Woke up at 4 am, just in time to setup and try to capture the Elephant Trunk before morning twilight. Got a total of 45 frames, for 7.5 minutes of integration, no filters.

The picture above is a 2x Drizzle stack done in PixInsight. Doing your own stacking and stretching definitely pays off. Compare with the JPG straight out of the Seestar S50 in the second image here. Night and day difference. But the raw JPG does show a hint of the dark clouds in the lower portion of the image.

I believe that the lumpy appearance of the diffuse background nebulosity is due principally to the low signal to noise in this image. We are bouncing noise, even after a fabulously effective NoiseXTerminator treatment. The NoiseXTerminator does a great job on higher spatial frequencies, but the residual is full of low spatial frequency noise.

I would expect that the noise spectrum, being one from nature, in the structure of the nebula itself, and the nature of physical measurement processes, would exhibit a 1/F spectrum. So the lumpy structure represents the lowest spatial frequencies, and strongest amplitude, of the noise with which NoiseXTerminator must contend.

In that sense, perhaps the Seestar S50 JPG is the most honest representation, choosing to ignore things below 3-sigma in significance. But we know from past experience that the grand structure of the Elephant Trunk is really there. So even though we captured it at low SNR, it seems a fair conjecture to present it. The lumpy structure in the background simply attests to the fact that we are conjecturing to some degree.

The other thing to note here is that there seems little difficulty showing the darkest clouds. Even the Seestar S50 raw JPG showed hints of them. We see much of the nebulosity as a psycho-visual effect of the absence of signal. The dark regions provide a context for our vision, demarcating the nebulous glow regions, helping us to discern their existence and shape.

The fact that the dark regions appear so prominent testifies to the fact that the instrument itself is not introducing excess noise, which would serve to wash out those dark regions. So the noise here, I would argue, is dominated by small-number photon statistics.

While the Seestar S50 would not be the instrument of choice for such DSO's, I think it does a credible job for a 2 inch telescope.