New Life Games Tech Forums

 :5-

I was playing several 3902 games and thought of something.     The first reel only spins for 10 or so spaces.   Is there an actual reel that someone could memorize and guess what is coming next? 

How many images are there ?      I never seem to see patterns repeat. 


I'm just thinking aloud.   :58-

Now serving #480.  Now serving #480...  (Munch, you're 481 :96-)


"...and several butcher's aprons. And now, BBC4 will explode." M. Python.



Now serving #481. :30-

I don't own a Game King, but I suspect that they only animate as much as is needed of the virtual strip. They don't have draw every symbol in order. They might also be using teaser strips for the spin.

Keep in mind that I don't know for sure, but I do know how to program animation.

This COULD be what they're doing:

Let's say that we number the symbols in reverse order, so that the numbers increase as the "reel" spins in the "window".
32
..
25  ]
24  ]  Center line in the "window"
23  ]
..
1

Let's say your last spin landed reel one of five on the center line symbol 24 of 32. They would draw symbols #25, 24, 23 in the window, as seen above.

Now let's say that the next spin requires symbol 14 of reel one to land on the center line. Instead of animating everything from 26 to 32 and then 1 through 15, they could jump straight to 6 and animate ten symbol changes.

So what you would see would be the following, ending with symbols 15,14,13 in the window. It would appear seamless to the player.

15   ]
14   ]
13   ]
12
11
10
9
8
7
6    < start new spin with this symbol
25  ]  <  top symbol on last spin.
24   ]
23   ]


They could also do the reverse: Animate symbols 26-32 and then 13-15. It wouldn't matter.
If they chose to use teaser strips, then they would just substitute the teaser symbols for positions 6-12 in my example.

Munch,
What I'd like to see is a video slow-mo of these spinning animations on youtube for us here!  :96-

Agreed. Especially if you can get a few dozen spins and if we can dig up the correct PAR sheet for the game and percentage you're running. :89-

What a great idea to slow down the video. 

I'd like to investigate this and see what's going on.    I'm also going to look in the play history.  It shows the last couple of spins with the virtual reel stops.  It may shed some light on what's going on.

To do it properly, you would want to speed up the video recorder during filming. It would then play back at normal speed in slow motion and would still have all of the detail. If your camcorder can't do that, then we'll have to settle for a slowed-down video.

Here's the best I can do in slo-mo

http://www.youtube.com/watch?v=KQ0rkIrzt_w (http://www.youtube.com/watch?v=KQ0rkIrzt_w)

Munch, the video is private. You can either pm me an invitation to view it or make it public. Also, I'll need to know the SG and SB chips you have and the percentage you've chosen.

Oh I'm sorry -   I fixed it.


I think the virtual reels are just so large that it's hard to know what's coming next. 

I can track from the video how many spaces the reel will move with each spin.  What I have to do now is look at the past play history and look at the virtual stops to see what's going on.  I'm going to guess that the random number generator picks what symbol appears on the reel. 

So in other words each time you play the reel strip changes. 

Reel 1

20

45

53

19

each time you play it generates a new strip.

Hey, jdkmunch I don't  if this helps, but if you open door and hit the white button to get to the menu screen then go to Diagnostics - GameTests - PaytableTest - then the  percent you have your machine set at you'll see that you can nudge the strips up or down one symbol at a time. Then you can see how many symbols are on each virtual reel, and what pattern there in.
Mike

FYI  Just checked my Tailgate Party and theres 46 symbols on each virtual reel.
Mike

Each time they stop, the music sounds like an introduction to a big movie scene from outer space!  :96-

That's not relevant. You should be writing down or otherwise documenting the spin data as I describe below. Keep in mind that there is more than a good chance (especially for the animation of the first reel) that they are skipping symbols to "get to the chase". They may also be using alternate "teaser" symbols during the animation.


COMPLETELY irrelevant.


Looking at the play history to document the final symbols (in order) that comprise each outcome would be useful.


Perhaps, but doubtful.  :103-
See below.


I don't understand these data.  :128-


(Continued in pt 2..)

(...Continued from pt 1)


It's possible, but are you sure?

---

First, to reiterate what I've said before, I don't own one of these slots. Further I haven't got a SINGLE PAR SHEET for an I-Game video slot, so I have nothing to work with.
S2000 and I-Game documents seem to be scarcer than hen's teeth. :37-


The video as uploaded is too blurry to see the symbols during the animation (spinning). If you can make them out in the original, you're set. Otherwise, you won't be able to assess what they're doing during the spin at all. i.e., you'll only be able to prove or disprove half of your hypothesis: whether they use a fixed strip order or randomize the symbols on each strip. You will not be able to determine how or what symbols are being skipped or substituted in-between.

I can also tell you that two minutes of video isn't nearly enough data. You'll probably need ten to twenty times that much unless the strip order is fixed and they are using the same strips for the spin animation, in which case you'll be able to figure it out much faster.

