I need to explore further but it's not in <RUSbase>, <MillBase> or even <ICOfY>.

I found it here and also here:

http://alawibaba.com/chess/games/ta...

The knockout tournament had slipped my mind, this was another example game from Lombardy - so I was thinking it was a one-off. If I was only a little bit sharper!

Anywho - I found the 42-move version on <ChessBase> online, which has all four <Tal // Gufeld> games under URS Cup 1970.

Plus, looking at <NIC> online for Tal w White against Gufeld there are 3 games from 1970.

<
PU 11.10 B09 Mikhail Tal Eduard Gufeld Dniepropetrovsk 1970 1-0

PU 11.10 B09 Mikhail Tal Eduard Gufeld Dnipropetrovsk 1970 1-0

KF 5.4 B06 Mikhail Tal Eduard Gufeld Dniepropetrovsk 1970 ½-½>

They don't tell you that one version of Tal's win has 42 moves (ending w 42.Ke3), while the other has 30 moves (ending with 30.Ne1).

Which is more likely correct?

It's also a little funny that the location is spelled slightly differently - given the first column suggests the same source.

Which is more likely? I think the 42 move version, with Gufeld bashing out moves to get past the time control. Kotov then truncates the game for Informator, because White is a piece up and the rest is embarrassing or not interesting. Chessbase has a different source for the game.

I'm afraid it's just speculative though, as I have no idea what sources any of these people had.

I recall you telling me exactly this before. It's just so unfamiliar a notation that it takes me a few exposures to internalize.

But it's the rarely used last column where NIC gives the source reference, not the first column. I remember now.

I agree with all your points - NICbase is quite useful, but still requires a certain level of care and feeding (like all db's).

I tend to agree with your assessment on the game too, but I haven't had time to replay it yet. It really would be a somewhat remarkable that Gufeld would continue against Tal a piece down, but "clock fever" may have been responsible.