f/0.7 is only 1 stop from f/1.0
It does however change the system noise:signal ratio as you would expect.
Remember though that NETD is quoted in an f/1.0 system regardless of lens fitted, so that '50mK f/1.4" is actually a 100mK system due to the lens fitted, so the f/0.7 lens only brings things down to looking like a 25mK f/1.0 system
As there's a lot more going on in the system, it is not that simple. the camera might not be running in the same state as gives the quoted NETD, then add in ADC noise, quantisations and display stretches which are more likely to limit overall system MDTD / sig/noise.
As for the experiment, yes (as I have all those pevicon lenses to hand). I have documented a thread (and now on my website) where I change a FlirOne lens for anything I could lay my hands on.
A complication is that Pevicon lenses are designed for what it essentially a 100um pixel 200x200 sensor. That means several things are rather different when on a 12um sensor.
The common focal lengths will be quite a narrow field of view
The resolution and depth of field will be rather poor
Poor edge performance will not matter - it is off the edge of the sensor
https://www.eevblog.com/forum/thermal-imaging/flir1-g2-lens-experiments/http://www.fire-tics.co.uk/project4/index.htmBill