-
Posts
1,166 -
Joined
-
Last visited
-
Days Won
7
stevedude2 last won the day on February 4
stevedude2 had the most liked content!
Recent Profile Visitors
12,325 profile views
stevedude2's Achievements
-
Any further progress with this @redbags or has it died on the vine?
-
This is probably my favourite game of all time. The replay-value and simplicity is hard to beat. The counters underneath that dictated what items appeared on levels was groundbreaking at the time, as were the score mechanics. Universal appeal, wholesome message at the end... they really piled so much into this. A fucking tremendous video-game.
-
Yeah that's it - you have to take the WWWWT hits off as well as the WWWWW ones
-
You need to be looking for all instances of - WWWWP - 3x2x2x2x1 = 24 WWWWT - 3x2x2x2x11 = 264 WWWWJ - 3x2x2x2x8 = 192 WWWWQ - 3x2x2x2x10 = 240 WWWWK - 3x2x2x2x10 = 240 WWWWA - 3x2x2x2x9 = 216 Add them up and you get 1176 hits for 4OAK wilds out of the total combinations - 1 line. WWWW(JP) pays £25 so that doesn't count as a 4OAK wild win, as that only pays £20. The others all pay equal or less, so they're paid as 4 wilds. And obviously WWWWW counts as a 5OAK wild win. In essence the formula is 3x2x2x2x(60-2-9). But by breaking it down like further up it's easier to follow and see if you've caught everything. There's no other symbols in the game to have on the end, so that would be all hits covered
-
Morning Here are the hits I'm getting for the v1 symbol distributions. W is Wild, P is Pot (scatter symbol) - I'm 99.9% sure I've calculated them correctly. The RTP% of the line wins comes in at 81.3036%. That's without the pots obviously, because I don't know what the pot seed values are, the fill rates or the weightings for winning one over the other and so on. These hits are based on 1 line, so just multiply them by 10 if your analysis has 10 lines in play Let me know how you get on
-
No worries If you fancy putting the award values up here I might do a spreadsheet for the v1 symbol distributions over the weekend if I get bored. Then we can compare our findings
-
You're on the right track but not quite... You need to be looking at what happens on all 5 reels with the win calculations, not just the first three, because some wins like W-W-W-JP-JP might be worth more than 3OAK wilds and won't count as 3OAK wild occurrences. Your calculation might be correct if the awards for 3/4/5OAK wilds are much greater than 3/4/5OAK JP symbols because that means you don't have to worry about what lands on reels 4 and 5 (as long as it's not another wild on reel 4, and then also one on reel 5). But even so, you need to calculate your hits of any win across all 5 reels and then divide by 60^5, not 60^3, even if you think the last two reels are irrelevant - for example, if there are no wilds on reel 4 and 5. Even though the wins you're trying to find are 3OAK, they are still across 5 reels so you need to run your calculations right to the last reel and then divide into the total combinations across those 5 reels (60^5 or 777,600,000). It's best practice and means even the really rare winning combinations will be picked up. If you cut corners and miss a few occurences they can be a right pain in the arse to find later. So for this game, if the 3OAK wild prize is greater than the prize for 5OAK JP symbols (and therefore you don't need to worry about filtering out wins that don't count as 3 OAK wilds), it would just be this for one line - Hits are 3 x 2 x 2 x (60-2) x 60 = 41760 Total Combinations across all 5 reels (60^5) = 777,600,000 777,600,000 / 41760 = 18621.69 Hit rate with all 10 win lines in play is 18621.69 / 10 = 1862.17 My result differs to yours because you're not accounting for the fact that a wild might land on reel 4 and then, rarely, on reel 5 as well, and those combinations will not pay as 3OAK wild wins. You're effectively doing this - 3 x 2 x 2 x 60 x 60 = 43200 Total Combinations across all 5 reels (60^5) = 777,600,000 777,600,000 / 43200 = 18000 10 win lines is 18000 / 10 = 1800 I always find the best practice is to get all your hits/occurrences of a win based on one line in play, then divide it into the total combinations to get your hit rate for one line. Then you can divide that by 10 or whatever according to how many lines the game is fixed to. That way you're not rounding numbers up or down until the end. The good news is that getting the wild win hits correct is the most difficult part. The hits for the rest of the symbols, and then the scatters is a piece of piss by comparison
-
There shouldn't be a difference in the number of hits between lines, no. If you're stepping through the reels so each of the 777,600,000 combinations are analysed once, each line would see the same winning combinations as the other. The win line arrays could be anything as long as they occupy one position per reel, and you should get matching results between each one. On a simple game like this anyway. I suspect it might be an issue with something that happens rarely, like when you get a combination like W-W-W-POT-A which has wilds and scatters in the same line win or something. It might not be seeing that in the right way. You should be able to calculate the hits for each win just using the symbol distribution, yes. That's usually how it's done. You don't need to worry about the layout of the reelstrips; all that affects is the profile of the game. You can bunch all the symbols together in stacks for a lumpy game, or spread them out so you get a high hit rate - none of that will affect the percentage return, just the volatility. You need the award values though in order to work out the percentage contribution of each win type.
