I also think that the 2500 is as reliable as a 2892 but the 2892 has a big advantage : its simplicity, and in mechanical stuffs simplicity means more reliability.
So, just to be clear, you are saying that they are both as reliable but the 2892 is more reliable because it is simpler?
Personally I think that simplicity is not obviously the key reason that something is reliable. Good, well executed design in high quality materials is usually the reason that something is reliable. Take a Russian Lada car engine. These are astonishingly simple and, in my experience, often have a total of
no moving parts. A Mercedes engine is significantly more complex and yet manages to be infinitely more reliable. Likewise, a quartz watch is a far more complex design than a mechanical (if a little more inscrutable) and yet...
As for the changes between the 2500C and the 2500D...
The 2500D/8500 escapement is even more complicated than the previous 2500 with added parts so I still ignore how the Co-ax would last longer than a Swiss lever which is 10 times simpler, escecially when the next generation of the Co-ax will all run in 28,800bph.
I'm not sure how the splitting of the functions of one quite complex wheel into two far simpler wheels with a single function each adds much to the complexity of the escapement. Personally I'd say that this simplifies it. More to the point, I'm not clear what the added parts are. Perhaps a list would be helpful as I can't think of any
added parts and there doesn't seem to be much of a change in the parts count.
The 2500D/8500 escapement is even more complicated than the previous 2500 with added parts so I still ignore how the Co-ax would last longer than a Swiss lever which is 10 times simpler, escecially when the next generation of the Co-ax will all run in 28,800bph.
I don't think it is more complicated and I certainly don't think that the coaxial is 'ten times' more complicated. A little bit more complicated sure, but mostly it's just different. I have explained how this difference makes the watch more stable elsewhere.
At least one thing for sure now is that the 3-tier escapement seems to have a better handling of the power flow so the stoppage and the flick-to-start could be avoided.
The need to put a little energy in the system to start isn't a fault in the first place, but it was 'solved' with the C revision. The 'stoppages', which now seem to be accepted as a reality, are still only evidenced by less than fifty complaints over a three year period on WUS. As several million 2500 movements were made in that period and WUS is around the top of google for this search term (and the majority of complaints came from new members) I'd say we are looking at a clear case of sample bias giving a misleading impression. (But I've said this all before).
Omega say that they moved the 2500 to a three tier coaxial wheel to standardise the escapement between the 2500 and 8500. However, as I have hypothesised elsewhere (and we have argued to death previously) there seems to be a possibility that oil may migrate from the leading edge of the pinion of the coaxial wheel (which powers the escapement) to the trailing edge (which takes power from the intermediate wheel) and cause an issue. This would explain why Omega are so precise about the amount of oil used and the application of an epilame. In the 2500D the potential issue with oil migration is solved as the trailing edge of the pinion wheel has no function as power is delivered through an ordinary cog elsewhere on the coaxial wheel. This cog would not be terribly sensitive to oil migration even if the oil migrated that much further, which seems unlikely. In short, I'm not sure there is an issue with the coaxial escapement of the sort imagined by those who are not statistically minded, but
if there were and
if this were the problem, then it is solved.
However, in my experience,any fault at all, from a magnetised hairspring to needing a service seems to have been included in the '2500 problem' - there seem to be a very small number of watches for whom oil migration may have been a problem and an awful lot of people who joined WUS Omega to vent and left again. There is nothing wrong with doing this, but when added to the usual issues any movement will have it has given the 2500C an unfairly poor reputation. As far as I am aware there have been
no complaints about the 2500B on WUS and less than ten about the 2500C in any other watch than the Planet Ocean.