Playtech rigged slots?

[Richas]

It seems to be generally accepted that when you play slots you are playing with unknowns. The only thing you will get is the paytable but you get no information about how likely it is to hit a certain combination or hit a bonus round. It would be a good service to actually list these probabilities and currently I know of only one software provider who has done this, and it's Galewind (see Link Removed ( Old/Invalid) ). In fact I believe they did this page based on feedback from people on this very forum. The wording of GRA terms is ambigious but it seems they don't require odds to be expessed to the player in this detail. It would be good for transparency if this was a requirement though.

I accept that the rule is not really very clear and I can see that publishing a full rtp is not an explicit requirement and maybe not part of the rule at all but this is the whole section

The published game rules and information should be sufficient to explain to
customers all of the applicable rules and how to participate. As applicable, game
information should include the following:
.......
h) Information about the likelihood of winning:
i) a description of the way the game works and the way in which winners are
determined and prizes allocated;

ii) For each game, information about the potential prizes and/or payouts
(including the means by which these are calculated) should be easily
available. This should include, where applicable:
(1) Pay tables, or the odds paid for particular outcomes.

(2) For peer-to-peer games where the prize is determined based on the
actions of the participants a description of the way the game works and
the rake or commission charged.
(3) For lotteries and other types of events where the potential amount or prize
paid out may not be known before the customer commits to gamble,
describing the way in which the prize amount is determined will be
sufficient.
(4) Displays of jackpot amounts that change over time (“progressives”)
should be regularly updated and as soon as possible after the jackpot has
been reset following a win.

Now to this bear of little brain it seems clear that the bold bit is additional to the bit underlined. I feel like I am so out of touch that I might come across as patronising or critical when I don't mean to but a general acceptance that all you get is the pay table is determined by what the casinos are providing, not what the rules say they are required to do. They have to have something for the bold bit that is supplemental to a pay table even if it is short of a full rtp disclosure.

If the play games and real games had something that meets the bold bit we could check if they play as described and that the descriptions match. It seems to me that we can only really see that the two "deviled" slots don't operate in the same way because the difference is visually explicit. There is no basis apart from huge sample sizes to check the rest of the stable of games - but there should be.

Clearly I am coming late to this particular party and I play poker and blackjack not slots but has anyone ever taken a sample game and made a complaint about the bold bit above not being met and seen what the GRA say about it? I guess I just prefer my "generally unknowns" to be the other players cards unless I force them to show me rather than pretty much the entire operation of the game.

 

[243 lines=daft]

It does seem that the regulation about free vs real gameplay has been breached here.

It did occur to me that playtech games can be played for fun in offline mode (unlike MG for example) and that might be a general reason for there being different sets of code for real and free?

A serious issue has definitely been unearthed here but I can't help feeling that a ongoing balanced analysis will serve the thread better than an all out presumption of outright rigging.

There is at the very least an important difference between this and the Finsoft 'thang', namely the 'virtual playing like physical' requirement (by which i mean we know how a 52 card deck should act in the real world but we can't say how an Ocean Princess turtle would act in the real world..) and also there's no suggestion of adaptive software here.

 

[binary128]

@Richas:

I agree that the "critical" word in the specification you've been referencing is "likelihood". So, when in doubt, check out Wikipedia. Not surprisingly,

You do not have permission to view link Log in or register now.

on this.

This article contains the sentence - "In non-technical parlance, 'likelihood' is usually a synonym for 'probability'." Whether these regulations were written "in non-technical parlance" is a real fence-sitter, but I would fall on the side that they are.

So, replace "likelihood" with "probability" and see where we stand.

Does the Casino/Software Provider have to provide "information about the probability of winning" (emphasis mine) or "the actual probability of winning"? The first, which is the actual regulation, is obviously not at clear cut as the second.

I know that the CM membership leans more to the slot player audience. And slot RTPs have historically been held "close to the chest" by the Casinos/Software Providers (although this has begun to loosen lately). In addition, slots are (pretty much anyway) the only type of Casino game where you can't go to a reference web site to get the game's RTP.

However, documenting non-slot RTPs can also be tricky because of the "House Edge" versus "Element of Risk" conflict. (There is a CM thread I started in late 2011 that goes into this whole thing in detail.)

The CM thread that I linked above arose from an internal objective at Galewind. Jufo's statement concerning the page to which he provided a link - "In fact I believe they did this page based on feedback from people on this very forum." - is true, as explained in this CM thread from early 2012.

