[Csnd] Distortion / Compression Modeling for Tape Delay
Date | 2017-10-24 03:37 |
From | Emmett Palaima |
Subject | [Csnd] Distortion / Compression Modeling for Tape Delay |
Hi, I was wondering if anyone could recommend some good resources on modeling the distortion and compression characteristics of magnetic tape. I am currently working on a digital emulation of the Echoplex tape delay, and have been using some samples I took of one of the original EP3 models. I have succeeded in replicating the frequency response of the tape using biquad filters.
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here
I am now moving on to modeling of the distortion characteristics. So far I've been experimenting with polynomial and tanh distortion, as well as a fairly basic compression algorithm. Does anyone have any addition methods / resources they could recommend? Thanks! |
Date | 2017-10-26 02:33 |
From | Steven Yi |
Subject | Re: [Csnd] Distortion / Compression Modeling for Tape Delay |
Hi Emmett, I don't have much experience with distortion algorithms, but I did use these two tanh based distortion: https://github.com/csound/csound/blob/develop/Opcodes/wpfilters.c#L701-L706 (discussed in Pirkle's App Note for the diode filter: http://www.willpirkle.com/Downloads/AN-6DiodeLadderFilter.pdf) Not sure if that's of any use. I'd love to hear how your project turns out! steven On Mon, Oct 23, 2017 at 10:37 PM, Emmett Palaima |
Date | 2017-10-26 12:35 |
From | Steven Yi |
Subject | Re: [Csnd] Distortion / Compression Modeling for Tape Delay |
I was also reminded off-list by Kevin that there was a UDO I did a while back of Tom Szilagyi's TAP Tubewarmth plugin: http://www.csounds.com/udo/displayOpcode.php?opcode_id=80 The description from Szilagyi's website (http://tap-plugins.sourceforge.net/ladspa/tubewarmth.html): "Tube preamps and reel-to-reel tape recorders have been around for a long time. Until the end of the seventies, the sound of almost every great recording bears the watermarks of these intrinsically nonlinear devices. In the eighties everything went digital, and the "harmful", "non-perfect" nonlinearities were greatly eliminated from the recording and mix-down process. However, it became clearly obvious shortly afterwards, that without these nonlinear artifacts recordings often sound dull, boring, colourless and thin, just like a piece of plastic. As a result, efforts have been made to bring the sound of analog equipment back to the studio -- with the power of DSP. This plugin offers the sound of these analog devices, but without the high-frequency roll-off of tube amps and the hiss of reel-to-reel tape machines. With this plugin applied and properly adjusted, many types of source materials will sound subtly richer, warmer, fatter, and closer." On Wed, Oct 25, 2017 at 9:33 PM, Steven Yi |
Date | 2017-10-26 22:33 |
From | Emmett Palaima |
Subject | Re: [Csnd] Distortion / Compression Modeling for Tape Delay |
Thanks so much for the advice. Tanh is something I was already making use of, that's kind of my go-to for basic soft clipping. UDO looks very intriguing, will definitely have to check that out in-depth. Another source I have found is the CCRMA article on Cubic Non-Linear Distortion using polynomials, which I like since there is more possibility for tuning than tanh: https://ccrma.stanford.edu/realsimple/faust_strings/Cubic_Nonlinear_Distortion.html Here it mentions using oversampling and a half-band lowpass filter to eliminate aliasing. I am wondering what that looks like in implementation. I've tried doing some research on half-band filters, but I couldn't find anything about how to set one up. What does a standard equation for a half-band lowpass look like? On Thu, Oct 26, 2017 at 7:35 AM, Steven Yi <stevenyi@gmail.com> wrote: I was also reminded off-list by Kevin that there was a UDO I did a |