Celestion Alnico Gold for gold diggers

Hey man, first of all, huge thanks for sharing all these captures! The core tone, speaker choice, and mic positioning across your uploads are genuinely fantastic—easily some of the best-sounding cabs out there. I wanted to give you a quick technical heads-up: I've gone through quite a lot of your cab uploads and different mic positions, and I noticed this exact same issue occurring consistently across every single file and version I checked (both IRs and NAM). When inspecting the waveforms and zooming in on the amplitude axis, there are distinct, recurring reflection spikes/artifacts sitting around -60dB to -70dB far down the tail (well after the natural 30–40ms speaker decay). Here is what happens as a result: In the IRs: At higher monitoring levels, these spikes act as audible slapback delays/flutter echoes. In the NAM captures: The neural net tries to model those late spikes and the noise floor during the decay, which causes an audible glitching/quantization-like breakdown in the tail as notes fade out. The great news is that the core captures themselves sound killer—you don't need to re-mic or recapture anything. A quick batch process applying a clean fade-out around the 30–40ms mark across the IRs (and training NAM on the truncated signals) would completely solve it and make this massive catalog absolute perfection. Really appreciate the huge amount of work you put into this, hope the feedback helps!