How much longer do counters realistically have?
Quote: AxelWolfI don't know how it's even possible for software to do something it isn't programmed to do. If it's learning, didn't someone or something programed it to do that?Quote: DRich
if any one of you can show me any computer software that does anything that it wasn't programmed to do,
link to original post
Yes, that is exactly my premise. Computers follow instructions, the programmer gives the software those instructions and it follows them.
Quote: DRichQuote: AxelWolfI don't know how it's even possible for software to do something it isn't programmed to do. If it's learning, didn't someone or something programed it to do that?Quote: DRich
if any one of you can show me any computer software that does anything that it wasn't programmed to do,
link to original post
Yes, that is exactly my premise. Computers follow instructions, the programmer gives the software those instructions and it follows them.
link to original post
And my premise is that AI has a huge hype about it. The latest AI. Ask yourself why? It's because it is a learning program.
And ask yourself what the point of programming and AI to learn what to do is if it only does what the programmer already told it to do is?
We actually wouldn't be discussing AI if all it did was a programmer told it 1+1=3. What actually is programmed is the the AI is told to scour the internet to understand what 1+1 equals and then it gathers false data and comes up with an answer it was never programmed to come up with.
If the AI comes up with answers it wasn't programmed for then the AI has not followed instructions properly. Yes, it followed instructions but it didn't follow them properly.
That's my premise.
These incorrect outcomes are so prevalent they have been named "hallucinations". And they are prevalent not because thousands of computer programmers don't know their jobs but because the nature of instructing an AI to learn what to do involves its accessibility to all data available... including false data... And an inability to differentiate what's correct and what's incorrect.
There is a video of an automated car driving on the highway that suddenly acts erratically and causes a crash. I don't think anyone would argue the programmers programmed it to crash.
Yes something went WRONG in it's programming but it certainly wasn't programmed to crash. In fact quite the opposite, it was almost certainly programmed not to crash.
Quote: darkoz
And ask yourself what the point of programming and AI to learn what to do is if it only does what the programmer already told it to do is?
In my opinion the benefit is just speed. A human could dig through the same data following the same steps but the computer is millions of times faster. It isn't doing anything a person couldn't do, it is doing it faster and with less errors.
Will it be useful to see right through sunglasses and track the movement of your eyes, dilation of pupils, microexpressions of your face. Maybe track whether your hands are warm or cold. (Your real hands not your cards).
When people talk about intuitively picking up cues from other people, I think maybe we all are accessing different senses. In the same way, computers will probably be able to use such things. Maybe they will be able to identify people even from the way they move.
I don’t know if this will eventually figure into catching cheats or APs but it may even catch people before they even reach the first table or machine.
The future!
Quote: rxwineEventually there will be better “vision” and “hearing” and physical “sensation”. (Sensory input) for computers. They can already see measure, hear and touch to a certain extent.
Will it be useful to see right through sunglasses and track the movement of your eyes, dilation of pupils, microexpressions of your face. Maybe track whether your hands are warm or cold. (Your real hands not your cards).
The problem is that these inputs are extremely weakly correlated with any specific behavior. Body language experts don't even claim to be able to tell if someone is lying more than about 55% of the time. If you see someone analyze a micro-expression on TV usually they have the advantage of hindsight ("The serial killer reveals his arrogant disdain for the interviewer momentarily...). That's very different from obtaining useful information going forward.
With AP's you are trying to separate people who have an edge with people who think they have an edge. That's way beyond any kind of tech we can even conceive of. The guy using a D'alembert system is accessing much the same parts of his brain as a card counter. Even with an MRI scan I doubt you could tell the difference.
I very much doubt it would ever be worthwhile for a casino to invest this even if costs are substantially brought down.
Quote: DRichQuote: darkoz
And ask yourself what the point of programming and AI to learn what to do is if it only does what the programmer already told it to do is?
In my opinion the benefit is just speed. A human could dig through the same data following the same steps but the computer is millions of times faster. It isn't doing anything a person couldn't do, it is doing it faster and with less errors.
link to original post
I suppose I should be clear I am working with Generative AI and what it can do is most certainly things no human could do.
It is doing a lot more than glorified research.
I highly doubt that.Quote: billryanI expect most casino table games will disappear in the next twenty or so years. Perhaps slots, as well as casinos, switch to internet games. People will no longer have to gamble in casinos.
link to original post
Online gambling has been around since the 90's and most people don't have to go into casinos if they don't want to.
People like going to casinos for many different reasons.
Quote: Archvaldor1Quote: rxwineEventually there will be better “vision” and “hearing” and physical “sensation”. (Sensory input) for computers. They can already see measure, hear and touch to a certain extent.
Will it be useful to see right through sunglasses and track the movement of your eyes, dilation of pupils, microexpressions of your face. Maybe track whether your hands are warm or cold. (Your real hands not your cards).