The AGCC (with whom Galewind Software is certified) has similar requirements in their Standards. However, our documentation in this area satisfied any possible interpretation of the Standard.

So, I'd bottom line it this way:

1. The GRA standard says that the Casino should provide information about the probability of winning (and not the actual, statistical, probability of winning) for all games.

2. The actual, statistical, probability of winning for some games (table pokers, for instance) is a contentious calculation.

3. Most people are not aware of point 2.

I would conclude that the GRA standard does not require the Casino to publish the specific RTPs for any/all games.

---

However, as to the issue of whether the Play-for-Fun and Play-for-Real games need to provide (essentially) identical game play, including RTPs - there was no room for interpretation of this within the AGCC regulations. This is absolutely true.

(That would be like a car dealer having a demo sports car with 520 HP, but the actual delivered car has 140 HP. Pretty clearly fraud.)

The only question I have on this one, to which I don't have an answer (because I didn't need to ask the question) - What about Play-for-Fun products that run totally on the client? Clearly, these games have a different RNG than the (server provided) Play-for-Real games.

In this case, however, there is no intent of fraud. And it may just fall into the "too trivial to worry about" category.

Chris

 

[Richas]

I would conclude that the GRA standard does not require the Casino to publish the specific RTPs for any/all games.

OK I can get that, I think that what galewind have done is a good example of meeting what the requirements are asking for and the sort of thing the regulator should be asking for from everyone but I can accept that for whatever historic and competition reasons (plus complexity) the sites don't want a simple RTP publication to be statutory (and it probably isn't) but at the moment we have a pay table listed as a requirement after this:

h) Information about the likelihood of winning:
i) a description of the way the game works and the way in which winners are
determined and prizes allocated;

There has to be something to cover the above and it can't be the paytable as that is part II of h) and is only the payout rates for events - at the moment there does not seem to be anything that most sites are doing to meet this additional non pay-table requirement and so they must be breaching it ........and the GRA is clearly not enforcing it..... and because it is not enforced it makes it impossible for us for most games to check if the play game matches the real money game.

For two games we can see that they operate differently - for the others they may or may not, we can't see a difference but we have no description of the way the game works and the way in which winners are determined and prizes allocated in order to make a reasonable assessment as to whether they are operating in the same way....or indeed if they are a game we consider fair or a game we want to play.

It would be nice to have a game description showing it is non adaptive for example. As it is they say nowt and the question of adaptivity is left to the same rules that say we get a description when we clearly don't.....it really doesn't inspire confidence does it?

 

[Richas]

However, as to the issue of whether the Play-for-Fun and Play-for-Real games need to provide (essentially) identical game play, including RTPs - there was no room for interpretation of this within the AGCC regulations. This is absolutely true.

(That would be like a car dealer having a demo sports car with 520 HP, but the actual delivered car has 140 HP. Pretty clearly fraud.)

The only question I have on this one, to which I don't have an answer (because I didn't need to ask the question) - What about Play-for-Fun products that run totally on the client? Clearly, these games have a different RNG than the (server provided) Play-for-Real games.

In this case, however, there is no intent of fraud. And it may just fall into the "too trivial to worry about" category.

Hmm I've been thinking about this. I am not sure we can say that just because there is another reason for making the play game run differently (offline availability and no strain on the providers systems) that intent to deceive the customer is ruled out, all we can say is that there is a motive for different game mechanism that might result in deceiving players accidentally unless the provider is very careful about how the different play game works. This is the requirement:

‘Play for Free’ Games
(1) Play for free games for no prize are not gambling but should accurately reflect any
„real-money‟ version of the game, and should not be used to encourage those under
18 to use licensed gambling facilities.
(2) In particular, such games should not be designed to mislead the player about the
chances for success by, for example, using mappings that produce different
outcomes than the cash game. Licence holders should be able to demonstrate this
equivalence to the Gambling Commissioner upon request.

The sites need to be able to demonstrate equivalence to the Gambling Commissioner on request. It does not really matter if they deceive accidentally or on purpose, and if they fail that test the supplementary question is whether it was designed that way. A lack of ability to demonstrate equivalence alone is a breach.

