Either the date, site name, or score is incorrect here.

<Smyslov> played <Geller> twice, once with each color, at the VORONOVO-GAGRA (USSR), 1952—Training Match Tournament.

<Smyslov> scored +0 -1 =1 in his two games against Geller.

Here are the game scores from that event:

[Event "Tournament"]
[Site "Voronovo (Russia)"]
[Date "1952.??.??"]
[Round "?"]
[White "Geller Efim P (RUS)"]
[Black "Smyslov Vassily (RUS)"]
[Result "1-0"]
[ECO "E14"]
[WhiteElo "0"]
[BlackElo "0"]
[Annotator ""]
[Source ""]
[Remark ""]

1.d4 Nf6 2.c4 e6 3.Nf3 b6 4.Nc3 Bb7 5.e3 Be7 6.Bd3 c5 7.O-O O-O 8.d5 exd5 9.cxd5 d6 10.e4 Nbd7 11.Nd2 Ne5 12.Be2 a6 13.a4 Re8 14.f4 Ng6 15.Bd3 Bf8 16.Qf3 Bc8 17.Nc4 Ra7 18.Bd2 Rae7 19.Rae1 b5 20.axb5 axb5 21.Nxb5 Nh4 22.Qg3 1-0

===

[Event "Tournament"]
[Site "Voronovo (Russia)"]
[Date "1952.??.??"]
[Round "?"]
[White "Smyslov Vassily (RUS)"]
[Black "Geller Efim P (RUS)"]
[Result "1/2-1/2"]
[ECO "B24"]
[WhiteElo "0"]
[BlackElo "0"]
[Annotator ""]
[Source ""]
[Remark ""]

1.e4 c5 2.Nc3 Nc6 3.g3 g6 4.Bg2 Bg7 5.d3 Rb8 6.f4 e6 7.Nf3 d5 8.O-O Nge7 9.Qe2 O-O 10.Nd1 b6 11.Nf2 Nd4 12.Nxd4 cxd4 13.Bd2 dxe4 14.dxe4 Bb7 15.Rfd1 Qd7 16.c3 Rfd8 17.cxd4 Nc6 18.d5 exd5 19.e5 Nd4 20.Qd3 Ne6 21.b4 Rbc8 22.f5 gxf5 23.Qxf5 Nf8 24.Qh5 Ng6 25.Ng4 Rc2 26.Nf6+ Bxf6 27.exf6 Qe6 28.Qf3 Qe4 29.Qxe4 dxe4 30.Bh6 Rdc8 31.Rac1 Rxc1 32.Rxc1 Rxc1+ 33.Bxc1 Ne5 34.b5 Nf3+ 35.Kf2 Nxh2 36.Bh3 h5 37.Ke3 Ng4+ 38.Bxg4 hxg4 39.Kd4 Kh7 40.Ke5 Kg6 41.a4 Bc8 42.Kxe4 Be6 43.Be3 Kxf6 44.a5 bxa5 1/2-1/2

Here is the cross table: http://al20102007.narod.ru/nat_tour...

===

<Di Felice "Chess Results 1951-1955," p.180,> confirms the crosstable and score between <Smyslov and Geller>.

VORONOVO-GAGRA (USSR), 1952—Training Match Tournament

This was actually a "team training event." Each member played a white and a black against each member of the opposing team.

TEAM 1: Smyslov, Botvinnik, Keres, Kotov

vs.

TEAM 2: Bronstein, Geller, Boleslavsky, Petrosian

########################################

The game on this page appears to have been played at

GAGRA (USSR) 1953-Training Match Tournament

This was not a "team training event"- it was a single round all-play-all tournament.

Here is the <Geller-Smyslov> game from that tournament, sourced to <chessbase>:

[Event "URS-training"]
[Site "Garga"]
[Date "1953.??.??"]
[Round "4"]
[White "Geller, Efim P"]
[Black "Smyslov, Vassily"]
[Result "0-1"]
[ECO "D18"]
[PlyCount "60"]
[EventDate "1953.??.??"]
[Source "ChessBase"]

1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 dxc4 5.a4 Bf5 6.e3 e6 7.Bxc4 Be7 8.O-O O-O 9.Qe2 c5 10.Rd1 Ne4 11.d5 Nxc3 12.bxc3 exd5 13.Bxd5 Qc7 14.e4 Bg4 15.h3 Bh5 16.c4 Nc6 17.Bb2 Rfe8 18.Ra3 Rad8 19.g4 Bg6 20.Re3 Nb4 21.e5 Bc2 22.e6 f6 23.Rd2 Bxa4 24.g5 fxg5 25.Be5 Qb6 26.Nxg5 Nxd5 27.cxd5 Qb1+ 28.Kh2 Bxg5 29.f4 Bh6 30.Rg3 Qf5 0-1

Here is the event crosstable: http://al20102007.narod.ru/nat_tour...

