Jump to content


Photo
* * * * * 5 votes

Fruit Machines Inside Out: Compensation


  • Please log in to reply
157 replies to this topic

#121 Twostroker

Twostroker

    Newbie

  • New Members
  • 129 posts

Posted 26 November 2017 - 08:45 PM

How does the jackpot frequency get determined and how does changing the frequency effect awards?

#122 edwardb

edwardb

    Junior Member

  • Regulars
  • 223 posts

Posted 30 November 2017 - 02:27 PM

How does the jackpot frequency get determined and how does changing the frequency effect awards?

Depends entirely upon how the jackpot is saved up for (if at all). Say you have a JP "pot" and you put 10% of your stake (after taking your payout % away) in. Well, divide the jackpot value by the amount you put in the pot per spin, and there's your jackpot frequency.....

 

But then we used to save up possibly multiple jackpots, or have multiple pots, one paying for jackpot + repeats, and/or single jackpots. There really are 101 ways to do it.

 

Also remember that if your main compensator/pot gets too high (owes the player money), you can drop in jackpots to bring the machine back towards even. 

 

Lo-tech games generally worked quite differently to hi-tech AWPs. Bar-X style games had many different pots to pay for wins of different sizes/streak/etc, whereas a pub AWP might have had just 2 pots, main and jackpot.

 

It all depends on the game and what you're trying to achieve.

 

 

Hope that helps......it might just confuse you more!



#123 Twostroker

Twostroker

    Newbie

  • New Members
  • 129 posts

Posted 02 December 2017 - 02:10 PM

That does help a lot it's all good to know, it is confusing but I suppose it has to be to confuse the players psychologically?

I like old scorpion 4 clubbers, these seem fairly hard to 'work out' there's the jackpot frequency hi - lo switch on the main board. I can understand why an operator might what to change the frequency due to what environment the machine is in e.g. local club or holiday park.

I dont really understand how it works or the numbers they are aiming at. I suppose these machines are working there % and pots over thousands of pounds where as some easy mpu5's ive played award jackpot at the first opportunity.

I like all these machines for what they are so it's always good to learn more about them and how they actually work

#124 edwardb

edwardb

    Junior Member

  • Regulars
  • 223 posts

Posted 05 December 2017 - 02:11 PM

No, there's never any psychological reasoning behind compensating a game in one way or another. It's just experience, going off of games that were popular and games that flopped.

Most fruit machine industry stuff is trial and error - and a hell of a lot of error, at that.

 

The old BFM clubs were great, and from memory (I could be wrong) I think they used to have a lot of compensators ("pots") - I think when we did Pacman Club (copy of Club Crazy Fruits) at Mazooma it had 12 or so. Lots of pots for smaller wins and a few pots for the big ones.

 

We used to play thousands upon thousand of manual games on a machine, with millions (if not tens of millions) of simulated plays on the same code - you can run it on a PC at warp speed. Look at the numbers, see how the game is playing, and tweak accordingly.

 

That's the black art of gaming. Any old muppet can make a game hit the correct % payout, getting it to play well is quite another thing.........



#125 CMR777

CMR777

    Newbie

  • New Members
  • 3 posts

Posted 20 June 2018 - 12:40 AM

On the subject of emptiers, Super Multinudge was a classic case of a deliberate emptier. Programmed by MW (I won't put his full name here), of BWB Monte Carlo Or Bust and MAB Bar X fame.

 

Supercharged and Thunderbird were other classic examples.

 

ps. Hello Edward -- long time since we last spoke!

 

pps. This is my first post to a fruit machine forum since Arcadia back in 2004 !!!

 

-Smarter Than The Average Bear hahahahaha-



#126 nails

nails

    The furniture

  • Regulars
  • 4578 posts

Posted 20 June 2018 - 08:24 AM


 
:)
 
What would you like to know about next?
 

 

 

If all DOND clones are the same machine but different alpha messages etc.   how do cheats accidentaly get in there?

 

i think the answer blows open the subject of coders going bad.

 

* in regards to the part where `we all have a good laugh about backing`  thats 50/50 bollocks IMO. If i was playing a RED back in the day, no way would i play it if was empty, but a fully hopper gave me a 50/50 chance its full or topped up. also hearing a £1 thud on a crazy amount of coins in the cash box was also a good pointer.  



