Tokyo Dawn Labs release the TDR Special Filters Bundle + TDR Prism a freeware spectrum analyzer
by FabienTDR   21st November 2023

Originally Posted by RJHollins ➡️

The other function, DYN BUMP.
This one has much promise ... but oft times, it just too much jumping around on the Master ... that's another one that would be great to have an 'Amount' control.

Maybe that would be consider for possible added function.

Originally Posted by ampetrosillo ➡️

Basically all analog equipment has some sort of lowpass filter to reject RF interference. In fact I don't see a pressing need for linear phase filters; if you are processing a track through a desk, outboard equipment, etc. you are going to incur heavy phase shift on the treble because basically everything introduces some. If you replicate modern ITB processing workflows (like, playing from tape, two or three compressors in the chain, eq in several points in the chain, colour boxes, etc.) the upper treble will likely be shifted by quite a few full rotations :D (hint: modern workflows aren't so "heavy" because digital processing is cleaner, it's just because it is essentially free). And several experiments have shown that we can't really hear phase shift that much (if at all), especially beyond a few hundred Hz.

Originally Posted by scolfaro ➡️

Definitely agree. Thank you for your thorough response up there, btw

Originally Posted by ozonepaul ➡️

In short:
the difference between
a.) recording to Tape and than transferring it to Digital vs
b.) straight Digital recording
was unquestionably audible - especially as far as transient smoothness/pokiness - but it was never huge. BUT: these differences were more and more audible
- the bigger editing moves I made (eg. more and more drastic EQ-ing, saturation etc.)
- the higher the original sample rate was before down-sampling.

Originally Posted by ampetrosillo ➡️

Basically all analog equipment has some sort of lowpass filter to reject RF interference. In fact I don't see a pressing need for linear phase filters; if you are processing a track through a desk, outboard equipment, etc. you are going to incur heavy phase shift on the treble because basically everything introduces some. If you replicate modern ITB processing workflows (like, playing from tape, two or three compressors in the chain, eq in several points in the chain, colour boxes, etc.) the upper treble will likely be shifted by quite a few full rotations :D (hint: modern workflows aren't so "heavy" because digital processing is cleaner, it's just because it is essentially free). And several experiments have shown that we can't really hear phase shift that much (if at all), especially beyond a few hundred Hz.

Originally Posted by 4fmb ➡️

I'm gonna try and get onna to that Time Machine!

Originally Posted by Rocket88 ➡️

Sounds like the material you're working on probably has a lot more dynamic range and complexity than the bass music I have going. You're right, more control would be better and make it more versitile. It sounds like you really like it in some places but it's just not working well everywhere you would need it to for this project. Did I read that correctly?

Originally Posted by ozonepaul ➡️

When Daft Punk's Random Access Memories came out I just couldn't believe how great it sounded. After it's release, there was an SOS article about the recording process:
https://www.soundonsound.com/people/...ries-daft-punk

In a nutshell, during preps for the record they made loads of different sample rate recording tests, loads of comparisons like
- Recording to ProTools (at different sample rates) vs recording to Tape;
- Recording to ProTools vs recording to Tape at different levels than moving the recorded material from tape to ProTools;
- etc.
At the end they decided to record everything parallel on Tape + 96k ProTools: "All the material was recorded on analogue tape into Pro Tools, which was running at 96k, where it sat side-by-side with the same digitally recorded material."

At the time I decided to experiment with these comparisons. Yes, I could hear slight differences (mainly as far as how the transients were sounding) but nothing super drastic. I felt like I got pretty similar recording results either way (96k digital rec. vs rec. to tape imported to my Reaper). BUT: as soon as I started to make similar edit moves (eg. some more drastic EQ-ing) and than down-sampling the result, bam, the difference became way more obvious.

So I figured, this section was very important in the "Random Access Memories" SOS article:
"We tried to get the sounds we wanted from the start - seconded Guzauski, - mostly by choosing the right microphones and mic positioning and mic pres. I used very little EQ while tracking, and also during mixing. We did not want the album to sound EQ'ed. We wanted it to sound as natural as possible. It's the old way of doing it, using EQ just to touch things up, and not doing anything drastic."
And I think this is the most important element, why "Random Access Memories" sounds the way it does. This is why they could smoothly edit their Tape recorded and ProTools recorded material together and just pick and chose whichever they felt served the given song or part better. By avoiding pretty much all non linear processing induced problems, they pretty much managed to avoid all "the rampant IMD fire". I'm pretty convinced that if the mixing process would have needed more drastic EQ/compressor etc. usage than it would have been a much bigger challenge to hide the differences between their Tape vs ProTools versions.