This crosstable is confirmed by <Di Felice "Chess Results 1951-1955," p.240>

Now, here is the game score on this <cg.com> page:

[Event "Gagra"]
[Site "Gagra"]
[Date "1952.??.??"]
[EventDate "?"]
[Round "?"]
[Result "0-1"]
[White "Efim Geller"]
[Black "Vasily Smyslov"]
[ECO "D18"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "65"]

1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 dxc4 5.a4 Bf5 6.e3 e6 7.Bxc4 Be7 8.O-O O-O 9.Qe2 c5 10.Rd1 Ne4 11.d5 Nxc3 12.bxc3 exd5 13.Bxd5 Qc7 14.e4 Bg4 15.h3 Bh5 16.c4 Nc6 17.Bb2 Rfe8 18.Ra3 Rad8 19.g4 Bg6 20.Re3 Nb4 21.e5 Bc2 22.e6 f6 23.Rd2 Bxa4 24.g5 fxg5 25.Be5 Qb6 26.Nxg5 Nxd5 27.cxd5 Qb1+ 28.Kh2 Bxg5 29.f4 Bh6 30.Rg3 Qf5 31.Qe3 g6 32.Bb2 Rxd5 33.Qc3 0-1

The games are clearly the same, except the <cg.com> version shows a resignation after three more moves 31.Qe3 g6 32.Bb2 Rxd5 33.Qc3 0-1

===

I would suggest a notation like <Primary Source> (De Felice) for explicit reference if possible. Is that reasonable?

(Some indication that we don't have a circular reference is needed imo).

Aside- It is interesting to see Geller-Smyslov hash the Slav Accepted out. There is URS-ch18 (1950) and Budapest (1952) where Geller doesn't play 5.a4. In URS-ch20 (1952) and thereafter he does play it. Like here (1953)!

But in the case in point for this one game here- There are at least two facts in play:

1. <Chessbase> gives a different date and three extra moves for the same <cg.com> game that is posted on this page.

(BTW- What do you mean by 1953 <Gagra training event> in 1952?)

As I said, Geller seems to have changed his approach to the opening around 1952 which adds some information (far from conclusive, but suggestive).

Sorting out RUSBase sources looks challenging. Does anybody correspond with Popovsky Alexey? I think I remember reading him thanking people for corrections at the new site.

I only have access to free collections and online sources, other than my Fritz12 database (which I thihnk is a subset of my Mill Base/Enormous/RUSBase collection).

I suspect that the current RUSBase has many corrections over RUSBase 1, and that the incorrect date/score is old, and comes from RUSBase 1. Since it's longer, automatic twin deletion will keep the incorrect version. These are just ideal speculations on my part. Just like I think after 30...Qf5 is a good place for White to resign.

<Geller indeed played a white against Smyslov in a 1953 <Gagra training event> in 1952, but both the score and the result is different than that reported in the <cg.com> game posted on this page.>

It should read

<Geller indeed played a white against Smyslov in a <Gagra training event> in 1952, but both the score and the result is different than that reported in the <cg.com> game posted on this page.>

I also thing the resignation after 30...Qf5 looks logical.

The thing is, we have no way of knowing the source that the original uploader of this game was using.

I'm not sure about your theory about RUSBase vs. RUSBase 1, since I have only been aware of this one here:

http://al20102007.narod.ru/

I've had to do a fair bit of chess database archeology while doing the research on tactical positions that brought me here in the first place.

One site cited extremely often was PITT (the Pittsburgh Chess Archive). It's currently an HTML site:

http://www.pitt.edu/~schach/Archive...

which links to the content on the original ftp site, which is now quite stale (I'm fairly sure):

ftp://ftp.pitt.edu/group/student-activities/ches-
s/PGN/

I would love to get the archived material if anybody has it (am willing to trade RUSBase SCID/PGN in exchange!). I really need the "92 Tal Memorial Quick Play" tournament the most...

Another old, and unfortunately now obsolete, database structure was the RUSBase I, II, III, IV structure. You can find it, still referenced, in the external links at the bottom of this Wiki page:

http://en.wikipedia.org/wiki/Baltic...

The old links are as follows:

RUSBASE (part II) 1970-1984
http://www.webcitation.org/query?ur...

RUSBASE (part III) 1961-1969,1985-1990
http://web.archive.org/web/20091027...

RUSBASE (part IV) 1938-1960
http://www.webcitation.org/query?ur...

RUSBASE (part V) 1919-1937,1991-1994
http://www.webcitation.org/query?ur...

The original RUSBASE site is archived (afaik) here:

Russian Chess Base:
http://www.webcitation.org/query?ur...

The old collections were "collected" together to make the "modern" RUSbase site you cite (and where I've downloaded and collated what I call RUSbase). Again, I'm guessing, since I'm doing this all forensically, from information available on the web currently.