The problem is that these inputs are extremely weakly correlated with any specific behavior. Body language experts don't even claim to be able to tell if someone is lying more than about 55% of the time. If you see someone analyze a micro-expression on TV usually they have the advantage of hindsight ("The serial killer reveals his arrogant disdain for the interviewer momentarily...). That's very different from obtaining useful information going forward.
With AP's you are trying to separate people who have an edge with people who think they have an edge. That's way beyond any kind of tech we can even conceive of. The guy using a D'embert system is accessing much the same parts of his brain as a card counter. Even with an MRI scan I doubt you could tell the difference.
I very much doubt it would ever be worthwhile for a casino to invest this even if costs are substantially brought down.
link to original post
OH. I would also argue weakly correlated events aren't worth much, but a collection of them can paint a more accurate picture, like clues for a detective. As to cost, well casinos have deep pockets. The same software could be useful to large corporations where such things might be needed with tweaking for particular usage. Alos a way to cut cost is a rental system. I don't have a money to buy a Rolls, but I should have enough to rent one for awhile. Software doesn't really wear out, just gets older and less useful usually.
Quote: rxwine
OH. I would also argue weakly correlated events aren't worth much, but a collection of them can paint a more accurate picture, like clues for a detective. As to cost, well casinos have deep pockets. The same software could be useful to large corporations where such things might be needed with tweaking for particular usage. Alos a way to cut cost is a rental system. I don't have a money to buy a Rolls, but I should have enough to rent one for awhile. Software doesn't really wear out, just gets older and less useful usually.
link to original post
That's a very shrewd observation. The compound probability of a series of inferences does add up. There's some really exciting possibilities there, though outside of national security I have seen little development in that area. I was sort of hoping someone would come up with an intelligent counter to my dull scepticism.
That said 55% is the upper boundary kinesic-types seem to be placing on the total box of tricks they possess. Given that most of detecting a counter is simply working out how much money is bet into positive counts the gains from visual/biological analysis are going to be incremental at best.
Quote: DRichI definitely wasn't trolling, if any one of you can show me any computer software that does anything that it wasn't programmed to do, I may acquiesce. Yes, we will need to see the source code to prove it.Quote: darkoz
I did feel DRich might be trolling but he has never seemed that sort and I didn't want to make the accusation. I'm glad now the thought wasn't completely out of left field.
Thanks again for your expert analysis.
link to original post
link to original post
FWIW, I agree with darkoz in that you never struck me as a trolling type so while a possibility mentioned for the sake of completeness, it didn't seem to me the most likely reason for the difference of opinion. That being said....
Back in the 90s I was having lunch with a bunch of colleagues and I got into a debate/argument with one about the pros and cons of the top-down (a.k.a waterfall) design methodology vs. the iterative approach. I was praising waterfall while my friend Mike was all-in for iterative. Now I had a lot of respect for Mike (and I assume vice versa) so we both really wanted to figure out why we didn't see eye-to-eye. We eventually realized there were a bunch of unstated assumptions (aka 'biases') we both had due to the nature of the projects we had worked on in the past. I had worked on a lot of large-scale distributed mission-critical control systems where it had to work right the 1st time or bad things happened. Mike's background was heavy on sensor systems where build-test-repeat was feasible.
My point here is that I suspect one reason you and I are viewing this issue differently is due to differences in our programming backgrounds. So before debating this point further it might help clarify our past experiences.. Mine is about 30% Assembler, 30% C, 30% Java, and the rest mostly Perl, Python, Fortran, and R. I would also say about 50% of my code had to deal to some degree with event-driven, near-realtime stuff (i.e., ASTs, interrupt handling, etc.) and very unreliable computing/comms environments (e.g., tactical networks). I would be interested in hearing where you are coming from.
Quote: Archvaldor1Yeah, it won't.Quote: TumblingBones
Quote: Archvaldor11. You can detect card counting with statistical analysis. The tech has existed for years. It is nothing to do with AI.
So what? I can hike from NYC to WDC. The necessary tech (i.e., shoes) has been around for millennia. But taking AMTRAK is faster and more comfortable. Bottom line is whether or not supplementing statistics with "AI" will reduce false negatives/positives and/or save money
Let me respond point by point...
Quote: Archvaldor1The vast majority of counters are detectable through bet variation. You look at how much money they are betting into positive counts. It shows up very quickly even with deep cover. Random players do not bet more into positive counts. They do not know what the count is. You can write ten lines of code which separate regular players from counters within a short timeframe.
I'm in partial agreement. A lone counter is probably an easy target to identify. But what about teams? Also, while the original question was specific to blackjack and counters, I'm looking at the potential for using AI to target other types of AP play as well (e.g., slot vultures).