Surely finding even one game that has been running for a long time in a way that can deceive is reason enough for the regulator to start asking for such demonstrations (or ask for more than they do now)- starting with the rest of these playtech games. Now maybe asking a licence holder to show this equivalence for some of their games every few months (or annually) would help? Who knows maybe they do this? Maybe the GRA should be asked if they do this and if not if they will consider doing it from now on and informing the players of how many games are being checked/rechecked each month/quarter/year - such a measure might improve confidence.

 

[binary128]

Richas,

I understand that you're still trying to clarify (or receive clarity on) exactly what these regulations really boil down to.

And I'm sorry to say that I've given you all that I've got.

I can say that the GRA and AGCC Standards are significantly the same. As you've seen, regarding this whole "likelihood of winning" thing we provided enough information to satisfy any interpretation of the Standard. Therefore, there was no need for us to "dig into the details" of this requirement with the AGCC. It was just "OK, check, next."

There is another part of the Standard - Game Malfunction - that might conceivably be involved here. (Another CM post in which we addressed this for the AGCC.)

Chris

 

[Richas]

Richas,

I understand that you're still trying to clarify (or receive clarity on) exactly what these regulations really boil down to.

And I'm sorry to say that I've given you all that I've got.

I can say that the GRA and AGCC Standards are significantly the same. As you've seen, regarding this whole "likelihood of winning" thing we provided enough information to satisfy any interpretation of the Standard. Therefore, there was no need for us to "dig into the details" of this requirement with the AGCC. It was just "OK, check, next."

There is another part of the Standard - Game Malfunction - that might conceivably be involved here. (Another CM post in which we addressed this for the AGCC.)

Chris

Thanks for this. i appreciate that you are trying to help and we definitely agree on loads, wanting a regulated industry that we can all have confidence in and as an industry insider I welcome your insight. I also think that Galewind (who you work for) have done a bit more than tick the box and move on, you read the question. I have doubts that others bothered.

Thanks again, I'll look into the malfunction bit when I get a chance but I wanted to thank you for responding.

 

[Betfred]

Hi all,

Our investigation of the two games in question (Ugga Bugga and Ocean Princess) shows that there was a difference in the RTP between the real play version and the free-play version. We have raised this with the supplier, Playtech, and they agree with our analysis and are deploying a fix, estimated for this coming Monday.

It is important that people understand how the differences arose and their significance. The fun play/offline version of these games are self contained software, and have the same math and RNG model as the real play version. However, because the fun/offline version does not interface with the gaming server (to record the bet) it did not receive an update deployed to the real play version of the game. As the forum pointed out, this meant a visual glitch was not corrected in the fun/offline version, which restricted the representation of certain reel combinations to fun play players. This would have created a slightly different game play and a different RTP between fun and real.

While this is not intended or acceptable, it must be stressed the real play version does play in the manner described and advertised in the help file. However, we will be withdrawing the fun play versions of the game until the fix has been deployed. As such, the effected games will not be available on the Betfred website until further notice.

Due to the way fun play is recorded, the exact difference in RTP of the free-play version cannot be determined. Analysis of real-money play shows very few player numbers, which would mean only a few players could claim to have been misled by this fault. If they believe they have, we will happily review their play and consider a refund.

The Gibraltar regulator is aware of this incident and is working with Playtech to resolve the problem, as it may affect other Playtech clients as well.

Finally, both the Gibraltar regulator and Betfred have asked that Playtech (and our other suppliers), in addition to our own testing programme, to commit to a full audit of their games to ensure this scenario isn’t repeated. As part of our commitment to Casinomeister and forum members, we will post any results and actions taken by Betfred here.

As always, I would like to thank the members for bringing this to our attention and to apologise for any inconvenience caused.

Kind regards,

Aaron

 

[KasinoKing]

Due to the way fun play is recorded, the exact difference in RTP of the free-play version cannot be determined.

This part of your otherwise very good post is, frankly BS.
Software suppliers know exactly what the T-RTP of any game is - they don't need people to play it so they can work it out from the results!
I don't know who is giving you this information, but if I were you, I'd be pretty annoyed with them

KK

 

[Betfred]

Hi KasinoKing,

Playtech and Betfred know the T-RTP for the games However the bug altered the performance of the game in fun play/offline mode and because these versions do not interact with the game servers, Playtech have no recorded information on play/payouts and can't therefore calculate how the bug effected the RTP. They could, of course, simulate the bug in a real environment, run the effected versions through millions of game rounds and record the results to determine it, but I think it makes more sense for them to concentrate on getting a fix deployed.