-
Calculating the number of hits for the pot symbols is a lot easier than the line wins - I'll explain why in a minute. I used the v1 symbol distributions from your pic above and got these numbers for the pots below (sorry - cell N32 is 664848) - The line wins can be a bit more difficult to do when you've got wild symbols on all reels. This is because when you're calculating the hits for a line win, let's say 4 of-a-kind K, you'll include the wild symbols in there as well, because wilds will substitute for K. This means you can end up with more hits than you've actually got, because if you look at the composition of those hits, such as KKKK-, or KWWK-, some of them will look like this - WWWK-, and also WWWW-. If the prize for a 4 of-a-kind wild line win is greater than that of four Kings (and it usually will be), you won't be paying that as a 4 of-a-kind K win, you'll be paying it as four wilds, as it's worth more. Those hits would be deducted from the number of hits for 4 of-a-kind K and added to the number of hits for 4 of-a-kind wilds. So you have to be able to compare the awards and filter out the combinations that are paid as wilds. If both awards are identical you can either count it as a wild win or a symbol win, but never both otherwise you'll have those hits appearing twice. You'll need the award values to be able to see what hits need to be counted as wild wins instead of normal symbols wins. The number 156005 for the 3 of-a-kind line win hits should end in a 0 if you've got 10 lines in play. A better approach might be to fix the game to 1 line and work out the hits that way. You'll find that when you add more lines the hits should increase linearly - extra lines are just extra ways to win, each line should return the same RTP%. There are exceptions; for example on a slot where there's a sticky wild in the middle of reel 3, any line going through that position would get more hits and return more money than ones not going through it. But this looks like a simple game and the number of lines will be irrelevant. If you use one line only you'll pick everything up in your analysis without having to worry about what symbols are outside of that line.
-
Maybe it's used for the win analysis? Like {K},{K},{K},{not_K,not_Wild},{anything} gives you all your hits for 3 of-a-kind Kings? Those win line arrays are unfamiliar to me. Usually you see the rows defined from top to bottom rather than each reel position, so a line that reads 0,1,2,1,0 would be a V-shape, 1,1,1,1,1 would be line 1 through the middle and so on. But then I'm not a software developer, never written a line of code in my life. All I've ever done in the past is hand over game maths, usually calculated on a spreadsheet (and the last time was a few years ago now), and then the dev imports it in whatever way they see fit. I'm assuming what you're working with is source code you've got from somewhere? You'd think there would be weightings for how often the reelsets get used really, because they could be absolutely anything - there's loads of ways of reaching the desired RTP% from weightings but they'll all play a different game in terms of profile/volatility and so on. Even knowing what the total of the weighting values added up to would help. Calculating the hits for the scatters is a piece of piss - (n = number of scatters on each reel | the *3 is because the reel window is three rows in height | 60 is the number of symbols on a reel) 5 of a kind - (n*3) x (n*3) x (n*3) x (n*3) x (n*3) 4 of a kind - ((n*3) x (n*3) x (n*3) x (n*3) x (60-(n*3))) + the other 4 reel permutations OOOXO, OOXOO, OXOOO and XOOOO, where X is the reel that no scatter landed on 3 of a kind - ((n*3) x (n*3) x (n*3) x (60-(n*3) x (60-(n*3))) + the other 10 permutations OOXOX, OXOOX... XXOOO Probably telling you what you already know but at least others can see as well to maybe stimulate some discussion. Best of luck with it
-
SymbolX is the scatter symbol I take it?
-
I've used Slot Designer in the past and that can spit out the correct RTP in seconds by stepping through each reel combination. It's ridiculous how fast it does it - must have taken a lot of work and refinement to get it to that standard. And that's through actual analysis, not simulation. Costs 10k Australian dollars for a year's license though. Or you could calculate the RTP on a spreadsheet. Seems a simple enough game - once you've done one reelset you can just copy the page 16 times and paste the other reelstrips in, and then put the weightings between each reelset in as well and voila! Maybe I read your post wrong though - do you have the RTP already and are simply trying to prove the maths via this program? It sounds easy what you're trying to do but I don't think it is. That's why simulation ends up being relied upon and not the true value.
-
Could have done with a background panel on those melon and grapes wins, especially the grapes. Also the hold buttons aren't quite centrally-aligned with the reels - 7/10. *runs*
-
No. Just found it - Cash n Grab
-
There was also a fruit machine game with a £1.50 jackpot that had skulls as the jackpot symbol. Anyone remember that? The name escapes me...