Quote: Archvaldor1Maybe a hypothetical AI could make inferences about a player's age, ethnicity, could assess the expectation gained from playing departures or decision-making frequency etc. None can do this currently with any degree of accuracy, If they could then you make it a lot easier for card counters who work out how the program works to evade detection. For what? Detecting a counter within 172 hands rather than 161? It isn't worth it and it isn't ever going to be.
Two questions:
First, re the accuracy statement, can you be more specific as to what you consider acceptable versus what you consider to be currently feasible? Second, I don't follow your logic when you state " If they could then you make it a lot easier for card counters who work out how the program works to evade detection.". Can you clarify?
Quote: Archvaldor1The subtext I get from what you are saying is "AI has super-powers which work in some vague and non-specific way". For you to code something effective you need to be able to break it down into actionable steps and you haven't really done that yet so I'm very sceptical.
link to original post
I can be less vague and much more specific. Before going down that path, however, I would want to know how familiar you are with things like knowledge graphs, clustering algorithms, digital twins, and similarity metrics. It's not required but it would save time.
Quote: TumblingBonesQuote: Archvaldor1Yeah, it won't.Quote: TumblingBones
Quote: Archvaldor11. You can detect card counting with statistical analysis. The tech has existed for years. It is nothing to do with AI.
So what? I can hike from NYC to WDC. The necessary tech (i.e., shoes) has been around for millennia. But taking AMTRAK is faster and more comfortable. Bottom line is whether or not supplementing statistics with "AI" will reduce false negatives/positives and/or save money
Let me respond point by point...Quote: Archvaldor1The vast majority of counters are detectable through bet variation. You look at how much money they are betting into positive counts. It shows up very quickly even with deep cover. Random players do not bet more into positive counts. They do not know what the count is. You can write ten lines of code which separate regular players from counters within a short timeframe.
I'm in partial agreement. A lone counter is probably an easy target to identify. But what about teams? Also, while the original question was specific to blackjack and counters, I'm looking at the potential for using AI to target other types of AP play as well (e.g., slot vultures).Quote: Archvaldor1Maybe a hypothetical AI could make inferences about a player's age, ethnicity, could assess the expectation gained from playing departures or decision-making frequency etc. None can do this currently with any degree of accuracy, If they could then you make it a lot easier for card counters who work out how the program works to evade detection. For what? Detecting a counter within 172 hands rather than 161? It isn't worth it and it isn't ever going to be.
Two questions:
First, re the accuracy statement, can you be more specific as to what you consider acceptable versus what you consider to be currently feasible? Second, I don't follow your logic when you state " If they could then you make it a lot easier for card counters who work out how the program works to evade detection.". Can you clarify?Quote: Archvaldor1The subtext I get from what you are saying is "AI has super-powers which work in some vague and non-specific way". For you to code something effective you need to be able to break it down into actionable steps and you haven't really done that yet so I'm very sceptical.
link to original post
I can be less vague and much more specific. Before going down that path, however, I would want to know how familiar you are with things like knowledge graphs, clustering algorithms, digital twins, and similarity metrics. It's not required but it would save time.
link to original post
I appreciate your response, you raise some interesting points.
I am aware you can add complexity to counter detection methods. The point is that in the vast majority of cases you can detect counters relatively easily with a pen and paper, Any gains over and above that are going to be very limited. I don't believe it is worth it implementing AI to catch a counter in a dozen or so less hands or whatever.
I would welcome the introduction of AI into casinos to detect ap's. The reason being that in practice whenever any one has attempted to introduce AI it usually results in lower efficiency, Corporations generally use it to reduce the amount of employees they have.
The one thing that really concerns me as an AP is someone with equivalent skills working against me. That is a real problem. Replacing competent people with AI makes my job easier.
With team play you still have one guy betting big into positive counts. It still isn't that hard to detect. If you mean more advanced methods there is a difficult question of what the AI is programmed to look for. I don't know how you would, say, catch a team that is trying to steer the dealer a bust card. There's an almost infinite variety of such plays.
I don't think we really disagree that much on this, it is more just a difference of perspective over what would actually happen rather than what could AI be used for.
Quote:..there is a difficult question of what the AI is programmed to look for.
Yup, that's always the stumbling block as perfectly illustrated by this example from a recent DARPA experiment. Of course, just because something doesn't work the first time doesn't mean giving up on AI technology. You just have to think outside the box ;)
Quote: TumblingBones
My point here is that I suspect one reason you and I are viewing this issue differently is due to differences in our programming backgrounds. So before debating this point further it might help clarify our past experiences.. Mine is about 30% Assembler, 30% C, 30% Java, and the rest mostly Perl, Python, Fortran, and R. I would also say about 50% of my code had to deal to some degree with event-driven, near-realtime stuff (i.e., ASTs, interrupt handling, etc.) and very unreliable computing/comms environments (e.g., tactical networks). I would be interested in hearing where you are coming from.