Here's what I would look for, YMMV.

• First, you will need the PAR sheet showing the strip table for the specific SB chip and payout percentage that you've set. If a complete PAR (and not all copies are complete) does not document any specific order for the symbols, but only lists the number of each symbol on each reel, then there might be something to your hypothesis that the symbol order is randomized for each spin. REMEMBER: you might not have a complete PAR sheet document, so the fact that they are missing won't conclusively prove this.
• Second, document the symbols & order for each final segment of strip seen at the conclusion of every spin for each reel. You you should be able to find the location of each segment on the strip listing in the PAR, and you should eventually be able to stitch the strip segments from the same reels together (although you won't need to because you'll have the PAR in front of you.)
• Third, separately document the symbols & order of the animated portion of your video, excluding the first and final symbols that comprise the previous and following ending positions. You will then try to find these segments on the PAR strip listing. This task should be completely separate from the documentation of the final outcomes. If you can find them, then you should be able to figure out exactly what is being selected for animation on each spin. You may need to look at several (or several dozen) segments and compare the positions within the PAR listing to the their previous and subsequent outcomes to come to a conclusion as to precisely what they are doing. If these animated segments are not on the PAR listing, then that would tell you that they are using either a substitute set of strips for the animation, or that they are randomizing the animation. You'll have to document a lot of spin animations to come to a conclusion. You might even be able to piece together the second strip set, if one exists!!
  

IF you can't map the FINAL OUTCOME segments to the PAR, then your statement that the symbols on each strip are being randomized for each spin may be correct.

It should be possible to scramble the order of the symbols without affecting the number of paying combinations on each line, provided that symbols involved in multi-symbol pays are left intact** and as long as the available number of specific symbols on each reel remains constant.

In other words, each of the three selected "stops" would be removed from the pool of available stops before selecting the next. So if there were only two red sevens on the first reel in the virtual table, then one would never see more than two red sevens in the window on the first reel.

Finally, if that were not done (that is, all symbols remain available and one could end up with three red sevens stacked on the first reel even though there were only one red seven on the virtual strip) then you would have an 8-liner, only with 15 reels instead of 9.




**They would certainly not want to scramble the location of the three individual wilds that form a triple stacked wild (such as seen in Wolf Run and others).


As an aside, if the symbols are weighted and fixed in position (as they are with physical strips) then the symbol order matters a great deal, and can get complicated enough to require a computer to determine the strip order. This becomes necessary to ensure that at the end of the day, the number of virtual stops for each type of symbol is the same on every pay line in every position.

---

Let me know how it goes. :89-


StatFreak :31-

Here's what started this:

I've played my Black Cherry for tens of thousands of spins.   Through all this play I now have a good idea of what combinations come out the most.  I can see the par sheet come to life and I enjoy it. 

I was playing the enchanted unicorn I-game and noticed that I couldn't remember ANY symbols on the strip.   No matter how much I played I could never match a pattern on the virtual reel strip. 

I then fumbled into the play history and saw this:  You have a history of every spin and the virtual stop number.  So from these two photos you can see that
the 1st reel went from virtual stop 29 to virtual stop 75 in just one spin. 

This led to the question what is a virtual stop?   It clearly can't be a position on the strip cause from the video you can analyze that in fact it only does move a few symbols.    Or could it? if the "stops" are not in order.   :103-

Anyway this is the stuff I think about when I play.    It keeps me from simply staring at the game and drooling on myself.   (although that can be fun sometimes too)

Stat - I am going to take your advice and document the reel.   

Okay, but read what I said from the beginning. You CAN go from virtual stop 29 to virtual stop 75 in just one spin and only show 10 symbols. I've already explained multiple ways that  this can be done.

Only the final five symbols that come to rest mean anything. There are two possibilities:
1) They are always in a fixed order that matches the PAR listing (e.g., stops 72,73,74,75,76 showing)
2) There is no fixed order virtual strip and the five symbols are scrambled at random, but according to probability,
     and without any stop being used more than once. (e.g. stops 18,20,74,2,41 showing)



That's Wrong. The strips can be in order and it can skip drawing the unneeded symbols in-between. Again, read my earlier posts.


 :184- :184- :127- :208- :208- :208-

Hey, me too. :71-  Before joining NLG (had to make my own PAR sheets), I had analyzed (correctly, btw) every game I had, including the re-spin games, with the exception of Haywire Deluxe. I was sure that the virtual stops dictated whether or not there was a Haywire, but I couldn't lock it down. So I gave up and decided to just try to figure out the odds of them happening. I printed a colored spreadsheet with every possible payout outcome on it and ticked off every single win in the appropriate row. I taped the paper to the award glass and did that for months (and several sheets of paper). If nothing else, if gave me something to do while I was trying to hit the jackpot. :5- :30-

I know that this is possible but is this in fact what's happening? 

