Hi everyone,

i only recently started using PD (MaxMSP user before) and rebuilt a vintage Vestax DDG-1 first for a start. Works nicely, but my MaxMSP version of it using their tapin~/tapout~ combination seems to have significant advantages over delwrite~/delread~/vd~in PD. The MSP version

1. behaves like a tape on increasing or decreasing delay time, in other words it really pitches the signal, like the original outboard delay is changing the sample rate then.
2. doesn`t produce clicks upon changing the delay time.

So question;: Has anyone attempted to build a feedbacking delay with tapread4~, susloop~ instead here, which theoretically could rather sound like a reclocked delay chip, i.e. tape speed change?

Any hints are very appreciated, thanx much in advance.

Have you tried feeding a [vd~] object with a [line~] with a ramp of e.g. 50 milliseconds. for instance, because then the delayline also starts behaving like a tape and you will get no more clicks. Something like this (i guess):

[delayLength\
|
[pack f 50]
|
[line~]
|
[vd~ myDelayLine]

Thank you, Domien yes i did experiment with line settings.

i am working with Feedbacks above 95% to basically create steady basstones from percussion on adc~, playable via midi keyboard. signals are getting limited all over the place to prevent internal clipping.
However, especially when i jump from say 20ms feedback delays to 1000 ms there is lots of clicks suddenly.
The DDG-1 is trickier. I has a dial to inc/dec the sample rate from 0.5-2 times the base rate, which obviously is responsible for the tape delay pitch effect.
For this dial it supplies 5 different delay ranges, 2, 8, 32, 128 and 512 ms, and it has a hold button, which shuts input off and reads the buffer only.
The total buffer has 512 ms always, but can be played back at variable sample rate between 256 and 1024 ms.
If i switch to a shorter delay range, say 32 and turn off HOLD, then only those 32 ms are being overwritten, the rest stays in the buffer until a longer option is selected.

I want to emulate this behaviour as well as the variable sample rate of the delay buffer.

The HOLD option is easy: Switch off delay input and turn feedback to 100 but i am afraid the rest of the buffer issues won`t work with delwrite~/vb~-

Any recommendations which way to go?

I'm not really sure how to solve the DDG-1 questions. Maybe you should work with 2 delay lines~ and program some logic to jump - in case of a hold - to an auxiliary delay which serves as a stutter buffer. About the resampling maybe you could have a look at the [block~] object, which allows to up- and down sample blocks (I couldn't tell you if this is the same as resampling).

But the fact that vd~ clicks during a jump from 20ms to 1000ms is strange. Maybe you should post your patch so we can all have a look?

Using vd~ is analogous to moving the playback head of a tape delay, or something. Pitch change in this case is proportional to the rate of change of the delay read. It's not the same as changing the speed of the tape or sample rate of the delay. For that you'd need to vary the speed of both the read and the write. I haven't tried it, but the new ipoke~ port might help...

I find this thread helpful: http://puredata.hurleur.com/sujet-6887-trying-tape-delay.

The tape delay patch on the bottom of the thread is a start for what i was meaning, using tapread4~.

I will further evaluate this. Thank you much for all your suggestions. As a newbie to PD i don´t quite understand yet, how to install other objects, than what is already contained in PD-extended (am on a Mac 10.7 with PD-0.42.5-extended) - so i´ll evaluate ipoke~ as soon as i understood this and found it, too