I would guess my past was probably 20% FORTRAN, 60% C, and a little of about 10 other languages.
I have no interest in debating credentials, but I will stick to the only point I was trying to make. Show me any computer program that does more than it is instructed to do. Obviously, disregarding compiler bugs which are just doing what they were programmed to do.
Quote: DRichQuote: TumblingBones
My point here is that I suspect one reason you and I are viewing this issue differently is due to differences in our programming backgrounds. So before debating this point further it might help clarify our past experiences.. Mine is about 30% Assembler, 30% C, 30% Java, and the rest mostly Perl, Python, Fortran, and R. I would also say about 50% of my code had to deal to some degree with event-driven, near-realtime stuff (i.e., ASTs, interrupt handling, etc.) and very unreliable computing/comms environments (e.g., tactical networks). I would be interested in hearing where you are coming from.
I would guess my past was probably 20% FORTRAN, 60% C, and a little of about 10 other languages.
I have no interest in debating credentials, but I will stick to the only point I was trying to make. Show me any computer program that does more than it is instructed to do. Obviously, disregarding compiler bugs which are just doing what they were programmed to do.
link to original post
Below is an image I generated. Hands and fingers are notoriously difficult for AI image generators. The below example (note the four fingered hand) is common.
I simply don't believe that the original programmers thought that was the expected output or programmed for four fingered hands.
The article below pretty much reiterates what I have been saying. The AI is trained from the internet on false or misleading data. Which results in the AI giving bad results. Nothing the programmers wanted is coming out in these results.
https://www.buzzfeednews.com/article/pranavdixit/ai-generated-art-hands-fingers-messed-up
Quote: darkozQuote: DRichQuote: TumblingBones
My point here is that I suspect one reason you and I are viewing this issue differently is due to differences in our programming backgrounds. So before debating this point further it might help clarify our past experiences.. Mine is about 30% Assembler, 30% C, 30% Java, and the rest mostly Perl, Python, Fortran, and R. I would also say about 50% of my code had to deal to some degree with event-driven, near-realtime stuff (i.e., ASTs, interrupt handling, etc.) and very unreliable computing/comms environments (e.g., tactical networks). I would be interested in hearing where you are coming from.
I would guess my past was probably 20% FORTRAN, 60% C, and a little of about 10 other languages.
I have no interest in debating credentials, but I will stick to the only point I was trying to make. Show me any computer program that does more than it is instructed to do. Obviously, disregarding compiler bugs which are just doing what they were programmed to do.
link to original post
Below is an image I generated. Hands and fingers are notoriously difficult for AI image generators. The below example (note the four fingered hand) is common.
I simply don't believe that the original programmers thought that was the expected output or programmed for four fingered hands.
The article below pretty much reiterates what I have been saying. The AI is trained from the internet on false or misleading data. Which results in the AI giving bad results. Nothing the programmers wanted is coming out in these results.
https://www.buzzfeednews.com/article/pranavdixit/ai-generated-art-hands-fingers-messed-up
link to original post
Ai been watching too many cartoons.
The programmers say go out and learn what a human hand looks like so it can be replicated correctly.
It finds human hands that are not correctly represented including cartoons and Rob Liefield drawings, etc
Having learned from false or incorrect data, it gives wrong and unintended responses.
Quote: billryanI've been told that AI has trouble with hands because they're rarely the focus of photographs. Human photographers don't center their photos on hands, and the fact that hands are complex, with five fingers having tens of thousands of possible combinations, makes it difficult to draw them properly.
link to original post
Correct. Hands are usually doing something that obstructs them partially such as holding objects, etc.
Quote: darkoz
I simply don't believe that the original programmers thought that was the expected output or programmed for four fingered hands.
I agree that wasn't the intention but without seeing the code we have to assume it did exactly what it was programmed to do. Intentions have nothing to do with how computers work.
Quote: DRichQuote: darkoz
I simply don't believe that the original programmers thought that was the expected output or programmed for four fingered hands.
I agree that wasn't the intention but without seeing the code we have to assume it did exactly what it was programmed to do. Intentions have nothing to do with how computers work.
link to original post
But there is no single source code to look at.
Midjourney creates images. Has problems with hands
Their competitor Stability creates images. Has problems with hands.
Their competitor Magnific creates images. Has problems with hands
Open AI in the USA, Minimax and Kling from China, Google Veo, the list goes on
EVERY IMAGE GENERATOR AI currently available has this problem.
All coming from different source codes generated in a multiplicity of different regions around the globe.
There is a reason why every single one is failing. Because as a learning tool it is gathering the same images from around the internet and the globe and that information collection is filled with false data leading to results not programmed by every single programmer currently in the entire world
EDIT: Openart actually has a dedicated AI called simply enough "Fix Hands" or something similar. The problem is so prevalent there has to be a dedicated AI just for that.
