Maybe. I'm looking into it. I'll know more in a couple of hours.

EDIT: This has nothing to do with the SSE2 limit. That limit was passed ages ago. The issue here is whether we've passed the x87 limit. Theoretically, we have about 8 million to go until we reach the limit, but it's expected that some specific tasks will fail earlier. We may have reached that point. If we have, it's time to turn off the GFN-16 CPU app.
____________
My lucky number is 75898⁵²⁴²⁸⁸+1

I ran a few tests, and the Windows version of Genefer is failing with the x87 transform on that particular number. The Linux binary does it correctly, but it's certainly going to start failing on other numbers soon, if it hasn't already.

This is not unexpected, but the exact timing of when this would happen is difficult to predict.

It's time to make GFN-16 GPU-only. You should select something else to run on that CPU.

Within the next few minutes I'll be shutting off GFN-16 CPU tasks.

GPU tasks are not affected.
____________
My lucky number is 75898⁵²⁴²⁸⁸+1