Of course, all the geocities webpages are gone, which means RUSBASE II-V aren't necessarily available. By this, I mean you might get the header page as stripped-down webcitation page, but you can't usefully go any further (i.e. to the download or crosstable pages)

If you want to find citations to sources on the web, however, you must reconcile yourself to RUSBASE II-V references, and hope they've been forwarded into the modern site.

That's been my experience, limited though it may be.

So ends RUSBase vs. RUSBase 1.

Now, I'm a little confused about the last paragraph. You say you "downloaded" the <chessbase> version from <RUSbase=http://al20102007.narod>... site?

What is the distinction between <chessbase> and <narod>? I think of <chessbase> as the commercial product, either online or on DVD. <narod> is a site with lots and lots of zipped pgn's that I put together to form RUSBase.

I suppose you mean that the <chessbase> version matches the <narod> (aka RUSBase) version with certain exceptions?

Anyways, my 1953 Garga games from RUSBase show a 11 man round robin (complete with the exception of 7 games involving Bondarevsky). Here is the list of tournament participants:

URS-training
Garga, 1953
Ti Age Nat
1: Smyslov, Vassily gm 31 RUS 7.5 / 10
2: Boleslavsky, Isaak gm 33 URS 6.5 / 10
3: Petrosian, Tigran V gm 23 URS 6.0 / 9
4: Averbakh, Yuri L gm 31 RUS 4.5 / 9
5: Kotov, Alexander -6 RUS 4.5 / 9
6: Taimanov, Mark E gm 27 RUS 4.5 / 9
7: Geller, Efim P gm 27 RUS 4.5 / 9
8: Keres, Paul gm 36 URS 4.0 / 9
9: Tolush, Alexander V 3.5 / 9
10: Ragozin, Viacheslav gm 44 URS 2.5 / 10
11: Bondarevsky, Igor gm 39 URS 0.0 / 3

48 games: +17 =18 -13

I generated this crosstable by just selecting the Geller-Smyslov 1953 game with SCID, and then edited it down to player/results.

Aha sorry for still more confusions. I think I can clear it up though.

You wrote: <Now, I'm a little confused about the last paragraph. You say you "downloaded" the <chessbase> version from <RUSbase=http://al20102007.narod>... site?>

This is what I actually wrote:

<I did in fact download the <chessbase> version of the game on this page from the http://al20102007.narod.ru/nat_tour... event, which is where I got the PGN that I posted above. As stated before, the PGN is sourced to <chessbase> and gives 1953 as the year.>

If you click on the link, you'll see the same tournament you listed a crosstable for: http://al20102007.narod.ru/nat_tour....

An 11 player all play all tournament.

You'll also see a game dowload link on the page: http://al20102007.narod.ru/nat_tour...

If you download those games, you'll find the PGN for the game that is obviously a version of the game here on <cg.com> page- with a different ending by three moves. I'll repost the PGN here again:

[Event "URS-training"]
[Site "Garga"]
[Date "1953.??.??"]
[Round "4"]
[White "Geller, Efim P"]
[Black "Smyslov, Vassily"]
[Result "0-1"]
[ECO "D18"]
[PlyCount "60"]
[EventDate "1953.??.??"]
[Source "ChessBase"]

1.d4 d5 2.c4 c6 3.Nf3 Nf6 4.Nc3 dxc4 5.a4 Bf5 6.e3 e6 7.Bxc4 Be7 8.O-O O-O 9.Qe2 c5 10.Rd1 Ne4 11.d5 Nxc3 12.bxc3 exd5 13.Bxd5 Qc7 14.e4 Bg4 15.h3 Bh5 16.c4 Nc6 17.Bb2 Rfe8 18.Ra3 Rad8 19.g4 Bg6 20.Re3 Nb4 21.e5 Bc2 22.e6 f6 23.Rd2 Bxa4 24.g5 fxg5 25.Be5 Qb6 26.Nxg5 Nxd5 27.cxd5 Qb1+ 28.Kh2 Bxg5 29.f4 Bh6 30.Rg3 Qf5 0-1

Note that the "Year field" says 1953, and the "Source field" says <Chessbase>.

That's what I meant by "downloaded from <Chessbase>.

I should have said "downloaded from http://al20102007.narod.ru/nat_tour..., which lists <Chessbase> as their source for this game."

###############################

I'm sorry but I did make another egregious error by labeling the <Gagra Training Tournament 1953> as the <GAGRA (USSR) 1953-Training Match Tournament>. They are of course two distinct events:

<Training Match-tournament- Gagra/Voronovo 1952> (A team match tournament): http://al20102007.narod.ru/nat_tour...

and

<Gagra Training Tournament 1953> (11 player all play all): http://al20102007.narod.ru/nat_tour...

At any rate- fascinating stuff!

Although I fear we may now be getting off topic since this game page only concerns one game from one event?