Recent Posts

Hi guys, I didn't know where would be appropriate place to start this topic. If it's wrong place  please move it to the right one.
Anyway,  could you tell me your opinion, explain some differences between chess.com, lichess and chesstempo in terms of tactics training? I mean, where are best tactical puzzles and why?  I know that lichess is most popular becaue it's all free. But, is the quality, level of tactics puzzles that good? Or maybe chesstempo will rise my skills way better?

Thanks in advance

1...e4 ends up with white having two pawns for the piece, compared to 1...Nf6 which is immediately resignable (which is exactly what happened after Kasparov played that move in the source game). Stockfish at higher depth sees that 1...e4 is eventually going to win, but it is a poor choice compared to 1...Nf6, and the extra white pawns should have been enough to have the player look for something better.

However, having said all that, the eval for both moves is a little closer than I'd like, so I've disabled the problem now

There is an isssue with mixed mode non-winning problems with their starting rating, a great many cluster in the 1500 rating, and because there are so many to choose from in that range, it ends up taking a LONG time for them to get a first attempt due to the way the user rating/problem rating matching algorithm works. If you are including 1500 in your rating range, you might want to use a range that explicitly excludes problems exactly rated 1500 (or perhaps more conveniently, use an attempt number filter to make sure they've had at least one attempt). We should probably do something to perturb the starting rating like we do with non-winning problems.

Just following up to say I made this custom set (non-winning, max five pieces) and it’s great! The weird thing though is out of the 21 problems I’ve been served so far, only one of them has ever been attempted by anyone else before, and that one only had two previous attempts. The rest of them I was the first person trying them. So all of them have been rated 1500 which makes it kind of unpredictable, but that’s ok. They’re actually crazy hard problems for the most part. Anyway, thanks for the idea!

please edit this problem, 1...e4 is -3 while Nf6 -4.8 - both are clearly winning. Thanks
https://chesstempo.com/chess-problems/54885

Thanks very much for your help. I'm really enjoying the puzzles here, you have an amazing website!

Not easily.

You could probably approximate this by choosing the rating range you want the 100 problems to range across, and then using the problem quality filter to reduce the number (by increasing the minimum quality) until the search results hits around 100 problems.

However this will return a relatively fixed set of problems each time if you use the same quality range. You could perhaps use a very narrow quality range to start with, say 4.12 to 4.15 and then increase to 4.16 to 4.19 on the next run through which will give you a few sets, but I'm not sure all this is worth the effort, and if you really want the behaviour of chess.com's puzzle rush, you'll likely be better served by just using their mode than trying to find a CT approximation, as I don't think any of the hacks you could come up with via custom sets will be very easy to work with, especially if you want to keep doing this in the longer term.

I really like solving puzzles in a format of ever increasing rating level, starting from trivially easy. Is there a way of creating a custom set of say 100 random puzzles sorted by rating level where the first puzzle is rated around 1000, and each puzzle thereafter increases in rating by around 25 points?
This would simulate chess.com's puzzle rush survivor feature.

Thank you for the kind words. Yes, the road is indeed endless, and it's good to be back on it.

You mean to play training games together? What is your rating?

Hehe, I play in zen mode also so don't really know - I haven't played regularly but want to do so and am looking for a playing buddy or buddies. I've played a few OTB games this season and rating of these would probably be just over 1600

Finally got on the scoreboard on Sunday morning with a win. I was again playing against a kid (older, and higher rated). He was pretty strong, but I don't think he'll make it as far as the younger kid I lost to the night before.

He played the more conservative Ne2 instead of Nf3, which allows him a reloader on c3 to defend the e4 pawn/square. (In some Nf3 lines, white tries a quick castle and sacs the b2 pawn in order to gain initiative, with lines like 5. Nf3 Nf6 6. e5 Ne4 7. O-O!? Nxc3 8. bxc3 Bxc3, but those aren't usually very scary for black and we have an extra pawn). I've been opting for the Nf6 lines more and more lately, because the d5 lines are sometimes a bit dull (White answers ...d5 with e5, and we're in a French advance structure where we try to play ...Ba6 to exchange our bad bishop in combination with the ...c5 break). I tend to prefer the more 'compact' setups where I don't block my long diagonal and play ...d6 instead of ...d5, which ressembles more a hippo.

