I can't really remember what it was like to play at 800, but at my current 1200 I'd say that knowing a little bit about the endgame (I read Silman's and Horowitz's books) has probably rescued more games for me than anything else.

As a novice in chess, I find that I have been enjoying the end game workouts designed for beginners.  I keep going through them and hopefully the lessons may actually sink in one of these games!    The bishop endings are very tricky for me at this time, but I hope to get a handle on these things soon.  Learning tactics is great, but it is nice to see things applied in end game situations.  The chess workouts I have used to this point are very educational, and do not seem to be too advanced for a new guy such as myself. 

I find that some of the computer workout beyond beginner level a little frustrating. For example I was playing against the one linked below, which I beat after about 10 goes but then struggled to beat again. I checked 'best move' which only gave a draw. I've since tried to beat it again and failed. It would be useful if the 'best move' actually showed to how to do it so you could check how to beat it and then learn the strategy behind it, instead I'm left with dancing kings.

http://www.chess.com/computer-workout/server/n-v-b-up-a-pawn

I ran the above position through and old Fritz engine and Houdini and they managed to convert a pawn after about 50 moves. Is there something obvious I am missing with this 'intermediate' position? I would love to crack the technique behind this one.

Obviously you have an extra pawn and a good knight Vs a poor bishop, but there seems no clear logica besides dancing your pieces around until black makes a poor move. Yes white has a nice square in c5 which ties down the bishop to defensive duties but even with this is not easy, or at least I am missing something. All the computer wins in this position have been completely different, so if there is a stronger player out there (or just one better at these sort of end games than me) I would love to here your thoughts on it.

Thanks.

http://www.chess.com/computer-workout/server/n-v-b-up-a-pawn

Yes but this position is driving me crazy.

If I remember right, trading the Knight for the Bishop leads to winning fairly quickly.

Yeah it does seem lame, but it was really driving me crazy too.

OK then, I am being dumb with it yes, a pawn up and a good knight for a poor bishop, but why does the computer best move only lead to a draw and it takes Houdini 50 moves or so to convert the pawn? It can't be just me. I must have done this one about 30 times and won once.

OK, this is one of the quickest wins from this position played by Houdini, converting a pawn in just 47 moves!! I've played it several times with different engines on different computers and the pawn promotion averages about 50 moves from the given 'intermediate' 'easy' ending. I must have played this position myself about 40 times and won only twice. If there is an easier win from this position that both myself and Houdini is missing I'd love to hear about it.

hmm good advice

wen is d rite tym fr opening learning?

OK then. I'll just keep trying with this one, probably the best way to learn it anyway. I have beat it but more by luck than skill. I need to be able to win positions like this one first time, every time. This is one thing absent I find from a lot of chess material for intermediates - winning 'won' games, converting the extra pawn etc. There's plenty of stuff on tactics and strategy, opening, basic endgames, brilliancies; but converting the extra pawn not so much.

Here's a typical attempt:

Oh fool. I should have just swung the knight around to c5 again and the bishop is overloaded with defensive duties. Hey, this thinking aloud works. I'm trying that now.

OK, let's try again:

Argghh.

OK, I'm going to have another go. I hope people don't mind  the double posts because I find this useful and it is therefore possibly useful for other others too. Here goes:

OK another draw.

I need to keep the threat alive on both wings I think. Let's try again.

7.Ke5 would win very easily (#17)

OK, thanks, I'm going to have a look at that suggestion now.

Thanks.