Tube rolling falls in two camps broadly: 1. The belief the operating point is critical for operation, 5% deviation is too much and all tubes sound the same when ran exactly the same in the same circuit as long as the specifications are respected. This is what the RT tube is, a basic 6N1P that in the Stratus circuit has the gain and current it should have. RT doesn't test the tubes exactly to the Stratus circuit but it's close enough. It's hard for me to believe they make such a difference, I've ran single plate output tubes, one pair testing over nos and another near emission failure. Couldn't hear a difference between them for casual listening. Still I have an open mind, maybe they do sound better, revelation it's not. 2. The tube roller, trying everything under the sun to add flavour/coloration. I'm very guilty of this. Trying the ECC40 in the Stratus won't hurt but the circuit is not designed with it in mind. Ideally you want 3 to 5 times plate loading resistance. With 6N1P the Stratus has 5 times, the ideal value. With ECC40 it's only twice, not enough by anyone's standard. In theory you should hear more distortion but in practice I prefer some of the earlier ECC40 to the 6N1P or 1539. Let's put it this way, the 6N1P family will sound to their full potential while the ECC40 will not. Be careful with ECC40 adaptors. I've had three tubes die on me because of them. The tubes are quite fragile and some NOS bases need a bit of force to get them in. In my case the glass would fissure but not break immediately. One tube lasted tree days in the adaptor before vacuum was lost while running it(not in the Stellaris thankfully). If you have to push it in you'll break it. Tungsram seem more tough than the Holland made ones.