After flashing 1.6.7, the signal to noise with AGC on does seem better, but I still find it easier to copy some weaker signals when I turn off the AGC and add some attenuation - to get the base noise level down into the -70 to -80 dB range.
I'm noticing some spurious 'signals' in SDR# (v1664) that seem to move with me as I tune up the bands though. So, for instance, I find a weak 'birdie' at 11.185MHz, and the same exact tone will follow me as I step up through the bands, tuning in 1MHz increments. It gets weaker, then stronger again, but it appears again at the same location, exactly 1MHz higher, at 13.185MHz, 15.185MHz, 16.185MHz, etc. The higher in frequency that I find it, the stronger it seems to get. By the time I get above 27MHz it is pretty strong. On some bands, I'm also seeing a 'phantom' signal that shows up exactly in the center of the spectrum display, and only appears while tuning up or down. I looked around the same frequency ranges in HDSDR and SDR-Console, but I don't see these 'spurs' there.
Could these remaining 'signals' be generated somehow from my PC, or be some sort of artifact from software? It doesn't seem to matter what spectrum width I use. I've most often been using 384kHz, or 192kHz.
I'm using a cheap 8" Windows 10 tablet as my PC - hoping to use it while portable. If it's generating the noise though, I may have to switch to something else. I'm using a 6' Anker USB cable with a ferrite choke in the center, with about 8 turns wound through it. That seems to get rid of most of the 'noise' from my computer. I tried adding snap-on ferrites, but they didn't seem to make any difference. Maybe I need a longer cable, and a second choke in line? Or could these somehow be IMD from local broadcast stations? I don't see any harmonic relationship between the locations, so it seems like an interaction with something local, or in my PC.
I'll have to try flashing older firmware again to see if anything about the signals changes.