#127 fuzion

fuzion

    Part of the furniture

  • Regulars
  • 1726 posts

Posted 20 June 2018 - 09:37 PM

Yeah a genuine drop onto a decent clash of coins is always a good way to tell if it's refloat day or not :)

 

....well that and every other machine dropping.

 

J


Edited by fuzion, 20 June 2018 - 09:37 PM.

// stumblin' in the neon groves


#128 fuzion

fuzion

    Part of the furniture

  • Regulars
  • 1726 posts

Posted 20 June 2018 - 09:39 PM

I've done it with a Grand Golden Game back in the day, tried one it was dropping...  tried the others just to make sure and of course all 3 are dropping down the back...    this was late afternoon as well so obviously they'd had no action. 

 

J


// stumblin' in the neon groves


#129 mrrix

mrrix

    Newbie

  • New Members
  • 78 posts

Posted 22 June 2018 - 12:19 PM

Really good thread! I would love someone to take one of these emulations, tweak it and leave us an easter egg to find! Maybe just a clue every week until someone finds it!



#130 edwardb

edwardb

    Junior Member

  • Regulars
  • 223 posts

Posted 10 July 2018 - 02:55 PM

On the subject of emptiers, Super Multinudge was a classic case of a deliberate emptier. Programmed by MW (I won't put his full name here), of BWB Monte Carlo Or Bust and MAB Bar X fame.

 

Supercharged and Thunderbird were other classic examples.

 

ps. Hello Edward -- long time since we last spoke!

 

pps. This is my first post to a fruit machine forum since Arcadia back in 2004 !!!

 

-Smarter Than The Average Bear hahahahaha-

 

Hello - long time!!

 

MW yes I know who you mean - not seen him for ages - is he still kicking around the industry?

 

 

 

 
:)
 
What would you like to know about next?
 

 

 

If all DOND clones are the same machine but different alpha messages etc.   how do cheats accidentaly get in there?

 

i think the answer blows open the subject of coders going bad.

 

* in regards to the part where `we all have a good laugh about backing`  thats 50/50 bollocks IMO. If i was playing a RED back in the day, no way would i play it if was empty, but a fully hopper gave me a 50/50 chance its full or topped up. also hearing a £1 thud on a crazy amount of coins in the cash box was also a good pointer.  

 

 

 

DOND clones, cheats etc probably put in the original game, and commented out for that game - then put in the clone for testing purpose and they forgot to remove/disable it before release.

I've seen some very sloppy coding. I always used to surround my test code with an #ifdef, e.g.

 

#if (RELEASE==0)
// Test code here..

if (read_switch(CANCEL))

{

    DoSomething();

}

#endif

 

Meaning that any test code cannot be in the final build...but others weren't so cautious. I never had an urgent ROM update on any of my games. For some programmers it was a matter of how many....

 

Re. "backing", well, given that all the stuff to do with hopper levels is buried deep in library code which (and I can only speak for Mazooma) no one touched, apart from those who *knew* what they were doing, there is no correlation between hopper levels and how happy a machine was. Changing that would reflect in every game currently in development, adjusting makefiles, etc......nah don't even go there.

 

I used to look after the system software for foreign games at Mazooma, and it (all Bell Fruit code) was a spiders nest - very very complex - you had to know it backwards else you'd cause nightmares.

 

But I love all this urban myth stuff, always have, as I used to believe it myself before starting in the industry,


Edited by edwardb, 10 July 2018 - 03:00 PM.


#131 edwardb

edwardb

    Junior Member

  • Regulars
  • 223 posts

Posted 10 July 2018 - 03:01 PM

Really good thread! I would love someone to take one of these emulations, tweak it and leave us an easter egg to find! Maybe just a clue every week until someone finds it!

 

There is a plan afoot for an emulator-only game, which I'd let forum members design, and myself code it up over time. Can leave a few things in, for sure.

 

Just need a bit of time to do it....



#132 Mavroz

Mavroz

    odd job man...

  • Layout Creator
  • PipPipPip
  • 882 posts

Posted 10 July 2018 - 03:17 PM

Good to see/hear you are back around.



