cheers

i actually hope im wrong....i learn more that way.

[akisD28]

Cheers, mate!

Do you want the impulses & the answers? If yes, gimme your email addy.

Originally Posted by tuRnitUpsuM

but i still make a mean pasta.

Not better than mine!!!

[akisD28]

Cheers for replying.

As always, if you want impulses+answers, send your addy.

[wwittman]

with all due respect, it just all sounds very "digital" to me.

for whatever reason, I might think the cymbal trail offs are a TINY bit smoother on A, but on the whole I don't find any of it sounds terribly "analogue".

maybe it's what you're doing to it AFTER the transfer, or it's the basic recording.

Not meaning to be critical.
Only to say that when I compare analogue to digital the differecnes are more striking.

If I weren't a trusting person, I'd almost think NONE of those were analogue.

[akisD28]

Thanks for your comments, William. :o)

One of the recordings is analogue, as it was made clear in the first post. However, all my mixing/summing was done in the box, as was also said in my first post, so perhaps that's what you hear?

PS: Just to clarify a bit more: as said in the original post, the mix was done using the recorded-straight-to-digital files and then they were replaced by the 2 other varieties for the other mixdowns to be exported. I wouldn't expect my mix to sound analogue just because the tracks were recorded to tape first; it's a digital mix. I just wanted to see what difference recording to tape first would make to the overall sound.

[akisD28]

Just to let you fellow slutz know that thanks to feedback received, I have improved my mix a bit and you can find all versions here:

http://akis.files-upload.com/10151/4416-2.rar.html

I've left the original links for the test intact in the first post, though, for consistency.

Thank you all for your feedback.

[s34nsm411]

could i get that impulse? my email is s34nsm411@yahoo.com

thanks

[akisD28]

Sure, I can send them to you, but can I please ask you to take the test, too?

[christowaffle]

1) i'd have to say that all around A was my favorite. fuller w/o being too crispy/bright and punchy w/o being spastic. and is it just me or is the image in B and C unnaturally wide while A is wide w/o sounding weird?

2) i guess A as tape. B digital. C digital w/ tape emu.

3) drums in A are nicest. fatter, etc. same w/ bass. also in A vox have nicer mids.

interested in hearing what's what! (and that impulse sounds good too!)

milo.open@gmail.com

[akisD28]

A big thanks to all who participated in this test.

The answers are:

A - digital

B - tape

C - tape simulation

Cheers,

Akis

[tuRnitUpsuM]

akisD28


wanted to ty again for the thread/test.....


I've been humbled ever since you emailed me the results.... and have actually considered many things i might not have since then....cheers


As well as really placed a hold on my decisions ive locked away about tape simulations..... fair to say ive had reservations about it.... and now have really been keeping my eyes open for 2-track open reel decks.... mind u i have not heard all of them...(tape sims).... but was alittle too easy to spot for my liking.

cheers again....

[Gunnar]

Hi I did'nt think the difference was that big. But I found Sample A to sound clearer and better.

I have heard many saying taperecording sounds better. But if it sounds like this I'm not that impressed. Maybee the taperecorder needs to be older?

[tuRnitUpsuM]

Hi Gunnar


Was kinda implying the let down...ive experienced and from this thread (supporting to an extent)... that TAPE_SIMULATION is very noticeable at being just that..... a simulation. Just doesnt have the same vibe as actually using tape and record-heads, playback-heads.....etc.

again...i duely note i haven't tried em all.

nothing to do with NEW vs OLD.

Better is subjective...and discussed ad nauseam.


cheers