"videos should be ignored for the following reasons:
Soto: An ice water bath is not a faithful simulation of realistic field conditions unless the hiking trip has gone terribly wrong
Yours: The fact that you refilled the canister flawed the test from the very beginning. Furthermore, your test doesn't control for other variables beyond temperature. "
Jim's test is actually very good. While most quality canisters have propane and possibly isobutane instead of N-butane. there are probably some cheep brands out there that are filled mostly with just N-butane. In most cases vendors do not list the exact contents of the cans and in fact the contents of the can may very over time due to canges in the price of propane, isobutane and N-butane. When Jim filled both his canisters with butane he eliminated the biggest variable in most other reviews and test.
One other point that should be made is that isobutane (found in MSR and snowpeak canisters) boils at 15F while the more common and lower cost N-butane boils at 32F. Both have the same number of carbon and hydrogen atoms. Only the arrangement of the molicules are different. So simply using using a isobutane/propane canister you will get much better performance regardless of the stove. The only variables in Jim's test were the stoves.
In Satos ice bath video they conveniently painted both canisters white. We have no idea what fuels were in the canisters or even if the canisters were fueled with the same fuel.
"Running with canister in ice water isn't fraud, just misleading
When I run stove at 40 F, ice forms on the outside of canister. If it was in ice water it would be warmer."
Acutally the temperature of the ice water depend on how much water there is. If it is mostly water the temperature will be close to ambient. However if it is mostly ice the temperature will be 32F which is the melting point of water. However if you take mostly ice, a little water, and add a lot of salt the temperature will quickly drop to approximately 10F or even colder.
Sato's ice water test would be valide if you know what the composition of the fuel is. We don't.
Paul, if you want to test your stove in cold conditions I would try to duplicate the sato test but do as jim did and use two canisters filled with N-butane (jim could probably help you with that). You should get the results Jim got. After that I would do the test again but this time fuel the Sato with N-butane and fuel the other stove with MSR or Snowpeak fuel.
"I suspect that the jet in the Soto is a larger diameter than most stoves so the same amount of gas is discharged into the burner head as a normal stove to overcome the issue of lower pressure supply in the Soto."
I would agree that the jet size of the Sato is probably larger. That might give a slight improvement in cold performance or make clogging of the jet less likely. however the difference is probably not noticeable. However under warm conditions say 80F the stove might consume more fuel than Other stoves. The warm weather fuel consumption problem can be solved by adding a regulator. In fact in one of the links I posted earlier there is clear evidence of the regulator kicking in warm conditions.
the other possibility is that Sato added the regulator to try to improve safety. If you use a bad windscreen design in warm weather you could get into a situation were the larger flame heats the canister, which causes an even bigger flame, causing more heat,…BANG! Adding a regulator would clearly limit the maximum size of the flame.