You'll have to run my experiment in reply 14 to find out.  :96-
Read it carefully.
You have two separate objectives. Don't confuse the two.

By the way, you got me thinking about what would happen if they scrambled the symbols on each reel instead of adhering to a specific order, given that they would still select the three symbols from a pool for each strip that had the same number of occurrences of each symbol (same weighting) as a specifically ordered virtual strip.

You know what happens?  :214-

It decreases volatility and increases hit frequency, while leaving the total payback percentage and the odds of hitting any individual line win unchanged !! 

To put it another way, you win more often BUT WIN LESS each time!

Does that sound familiar?  :118-


See retraction post below.

So, now you've got me wondering. :103-
Of course, they can build the same design into a fixed strip, too.

Stat :31-

 :186- I really find it interesting.  

Now for the important question -  it's 8AM here  -5AM by you?

Are you just waking up or going to sleep?   :97-

Neither one. I posted the long-winded stuff, went to bed, and then started thinking about what would happen if the symbols were randomized. That kept me from going to sleep. After I ran the scenario in my head in the dark :149- :30-, I decided to get up and post my discovery; then I found you here. I suppose I'll go back to bed at some point.  :200- :25-

Hey try the above I believe it will answer some of your questions. Also based on your last picture I can tell you you have at least 97 symbols on each virtual reel. And In addition to Paytable test there is a Reel strip test. Hope this helps.
Mike

Hi Guys

Have a look at these - May prove usefull.

Ian

That answers alot if questions ! Thanks

So there is indeed a true 128 position virtual strip. Based on this stat is 100% correct that the animation
is just for show. Even though it only moves 5 positions the
machine just hides symbols to arrive at the generated stop point.
 

Hi

Had a long reply with the attachment but internet flaked out on me. Lets try again.

The info supplied is for the International version of Stinking Rich - My guess would be that the Math for IGT domestic products would be the same. If you can get the paytable ID from your machine you can confirm this. I think this is displayed under program version information in the test menu.

I had never given it much thought until you asked the question and this info does not conclusively prove that the five symbols on screen are consecutive numbers of the virtual strip. I have always just assumed that they are and like Mike said if you go into paytable test you get the same impression. Now that you have the info its pretty easy to confirm if your paytable Id matches that of the info. Just play a few games and confirm that the symbols displayed on screen match the consecutive symbols somewhere on the virtual strip.

Stat - Another aspect of video games that I have been wondering about. Conventional wisdom says that you you should always cover all the lines and reduce your credit bet per line as opposed to play max bet per line on fewer lines (That the advice I used to give to customers anyway) but that may be another false assumption since i have never actually tried to prove this mathematically. What I am interested in is how you would calculate how much better the RTP is if you do cover all the lines.

Ian

TZ, first, thanks and K+ for the PAR sheet.  :3- :3-
It certainly suggests that they are using standard strips with no randomization of symbols.



Mike, even though you can step through the strip order in test mode, it doesn't conclusively prove that they're not randomizing the strips. Only comparing live spin data to the strip listing can do that. However, I doubt that they are after seeing the PAR. It's a standard issue PAR, with no mention of randomizing the strip symbol order.

I believe that they would have to disclose any such actions. Besides, randomizing the symbol order exponentially increases the number of combinations, which would have to be disclosed on the PAR.



TZ, from what I remember of the discussions on the board a while back, some of the older video slots required a bet on every line to activate scatter pays and perhaps some other bonuses. I've seen newer games that no longer have that requirement. As long as every payout is proportionately available to the bettor, then there is no advantage to playing max lines.

You'll want to be sure that the scatter pays will activate anywhere in the window (i.e., on un-played lines). They should pay a multiple of your total bet, even if the bet is not a multiple of the number of lines. The same applies to bonus rounds. If there are buy-in bonuses (extra button), the same applies. The for-pay bonus must be available for less than max line bets and in proportion to the size of the bet.

Line bonuses should not make a difference, as long as they pay a multiple of the number of coins bet on the line they hit.

Disproportionate top jackpots matter. Of course, it's up to the player to decide if they're worth a max bet.

Keep in mind that given the same sized bet, more lines at fewer coins decreases volatility and increases hit frequency. Fewer lines at more coins does the opposite.

<EDIT> Correction.  I did some more figuring in the light of day and realized that this isn't always true.  :25- Fixed symbol locations can be designed to fall on either side of the volatility of a random outcome. Basically, there is no motivation to randomize the strip order, because doing so can only return one specific hit frequency and volatility index for a given weighting of symbols, and that would limit the designers' creativity to arrange the symbols to carefully craft the ebb and flow of game play that they desire for the theme.

Sorry about that. I can screw up with the best of 'em!  :200- :30- :30-

Munch, K+ for the question. :3-  I had never really thought about symbol randomization before. :71-

Thanks!   For me it makes the hobby that much more fun.

I love to know what makes things tick.