I look around me and I see it isn’t so (Apologies to Paul McCartney)
It’s quite likely that the weight of the number of books written on chess tactics would be enough to crush and kill and a man of medium build – instantly. Even the collection of books I have on the subject might be enough to send him to the hospital; in fact, I’m a little tired of writing reviews of tactics books. Does anything more noble than the desire to make a buck (or pound, euro, ruble, etc.) justify this unending crime against treedom?
Most of the time, the answer, in my opinion, is a resounding “no”. Rehashing old material is terrible, as long as the original is still in print, while putting out new work solely because it’s new is also senseless, if it is intended for the same audience as the older work. The point is that if the student needs to grasp the ideas, the older work will generally succeed as well as the new material. Because he’s not familiar with the ideas that the old material covers, it’s as good as new to him. So why chop down more trees and jack up the price when no new benefit is conveyed by doing so?
Why? Because people buy, that’s why. If people would simply master the tactics books they have and then buy one that challenges them a level up, they would get the full benefit and not waste their money. But alas: hope springs eternal, and discipline often springs away like a scared rabbit. (Have I insulted everyone yet? Don’t forget the second sentence of the first full paragraph; I’m there with you, my brothers and sisters!) So let’s be more disciplined, get the most from the books we do have, and send our implicit statement of ridicule when greedy publishers and lazy authors try to make a quick buck by reinventing a perfectly good wheel.
And now I turn to the book under review, a collection of (about) 1,000 puzzles split up into various historical eras. There are fourteen historical chapters, going all the way back to the ancient era of Shatranj (an earlier version of the chess we play) and continuing up to the year 2000. Each chapter starts by recounting the chess history of that period, highlights several fine combinations from the era, and then concludes with a number of exercises for solving, a la the traditional puzzle book.
Now, it should go without saying that this is not a full-fledged book on chess history by any means. I read another review which complained a little that compared to Kasparov’s My Great Predecessors series, the history in Chess Gems was rather thin. This is true but highly irrelevant: Kasparov’s books comprise about 2000 large-size pages compared to Sukhin’s 300, and Sukhin’s book is explicitly centered on combinations; he doesn’t pretend to offer a historical tome. (Conversely, one could also attack Kasparov’s books as chess history, as there are large biographical volumes devoted to players he deals with in comparatively few pages of prose.)
On the other hand, while the Kasparov books win in the stories and deeply annotated games departments, Chess Gems lets us see a greater overall quantity of fragments, and top players outside the highest circle are included as well. In this respect, one can see the books as complementary rather than more or less thorough means to the same end.
Considering Chess Gems as a puzzle book, which is what it's primarily intended to be, it’s pretty good for club players under about 1800. It’s not a structured work – it doesn’t separate based on particular themes or difficulty. That makes it unsuitable for novices – the the structured approach is best for learning the ideas in the first place – but for those who do have the ideas the unstructured approach is superior.
So maybe it shouldn’t be one’s first tactics book, and if you’re seeking a Ph.D. in chess history, you don’t need this in your library. But for the average club player whose shelves aren’t groaning under the weight of other unused tactics books, it’s not a bad buy. The snippets of history covering the 20th century are pretty thin, but the chapters on previous centuries – the earlier the better – are surprisingly rich.
Let’s look at some problems from a couple of the early chapters, starting with the shatranj era. Shatranj, for those unfamiliar with chess’s “great predecessor”, had some differences with the modern game. Knights, rooks and kings moved the same way, but there was no double pawn move, the queen could only move diagonally and just a single square at a time, while bishops moved only and exactly two squares diagonally, jumping over pieces if it wanted to. There was no castling, and a game could be won by not only by checkmate but also by stalemating the opponent or taking all his pieces.
Many of us have seen this ancient problem. It’s a shatranj problem, but since in this position all of the pieces have the same capacities as in modern chess, it’s right at home for us, too.

Abu Naim Al-Khadim (9th century), White to move and win
This position still shows up from time to time in contemporary tactics books. However, it's about the only position from that time that does, probably because it doesn't include any distinctive shatranj features. But Sukhin has included ten more fragments, including these:

Al-Adli (9th century), White to move and win (Remember that Black's king isn't in check, as shatranj queens only move one square diagonally.)

Firduosi at Tahitala Manuscript (1501), Black to move and win
(The solutions to these three positions can be found at the end of this review.)
And here are three positions from the 18th century Syrian player Philip Stamma’s 1737 book (itself in part a puzzle book – you see how old this genre is?):

Stamma (1737), White to move and win

Stamma (1737), White to move and win

Stamma (1737), White to move and win
Finally, here's a position from the last chapter of the book. Since we've mentioned Kasparov several times already, let's give him the last word, chessically speaking:

Gelfand-Kasparov, Novgorod 1997; Black to move and win
(Solutions to the Stamma and Gelfand-Kasparov positions here.)
In sum, it’s a fun book; not necessary, but a decent tactics text for the club player, and an enjoyable way to catch up on some of the great combinations in chess history. (Those interested can find the book here and here, and no doubt elsewhere.)
Solutions to the shatranj puzzles: