Martin, I get it, the “getting your money’s worth”
But mathematically, it doesn’t stack up. The noise contribution from the temperature is already really small at 5C, to the point that in most scenarios it doesn’t significantly impact your total noise.
All I’m saying is, the great performance is from the sensor itself, and not just the cooler. The cooler is almost incidental, the noise levels are so good, but useful for keeping it consistent to make calibration viable. These sensors are such a step up from the previous generation (ASI1600 etc) and makes them look noisy by comparison.
Living in SE QLD, the summer nights are frequently “feels like” 30C or thereabouts, so what I use works well all year round, and I don’t have to mess around with different sets of darks
Anyhow, like I said, just my 2c…