But we have charts!

Of plugin X vs plugin Y anyway.

Now, if an actual tape deck was included in any of these measurements, I seem to have missed it. What would happen if a plugin measured all wrong with weird anomalies, but the hardcore tape aficionados thought that was the one that sounded the "most like real tape"?

I personally have never been into 'nailing', but for those who are, what value comes from these measurements?

I spent years tracking to 2” 16/ 24 trk.

We did as much as possible to minimise the side effects of bits of rust stick to plastic tape dragged over a magnetic head.

Now, it’s possible to get those few positive side effects of tape with other gear.

Saturation and compression free of time domain.

But the biggest positive you’ll never get.

Tape isn’t digital.

I know that’s an abstract thing - but it’s a fundamental.

Change the recording medium and you need a different approach and you get a very different result.

A few years ago I tracked drum and bass into 2” 16 trk (my favourite tape format) it was a big WOW …. a “tape plugin” doesn’t approach that sound even remotely - it’s so fundamentally different in nature.

Just my 2 cents.

Clearly ymv.

The truth....
Example...there's a tape plugin thread here where some people have questioned some of the sound/behavior of the plugin, only to be met with responses like "obviously you never used tape before" or "that's just how analog is" or "Stevie Nicks hairdresser saw them use the stephens on a lunch break" etc...cool story, bruh, but we're talking about a plugin here lolz
I investigate the claims and try to get reel answers and share them with the inquiring minds...and then get asked to please leave and take my discussion here (by the developer), and then have all my posts deleted and then get banned from thread shortly after.
So...while some may be into mythology and magical thinking, I'm into measurements, because they are reel, and can sometimes give us answers to questions we have.

One of the "complaints" about the p821 was that it didn't sound good when pushed, maybe even digital sounding or whatever...
The reason given by the developer was something like the Stephens was solid state and that clipping sounds like ass... again, that's great and all, but we're talking about a plugin and just taking someone's word that it's an accurate model or whatever.
I looked into it and found that the EQ curve has a huge impact on the sound of the clipping...obviously. And this is not part of the electronic circuit, it's from the tape section. The "thru" portion is flat.
To add to that, I compared the "thru" mode, which is supposedly just the electronics, vs. the tape mode, both clipping, and I found with tape mode there was significantly more inharmonic digital distortion. (esp when dynamics trigger, as seen in a prev post).
So that, combined with the fact this has a dynamic EQ curve which greatly effects the clipping, would lead me to believe that the "bad" clipping sound has nothing to do with Stephens solid state anything...
I casually suggested to listen to just "thru" mode clipping, to see if the clipping itself was in fact the reason, and crickets....

Plots: blue = tape, yellow = thru
Clipping comparison & zoom on harmonics

here is a sample to show the dynamic filters in p821...
i use an amplitude sweep of a sine and you see the different filters kick into action at different thresholds...
again, this isn't necessarily accurate in the time domain but you can clearly see what's going on...



(*the peak filters are not the same exact shape because in the frequency plot they are being triggered by a sine wave in band but in the amplitude sweep they are being triggered by an out of band sine wave, so the gains of the filters are different.)

If you really want to be scientific, the comparison and analysis really ought to include real tape as a control.

Don't get me wrong - I personally don't give a rat's ass about 'authenticity'. I still have my tape decks in working order, and I sometimes still use them to record and play back sound. But I never use them as an "effects processor". But for those who are after authenticity - and those simply seeking scientific knowledge - a plugin-vs-plugin only set of tests is very obviously and painfully missing that essential baseline for comparison. It's taking place in a vacuum.

Is there something specific you wanted info about ? I have the demo and can check a few things in my spare time...

And not directed at you, but general public...is there a link to a formal/technical answer from IK about the "pre-ringing" issue with their tapes ? I google and it takes me here lolz...
I'm a little late to that party, but the curiosity was killing me, so I got the demo and I have some questions...maybe already answered.

*edit: is there a known PDC issue with the ik tapes ?

Just if there's any aliasing happening, tbh. Thanks if you can find anything about that.

the aliasing for the arturia j37 doesn't seem to be an issue until you start to drive/abuse it. i attached a few random screenshots of different tones so you can get an idea. here i show "nominal" levels, full scale, and the two different oversampling options. you can see "render" quality OS is a significant improvement.