Anyway, even though probably I disregarded some other variables, still, based on my own comparison tests I found that similar EQ/compression moves on recordings that
a.) I made on Tape and than I imported into Reaper vs
b.) I directly recorded into Reaper
sounded more and more different (especially after down-sampling) the higher the original recording sample rate was.
In short:
the difference between
a.) recording to Tape and than transferring it to Digital vs
b.) straight Digital recording
was unquestionably audible - especially as far as transient smoothness/pokiness - but it was never huge. BUT: these differences were more and more audible
- the bigger editing moves I made (eg. more and more drastic EQ-ing, saturation etc.)
- the higher the original sample rate was before down-sampling.

So after my tests, I always theorized in my head that at high sample rates (eg. at 96k), if my similar edit moves make the audible differences (between the tape recorded vs the straight to digital recorded material) bigger, than there must be some different "junk" somewhere that
a.) I just can't hear,
b.) but with my EQ (or any other non-linear editing) moves somehow I "excite".
Now this "junk" clearly breaks into the audible after down-sampling. The higher the original sample rate was, the bigger the audible difference was after non-linear editing and down sampling.

So reading Ghost's above "I know of some mastering engineers that do not like mastering at 96k" post... the more drastic edits they have to make in a song, the more I understand them

Originally Posted by FabienTDR ➡️

Guys, don't forget to check Elliptical's stereo to mono preservation/compensation features.

Originally Posted by ampetrosillo ➡️

This high opinion of Daft Punk's RAM makes me smile every time, because I think it sounded like utter shite. :D What's this wimpy, silky, smooth polished aura that pervades the album? Is it supposed to be disco (or disco adjacent) or is it supposed to be "easy listening" tripe? Yawn.

Originally Posted by ozonepaul ➡️

...RAM sounded like utter shite...
OK, that's just plain silly.

Originally Posted by David Rick ➡️

If I could make one feature request to @ FabienTDR : Please add an option in Ultrasonic to show the LPF curve reflected from Fs down into the passband. This is a quick way to see how much junk can (potentially) alias into audibility.

Originally Posted by soupiraille ➡️

Ultrasonic is not an anti-aliasing filter, nor a SRC filter. It’s just a LPF

Originally Posted by David Rick ➡️

In the digital domain, any time you create harmonics that reach past the fs/2, they are going to alias. You don't need to be doing SRC for that to happen. It's perfectly reasonable to ask "How large do those harmonics need to be before their LF images become significant relative to the noise floor in the audio band." I find the kind of graph I requested to be helpful in thinking about that.

Originally Posted by David Rick ➡️

It's perfectly reasonable to ask "How large do those harmonics need to be before their LF images become significant relative to the noise floor in the audio band." I find the kind of graph I requested to be helpful in thinking about that.

Originally Posted by jih64 ➡️

Originally Posted by jih64 ➡️

You got a little over 5 hours from time of this post

Originally Posted by jih64 ➡️

If you missed out, never fear . . .






For those interested, it must be a long 5 hours because its still going

Originally Posted by Calagan ➡️

I'm still convinced that Tokyo Dawn labs could offer an awesome EQ/transient plugin in the vein of SplitEq, Spiff or Crave transient EQ.
I was searching for a cheap and good one these last days and there is a place to take.
Given how good and useful are Arbiter and Infrasonic, I'm pretty sure TDR can nail the transient EQ plugin perfectly.

Originally Posted by Ribbonmicguy ➡️

Bought it 5 minutes ago.

Originally Posted by David Rick ➡️

In the digital domain, any time you create harmonics that reach past the fs/2, they are going to alias. You don't need to be doing SRC for that to happen. It's perfectly reasonable to ask "How large do those harmonics need to be before their LF images become significant relative to the noise floor in the audio band." I find the kind of graph I requested to be helpful in thinking about that.

Originally Posted by AudioKemestry ➡️

I think this type of aliasing display that you're requesting for showing harmonics above the 'Nyquist' frequency would be very interesting/useful and potentially novel, as I don't recall seeing this done in any plugins yet. Not sure how you could pull this off though.

Originally Posted by FerrousMeuhler ➡️

Hi Fabien, great set of tools, loving these immensely. I noticed however that the GUI's for all four new plugins are showing up black on my Audiogridder local server (windows, Crossover) on my Catalina 10.15.7 OSX setup. All other recent PC plugins I have installed are fine. Was wondering if anything major has changed in the plugin framework that would cause this, and if there is any workaround. Possibly a windows 7 vs. 10 issue ? Any help would be appreciated. Cheers in advance.