#133 mort

mort

    Layout Designer

  • Layout Creator
  • Pip
  • 354 posts

Posted 12 August 2018 - 02:06 PM

Hi edwardb,

 

This has been a fascinating thread to read (I have been dis-interested in FME for a while and only recently had the urge to see what's been going on). I have been hoping to get insight from real programmers for ages. I have always been particularly interested in understanding the logic code behind the machines. We, as the players, interface directly with the visual elements of course, which as you make clear is actually very separate from the logic running "behind the scenes".

 

I would be really interested in hearing more from you particularly to see source code, or even pseudo code, in which you can make clearer to us non-coders how things work. The insight on the compensators has been brilliant.

 

There is also a very important bit I picked up on in this thread, which is this section here:

 

fruitman69: "Not really, anti force code normally will monitor certain variables, if forcing is detected then it would set a flag"
 
edwardb: "Actually, in my experience, what people termed anti-force code actually wasn't. It was anti-stats testing code! Manufacturers used to buy other manufacturers machines and play them, log every spin, and work out what the game was doing. I remember visiting Maygay in 1996 and seeing a Barcrest 10 out of 10 being stats played. So what everyone did, was monitor the last, say 200 coins to go in - if they were all £1 coins then you mangle how the game plays. Still hits % but plays differently. Obviously in a pub you'd get other coins inserted, so this would never happen."

 

I have been hanging around FME for a good few years now, and one thing that I, and others, would comment on occasionally was that some machines did not play as expected. We put this down to a number of factors such as:

 

1. Many of our ROMs had come from BBS downloads, so they would be later revisions with things like emptiers, tricks and certain ways of playing changed/patched out etc.

2. It's hard to try and play a machine in the emulator in different ways yourself, you would tend to just 'force' all the time. In the real world the machine would see all different forms of play from many different players. So a long-term anti-force flag or series of flags would have less chance of being set compared to playing the emulation.

3. Something was not quite right with the emulation.

 

Many of our emulated layouts only accept £1 coins, and even if they have the buttons/shortcuts programmed for different coins, you would expect most players to just keep hitting 0 to enter £1 coins (I know I do!)

 

This would mean, then, that if a machine has "anti-stats" mode, we must be spending most of our time within that mode when playing on the emulator !

 

I would really like you to expand on "anti-stats" mode and provide more insight on what it would do. You say it would still target the same % as set, but would go about doing it in different ways. What would be the purpose of this ? Is this a method of trying to hide gameplay from competitors ? After all, it's gameplay that's the hardest thing to get right and the very thing that make's the machine a success.

 

Please any more info on anti-stats mode would be brilliant.

 

Mort.


Edited by mort, 12 August 2018 - 02:12 PM.


#134 mort

mort

    Layout Designer

  • Layout Creator
  • Pip
  • 354 posts

Posted 12 August 2018 - 05:08 PM

Having re-read my post above I wanted to make it clear that I am only interested in the emulation side of things here. I have zero interest in playing machines in real life. For me it's all about nostalgia, and emulation provides that hit. 

 

But I also have OCD with regard to getting things correct and understanding how things work. Emulation across many systems, not just Fruit Machines, is very important to me.

 

Whilst the majority are very happy to get an emulated machine to play and that's enough (which I totally understand) I have always  been interested in understanding how the very method of playing an emulation may actually impact that gameplay experience we remember and more importantly being aware of what we may need to adjust via our way of playing emulated machines to ensure we get that same experience we would have had in real life.

 

Following on from my post above I decided to run a favourite machine on mine, Club Bullion (MPU4), using Auto Play with MFME 6.1.

 

I am running two instances of the machine both with the 1.5 program ROM, both from a factory RAM reset.

 

The first instance is auto playing and entering £1, 20p and 10p coins.

 

The second instance is auto playing only inputting £1 coins.

 

Already the first instance has a cashpot value of £16 and a reserve value of £3 (IN= 2242 | OUT=0 | VTP=89330)

 

The second instance has only managed to get to a cashpot value of £2 and a reserve of £2 (IN=2046 | OUT=0 | VTP=89134)

 

I appreciate I would need to do a bit more testing on this, but it shows a difference.

