Proof Game in 4 moves

The book only gives PGs in 4.0 moves, so I don't have any longer PGs.

Ah, I see. Make your own, like I did above.

Devious pawns.

The length of a PG has little to do with its difficulty.

If anyone's interested in a thread with PGs longer than 4.0 moves, I can post several.

I'll take some longer PGs.

OK - started a new thread: "Proof Games and more Proof Games".

There is a different flavor between short PGs and long PGs, at least from what I see. Long PGs usually involve inventory tracking, move planning, etc, which looks rather confusing to a beginner. I also find short PGs elegant, being able to hide an evil trick with just so few moves, although I should actually try long PGs before making this claim. I don't really talk about difficulty; only that 4.0 is kind of brute force-able while 7.0 is not really (just see, 4.0 doesn't allow promotions) so that 7.0 is generally harder than 4.0. I'm sure a 7.0 can be as hard as a 27.0, but mostly being difficult in a certain flavor, while the long game is difficult in a different flavor.

That's my opinion, at least...

4.0s are the right length to get people interested in PGs - Tibor Orban's is enough to confound beginners but not put them off, for instance. 7.0s can be far meaner, since it's easier to hide cross-captures and promotions (promotions take 5 ply, and at least 1-2 more to disguise.) They're still not too daunting, still solvable by forward play brute force, and will quickly get people used to seeing promotions, castling and cross-capture shenanigans.

Long PGs are a different ball game. The typical ~20 move PGs usually require a great deal of retroanalysis even before moving the pieces, and then the move-order finesses come into play. Trial-and-error from forward play alone won't get you far - too many possibilities, which need to be cut down by judicious retroanalysis. However, the compensation is the quality of the encoded content; longer, more complex manouevres can be woven into the PGs than the simpler "shorties", which typically offer but a single theme each.

But I digress. Putting this link up for whoever's interested: a booklet by Alain Brobecker et al, translated by Gianni Donati. Featuring a pedagogical collection of 25 short PGs, solutions and some commentary on recurring themes. Includes the Orban, Hoffmann, Kiesow, and Schweig compositions posted earlier in the thread.

That's more or less what I'm trying to say, although you raised a few good points I missed or opposite of my prior view. After actually trying to solve one long PG (that one), I'm more sure that the two have different flavors indeed, and now I actually think longer is in general more difficult; most 5.0 is way easier than most 25.0, although of course some 5.0 can be harder than 7.0.

By the way, because you provided that link, I also provide a link to Proof Game Shorties by Mark Kitley, hosted on Retro Corner.

For those who want more short PGs, here's one:

Little evil pawn.

Inspired from the one on this thread.

chaotic_iak, 2014
PG in 4.0 (verified by Euclide)

Note: This might have been done by someone else, considering it's pretty easy to come up with the idea. If it has been done, then oops.

Answer in white:

1. g3 b6 2. Bg2 Ba6 3. Bb7 Qc8 4. Bxc8 Bxc8

Note: You must solve the puzzle with exacly 4 moves

Try this one:

Not unique. Ng8 has many ways to go to Ne5, the move order for e2-e3-e4 and Ng1-f3xe5 are not determined...

How about this

EDIT: Nevermind, it's cooked.

Well, I tried. First attempts at making any sort of PGs. Pretty sure all unique, but very simple. All PGs in 4.0 of course.

Answers in white:

1. Nc3 d5 2. Nxd5 f6 3. Nxf6+ Kf7 4. Ne8 Kxe8

1. d4 Nc6 2. Bf4 Nxd4 3. Bxc7 Nc6 4. Bb8 Nxb8

Remellion's second puzzle is cooked.

Alternative solutions in white:

1. d4 Nc6 2. Bf4 Nxd4 3. Bxc7 Nc6 4. Bb8 Nxb8 (Frankwho)

1.d4 c5 2. Bh6 cxd4 3. Qxd4 Nxh6 4. Qd1 Ng8 (caveatcanis)

Remellions third puzzle is also cooked, but in a less interesting way:

1. d3 c6 2. Be3 Qc7 3. Bb6 Qxb6 4. d4 Qd8

1. d3 c6 2. Bf4 Qb6 3. Bc7 Qc7 4. d4 Qd8