mjrgh

> But then again, i just checked the MAME codebase, and they do not seem to use any DCS filtering at all AND have sound without buzz/noise??? Keep in mind...

> And the sample rate converter is basically one monolithic code path, Just to clarify, I didn't mean that there was necessarily a bug in the sample rate converter code...

Are there tags or commit numbers for the 2.7 and 2.8 releases? I don't see them in the tag lists. I was thinking I could try diffing the two versions...

Here's the diff, for anyone else who wants to try finding it by inspection: https://github.com/vpinball/pinmame/compare/f2476cbb361068575146b505c5f4d7d2473ffbce..2ddfb10e4f60c9b0ed7122ca655e36b30bc27bbc It's a large diff, but I suspect there are only a few files that are...

@toxieainc, one thing that might be worth a quick try is reverting the changes at the following lines in src/wpc/wmssnd.c (line numbers reflect old/new sides in the diff linked above):...

> but the issue also happens if one disables the WPCSPEEDUP thingie.. Ah, okay, that rules out that part of the code, at least. > Couldn't it rather be the...

> reverting the whole wmssnd.c and patching it up to work again doesn't change anything > same for all of adsp2100 [and the resampler] That's pretty perplexing. I'd propose trying...

Same thing happened to me a few days ago. I didn't make any changes in my local setup at all, so my guess is that Amazon pushed an Alexa update...

> It does make me wonder if it is related to an update Amazon pushed out. There's nothing to wonder about - that's all but certain, based on my setup....

> I suspect that there is $$ motivating this maybe since those vendors that > provide Skill's might pay a fee and Phillips took some action. That seems unlikely given...