After 8...Be7 (as e5 was a threat, and the bishop is no longer needed on b4 anyway)  9. e5 Nd5, exchanges are forced which helps alleviate my lack of space due to the e5-d4 pawn duo. I took back on e7 with my knight, as I was afraid of Be4 pinning me (since the Bb7 is undefended), but that wasn't necessary since Be4 is answered with ...Nxc3 and the white queen hangs, so no time for Bxb7. A sad failure to calculate properly a 2 move combination. Still, it seems Stockfish prefers ...Nxe7 anyway, maybe because it allows ...Nf5 later, which led my opponent to start burning a lot of time.

I gambled with 14...Nxd4. I was sure my opponent had seen the disrupting 15. Nf6+!! move. I had calculated the (I believed) strong answer 15... Kh8 (since 15...gxf6 looks suicidal), where there's an attack brewing for White (4 pieces are closing in on my king), but nothing looks decisive yet. Of course, Stockfish sees right through that, and probably a stronger player like an expert or even an 1800+ would have played this move, and mated me on the kingside, so I think got lucky.

Much better to keep the game even would have been 14...Nd7. Once again a simple chess principle. Get your pieces developped. The Nb8 is often very difficult to get into the game because he'll block the queen's view of the d-file, or the Bb7's long diagonal, and maybe allow d5 to be played by White. Yet the simple 14...Nd7 15. Rf4 c5 and the game is balanced and both players have chances.

I think this is one of my weaknesses. Just as in the round 2 where I never castled and ended up in a very contorted and uncomfortable position, here again I went for an adventure, and was a little too greedy by grabbing on d4! I just lost a 15+10 game this morning by making a silly pawn grab in a completely winning position (and instead got mated a few moves later). This will be one of the challenges this year: to correctly evaluate pawn grabs. In both these games (at the tournament and this morning's), my evaluation was dead wrong.

After 15. Rf4?? however, I easily liquidated into a winning endgame with 15... Nxe2+ and a queen exchange. The game ended abruptly after that. In the post-mortem, I pointed out to my young opponent that he could have put up a much longer resistance if for example he got into a setup with the bishop on d3 defended by the c2 pawn (blocking the d-file). As it played out, I just took over the file, easily defended the threats on f7, and converted a few moves later. I was happy to see I played all the top computer moves in the final few moves, never giving my opponent a chance (the Bxc6 exchange was terrible: I'm up a pawn, I have a BvsN with pawns on both wings, e5 is terminally weak, and I own the only open file; too many advantages for Black).

The game was pretty short (about 2 hours) since my opponent used alot of his time, and I only used about 30 minutes for the entire game. After Nxe2+, my position plays itself, I was using ~15 seconds per move.

( ) vs ( )

Date:

Event:

Site:

Round:

Result:

[Event "47e Championnat de Mauricie"] [Site "Trois-Rivières"] [Date "2024-05-12"] [Round "4"] [White "L.L."] [Black "kharv"] [Result "0-1"] [WhiteElo "1628"] [BlackElo "1684"] [Variant "Standard"] [TimeControl "-"] [ECO "A40"] [Opening "English Defense"] [Termination "Normal"]  1. d4 b6 2. e4 Bb7 3. Nc3 e6 4. Bd3 Bb4 5. Ne2 Nf6 6. O-O O-O 7. Bg5 h6 8. Bh4 Be7 9. e5 Nd5 10. Bxe7 Nxe7 11. Ne4 d6 12. f4 Nf5 13. Qd2 dxe5 14. fxe5 Nxd4 15. Rf4 Nxe2+ 16. Bxe2 Qxd2 17. Nxd2 Nc6 18. Bf3 Rad8 19. Bxc6 Bxc6 20. Rf2 Rd7 21. Raf1 Rfd8 22. Nb3 Bb5 23. Re1 Rd1 24. Rff1 Bxf1 0-1