1 > two tones, -15 dbfs pk (-18 dbfs rms = 0 VU)
2 > studio vs render OS, 0 dbfs pk
3 > studio vs render OS, -15 dbfs pk, drive = +24 db (*no aliasing under the overlap/1kHz with this tone in render quality pic)

imo the aliasing shouldn't really be an issue if operating at nominal levels, and if using the drive, the "render" OS option seems adequate. i guess you could also just turn the noise up if the aliasing becomes an issue

Thanks, this looks pretty good.

Thanks, this looks pretty good, especially when using Render settings.

...and here are a few arturia j37 frequency responses from a sine sweep input...
green = 0 dbfs pk, red = -15 dbfs pk

(*aligned @ 1kHz, these are not actual output levels)

*edit:

I didn't read through this yet, but the pics are cool...
https://hometheaterhifi.com/blogs/jj...g-and-repairs/

...and might as well have this here too...(maybe someone linked already)
https://endino.com/graphs/index.html

i haven't really seen this too much...
it looks like the noise floor for the ik tape 80 (and others) follows the input level...(see .gif)

...does tape do this ?

i would guess that if you drive it hard, you would get more noise along with everything else...

(amplitude sweep -48 > 0 > -48 dbfs)

What strikes me most is that this plugin is supposed to be subtle, it's even it the opening post: "Engineers loved it for its ability to enhance without altering".
But it's not subtle at all, the frequency response seems to be all over the place, especially in the low end.

Since you did a lot of tests and measurements, have you found settings for a rather flat frequency response?

Problem is that mr. HighKoalaTea making almost useless "measurements", and i've told him before, that one cannot measure freq response of a highly nonlinear processor this way and get meaningful results... There are special tools made for this. But, i guess, any guy with an oscilloscope is declared making a "scientific measurements" nowadays.

It is measurably the most colored "tape" plugin I've ever tested...I don't even consider it tape, it's a dynamic EQ with a tape skin. Tape sort of acts like a dynamic EQ, but nothing like p821. Honestly I thought it was an April fools joke.
The shelves/bias can take the smiley curve down, but I don't think it can ever really get flat. I didn't spend too much time with those features, so I don't have any figures to share. The best bet (imo) is to keep peak level <= -20 dbfs +/- if you want to avoid the exaggerated EQ curve.

Please stop...I know what I'm doing. What's your problem ?

Then again, simply repeating over and over that he's doing it wrong without posting your own measurements done with the "correct" method doesn't prove anything either.

I don't know how to read most of these graphs, but I know from testing the plugin myself that it has a huge impact on the frequency response.

Ok thanks. I wondered if I should test it again now that you can disable the widening. But fighting the eq moves this plugin does out of the box seems tedious and you basically have to commit to use this as your first insert before doing anything else.

look at the screenshot here...this is real music, an orchestral piece, good dynamic range. i gained it up until it hit the zero on the p821 meter and it was bouncing in the red pretty good.
i lined the in/out spectrums up at around 400 Hz because that's where it seemed to do the least EQing...but regardless how or where you line them up, the frequency response is drastically different than what went in.
Span set to "max" (peak hold) setting with 1/12 oct smoothing, but you get the idea...

red = input
green = output

...that pretty much coincides with the frequency response (456/30). boost @ 85, 800, 2.5, 9.5 Hz (+/-)

*I don't see (or hear) what's "wrong" with the frequency responses

It's good you know what you're doing, but the clearer you can get your points across in order for others to understand your message will be received. You'd also empower others to make their own experiments and tryouts when they better understand what you're doing, and not just you. You know.

well...p821 doesn't work in plugin doctor for me, or maybe it does but i just hate that app so much i don't care to try again, so i can't give you a hammerstein plot (i honestly never bother with them anyways, bunch a squiggly lines), but i (tried to) made a video of a sine sweep viewed on Span and you can see the harmonics & dynamic filters etc. the first 10 Hz are super boring and low res, but i promise it gets better. it's my first davinci vid edit so bear with me (PC noob, ex iMovie pro).

When I was messing with P821 I came up with this preset. It's the flattest I could get it, and you have to keep the black vu meter around -18. You can push it a little to -12, but more than that starts to change the sound too much for my taste.

I havent messed with Arturia's plugin much, but want to give it a spin. Pretty recently Eric Valentine tested it against a handful of others, and said it was the most "Makes things sound musical, like a real tape machine" plugin he's heard. I'm paraphrasing, but he really liked it.