Somebody wrote:

...

The way I remember it, anything to do with tables or other allocated memory can break with -nogui. [tabwrite~], [tabread~], [delwrite~], [delread~], [vd~] etc etc....

I'm very curious about this issue and quite surprised, since I have been using -nogui for a long time, with quite a lot of tables and [tabread/write~]s, and also some [delread/write~]s (no vd~ though), and I never experienced any such issue...

May any of the following things be the reason why I don't experience this issue?

1. I work under Windows
2. The patch running in -nogui has almost no gui objects (maybe a couple

of numberboxes and toggles or so) and no gop-enabled subpatch or abstraction at all. 3) _Every_ single abstraction that contains any dsp objects (or some "ascendant" of it), always contains a [switch~], which is loadbanged to off, and later explicitly turned on when needed. 4) I also turn dsp off and on with [; pd dsp 1/0( messages. I think dsp is always turned off and on at least once before "showtime". 5) All is controlled via [netreceive] from another patch, running on another instance of Pd, which has gui and doesn't process any sound (it is a "gui-engine" architecture)

I guess (3) is the best candidate...?

None of the mentioned things was done with the aim of avoiding this issue since I didn't know anything about it...