Yes maybe sawtooth like is a better word

To create a true square wave you would need infinite bandwidth. That is of course impossible. Also most compressors don't allow the ridiculous settings that would be needed to create a square wave.

That said, TrackComp allows some pretty ridiculous settings that do a fair job of squaring the wave:



Or if that still sounds too musical, you can mangle even further with DMG Compassion:



I think I'm going to put that on some supersaw synth to create my next EDM mega hit.

Alistair

Yes I mentioned DMG Audio. I'm a big fan. I like all these independent developers that go the extra mile!

And yes, of course you can link any of these graphs or posts. But do keep in mind that all these tests are theoretical and incomplete. Some of these things really do matter sonically. Some of it is just extreme settings that no one would use in practise. And sometimes there are good reasons for the behaviour but might not be obvious at first glance (like the RComp soft-knee). That is why I specifically wrote "For entertainment purposes" in my other post. I meant that.

Alistair

I also want to say something on the positive side for Logic.

The reverbs are golden. Both the newer Chromaverb and the updated Space Designer have quality options which stomp aliasing, and the other verbs pass without issues.

It’s survivable terrain if we know where the landmines are, but just keep in mind they are pretty much everywhere.

Also, different quality settings are built into the included patches, so it isn’t a set it and forget it situation.

I’m going to deep dive on the instruments today and see where they fall. Most of them feel ok to me at 96k, but third party options are noticeably better.

To my ears the compression action is what i like about it. The tone is ok. There are a lot of plugins without aliasing that sounds bad also. I mean, its there i bought, i liked it. Then use it make music. We can email the dev about this too!

Based on how I understand compressors I think you can't rectify the curvature of a sine wave and make it a straight vertical line, you only can compress more and more till you reach a flat, horizontal, line. Compress it to death, in other words. No signal, no wave at all.
Yet, on the other hand, given the threshold range needed, and precise in a way that one can target individual code values, you simply punch in the first code value that isn't zero, attack and release to zero, ratio to infinite. Then you gain it up and - voila! I also could be wrong, of course.

No, you are right.

I didn't take the term "square wave" to mean a wave with an instant rise and instant fall. (Which indeed would be needed to create a true square wave). In this context I see it as meaning a flat horizontal top with infinitely sharp edges. Or at least as close to that as we can get given the limitations.

Alistair

I purposefully use the dynamics section in an old digital mixer on the regular for some square wave clipping. Those 0 attack 0 release setting can be fun.

Must be that damn "language barrier" thing again then...

Hornet tape has a little fold back aliasing going on?

Stop massively clipping.

Hornet Tape doesn't clip. It does fully alias though...

Alistair

Look, set to -1db which is a quite normal peak to have. So, not clipping at all. Look at that fold back aliasing!!

It's not this individual plugin per se, but the combination of all sounds and aliasing plugins together. It all adds up.

Colored EQ
Pre-amp
Distortion/saturation
Channel compression
Drum bus compression
Tape plugin or glue compression
Soft clipping
Hard limiting

All this together can be a big fat aliasing fest.

Not that bad actually. Cannot see if this has fold back aliasing. I think this is IMD, which can be filtered out with TDR USF.

Also, not that bad for a tape plugin that includes saturation and compression.
Turned off the wow/flutter, noise and asperity.

Reelbus introduced a HQ mode, which cleans up a little.
"New high quality (HQ) mode using extreme oversampling"

All Toneboosters v4 series looked good to me, using the HQ mode when necessary.

The easiest way to tell if it is aliasing artefacts or not is to hover your mouse over the frequency slider in Plugin Doctor and scroll your scroll wheel. Then look at how the frequencies react. If they start moving to the left while you increase the frequency, that is aliasing.

Alistair

OH yes, there is no oversampling inside Tape

look at the aliasing level, it's below -70dBs, so it's usually mostly covered by the hiss generated by the plugin, then again if you turn off hiss it's another story but isn't hiss a part of the tape recording experience??

Saverio

Also not in Hornet Angle

Can't this be easily fixed by building in a steep 20k LP filter?
Or 90k when working at 96k?

Starts aliasing when using the Harmonics slider or increasing the Input drive.

And yet Serban Ghenea and John Hayes use it on global hit records

Pff...Amateurs

Some limiters do a better job than others: Digital Limiter Quality Test

In German, we have a saying "eat **** - billions of flies can't be wrong"

This is actually a nice little EQ/Compressor combo: Analog Obsession STEQ

EQ and Comp turned off generates odd harmonics.
EQ and Comp turned on does not alias or create IMD.

It is not normal to hit any tape machine with -1dbfs signal. That’s often between +17 to +23 dbu.

A plugin is never a real tape machine.

And if it were modeled exactly after a real tape machine, it should certainly not create fold back aliasing. That's just something that doesn't exist in the analog domain.

You’re not even measuring the foldback, you’re showing the imd from overloading the plugins. you just overloaded the waves Kramer tape (1960s tube ampex) instead of gainstaging it so that peaks of the tracks hit under 0 vu like the manual tells you to. Yes there is foldback but you’ll never hear it when it’s well under the noise and distortion. It’s a cool plugin but of limited use to modern productions when compared to U-he Satin, UAD Ampex, and Fuse TCS-68.

*********** had decent measurements but again I don’t know how carefully he gainstaged it and he didn’t turn the noise on: https://***********/waves-kramer-tape/

First off, this was my conclusion regarding Kramer Tape:
"Not that bad actually. Cannot see if this has fold back aliasing. I think this is IMD, which can be filtered out with TDR USF."

Second, the test that i've done was based on minimum input level in the kramer tape. The VU meter is at -1, so not in the red zone. How is that overloading?