@1
"it's best to set-up the computer strength to be only a few 100 elo rating points above my own."
++ Set it as high as you can. You will need less repetitions.
How best to train with old chess computers?
Sort:
Disagree. Research has indicated the most optimum level for training games in terms of speed of improvement is a 25%/75% win/loss rate. People soon lose interest in chess (or playing chess computers/bots) if they have little chance of winning - many old chess computers are in pristine condition after decades simply because of this characteristic.
My method is to play a game, do a Stockfish (SF) analysis to find my first poor/weak move, then re-run/set-up the chess computer to the defective move and play again with the improvement, rinse and repeat. Eventually you get a very fine and memorable game, in my case this can take 3 or 4 repetitions. I think it's best to set-up the computer strength to be only a few 100 elo rating points above my own. Many of the old chess computers make this whole process quite trivial.
Note, if SF recommends a move/line I would have no hope of playing out I ignore it for a more sensible / human one. Simple fact is at my level, c. 1500 OTB, this is good enough giving my likely opponents. I doubt I can reliably play more than 4 to 5 ply ahead as at least half my poor play comes from poor candidate move selection.
It's a shame that the benefits of these old chess computers appear to be completely ignored nowadays.