One problem with allowing an engine 12 seconds to look at each position is that if you come up with a brilliant sacrifice that leads to a forced mate quite a few moves down the line, the engine will probably flag that move as a blunder. It is always better to spend a little more time on your analysis just to make sure which moves are blunders and which moves are brilliancies.
Chessbase 10 vs Chess Assistant 10
Sort:
Sep 30, 2009
0
#23
rigamagician wrote:
One problem with allowing an engine 12 seconds to look at each position is that if you come up with a brilliant sacrifice that leads to a forced mate quite a few moves down the line, the engine will probably flag that move as a blunder. It is always better to spend a little more time on your analysis just to make sure which moves are blunders and which moves are brilliancies.
By the way, who's the guy at Your avatar?
Sorry, but I had to come up with that one. As a true Tal-fan I don't believe in correct or incorrect sacrifices but those which work and those that don't
Sep 30, 2009
0
#24
rigamagician wrote:
One problem with allowing an engine 12 seconds to look at each position is that if you come up with a brilliant sacrifice that leads to a forced mate quite a few moves down the line, the engine will probably flag that move as a blunder. It is always better to spend a little more time on your analysis just to make sure which moves are blunders and which moves are brilliancies.
I doubt that very much. Most modern engines - Toga with SCid, new fritz , Rybka do find most of the spectacular combinations in GM games. I have tried this with many of the game of the day puzzle at chessgames.com. The engines need only about 5-10 seconds max. And since I am not anywhere as good as a GM - 12 seconds is adequate..
Note - the auto-annotate is useful when you DONT HAVE much time.
Obviously, I am not going to call my games brilliancies when I haven't spent time on it ...(I do however flag them as such if the engine found some beautiful forcing combination )
I have been an on/off player for a some years now and I just dont dedicate much time to it overall - though in certain periods I feel I am spending too much time on chess:)
At one point i never used a computer at all for chess. It seemed to me more natural to just go over mistakes that can be corrected by the engine rather than compare/analyse opening lines deeply.
As far as sacrifices incorrect/corrrect. I find that in my games I make incorrect sacrifices and win because of the pressure on my opponent (not cc games of course) , but I have found some incredible but correct according to engine combinations that I missed that i wished i had played
Interestingly, there is no chess database software that is compatible with chess960 (fischer random) castling rules.
Except Arena which doesn't work for me, it always crashes when i try castling when playing against engine.
I purchased CA 10 and borrowed CB 10 from a friend.
Here are my observations (I just started out on CB 10, so excuse me for any errors)
Searches
Generally, CA 10's searches are much faster, in fact, way faster than CB 10's. The criterias for searches between the two databases seems to be the same, both have the same criterias the user can choose from. The only difference is that CB has the 'medals' criteria which Ca 10 does not, because it does not have the medal feature at all.
Openings
The most notable defect in CA 10 is that it does not have a 'repertoire' base, which CB 10 have. However, CB 10 does not have 'Opening Tests', a function of CA 10 which I have used on more than a few occasions.
CA 10 comes with Comprehensive Chess Openings 2008 by GM Kalinin. I am not sure if Chessbase Starter package comes with a similar openings manual because I only have CB 10, not CB Starter Package.
Both programs have Opening Report, and Find Novelties feature which I have yet to test.
I am not sure if there is a 'prepare for your opponent' feature in CB, but in CA 10, it was a very detailed report.
Graphics
In general, I like CB 10's board much better than the board in CA 10. You have to hide all the game info to make the CA 10 board bigger, while CB has the game info to the side instead of below the board.
I also found CB's fonts more appealing.
Analysis
I think it is the analysis, as well as the speed of the search which CA 10 is much better in.
There is no game analysis in CB, a feature which I am sure will be useful to many people.
Tree
Looking at the tree alone, both programs to a very good job, although (excuse me for repeating this) CA 10 is once again much faster. However, that must be weighed against the fact that (again) CB has the bigger, and consequently, the more appealing board.
Other
CB 10 can create a tournament crosstable, something which CA 10 cannot do (or something which I have not found out where it is yet).
The fact that every thing which CB opens it opens in a new window is quite annoying, and I much prefer the CA way.