"Fix Fingers" to be specific!
Quote: darkozQuote: DRichQuote: TumblingBones
My point here is that I suspect one reason you and I are viewing this issue differently is due to differences in our programming backgrounds. So before debating this point further it might help clarify our past experiences.. Mine is about 30% Assembler, 30% C, 30% Java, and the rest mostly Perl, Python, Fortran, and R. I would also say about 50% of my code had to deal to some degree with event-driven, near-realtime stuff (i.e., ASTs, interrupt handling, etc.) and very unreliable computing/comms environments (e.g., tactical networks). I would be interested in hearing where you are coming from.
I would guess my past was probably 20% FORTRAN, 60% C, and a little of about 10 other languages.
I have no interest in debating credentials, but I will stick to the only point I was trying to make. Show me any computer program that does more than it is instructed to do. Obviously, disregarding compiler bugs which are just doing what they were programmed to do.
link to original post
Below is an image I generated. Hands and fingers are notoriously difficult for AI image generators. The below example (note the four fingered hand) is common.
I simply don't believe that the original programmers thought that was the expected output or programmed for four fingered hands.
The article below pretty much reiterates what I have been saying. The AI is trained from the internet on false or misleading data. Which results in the AI giving bad results. Nothing the programmers wanted is coming out in these results.
https://www.buzzfeednews.com/article/pranavdixit/ai-generated-art-hands-fingers-messed-up
link to original post
Also note how her left hip looks incomplete, at least at first glance, due to being close in color to the back of the seat. It’s almost as if the fabric in that region is reflecting what’s behind her…
ETA: She looks hot! Can you introduce us? I really don’t mind the disfigurement in her hands. It’s a bit endearing… lol
Quote: camaplQuote: darkozQuote: DRichQuote: TumblingBones
My point here is that I suspect one reason you and I are viewing this issue differently is due to differences in our programming backgrounds. So before debating this point further it might help clarify our past experiences.. Mine is about 30% Assembler, 30% C, 30% Java, and the rest mostly Perl, Python, Fortran, and R. I would also say about 50% of my code had to deal to some degree with event-driven, near-realtime stuff (i.e., ASTs, interrupt handling, etc.) and very unreliable computing/comms environments (e.g., tactical networks). I would be interested in hearing where you are coming from.
I would guess my past was probably 20% FORTRAN, 60% C, and a little of about 10 other languages.
I have no interest in debating credentials, but I will stick to the only point I was trying to make. Show me any computer program that does more than it is instructed to do. Obviously, disregarding compiler bugs which are just doing what they were programmed to do.
link to original post
Below is an image I generated. Hands and fingers are notoriously difficult for AI image generators. The below example (note the four fingered hand) is common.
I simply don't believe that the original programmers thought that was the expected output or programmed for four fingered hands.
The article below pretty much reiterates what I have been saying. The AI is trained from the internet on false or misleading data. Which results in the AI giving bad results. Nothing the programmers wanted is coming out in these results.
https://www.buzzfeednews.com/article/pranavdixit/ai-generated-art-hands-fingers-messed-up
link to original post
Also note how her left hip looks incomplete, at least at first glance, due to being close in color to the back of the seat. It’s almost as if the fabric in that region is reflecting what’s behind her…
ETA: She looks hot! Can you introduce us? I really don’t mind the disfigurement in her hands. It’s a bit endearing… lol
link to original post
Her left hip is indeed incomplete. That's because of this grueling process on this project and I grabbed an image in the middle to show hands as the issue.
The original "live" image is too large for me to upload to WOV.
The original is actually a sketch. see below.
Note the shot involves not just a close up of a woman's waist but another person sticking their hand in with a knife. That arm caused the AI to mis-read and I had to use another AI tool to remove the mis-read, hence the distorted hip. That at least is explainable.
Next I had to generate the arm and knife separately. Note the hand without the knife is not holding a proper image of a tissue either. That also had to get "fixed"
And then, after using AI to "fix fingers" I used another AI to create a Blendboard.
and the final result is...
Since you find her hot, she is an AI creation and I am working on a project with literally hundreds, close to a thousand of these "shots" It's a month's long grueling process. Probably should have its own thread. But here is what she looks like.
Quote: darkoz
But there is no single source code to look at.
I am not sure what you mean by that. If they were running a program there is source code that comprises that program. For example, ChatGPT is a program made up of source code. I am sure it is huge, but there is definitely defined source code that makes ChatGPT what it is.
Most software is made up of multiple source code modules.
Quote: DRichQuote: darkoz
But there is no single source code to look at.
I am not sure what you mean by that. If they were running a program there is source code that comprises that program. For example, ChatGPT is a program made up of source code. I am sure it is huge, but there is definitely defined source code that makes ChatGPT what it is.