Cheers,

Aaron

 

[P.V.]

Hi all,

Our investigation of the two games in question (Ugga Bugga and Ocean Princess) shows that there was a difference in the RTP between the real play version and the free-play version. We have raised this with the supplier, Playtech

Kind regards,

Aaron

Sounds like a lot of issues going on between free play and real play online as Aaron has pointed out.

The heads up is don't play free versions, think you're playing the same game while betting real money at OC's.

 

[vinylweatherman]

So, we have the same game with two different RTP settings in a software that was previously thought to be like MGS in that all games were using the same logic with all operators. A visual glitch would NOT alter the RTP in any way, it would merely mis display the result. The mere existence of this glitch implies that this is not a simple weighted slot, but some kind of fixed probability table of results, which are then represented by creating "eye candy" effects as though it were a slot. It's not so much a slot game, more a scratchcard that represents it's results as a slot.

The update may well have been a fix for a glitch that meant one particular layout never got selected to display a certain result, but it could be far more serious. It could be evidence that like RTG, Playtech slots have operator configurable RTP settings, and this glitch is similar to the "two pears" glitch on the RTG game Fruit Frenzy that showed players proof that different RTP settings were in effect.

The test for this would be to see whether this game operates differently between different operators in real play mode. If some have the glitch spotted in free play, whilst others don't, it could show that Playtech games are indeed running at different RTPs among different operators.

You had better hope that what you were told by Playtech about the two running at different RTPs because only one got an update is BS. Playtech should realise that they have more or less admitted that their games CAN run at different RTP settings, which are altered by rolling out updates to the games. Which is higher we don't know, and we don't know whether any such settings are in use by operators.

Unfortunately, we were misled by the RTG statement that the 91.5% setting was "only for land kiosks" when it was revealed that the "two pear" setting on Fruit Frenzy was the 91.5% one, rather than the top 97% one, with the explanation changing from "land kiosks only" to "operators may make the business decision to operate on 91.5%".

This is worse for Playtech, since they don't state RTP in the help file, so can't try and hide behind the "wrong helpfile" excuse, nor can they claim that since the game operates to the RTP stated in a help file, there is no issue to address.

It seems the earlier case has caused some to look at other games from other softwares, and other examples are being revealed of "glitches" in games that mean that free play modes do not play the same as real money modes.

 

[Balthazar]

Playtech and Betfred know the T-RTP for the games However the bug altered the performance of the game in fun play/offline mode and because these versions do not interact with the game servers, Playtech have no recorded information on play/payouts and can't therefore calculate how the bug effected the RTP. They could, of course, simulate the bug in a real environment, run the effected versions through millions of game rounds and record the results to determine it, but I think it makes more sense for them to concentrate on getting a fix deployed.

So in other words:

-There was initially a bug with the slot that affected the RTP (in both real money and free money modes)
-The slot got patched in real money mode
-The slot didn't get the patch in free money mode
-They don't know what was the RTP of the bugged (unpatched) version of the slot
-Therefore, there's no way to know the RTP in free money mode unless they simulate millions of spins.

Am I correct?

 

[E-Gambler]

Hi KasinoKing,

Playtech and Betfred know the T-RTP for the games However the bug altered the performance of the game in fun play/offline mode and because these versions do not interact with the game servers, Playtech have no recorded information on play/payouts and can't therefore calculate how the bug effected the RTP. They could, of course, simulate the bug in a real environment, run the effected versions through millions of game rounds and record the results to determine it, but I think it makes more sense for them to concentrate on getting a fix deployed.

Cheers,

Aaron

perhaps they could concentrate on both ?

 

[Richas]

Hi KasinoKing,

Playtech and Betfred know the T-RTP for the games However the bug altered the performance of the game in fun play/offline mode and because these versions do not interact with the game servers, Playtech have no recorded information on play/payouts and can't therefore calculate how the bug effected the RTP. They could, of course, simulate the bug in a real environment, run the effected versions through millions of game rounds and record the results to determine it, but I think it makes more sense for them to concentrate on getting a fix deployed.

Cheers,

Aaron

Aaron you need to get on to Playtech and whoever is in charge today at BetFred. You have just admitted that you can't meet your regulatory requirements.

