Airwindows ToTape7: Free Mac/Windows/Linux/Pi CLAP/AU/VST3/VST2/LV2/Rack

[holmer]

Chris, do you have a plugin that emulates that crunchy old school 90s aliasing?

[El°HYM]

https://www.airwindows.com/derez3

[WackyZoundz]

As already pointed out: If you want less aliasing, set your samplerate higher or use a stock/third-party oversampling wrapper, problem solved. There's no need to put oversampling into all the plugins.

[martiu]

[El°HYM]

When using samplerates above 500kHz almost all Aliasing suddenly starts to disappear.

[EfreetiSultan]

You noobs still using 44k in your hosts?? What is this, 1991??
Just use 384khz, and band limit everything with a 12db/o linear phase low pass. Oversampling is the work of the devil.
192khz with a 12db lp filter also works,
Or 96khz with a 24db lp filter, increasing aliasing with each step down.
But bandlimit you must. Oversample you must not!!

Applaud to Chris for understanding this very simple philosophy.

[BackInCheck]

Oversampling is mandatory. It's still very common to work in 44.1k or 48k. Lower CPU tax. To have an oversampling option, that can be turned on when exporting / rendering is pretty standard these days.

[funky lime]

Meh, I can't hear anything over 10Khz anyway. This plugin is lots of fun though, thanks Chris from Airwindows!

[WackyZoundz]

You noobs still using 44k in your hosts?? What is this, 1991??

I use 44.1 kHz as host samplerate for several reasons. First, there is no need to run each plugin at a high samplerate, that's just a waste of CPU and RAM resources which I rather use for more plugins. Second for compatibility reasons - I still use some old plugins which don't support high samplerates (They don't need it anyway). Third, some (Well, actually quite a lot) plugins don't work properly at samplerates higher than 44.1 kHz because they're poorly written, introducing plenty of issues like filter warping. Fourth, I prefer the what-you-hear-is-what-you-get approach. By working with the target samplerate I don't have to deal with the usual surprises caused by faulty conversation and there is no need to convert anything later either which saves time. If a plugin needs a higher samplerate I usually put it in a wrapper.

Just use 384khz, and band limit everything with a 12db/o linear phase low pass.

In this case IIR filters are superior because they're a lot faster (demand less CPU resources) and don't introduce pre-ringing (which also folds back). You also need more than 2nd order filters.

Oversampling is the work of the devil.

What do you think you're doing when you use a high samplerate? You oversample.

[rafa1981]

In this case IIR filters are superior because they're a lot faster (demand less CPU resources) and don't introduce pre-ringing (which also folds back). You also need more than 2nd order filters.

- IIR filters are not always faster. On current CPUs they are actually slower for brickwall filters. It actually may make sense to implement minphase filters as a FIR.
- Pre-ringing is not a problem when the cuttoff is on the 19Khz+ area. It can't be heard. Pre-ringing might be a problem when the cutoff frequency is low enough.

[WackyZoundz]

Which plugin features such filters? So far I've only seen them in standalone applications or hardware units.

[EfreetiSultan]

You noobs still using 44k in your hosts?? What is this, 1991??

I use 44.1 kHz as host samplerate for several reasons. First, there is no need to run each plugin at a high samplerate, that's just a waste of CPU and RAM resources which I rather use for more plugins. Second for compatibility reasons - I still use some old plugins which don't support high samplerates (They don't need it anyway). Third, some (Well, actually quite a lot) plugins don't work properly at samplerates higher than 44.1 kHz because they're poorly written, introducing plenty of issues like filter warping. Fourth, I prefer the what-you-hear-is-what-you-get approach. By working with the target samplerate I don't have to deal with the usual surprises caused by faulty conversation and there is no need to convert anything later either which saves time. If a plugin needs a higher samplerate I usually put it in a wrapper.

Just use 384khz, and band limit everything with a 12db/o linear phase low pass.

In this case IIR filters are superior because they're a lot faster (demand less CPU resources) and don't introduce pre-ringing (which also folds back). You also need more than 2nd order filters.

Oversampling is the work of the devil.

What do you think you're doing when you use a high samplerate? You oversample.

The 384khz bit was satire.. Noone actually uses that unless very specific use. But still accurate..

Why are iir filters superior because they are faster and uses less cpu? In fact that sounds like a quality decrease..

Why would you need steeper than 2pole filters for AA if the samplerate is so high?

No, oversampling is Upsample + Very steep filter + downsample. Absolutely NOT the same as simply working in a higher samplerate. If you use say 3 plugins with oversampling that is 6 times the signal gets resampled and filtered, vs none simply using a high samplerate from the start.

[El°HYM]

Meh, I can't hear anything over 10Khz anyway. This plugin is lots of fun though, thanks Chris from Airwindows!

Well, as long as you can hear 2-4kHz, aliasing can still be your friend.

I love KVR because the people here are nice and full of audio wisdom (actually not his Quote).

Would love to hear some thoughts on this from Fabien, if he wants to chime in? As we all know TDR - Plugins have their very own unique way of dealing with Aliasing and the use of Oversampling, or just Sampling in this case. The use of their Filter - Bundle is also very related to this Topic. viewtopic.php?t=603998

[ccv394]

Always love a new Airwindows plugin and the tape stuff is fun to explore.

[QCGroove]

Aliasing is very good, people should use it way more.

Honestly this. I know there’s more people out there who actually prefer audible aliasing in a mix. I think it’s the special sauce of 97’ to 06’. Music from that era sounds so crisp and punchy. I think it has something to do with aliasing the transients and changing the phase response of the attack.