Most software is made up of multiple source code modules.
link to original post
I thought I was clear on it but what I am saying is every source code by every company in the entire globe has the same problem.
It's like if suddenly Coca Cola, Pepsi, RC Cola and every soda manufacturer was unable to make properly tasting soda and you asked for the source code.or formula.
The formula for Coke is different than Pepsi which is different from RC Cola. There is no single formula. Yet they all come out exactly the same problem.
You want to examine the source code for Minimax. Sure, it's going to be different than the source code for Kling. Which is also different than the source code for Runway. Which will be different than the proprietary code for Google Veo. I can run down the list you will have NO SINGLE SOURCE CODE but EVERY SOURCE CODE from every company to examine...... And they all have the same problem.
The same!!!!!!!!
They can't do hands properly.
I'm also not certain why you keep insisting we can't look at the source code. Several of the AI codes are open source on GitHub and you can examine them to your heart's content. Like I said above, you will be examining all of them because they all manifest the same problems. They have been examined by many many experts, I am sure
EDIT: This is a billion dollars business. With issues that repeat the same mistakes for EVERY SINGLE COMPANY. Just the concept that every computer programmer able to put together billions in software can't generate proper source code is ridiculous. The reality is the AI gets it's data after programming and that data, especially if it's incomplete or false, causes results not programmed for originally. And no one yet has figured out how to get around it.
Quote: darkoz
I thought I was clear on it but what I am saying is every source code by every company in the entire globe has the same problem.
My only point is that the software that generated that image exists and without being able to see the source code one will never be able to determine if it did what it was instructed to do.
As I have said multiple times, I have never heard of any software doing anything more than what it was programmed to do.
Quote: DRichQuote: darkoz
I thought I was clear on it but what I am saying is every source code by every company in the entire globe has the same problem.
My only point is that the software that generated that image exists and without being able to see the source code one will never be able to determine if it did what it was instructed to do.
As I have said multiple times, I have never heard of any software doing anything more than what it was programmed to do.
link to original post
Lol, you have heard of it
We are discussing it right now
You just refuse to believe it.
Which source code that makes hands imperfect would you like to examine? It's all of them. And at least some of them are open source.
The one I posted above was using Prome.
The one below used Midjourney and the one below that used Dall-E.
Each one has a different source code. But same problem.
Quote: darkozQuote: DRichQuote: darkoz
I thought I was clear on it but what I am saying is every source code by every company in the entire globe has the same problem.
My only point is that the software that generated that image exists and without being able to see the source code one will never be able to determine if it did what it was instructed to do.
As I have said multiple times, I have never heard of any software doing anything more than what it was programmed to do.
link to original post
Lol, you have heard of it
We are discussing it right now
You just refuse to believe it.
Which source code that makes hands imperfect would you like to examine? It's all of them. And at least some of them are open source.
The one I posted above was using Prome.
The one below used Midjourney and the one below that used Dall-E.
Each one has a different source code. But same problem.
link to original post
This will be my last post on this topic.
If you can show me one line of computer software that does something other than what it is instructed to do, I will aquacise. I don't care if it is AI or anything else. just show me one line of code, you can reference the tens of billions of lines of published code or create your own. It does not happen. Software can only do what it is instructed to do.
What will it's programing tell it to do on two different days with the same question? I tried, "what are the physics of a snail climbing a hill." That didn't meet the requirements. First tried motorcycle, but that was worse as far as more links at bottom of answer.
Quote: DRichQuote: darkozQuote: DRichQuote: darkoz
I thought I was clear on it but what I am saying is every source code by every company in the entire globe has the same problem.
My only point is that the software that generated that image exists and without being able to see the source code one will never be able to determine if it did what it was instructed to do.
As I have said multiple times, I have never heard of any software doing anything more than what it was programmed to do.
link to original post
Lol, you have heard of it
We are discussing it right now
You just refuse to believe it.
Which source code that makes hands imperfect would you like to examine? It's all of them. And at least some of them are open source.
The one I posted above was using Prome.
The one below used Midjourney and the one below that used Dall-E.
Each one has a different source code. But same problem.
link to original post
This will be my last post on this topic.
If you can show me one line of computer software that does something other than what it is instructed to do, I will aquacise. I don't care if it is AI or anything else. just show me one line of code, you can reference the tens of billions of lines of published code or create your own. It does not happen. Software can only do what it is instructed to do.
link to original post
LMAO so how would I know you aquiese if this is your last post on the topic?
The fact is there are plenty of AI open source codes for you to choose from that show AI unable to do what it was programmed to do. You refuse to look because you are convinced of your own position.
There are dozens of articles on AI failing to do what it's programmed to do from hands being wrong to "hallucinations ".
No one, not even billion dollar software teams at Google can program correctly is your position. Because if they could then how could there be AI hallucinations? Everything does what it's programmed to do is your position.
So be it.