Here are the rules:

Play for Free’ Games
(1) Play for free games for no prize are not gambling but should accurately reflect any
„real-money‟ version of the game, and should not be used to encourage those under
18 to use licensed gambling facilities.
(2) In particular, such games should not be designed to mislead the player about the
chances for success by, for example, using mappings that produce different
outcomes than the cash game. Licence holders should be able to demonstrate this
equivalence to the Gambling Commissioner upon request.

The fundamental design of likely all the play money games means that you as the licence holder can't meet the requirement in bold. My advice - pull every play money game you have until you can meet the regulatory requirements for them. Like the scene from Aliens when they decide to take off and nuke from orbit "because its the only way to be sure" pulling the games is the only way to be sure (for the avoidance of doubt you are Hicks in this analogy - you know the corporal "He's just a grunt, he can't make that sort of decision".

 

[Betfred]

Hi Balthazar,

Yes, correct.

Ref vinylweatherman's post: I know of no ability to set RTP - we have absolutely no control over the algorithms of games. The fun and real play versions had the same RTP initially but that diverged with the update, which contained visual and RTP changes. Regrettably, the same changes were not passed on to the fun version. This is what has a) been detected in our investigation and b) what we've been told by Playtech.

Playtech are aware of the thread, as are the GRA. If there are further questions raised on the back of comments here, Betfred will assist all parties to ensure they are answered.

Kind regards,

Aaron

 

[Richas]

Hi Balthazar,

Yes, correct.

Ref vinylweatherman's post: I know of no ability to set RTP - we have absolutely no control over the algorithms of games. The fun and real play versions had the same RTP initially but that diverged with the update, which contained visual and RTP changes. Regrettably, the same changes were not passed on to the fun version. This is what has a) been detected in our investigation and b) what we've been told by Playtech.

Playtech are aware of the thread, as are the GRA. If there are further questions raised on the back of comments here, Betfred will assist all parties to ensure they are answered.

Kind regards,

Aaron

What changes were made to the information available to players of the real money game when changes were made to the RTP? That is a change to the likelihood of winning and the way the game works and the way in which winners are determined and prizes allocated; which you are obliged to provide information about under the GRA rules.

 

[Richas]

. If there are further questions raised on the back of comments here, Betfred will assist all parties to ensure they are answered.

Did the change in RTP of the real money game increase or decrease the return to players?

 

[Jufo]

Hi KasinoKing,

Playtech and Betfred know the T-RTP for the games

Aaron, could you tell us what is the T-RTP value for Ocean Princess at Betfred? Boyle casino manager has confirmed the following information:

The Ocean Princess slots game can be found at Playtech casinos. Playtech do not publish the house edge of their games, however several established Playtech casinos have started releasing this information and on the CasinoMeister forum, the Boyle Casino Manager confirmed that Playtech slots games are not configurable at the operators end. As such, it is possible to identify that the house edge of Ocean Princess is; playing max coins using the optimal strategy the house edge is 0.93%, playing less than max coins with optimal strategy the house edge is 1.26% and the average player strategy generates a house edge of approximately 3% and should be valid at all Playtech casinos.

Source:

You do not have permission to view link Log in or register now.

Therefore, Aaron, could you confirm for us that the T-RTP of Ocean Princess is 99.07%. If it's not this value, then either you run the game at different RTP setting than Boyle, which should not be possible, or you have made an update to this game in secrecy to lower the T-RTP without letting your customers know.

 

[Jufo]

Did the change in RTP of the real money game increase or decrease the return to players?

I have information that the demo version of Ugga Bugga pays 99.63% with optimal strategy. The real money version should pay 99.07% with optimal strategy. I would like Aaron to confirm the latter value.

 

[Richas]

If there are further questions raised on the back of comments here, Betfred will assist all parties to ensure they are answered.

Did the visual and RTP changes included in the upgrade of the real money game alter the optimal strategy for the game?

 

[ThePOGG]

While this is not intended or acceptable, it must be stressed the real play version does play in the manner described and advertised in the help file. However, we will be withdrawing the fun play versions of the game until the fix has been deployed. As such, the effected games will not be available on the Betfred website until further notice.

Hey Aaron,

First off i'd like to say how glad i am to see you back at the table with regard to these issues. I understand that you may have been told to refrain from commenting regarding the FinSoft issue for now and that your hands may be tied, but i can't say i think that policy is benefitting Betfred right now.

