Spin Semiconductor

Hello,

when the fv1 starts up and a long reverb is selected the internal mute is not long enough to suppress the complete reverb tail. So the units start up with a noise on the headphone output. This can be easily tested with the dev board but some FV1 samples are much noisier than others. Our dev-board is of the quiet ones.

The plan is to skip output for several seconds.
due to the lack of integer arithmetics it seems one has to:
- setup one (e.g. reg1) register with 1.0
- setup another register (e.g. reg 2) with 1/96000 for 2 seconds mute.
-skip these steps with skp run ...
-rdax reg1
-rdfx reg2
-wrax to reg1
-skp gez the useful code.

Does anybody know of a shorter way to do this?

Actually the long reverbs use up the whole memory and we would have to modify them.

I'll be offline till end of next week.

Thanks

Herbert K

Could you give me a better description of the problem? When the FV-1 starts up the internal clearing algorithm clears the entire memory before loading the program and unmuting the chip.

If this does not appear to be happening in your case please check that you have the capacitor on the xtal pin as indicated in current versions of the datasheet. We have discovered some crystal makers cut the crystal in a manner that can cause it to come up in a high order harmonic initially. The capacitor will suppress this and cause the crystal to operate at its fundamental frequency.

Hello,
[quote="frank"]
When the FV-1 starts up the internal clearing algorithm clears the entire memory before loading the program and unmuting the chip. [/quote]

I guess it does but I guess the input pulse which makes the clip led flash on power up (in our design as well as in the dev board) has not completely subsided when the memory is cleared.

So a spike is put into the effect while the output is muted. Depending on the regeneration of the effect the unmuting happens too early.
As I said, take a dev board with a high impedance amplifier behind it and power up the board. All effects but the long reverb are quiet.

[quote="frank"]
If this does not appear to be happening in your case please check that you have the capacitor on the xtal pin as indicated in current versions of the datasheet. We have discovered some crystal makers cut the crystal in a manner that can cause it to come up in a high order harmonic initially. The capacitor will suppress this and cause the crystal to operate at its fundamental frequency.[/quote]

Thanks I'll try that and come back.

Herbert

[quote="frank"]
If this does not appear to be happening in your case please check that you have the capacitor on the xtal pin as indicated in current versions of the datasheet. We have discovered some crystal makers cut the crystal in a manner that can cause it to come up in a high order harmonic initially. The capacitor will suppress this and cause the crystal to operate at its fundamental frequency.[/quote]

[quote="herbert"]
Thanks I'll try that and come back.
[/quote]

Unfortunately that did not work. And I'm quite sure my assumptions are correct. Anybody can try it with the built in effects 6 and 7 on the dev board. The muting is clearly audible :-)) before the decay of the reverb sets in.

So back to my original question:
Is there a shorter way to skip effect processing for the n first cycles than what I proposed?

Thanks,

Herbert K

hkoenig wrote: So back to my original question:
Is there a shorter way to skip effect processing for the n first cycles than what I proposed?

Not that I can think of. I suppose (if your code is written in a manner that this would fit) you could try ramping up an input coefficient, so it is 0 when the program starts and ramps up to 1.0 after some time. May result in less code overall.