Lucas et al,
I’ve commented on chess engines on this list before and how and why you should
use them. The purpose of them needs to be kept strictly in mind when you use
them. They are a tool and should only be viewed as such. They are not an
implement to improve chess at a radical level and ought only be there to check
blunder-analysis.
Now, chess databases are a different topic, as they can be used to check
opening theory, opponent information etc. That is very much a different story.
For such things like databases, then I suggest subscribing to such opening
references sites as Chess Publishing and investing into something like
Mega-Base.
For the practical usage of chess engines, you need to find something that is
accessible. In that sense, an engine can be used with a different interface,
which is important. Again, you need to decide what do you want to extract from
the engine.
If you wish to play an engine, then, any chess engine is accessible and
interaction is possible with the majority of screen-readers, such as jaws and
those on the I-phones etc. Win-Board is OK for opponents although you do not
get a high-level of opponent on there.
If you want an engine to help you analyse a position, to check for blunders or
extract data for an opening reference, then, that is different.
First, get a platform that is user-friendly for a VI player. Fritz is currently
the best platform, but the earlier versions. Fritz 9 is the most compatible and
some have argued that Fritz 11 is somewhat so, but takes more time and energy.
Chess-base itself is OK, but not the best and requires some learning.
Once you have that interface, you need to acquire an engine. Most engines are
fine. Personally, I use Hiarcs and you can load an engine into any platform you
decide to use. You must not use the engine that came with the chess application.
Then, once that is installed, you have two ways to treat the analysis. You need
to input the game by using infinite analysis and then save the game, best is in
PGN format.
To check blunders then, I would simply re-play the game and at the points where
you want to check errors or for blunders, hit space-bar and make the engine
play. Using the Jaws cursor is possible and you can review the options and
values of the moves, but this takes time and is quite annoying.
If you want a more comprehensive analysis, you need to give the engine time in
the full-analysis mode, say about 45 minutes or over an hour. Give it a high
threshold of moves to check. Allow it to run that full analysis and once
complete, save that analysed game in a different PGN file.
Once saved, it is then best to open up Windows Explorer and either use Notepad
or Windows WordPad and view the game in that way, using a WP package to slowly
go through it. Which is laborious, but it actually is very effective.
You need to learn all the value numbers, $0 up to $20 and then obscure ones,
but you can Google a list of those very easily.
For opening references, you need to have this checked in the option when you go
through the full-analyse mode.
Again, search the databases with Mega-base etc. will also help. Also, updating
your database with something like The Week In Chess is advised.
Incidentally, I hacked up Mark Crowther, the author of The Week in Chess
disgracefully last weekend with the black bits!
Take it easy
Chris
**********************************************************************************
Chris Ross
BCA website coordinator
Home:
83 Far End,
Northampton
Northamptonshire,
NN5 5FN
UK.
Tel: +44 (0) 1604 750 856
Mobile: +44 (0) 7786 064 537
email: <mailto:c.ross@xxxxxxxxxxxxxxxxxxx> c.ross@xxxxxxxxxxxxxxxxxxx
work:
Chris Ross
Inclusive Learning and Teaching Officer,
King’s Learning Institute,
Students and Education Directorate,
Room 5.19 Waterloo Bridge Wing,
Waterloo Road,
London,
SE1 9NH
UK
tel: +44 (0) 207 848 3081
Email: <mailto:chris.ross@xxxxxxxxx> chris.ross@xxxxxxxxx
*********************************************************************************
From: usbca_chess-bounce@xxxxxxxxxxxxx
[mailto:usbca_chess-bounce@xxxxxxxxxxxxx] On Behalf Of Lucas Radaelli
Sent: 11 February 2016 14:49
To: usbca_chess@xxxxxxxxxxxxx
Subject: [usbca_chess] Re: Annotated game: Milborrow-Ngozi
I am going to take a look on them.
If Chris could comment a little bit how he does the analysis with screen
readers using chess engines that would be nice.
Crafty works kinda well in terminal mode, and it is what I have been doing so
far, but probably others are doing it differently.
On 11/02/2016 12:40, mordue andrew wrote:
Lucas et al.
The engines that I use for analysis are Deep Rybka, Fritz and ChessMaster 9000.
Basically I set them for 'infinite analysis' when looking at a game and the
window for this is usually towards the bottom on the page. However, this may
vary according to which engine you use. At the end of each horizontal line of
analysis there is an evaluation, say +1.67 or -0.82, and the higher the figure
the more it is in White's favour. 0.00 is equal and minus numbers are in
Black's favour. Anything near plus or minus one is roughly the worth of an
extra Pawn. Anything in excess of plus or minus five (the value of an extra
Rook) and the computer wonders why you (or someone) are playing on!
I'm not technically savvy with VH software as I'm fully-sighted. If you need
advice that's more particular than this please check with other group members
especially Chris Ross.
Crafty is an engine that I'm unfamiliar with so I can't comment. However, in
Milborrow-Ngozi Black leaves theory with 3...Bd6. It's a poor move and has not
been played in master chess according to my databases. Chess engines can prove
it is a poor move but there will be no game references to find.
