Captain Custard,
I think many controllers would agree whole heartedly that certain non-radar separation standards are too large. The situation you allude to where a confliction in class C is "resolved" by forcing one or both aircraft to descend OCTA has frustrated me throughout my career.
My feeling is that the GPS derived sep standards are far too conservative, pilots appear to have a high degree of trust for GPS and our sep standards should reflect that.
The bizarre aspect of this debate is that we are apparently a "world leader" in application of RNAV seperation, ICAO are even more conservative and one of the most outspoken opponents to further reductions in RNAV derived sep standards is IFALPA! I suggest you and other pilots make representions to ASA through your companies and AFAP that the existing GPS standards are too conservative.
I believe that with a more realistic figure for GPS Navigational Errors, we could roll out more class E, and seperate everyone positively while minimising unecessary interventions. Then we could consign this bodgy see and avoid nonsense to the same bin as the G trial and AMATS and Airspace 2000 etc