I notice that I must express myself a bit more clear.
I have done the 7 circles with TCT (1500 problems) and with George Renko's ICT I (1300 problems). I have heard that the difficulty of ICT I is comparable with CT-art. These circles have given me a good improvement, average to what most knights obtained from it. So no refutation of the circles here.
But then I formulated the hypothesis that in order to get better at complex tactical combinations I had to learn to solve lots of simple combinations a tempo first. So I started to do the 7 circles with 10,000 problems from CTS, which are very low level. What I say now is that this experiment has refuted the hypothesis. I have not become better in complex tactical positions.
Well that is not quite true. I have become better in positions where there are a lot of parallel tactical elements which are shallow. But with the problem I do now there is a long sequential series of 9 basic tactical elements, and there the benefit from CTS is close to zero.
Now I'm experimenting how such long lines should be mastered.
6 hours ago