Conclusion
CA 10 is faster, and has more analysis features, while CB 10 have the better graphics. My personal opinion is that 'faster searches' and 'game analysis' outweighs CB 10's better graphics, but that is just my opinion. Perhaps you would much prefer CB 10 if you already have an analysis program such as Fritz.
Dec 28, 2009
0
#26
I've had a little more chance to try out different aspects of ChessAssistant lately, so I now have a greater appreciation for its strengths. One feature I like is that like Fritz, it has a sparring mode where you can limit the strength of any given engine, and play against it. Like Chess Openings Wizard and Aquarium, it has backsolving and batch EPD analysis features for people who want to analyze particular opening lines in depth, a feature that neither Chessbase nor Fritz has.
ChessAssistant downloads new games from their server automatically, which is cool in a way, but the first time it happened I was in the middle of playing a game, and found that I could no longer move. I clicked on "continue in background," but it still wouldn't let me get back to the game until the download was complete. Chessbase has one click download from TWIC, and you can connect to Chessbase's own database of games to search for particular positions or players' games, but I think you decide when to start it up.
One feature I really miss in CA is the lack of opening keys. Some people go on about how you can use classifiers to accomplish something similar, but it seems to me that the name of each classifier has to be fairly short, so it is mainly a way of naming variations, more than a way of building a transpositional repertoire as it is in Chessbase. Chessbase also gives you one click access to statistics on any given position key, something I haven't figured out how to do in ChessAssistant.
I also miss Chessbase's Reference Pane which gives you a list of games stemming from any given position with moves that follow, that you can sort with a single click by player Elo, name, date, result, etc.
ChessAssistant definitely is much cheaper. Chessbase offers an Openings Encyclopedia, but you have to buy that seperately, or else just analyze the games that come with Bigbase on your own.
The prepare for opponent feature is called a "Dossier" in Chessbase, and can be accessed by right-clicking on a player's name in an open database.
Rigamagician,
If memory serves me correctly, there is an option in CA in which you can choose when to automatically update.
I just started to look at CB, so I havent really used the reference pane yet, but it will definately be something I will begin to experiment with.
Thanks for the information on the dossier by the way.
I just use online databases. For software Rybka 3 Aquarium is brilliant if you're prepared to spend. If you're short on money then I highly recommend SCID.
I can't speak for CA, but I do have both Aquarium 2010 and ChessBase 10.
Aquarium is *way* faster than CB10 and has more analysis features, but CB10 is *much* better for managing and using databases.
zxzyz your right, something is better than nothing, but seen blunders will not teach you to play better, that is the problem. For example how you know which thought algorythm of yours failed to spot that blunder and how should you think next time, why this or that line you evaluated has wrong evaluation and etc. I tried another approach, maybe you will like it - i analyse games partially, in each game just one or several segments which i need to understand and learn, while other part of the game i left intact to save time. Even with that i barely have time for analyses.
I disagree. Many of the blunders caught are of the " puzzle type category". That is spot the skewer, fork mate in 4 etc. are essential to be a good chess player. And knowing how its done will help than not knowing at all!
I pay attention to the more tactical situations and see how I went wrong. Because of the auto-annotation I get to a key point in the game where I blundered very quickly later on at my own lesiure. Many times I was playing well until this point when I made a mistake. A few times I have missed some beautiful checkmate opportunities. When I find these I flag these games with the tactics flag. One I am specially interested is a "brilliancy" mainly because of a nice combination ..
Now if I really want to find out what happened in the game i just played and I HAVE TIME. I will quickly run through the game right away with the analysis on, and find this point. All this is for blitz games mind you, I dont study that much and I haven't analysed the slower games I've played here.
The thing is if I play about 10-15 blitz games - a lot of them are probably junk but when i come back to it another time - I find some interesting annotations/score on the games that change my view of many of them.At that point only, I can decide to analyze the game in segments like you suggest. But its really quick to just import into db and run auto-annotate.
I believe if your FIDE Elo is lower than 1900 then there really is no need to study openings in details and being so up to speed with the latest Master games . It just does not make sense to do so if you are going to hang a piece after playing a "perfect" opening.