The back and forth between DRich and myself should really be the way arguments happen here at WOV. Strongly worded but without devolving into taunts and insults.
If any of my posts did do that I apologize. Sometimes I reread my own past posts and am shocked as well.
Whatever disagreement on the matter DRich knows quite a bit on computer software! We can both move on.
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
Quote: SOOPOOThank God. If I had to read another DRich or DarkOz post regurgitating the same point….
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
link to original post
You would have to pay me more than that to watch people play blackjack eight hours a day.
Quote: SOOPOO
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
no
everybody is pretty much missing the point
card counting provides romance to the game of blackjack
the belief is out there that you can win and win big and just keep on winning
the casinos would never want to completely eliminate the possibility of if
that would cause them to lose a great amount of their profits as many would stop playing bj
if they wanted to they could completely end it with continuous shuffling machines
most wannabee card counters will fail for various reasons such as an insufficient bankroll and many will lose big
a few will win but most won't win big - they will be stopped before that
it's a delicate dance - the interactions between casinos and counters
casinos want to keep the dream alive and stop those few who have the ability to win big before they win big
many, many times they know the player is a counter, and they will allow him to play and only take action after his big bets come out
they will allow him to play at their tables for an hour but not for 5 hours
they don't want to completely stop it
.
Quote: lilredroosterQuote: SOOPOO
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
no
everybody is pretty much missing the point
card counting provides romance to the game of blackjack
the belief is out there that you can win and win big and just keep on winning
the casinos would never want to completely eliminate the possibility of if
that would cause them to lose a great amount of their profits as many would stop playing bj
if they wanted to they could completely end it with continuous shuffling machines
most wannabee card counters will fail for various reasons such as an insufficient bankroll and many will lose big
a few will win but most won't win big - they will be stopped before that
it's a delicate dance - the interactions between casinos and counters
casinos want to keep the dream alive and stop those few who have the ability to win big before they win big
many, many times they know the player is a counter, and they will allow him to play and only take action after his big bets come out
they will allow him to play at their tables for an hour but not for 5 hours
but they don't want to completely stop it
.
link to original post
Very interesting take! I wonder if it’s true? I’d doubt it. But we have ZCore who works for the other side, and I guess the Wiz who worked for the evil Venetian…. Maybe they can answer?
who would seriously believe that the casinos could not easily and completely end profitable card counting if they wanted to
it's just not logical to believe that
a great many step up to the bj tables thinking they can beat the game and in reality they cannot - they overestimate their own abilities -
and those are the players the casino wants at their tables
.
Quote: SOOPOOQuote: lilredroosterQuote: SOOPOO
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
no
everybody is pretty much missing the point
card counting provides romance to the game of blackjack
the belief is out there that you can win and win big and just keep on winning
the casinos would never want to completely eliminate the possibility of if
that would cause them to lose a great amount of their profits as many would stop playing bj
if they wanted to they could completely end it with continuous shuffling machines
most wannabee card counters will fail for various reasons such as an insufficient bankroll and many will lose big
a few will win but most won't win big - they will be stopped before that
it's a delicate dance - the interactions between casinos and counters
casinos want to keep the dream alive and stop those few who have the ability to win big before they win big
many, many times they know the player is a counter, and they will allow him to play and only take action after his big bets come out
they will allow him to play at their tables for an hour but not for 5 hours
but they don't want to completely stop it
.
link to original post
Very interesting take! I wonder if it’s true? I’d doubt it. But we have ZCore who works for the other side, and I guess the Wiz who worked for the evil Venetian…. Maybe they can answer?
link to original post
I am aware of casinos that have knowing let card counters play if their bets are small enough. It is not unusual to see a counter learning or practicing betting $5-$25. With that small of a spread many casinos won't bother you.
Quote: lilredrooster.
a great many step up to the bj tables thinking they can beat the game and in reality they cannot - they overestimate their own abilities -
and those are the players the casino wants at their tables
.
link to original post
Many players may overbet their bankrolls and have a marginal edge. However you have to be unrealistically terrible to actually have a negative expectation as a card counter. You'd have to do something like continuously reverse the sign of the count.
Even a bad counter is generally going to be significantly less profitable to a casino than a regular player.
