So when do I resign? When I'm down a lot of material and can't see any move I could make that would do anything productive. This means that if I have a king against a king, bishop, rook, and pawn, I'd resign. With those pieces, it's very unlikely that I could even pull off a draw. There's really no reason to make the other player prove he knows how to mate (unless he's actually a beginner and needs to learn).

!! With skills like that, it's no wonder you're a master.

You ask a very relevant question: "What is better, 4.d3 or is the <apparently more aggressive> play of 4.Ng5 best here?"

4.d3, is a simple, safe and non-confrontational move ... and it has been adopted by dozens of masters. (It is probably good for a small edge for White.)

4.Ng5, which some masters have branded as premature, is very radical, and puts a ton of pressure on Black. (It literally forces the second player into a gambit.)

I annotated quite a few games for my column in the Florida chess magazine, whether or not all of them gets used is anyone's guess. One of them was: S Kriventsov vs J Becerra-Rivero, 2006, it was an extremely complicated brawl that ended in a draw of mutual desperation.

This brings us to perhaps the final point between the two moves. after 1.e4, e5; 2.Nf3, Nc6; 3.Bc4, Nf6; 4.Ng5, these lines are DEEPLY analyzed ... entire forests have been sacrificed for the variations that now occur. Memory plays a huge role here, (one slip - for either side - is fatal!); and maybe that is why few players play either side of these lines nowadays - this continuation is mostly seen in correspondence chess.

Did I sufficiently answer your question, or was I obtuse?

I meant to post this earlier, but you were curious about the lines of 4.Ng5, and how they compare to the slower 4.d3, (of this game). Check out Estrin vs Berliner, 1965, for one example of how wild the play gets after White launches his Knight to g5.

More than one White player (including me), has stuck his Knight on g5 ... and then went astray in the ensuing complications. I would recommend playing through a few lines in MCO-14, if you have it, to get a feel for these tactical variations.

And just so you don't get the idea that Black wins all of the lines with 4.Ng5, check out Radjabov vs Naiditsch, 2005. Rady's approach is very logical, he solves the problem of his QB by an aggressive fianchetto. (Then after completing his development, White takes over the game, and gets both of his Knights to fantastic squares.)

And one other note. To see how much "book" knowledge White must have, the first player must be prepared to meet dozens of wild lines when you venture into the 4.Ng5 territory. (Check out the line with 1.e4 e5; 2.Nf3 Nc6; 3.Bc4 Nf6; 4.Ng5 Bc5; see the systems in the Opening Explorer. One game in this line would be Shabalov vs Igor Ivanov, 1996; a sharp draw.)