What i do want to point out to you however is the lack of consistency that's being shown here. You've pulled both the Playtech games that have been shown to be operating differently in free play mode to real play mode, but the FinSoft game that's doing exactly the same thing is still available!! thelawnet looked at the code weeks ago now and detailed their findings in the original thread and i've conducted testing on this game which you can see at https://www.casinomeister.com/forums/threads/finsoft-spielo-g2-games-issue.54475/. Aladdin's Treasure clearly does not function the same in free play as it does in real play - why hasn't it been pulled?

 

[binary128]

Some (Detailed) Data on System Audits

I'm posting this because I believe, based on a documentation comparison, that both the GRA and the AGCC do very similar things, in very similar ways. Obviously I can't definitively state that what follows applies to the GRA, but I believe this to be true.

Also, I'm only going to focus on the audits of the games, and not of the Administration and Control systems.

What the AGCC (and as I said the GRA and others) does in performing a System Audit of casino software is that an AGCC-certified Auditor takes the AGCC's published requirements, various casino documentation (help files, game source code, etc.) and (some of) the people that wrote all this casino stuff into a room. (Coffee is provided.)

The Auditor then starts at the top of their documentation, going through each section (and sub-section, and sub-sub-section) of the Standard, asking questions, receiving verbal answers or references to printed answers. If all of the answers that they receive to a particular question, or set of questions, are logical, consistent and, in their opinion, satisfy each of their "should" statements, then they move on to the next section of the Standard.

If, however, there is confusion or inconsistency between the "should" statement and the provided answer(s), then more (and more, and maybe more) questions are asked. Sometimes, this ultimately leads to an "OK, let's move on". Sometimes, this leads to "What you do, or say, does not meet the requirements of the Standard". A "Corrective Action" is thus assigned. Usually, this "Corrective Action" is weighted as to its severity - Critical, Major, Minor, Suggestion.

In a previous life I worked as the Director of Quality Assurance for a large North American manufacturing company. I had 10 years of experience in building a Quality System into a manufacturing infrastructure. I also had 10 years of experience in reading Quality System Standards, and in going through audits based on those Standards.

In addition, we also had our RNG/Scaling/Mapping system audited and certified by iTech Labs. And finally, we had 3 years of statistical analysis reports for all of our games available from Certified Fair Gambling.

I thus felt that we were very well prepared when it came time for our AGCC audit.

As I've said, for many sections of the Standard it was "OK, looks good, let's move on." However, even though we were well prepared, we still received about 15 "Minor Corrective Actions".

One of the sections of the Standard that took awhile to address was the "Play-for-Fun versus Play-for-Real" games. This part of the Standard was dealt with not only by the AGCC Auditor, but also by the Approved Testing Lab that had been assigned to our application (a company named SQS). Here, the Testing Lab spent a lot of time:

1. Comparing the actual game play in the Play-for-Fun and Play-for-Real casinos. (It's pretty easy to have one instance of each product up on the screen side by side.) This also included comparing the Help file documentation.

2. Comparing parts of the source code one to the other. For example, they compared the "reel strips" of all of the slots, the "hand scoring and payout calculator" for all of the Video Pokers, the "collapsing tube" of our Keno game, the "Deal code" for our Blackjack game, and so forth.

(Source code was the only way to compare certain aspects of the "Fun versus Real" products because we don't store game hands for the Play-for-Fun casino in a database. So, there is no other way to verify that the Play-for-Fun product's actual RTP is the same as its published Theoretical RTP, or the same as the Play-for-Real product's RTP values.)

As I've said, I have 10 years of experience in building, and documenting, Quality Systems. Even so, we wound up with 15 "Minor Corrective Actions" from the audit. (I'm not positive it was exactly 15, but that is a pretty good estimate.)

Does all of this mean that something still didn't "slip through the cracks". Certainly not. I/we had nothing to hide but we still missed 15 "things". There is no reason to believe that the AGCC/SQS Auditors didn't miss stuff as well.

I'd say the bottom line - people make mistakes. Judgement can be applied to that, but perhaps more important is the judgement we apply to the response to those mistakes.

---

Sounds like a lot of issues going on between free play and real play online as Aaron has pointed out.

The heads up is don't play free versions, think you're playing the same game while betting real money at OC's.

P.V. - I confess that I was stunned by your conclusion here. Given the amount of time that we and other casino software providers spend in making sure that the Play-for-Fun product is an exact replica of the Play-for-Real product, your conclusion was very deflating. It was almost like "Well, the Players won't believe it anyway, so why bother?"