 

Maybe it's as simple as the game being aware of the amount of physical coins IN to increment the cashpots. I'm going to do a bit more testing as I know the v1.6 ROM fills the cashpots much quicker than v1.5 anyway. 

 

 

Mort.

 

 


Edited by mort, 12 August 2018 - 05:17 PM.


#135 mort

mort

    Layout Designer

  • Layout Creator
  • Pip
  • 354 posts

Posted 12 August 2018 - 05:20 PM

attached image, with reference to my previous post.

 

From previous experience:

 

It took over £6000 IN to get the cashpot to £200 with the v1.5 ROM only inputting £1's via autoplay. 

 

It took over £2000 IN to get the cashpot to £200 with the v1.6 ROM only inputting £1's via autoplay.

 

I will see how this may change when I set autoplay to use different coin inputs.

Attached Files


Edited by mort, 12 August 2018 - 05:28 PM.


#136 stevedude2

stevedude2

    Senior Member

  • Regulars
  • 1043 posts

Posted 12 August 2018 - 06:26 PM

One thing that would definitely change the way certain games play in MFME would be the addition of a token input, because lots of games back in the token days had a token % and a cash %.  One of the main reasons for this was to stop people dumping tokens in one machine in particular and playing for easy cash wins.  If someone started dumping tokens in a Big Breakfast or one of the later Maygays on £8 tokens, the token percentage would go through the roof and you'd get nothing for ages before the jackpot just rolled straight in.  As a kid I used to love building up my tokens throughout the weekend on some of the 5p £6 Barcrest stuff like Hyper Viper and Viva Las Vegas and then pinching cheap features on stuff like BFM Around the Town and early Maygay stuff like Eastenders and Monopoly.  If I had an abundance of tokens then I'd dump them in a Labyrinth or JPM Fairground.

 

Of course the latest MFME allows for token switches now, so some games may well play much differently if that gets added and the game played with tokens.


Watch out! There's a SIG thief about...

#137 fuzion

fuzion

    Part of the furniture

  • Regulars
  • 1726 posts

Posted 12 August 2018 - 07:45 PM

,,,


Edited by fuzion, 12 August 2018 - 07:46 PM.

// stumblin' in the neon groves


#138 Twostroker

Twostroker

    Newbie

  • New Members
  • 129 posts

Posted 14 August 2018 - 12:38 PM

Regarding Anti-stats- I'm sure club machines could often take £200-£300 of £1 coins straight in? So on these do you think it is more really like a force blocker?

If you were testing another companies machine would you not just cash out a certain amount of times and simulate other coin denominations to try and remove the blockers? Presumably you'd have to do that when testing or simulating how your own machine plays-to get passed your own 'Anti-stats' ?
Just asking out of general interest as this is fascinating :)

#139 fuzion

fuzion

    Part of the furniture

  • Regulars
  • 1726 posts

Posted 14 August 2018 - 01:10 PM

attached image, with reference to my previous post.

 

From previous experience:

 

It took over £6000 IN to get the cashpot to £200 with the v1.5 ROM only inputting £1's via autoplay. 

 

It took over £2000 IN to get the cashpot to £200 with the v1.6 ROM only inputting £1's via autoplay.

 

I will see how this may change when I set autoplay to use different coin inputs.

 

Hey Mort....  might be worth doing a test using same programme when testing the multi-coin approach, like you say above version 1.6 fills quicker so you'd expect it to fill quicker regardless of coins entered.   Interesting stuff this, really enjoying the autoplay feature in mfme :)

 

J


// stumblin' in the neon groves


#140 fuzion

fuzion

    Part of the furniture

  • Regulars
  • 1726 posts

Posted 14 August 2018 - 01:16 PM

I have had the infamous jackpot 777 roll in on Maygay Monopoly.  I was using a £6 IOM all cash version.   It did seem difficult to get it to drop on the token version, but I did try in Deluxe Monopoly (early £4.80 token programme) using tokens and it did drop it in using mfme 5.   Token input is possible from mfme v5 onwards so is interesting that token input might affect this 777 roll in.   Saying that I have had it in the wild when 20p tokens input was disabled.   I'm not sure if there are dip settings for higher % of token wins, it all makes a difference.

 

J


// stumblin' in the neon groves





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users