When playing forward, both full and abbreviated notation are used with equal success. Decoding the abbreviated notation does not cause any problems for the user.

Why can't you do the same in retro recording (except for moves with capture and pawn transformations, of course?).

Pascal Wassong (version by Gerd Wilts) Dedicated to Babette

7B/b1pp2p1/N1p1p3/2K5/PP6/1Pk1Pp2/QRPNrP2/nRqrB3 w - - 1 13

14+12. Last 55 single moves?

Full record of the solution

The last moves were: -1. .. Bb8-a7+ -2. Qa3-a2 f4-f3 -3. Ra2-b2 Qb2-c1 -4. Rc1-b1 Qb1-b2 -5. Rb2-a2 Qa2-b1 -6. Rb1-b2 Qb2-a2 -7. Qa2-a3 Qa3-b2 -8. Rb2-b1 f5-f4 -9. Qb1-a2 Qa2-a3 -10. h7-h8=B Qa3-a2 -11. Ra2-b2 Qb2-a3 -12. Ra3-a2 Qa2-b2 -13. h6-h7 Qb2-a2 -14. Qa2-b1 Qb1-b2 -15. h5-h6 Qb2-b1 -16. Rb1-c1 Qc1-b2 -17. Rb2-b1 Qb1-c1 -18. h4-h5 Qc1-b1 -19. Qb1-a2 f6-f5 -20. Ra2-b2 Qb2-c1 -21. Qc1-b1 Qb1-b2 -22. Rb2-a2 Qa2-b1 -23. Rb1-b2 Qb2-a2 -24. Ra2-a3 Qa3-b2 -25. Rb2-a2 a2-a1=N -26. Ra1-b1 f7-f6 -27. Rb1-b2 Qb2-a3 -28. h3-h4 (not h2-h4!) a3-a2 -29. a2xBb3 (this is no longer unique, since -29. Ra2-a1 Qa1-b2 -30. Rb2-b1 Qb1-a1 -31. Ra1-a2 Qa2-b1 -32. Rb1-b2 Qb2-a2 could've happened too) Bc4-b3 -30. b3xNa4 Nb6-a4+ -31. h2-h3 (this tempo is needed, so retracting h3-h4 earlier is necessary!) Ba7-b8 -32. Nb8-a6 Bb5-c4 -33. Na6-b8 a4-a3 -34. Nb8-a6 a5-a4 -35. Na6-b8 Qa3-b2 -36. Nb8-a6 Qa4-a3 -37. Qa3-c1 a6-a5 -38. Rc1-b1 Qa5-a4 -39. Qa4-a3 Kb2-c3 and finally the position is unlocked.

It's in an abbreviated form:

The last moves were: -1. .. Bb8+ -2. Qa3 f4 -3. Ra2 Qb2 -4. Rc1 Qb1 -5. Rb2 Qa2 -6. Rbb1 Qb2 -7. Qa2 Qa3 -8. Rb2 f5 -9. Qb1 Qa2 -10. h7-h8=B Qa3 -11. Ra2 Qb2 -12. Ra3 Qa2 -13. h6 Qb2 -14. Qa2 Qb1 -15. h5 Qb2 -16. Rb1 Qc1 -17. Rb2 Qb1 -18. h4 Qc1 -19. Qb1 f6 -20. Rba2 Qb2 -21. Qc1 Qb1 -22. Rb2 Qa2 -23. Rb1 Qb2 -24. Ra2 Qa3 -25. Rab2 a2-a1=N -26. Ra1 f7 -27. Rbb1 Qb2 -28. h3 (not h2!) a3 -29. a2xBb3 (this is no longer unique, since -29. Ra2 Qa1 -30. Rbb2 Qb1 -31. Ra1 Qa2 -32. Rbb1 Qb2 could've happened too) Bc4 -30. b3xNa4 Nb6+ -31. h2 (this tempo is needed, so retracting h3-h4 earlier is necessary!) Ba7 -32. Nb8 Bb5 -33. Na6 a4 -34. Nb8 a5 -35. Na6 Qa3 -36. Nb8 Qa4 -37. Qa3 a6 -38. Rc1 Qa5 -39. Qa4 Kb2 and finally the position is unlocked.

What cannot be deciphered in an abbreviated entry?

  • 2
    My two cents is that it's simply friendlier to many to just list the entire solution, for those are are new or in passing or some other reason, to try and wrap their heads around rather than decihoper the shorthand. Of course, using the short form makes sense to do for easier/shorter problems, but the longer the problem is, the harder it is for some to understand the short form. Commented Jun 25 at 0:29

1 Answer 1


Following retro moves by eye does my head in, frankly :-) I estimate that I make at least 1 error in every 5 moves. The extra redundancy provided by verbose notation is invaluable to keep me pointed in the right direction. Surely I'm not the only one to appreciate this.

This is even more true if I am writing a retro solution for others to read. I also suffer from some kind of chess dyslexia (the cause might fascinate a neuroscientist) whereby I will blithely write e.g. c3, c6 or f6 instead of f3. I think this originated in the change from descriptive to algebraic notation aged about 9. At some deep level, as a tot, my spatial mapping to the board was hardwired to the descriptive model.

Solving engines also presume a verbose form, as a kind of highest common multiplier. Even with forward moves, there are issues in disambiguation that can be hard to spot if not tracked automatically, and in retro moves can be perplexing indeed since the root cause can far from where a contradiction appears.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.