Software
rollem411 said:
Anyone use Blackjack Pro 2021? I only tried out the demo, but it looks like the perfect program to practice card counting techniques. It's a lot faster than myself dealing and handling everything too.
http://www.123-free-download.com/download/cards-casino-lottery/blackjack-pro-2021/24246.html
Blackjack 2021 claims to calculate exact EVs, but it doesn't.
You might want to download (Dead link: http://www.bjstrat.net/cdca_demo_download.htm) _ the single deck demo of my program, cdca_ for reference. You will be able to compute the overall composition dependent EVs for various rules for any single deck composition. You will be able to compute the the overall EVs given player's first card is any card and also the EVs given player's first card versus any dealer up card. You will be able to compute the EVs for any given player hand versus all dealer up cards for any single deck composition. If you have Blackjack Attack 3, check the full shoe program EVs with the comp dependent EVs in BJA3 and I think you'll find agreement except possibly some minor differences for some multiple splits. I do not own BJA3, but the EVs in it were furnished by Cacarulo and his EVs are virtually always right on.
One place the EVs won't agree is for up cards of ten or ace. The reason for this is that my program computes EVs unconditionally assuming player loses to dealer's blackjack, whereas BJA3's EVs will be conditional, assuming dealer has checked for blackjack and doesn't have it. There is a simple mathematical formula to convert unconditional to conditional. I am working on a newer, total dependent program and in it the user has the option the display EVs either conditionally or unconditionally. I should have provided this option in the comp dependent program and I didn't. This option would be relatively easy to add, but in the meantime I'd be happy to furnish the conversion formula. The correct strategy can be seen, though, using either conditional or unconditional expected values.
You could use the program as a practice program, but the user interface would be a bit rough. Let's say you were using an online game as the source of hands. Let's say your hand is 10-2 and dealer's up card is 9. You would input 10-2 for player and click "Compute" to get EVs for 10-2 vs all dealer up cards. Now let's say player hits and draws a 4. You would add a 4 to player's hand and again click "Compute" to get EVs for 10-2-4 vs all dealer up cards. Finally player draws a 2, clicks "Compute," and finds it is best to stand. Dealer draws 5, then 3 for 17. At that point you would manually need to remove 10,2,4,2,9,5,3 and reset player by clicking appropriate buttons. At this point, you could click "Compute" with no player cards at all input and get the overall comp dependent expected value for all possible hands prior to the next deal. Splits would get messy because you would have to somehow keep track of all cards drawn, probably writing them down, so they could be input at the conclusion of the hand. After each hand the process would be repeated.
I wrote cdca mainly as a reference program, not a practice program. I am thinking of releasing a version of it that computes any number of decks, for a nominal price if there is any interest. My newer program has the user interface to be used versus an online game with manual input or it can deal randomly. In either mode it can compute pre-deal overall total dependent EVs as well as pre-draw EVs for a player hand. The default is to compute EVs using full shoe basic strategy, but the program is capable of computing any total dependent player strategy for any player hand, allowing the user to manually override basic strategy. It displays card images, wheras the comp dependent program doesn't. It also dispalys up to 2 counting systems at a time and the counting systems can be swapped with any other at any time. I am looking for a venue for it and would consider offering through this site if Ken is interested.
One other thing. I have tried my newer program on Windows Vista and it works OK. The comp dependent program works on Windows XP if you follow the download site instructions. It probably won't work on Windows Vista without adding some system support files. The program is written in Visual C++ using Microsoft Visual Studio 6 and Windows Vista doesn't support the C++ version of Visual Studio 6.
I would appreciate any comments or feedback. I have spent a lot of time writing these programs mainly to satisfy some of my own curiosity about blackjack. I know it's hard to believe, but these programs are exact calculations and not sims. When I first started, I couldn't ever imagine that I would have progressed this far.
k_c