I certainly agree with your first sentence - who wouldn't. But the purpose of the Standards is to weed out the "mistakes" (or the "bad apples"). I'd say that your conclusion here has the baby exiting with the bath water.

Chris

 

[Richas]

As I've said, for many sections of the Standard it was "OK, looks good, let's move on." However, even though we were well prepared, we still received about 15 "Minor Corrective Actions".

One of the sections of the Standard that took awhile to address dealt with the "Play-for-Fun versus Play-for-Real" games. This part of the Standard was not dealt with only by the AGCC Auditor, but also by the Approved Testing Lab that had been assigned to our application (a company named SQS). Here, the Testing Lab spent a lot of time:

1. Comparing the actual game play in the Play-for-Fun and Play-for-Real casinos. (It's pretty easy to have one instance of each product up on the screen side by side.) This also included comparing the Help file documentation.

2. Comparing parts of the source code one to the other. For example, they compared the "reel strips" of all of the slots, the "hand scoring and payout calculator" for all of the Video Pokers, the "collapsing tube" of our Keno game, the "Deal code" for our Blackjack game, and so forth.

(Source code was the only way to compare certain aspects of the "Fun versus Real" products because we don't store game hands for the Play-for-Fun casino in a database. So, there is no other way to verify that the Play-for-Fun product's actual RTP is the same as its published Theoretical RTP, or the same as the Play-for-Real product's RTP values.)

As I've said, I have 10 years of experience in building, and documenting, Quality Systems. Even so, we wound up with 15 "Minor Corrective Actions" from the audit. (I'm not positive it was exactly 15, but that is a pretty good estimate.)

Does all of this mean that something still didn't "slip through the cracks". Certainly not. I/we had nothing to hide but we still missed 15 "things". There is no reason to believe that the AGCC/SQS Auditors didn't miss stuff as well.

Thanks for this. Very helpful. Could you describe one of the 15 so that we can see what a minor corrective action is? I am guessing that what we have been discussing here is not in that lowly category but it would give us a better idea of the review process.

The Approved Testing Lab that certificates an application. Once it has issued a certificate what level of change to the game requires recertification? At first glance it would seem that changes that I would assume should affect that external certification have been happening without new certification (changes to game play/rtp).What is the trigger?

 

[binary128]

Thanks for this. Very helpful. Could you describe one of the 15 so that we can see what a minor corrective action is? I am guessing that what we have been discussing here is not in that lowly category but it would give us a better idea of the review process.

The Approved Testing Lab that certificates an application. Once it has issued a certificate what level of change to the game requires recertification? At first glance it would seem that changes that I would assume should affect that external certification have been happening without new certification (changes to game play/rtp).What is the trigger?

Here are 2 examples of the "Corrective Actions" that we received during our AGCC audit.

1.) When you "mouse over" a bet "hot spot" on any of the Roulette tables, a "tooltip" appears giving the "name" of the bet as well as the current bet amount. For the combination bets at the top of the bet table (any combination involving 0, 00, 1, 2 and/or 3) this "tooltip" did not appear. (We added them.)

2.) A few of the games displayed a currency symbol with the bet amount. The majority of the games did not display a currency symbol. The AGCC viewed this as a point of confusion for the Player. (We decided to remove the currency symbol from those games that used it.)

---

Here are some examples of game changes which required re-certification.

1.) I started a CM thread asking for Player input into our slots and video pokers. For the slots, the "min coin amount, min coin count, max line count" change required re-testing/re-certification. For the video pokers, the "min coint amount, max coin count" change required re-testing/re-certification.

2.) Changes to our slot console for available bet options, about which discussion was started here and then implemented here were then modified here. The original implementation was used for the first round of game certification. The modifications that are discussed in the third post then had to be re-certified by SQS.

---

Another example of re-certification, this one involving our RNG.

Our RNG system was first certified by Certified Fair Gambling on April 8 2009.

We modified the RNG (going from a single-threaded executable with mutex to a multi-threaded executable with mutex), and decided to change our Certification source to iTech Labs (because of international accreditations). We received our new RNG Certification on July 6 2010.

We then made a change to what is called the "background cycling system" of the RNG process, which required re-certification. We received our current RNG Certification on October 3 2011.

---

The issue with certifications is that they take time, and they also take money. (In our case, and on our budget, a lot of money.) The same issues exist for re-certifications - time, and money.


Chris