Ballys used to have mirrors on the ceiling so they would put another mirror on the pit stand and the counter catcher could be looking down into the mirror to count the cards on a specific table.
what part of it do you doubt?Quote: SOOPOOQuote: lilredroosterQuote: SOOPOO
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
no
everybody is pretty much missing the point
card counting provides romance to the game of blackjack
the belief is out there that you can win and win big and just keep on winning
the casinos would never want to completely eliminate the possibility of if
that would cause them to lose a great amount of their profits as many would stop playing bj
if they wanted to they could completely end it with continuous shuffling machines
most wannabee card counters will fail for various reasons such as an insufficient bankroll and many will lose big
a few will win but most won't win big - they will be stopped before that
it's a delicate dance - the interactions between casinos and counters
casinos want to keep the dream alive and stop those few who have the ability to win big before they win big
many, many times they know the player is a counter, and they will allow him to play and only take action after his big bets come out
they will allow him to play at their tables for an hour but not for 5 hours
but they don't want to completely stop it
.
link to original post
Very interesting take! I wonder if it’s true? I’d doubt it. But we have ZCore who works for the other side, and I guess the Wiz who worked for the evil Venetian…. Maybe they can answer?
link to original post
I was involved with detecting card counters , what would you like to know
Quote: Hunterhillwhat part of it do you doubt?Quote: SOOPOOQuote: lilredroosterQuote: SOOPOO
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
no
everybody is pretty much missing the point
card counting provides romance to the game of blackjack
the belief is out there that you can win and win big and just keep on winning
the casinos would never want to completely eliminate the possibility of if
that would cause them to lose a great amount of their profits as many would stop playing bj
if they wanted to they could completely end it with continuous shuffling machines
most wannabee card counters will fail for various reasons such as an insufficient bankroll and many will lose big
a few will win but most won't win big - they will be stopped before that
it's a delicate dance - the interactions between casinos and counters
casinos want to keep the dream alive and stop those few who have the ability to win big before they win big
many, many times they know the player is a counter, and they will allow him to play and only take action after his big bets come out
they will allow him to play at their tables for an hour but not for 5 hours
but they don't want to completely stop it
.
link to original post
Very interesting take! I wonder if it’s true? I’d doubt it. But we have ZCore who works for the other side, and I guess the Wiz who worked for the evil Venetian…. Maybe they can answer?
link to original post
I was involved with detecting card counters , what would you like to know
link to original post
Just I think it would be quite easy to identify card counters with not a lot of work. Why isn’t it done more often?
You are correct it’s very easy to identify most card counters. The place where I worked initially accepted my determination if someone was counting or not . Later on when management changed they wanted documentation from the program that they started to use.Quote: SOOPOOQuote: Hunterhillwhat part of it do you doubt?Quote: SOOPOOQuote: lilredroosterQuote: SOOPOO
Anyway…. card counting…. I’m actually surprised casinos don’t use a ‘guy’ to stop them. Hire me or Wiz or Axel or (pretty much anyone here) and just have us watch a table for a while. It won’t be hard to identify the counters. Exactly how much evidence you’d want would be up to management. Pay me $1000 for one 8 hour shift and you don’t think I can save the casino some large multiple of that?
no
everybody is pretty much missing the point
card counting provides romance to the game of blackjack
the belief is out there that you can win and win big and just keep on winning
the casinos would never want to completely eliminate the possibility of if
that would cause them to lose a great amount of their profits as many would stop playing bj
if they wanted to they could completely end it with continuous shuffling machines
most wannabee card counters will fail for various reasons such as an insufficient bankroll and many will lose big
a few will win but most won't win big - they will be stopped before that
it's a delicate dance - the interactions between casinos and counters
casinos want to keep the dream alive and stop those few who have the ability to win big before they win big
many, many times they know the player is a counter, and they will allow him to play and only take action after his big bets come out
they will allow him to play at their tables for an hour but not for 5 hours
but they don't want to completely stop it
.
link to original post
Very interesting take! I wonder if it’s true? I’d doubt it. But we have ZCore who works for the other side, and I guess the Wiz who worked for the evil Venetian…. Maybe they can answer?
link to original post
I was involved with detecting card counters , what would you like to know
link to original post
Just I think it would be quite easy to identify card counters with not a lot of work. Why isn’t it done more often?
link to original post
Soopoo:
why do you think casinos allow advantage play on some slot machines________?
they could easily completely 86 that if they wanted to
hint: for the same reason casinos don't completely 86 card counting
https://www.machinepro.club/advantage-play-slot-machines
.
Quote: lilredrooster.
Soopoo:
why do you think casinos allow advantage play on some slot machines________?
they could easily completely 86 that if they wanted to
hint: for the same reason casinos don't completely 86 card counting
https://www.machinepro.club/advantage-play-slot-machines
.
link to original post
I think you have to separate slot advantage play into two distinct categories.
Multicarding like DarkOz does is a direct hit to the casino’s coffers. I’d call it ‘free play’ abuse.
Vulturing is not a direct hit. Someone has to hit that must hit progressive. The coin in to get there is a positive for the casino.
I believe it does have a negative impact, as real APs take that money away from the casino, while ‘regular’ players ‘recycle’ it back to the casino in most cases.
If I’m the casino, I use a bunch of resources to stop the free play ‘abusers’ like DarkOz, but little to stop vulturing.
Obviously they don’t use enough because DatkOz can make $20k a week (not every week!).