Regards,
Tyson
On 11 February 2016 at 12:39 Lucas Radaelli <mailto:lucasradaelli@xxxxxxxxx>
<lucasradaelli@xxxxxxxxx> wrote:
Thanks for this commented game, I will reproduce it in my board as soon as I
get home.
Has anyone written something about using chess engines to analyze games?
You mention several times possible suggestions made by the chess engines, could
you point to links of the ones that you are using? Do you use them with
winboard or via the terminal?
I tested here crafty, which was interesting to get next moves and moves that
were part of opening books, but the analyze mode kept spitting possible moves
while he calculate stuff and was a little bit hard to follow the notation.
On 10/02/2016 18:26, mordue andrew wrote:
Hello all.
Another annotated game for you. The critical point occurs at Black's 23rd. Have
a close look and decide what you would play for Black. I'll tell you now that
Black chose the very obvious move and regretted it. You'll have a treat if you
stop and work out what happened afterwards without looking at the game first!
Best wishes,
Tyson
(6) Milborrow,A – Ngozi,K [C65]
22.10.2015
[Mordue]
1.e4 e5
2.Nf3 Nc6
3.Bb5 Bd6?!
[This is a clumsy move and is not theoretically recommended. The Bishop does
protect the e5 Pawn but otherwise it is poorly placed, has little scope and
obstructs the d7 Pawn . Moreover it makes a statement that the player lacks
opening knowledge. Viable moves are 3...a6, 3... Bc5, 3...Nd4, 3...Nge7, 3...f5
, 3...g6 and the currently fashionable 3...Nf6, the Berlin Defence.]
4.0–0 Nf6
5.Bxc6
[I wouldn't rush to play this move even though I do play the Exchange Variation
with 3...a6 4 Bxc6. The immediate 5 d4 threatens 6 Bxc6 followed by 7 dxe5
forking two pieces, another reason why 3...Bd6 is a poor move. However, in
reply to the text Black does better to go 5...dxc6 opening lines for his Queen
and other Bishop. By answering 5...bxc6, even though recapturing towards the
centre is strategically recommended, Black remains with a cramped position and
his Queen's Bishop has no prospects. The difference is enough for the computers
to evaluate the position after 5...dxc6 as +0.05 and after 5...bxc6 as +0.50!]
5...bxc6?!
6.d4 Qe7
[The alternative is 6...Nxe4 but after either 7 Qd3 f5 (not 7...Nf6 8 dxe5 with
the afore-mentioned fork) 8 dxe5 White is doing well. Note that 7 Qd3 controls
a6 preventing Black from developing his Bishop there. Also 7 Re1 is possible
but after 7...f5 8 dxe5 Bc5 White has the attack on f2 to sort out. and he may
lose the initiative while he does so.]
7.Bg5
[The engines prefer 7 Re1 and 7 Nbd2 but the text adds more pressure on to the
opponent. Indeed it promptly produces an error. Another alternative is 7 dxe5
Bxe5 8 Nxe5 Qxe5 when White could try 9 f4 pushing the Queen around - 9...Qxe4
loses to 10 Re1 – but 9...Qc5+ followed by 10...Nxe4 may be okay for Black.
Hence the preliminary 9 Nc3 before 10 f4 bearing mind that Black can't take on
e4 because of the pin with Re1.]
7...Qe6?!
[An awkward looking move. I can understand the desire to unpin, and indeed I
heartily endorse unpinning as soon as reasonably possible. However, I don't
think Black has anything specific to worry about on the h4-d8 diagonal
particularly as he has a Pawn on c6 preventing Nc3-d5.
Instead Black would do better to get on with his development. I suggest 7...0–0
and Deep Rybka mentions both 7...Ba6 and 7...Rb8. I like the latter because it
hits the b2 Pawn while the former allows immediate Queenside castling. Whether
Black wants to go Queenside with his King is up for debate, but I will add that
in the game Black's Queen's Rook and Bishop make precisely one move between
them – the losing move!]
8.Nc3 0–0
9.Qd2
[I prefer the line mentioned previously, namely 9 fxe5 Bxe5 10 Nxe5 Qxe5 11 f4
and if Black goes 11...Qc5+ 12 Kh1 he has to concern himself about 13 Bxf6. If
Black had played 5...dxc6 then he would have 12...Ng4 threatening 13...Nf2+ in
this variation and with some activity.
However, there is nothing wrong with the text move. Maybe White could consider
going a square further with 9 Qd3 but after 9...a5 Black threatens 10...Ba6
skewering.]
9...Re8
[Previously Black playing ...exd4 and taking the next move on e4 resulted in
losing material after White's Re1. Here 9...exd4 10 Nxd4 Nxe4?? leaves both
Queens attacked but after 11 Qe3 Black is losing a piece.]
10.Rfe1 Qg4?
[It rarely pays to wander with your Queen when you haven't completed
development. I can't even guess what Kwesi 's idea was here. As a possible
refutation I pointed out 11 h3 Qh5 12 g4 Qg6?? 13 Nh4 trapping the Queen. Of
course Black has 12...Qxh3 but my point is that White can gain useful tempi on
the errant Queen in some lines.
