Here is chart with my supplemental training (set G/H) combined with my original revision.
Much like my experience with the Bain supplement, I had great success with these difficult problems, achieving over 90% of problems completed in under 15 seconds. I was only able to achieve ~75% success rate on my initial revision of Woolum.
Here is the full graph for revision #2.
Problems completed in less than 15 seconds.
I improved slightly in this category from ~74.5% in revision #1 to ~78% in revision #2. I experienced ~2% decline from revision #1 to #2 with Bain. The improvement with Woolum is likely due to the increased amount of supplemental problems.
Side-by-side comparison of my last pass in revision #1 (Pass 6) with my single pass in revision #2 (Pass 7) in under 15 seconds.
Oddly enough, I did slightly worse on the sets I had completed more recently, although that could be the fatigue factor. 4 hours sleep and 700+ tactics problems in a row might wear on you a bit.
Chess Hero seems to be having a good effect on my solve times. My supplemental set for Heisman will be in the ~5% of problems range and my 15 seconds or less solve rate will be ~90% or better. Hopefully this is due to better retention then just being faster than looking at a flashcard and clicking next.
I will finish my final set of Heisman on 4/14 then start on Polgar - Chess Tactics for Champions on 4/19. I haven't even started to convert the diagrams into a pgn file for Chess Hero yet, so I need to get started on that ASAP. I wish that sort of thing came standard with tactics books these days. I know I could go with some tactics software, but I want to go through the same progress Bright Knight did with his tactics training. After I finish Blue Coakley, my tentative plan is to move on to CT-Art 4.0 followed by Chess Tactics for Intermediate Players.
Would be interesting, if you get better at problems you have never seen before. All slow repetitions of Chesstempoproblems made me slightly worse in problems i have never seen before.
ReplyDeleteI would like to know that too. I have an idea for a chess tactics book. A series of 25 question tests of similar difficulty that the student takes before undergoing a tactics training regimen and taking a subsequent test every 2-3 months after completing a course of study IE finishing a spaced repetition with a specific tactics book.
DeleteIf you get better and faster each time, it should be because of the tactics training.
there is a possibility to see if you get better in tactics: chesstempo. The first ~1000 attempts will make you familiar to the system, handling, pieces, timing...
DeleteNow you can measure your progress. Make your training and redo a few hundreds chesstempo puzzles to see an effect of your training.
You can repeat that method until you reach about 4000 attempts at chesstemp, then you will start to see many duplicates
Then you may change to the chess tactics server ..
An alternative : Chesstempo provides for premium users a "fide rating estimate" which is based on puzzles you never have seen before ( there are always new puzzles getting into the chesstempo database ). So premium user can simply use the fide rating estimate to recognise positive or negative effects
the same 25 puzzles doing more than 1 time dont work propper. I solved the puzzles of the chessexam once and more than 1 year later i still did remember a few puzzles of that book