Now Andrew begins a series of obvious exchanges that leave White with a space
advantage and active pieces, and leave Black with weak Pawns, exposed major
pieces and a draughty King. A high price to pay for one poor move.]
11.dxe5 Bxe5
12.Nxe5 Rxe5
13.Bxf6
[Here 13 Bf4 is a simple skewer winning the c7 Pawn, but I'd play the text as
well. Extra Pawns are nice but exposed Kings create psychological pressure for
the defender. Also after 13 Bf4 Re8 14 Bxc7 Black has no doubled Pawns but
after the text he has two pairs of them!]
13...gxf6
14.f4
[Harassing the enemy Rook and ruling out ...Re5-g5 as a defence to the
forthcoming Rook lift. Black's Rook has a choice; retreat down the e-file or go
left or right and risk being out of play. Kwesi chooses to go left but his Rook
seems to be exposed here as well.]
14...Rh5
15.Re3 Qe6
[Losing two tempi but better than 15...Qxf4?? 16 Rg3+or 15...Qh4?? 16 Rh3 Qg4
17 Rg3 which both lose the Queen. Black can move his King first but delays the
decision until he has some idea of which direction is preferable.]
16.Qe2 Rh6?
[Deep Rybka prefers this to 16...Rh4 but to me it seems better to control g4.
The text allows Andrew to easily penetrate down the g-file.]
17.f5 Qe5
18.Qg4+ Kf8
[Black can play 18...Kh8 because he can defend the mate threat on g7 after 19
Rg3 with 19...Qc5+ 20 Kh1 Qf8. However, he remains very passive and it's
difficult to see how the h6 Rook ever gets back into play.]
19.Rg3 Ke7
20.Qg7
[A bit precipitate. 20 Rd1 first is probably sensible.]
20...Qd4+
21.Kh1 Rh4
22.Rd1 Qf2?
[This psuedo-active move loses control of lots of important squares. Better
were 22...Qc5 or 22...Qe5. The latter begs the question whether the Queen
should have gone to d4 in the first place.]
23.Qg8?!
[White should play 23 e5! If 23...Qxf5 24 exf6+ Qxf6 25 Re3+ wins the Queen,
while 23...fxe5 24 Qxe5+ Kf8 25 f6 threatening 26 Qe7 mate is even worse.
Andrew's actual choice has a point though as he threatens 24 Rg7.
Time for a deep think before reading on.
Now Kwesi could justify his previous play in one of two ways. Firstly, 23...Qf4
to meet 24 Rg7?? with 24...Qxh2 mate. Secondly, the spectacular 23...Ba6
because 24 Qxa8 is met by Qxg3. After 25 h3 Black has an immediate draw by
perpetual check with 25...Rxh3+ 26 gxh3 Qxh3+ 27 Kg1 Qg3+ when the White King
can't go to f1 because of the Bishop on a6!
Now the latter line involves a little bit of imagination but clearly Black
posted his Queen on f2 hoping to take advantage of the pinned h2 Pawn. So by
the same token he should be asking himself "Why has my opponent allowed me to
play 23...Bb7 with a free tempo gain on his Queen?" Once you find the answer
you realise that 1) you should have trusted your opponent and 2) it was worth
doing deeper analysis and discovering the vital difference between 23...Bb7 and
23...Ba6.]
23...Bb7??
[If you still haven't worked out what's going on here yet then stop and start
analysing right now!]
24.Rxd7+!!
[It's a forced mate in ten moves including this move].
24...Kxd7
25.Qxf7+ Kd6
[The vital difference between 23...Bb7 and 23...Ba6 is that in the game after
25...Kc8 there's no flight square on b7 so 26 Rg8 is immediate mate. Now the
King has to run the gauntlet and no Black piece contributes to the defence.]
26.Qxf6+ Kc5
[Going backwards just runs into mate in two after 27 Rg7+.]
27.Qe5+ Kb6
[A bit of calculation is needed here. If 27...Kb4 28 a3+ Kc4 29 b3 mate.
Alternatively if 27...Kc4 28 b3+ Kb4 29 a3+! Kxa3 30 Qa5+ Kb2 31 Qa2+ Kc1 32
Qb1+ Kd2 33 Qd1 mate. It would have been wonderful to see this!]
28.Na4+ Ka6
29.Nc5+ Kb6
[If 29...Qxc5 30 Qxc5 - the threat is 31 Ra3 mate – 30...Bc8 31 Qxc6+ Ka5 32
Ra3+ Kb4 33 Ra4 mate. Or 29...Ka5 30 Qc3+ Kb5 31 a4+ Kb6 32 Qb4 mate. Finally
29...Kb5 runs in the same reply as in the game.]
30.Rb3+ Ka5
31.Nd3+
[Better moves were 31 Qxc7 mate and 31 Qc3 mate. White is still mating shortly
after the text but at this point Black threw in the towel.]
1–0
Text ends.
