EAB/Lemon Super League 2019 Round 13: Minskies Furballs AGA

[john4p]

Ah, that explains it. So speed should be fine on a standard Amiga 1200.

[malko]

Originally Posted by Shatterhand

I just noticed all the discussion about playing AGA version vs ECS.
I read the game loads better gfx on AGA machines, but its the exactly same game.
I was indeed playing on real hardware, but an A600... so, ECS version. So I guess my score is not valid, hehe
But this may explaing the hard fluctuation on the framerate... maybe the game was designed around 020 machines and it was just "hacked" to work on ECS, but maybe very slowly/jerky on a 7mhz 68000.

Would you mind comparing the game speed between your real A600 & an emulated WinUAE A600 ?
Out of curiosity, would like to know if both have the same speed.

(this comment apply under emulation, not tested on real HW) When I played the ECS I didn't noticed any jerky animation but found that it had a better difficulty (linear progression spread over the 11 levels) in opposite to the AGA that has a higher difficulty already in the beginning. Thus, from this perspective, I would say AGA is a less well programmed game and may quickly become a frustrating game for players not used to such kind of game.

PS: your picture on this post is not displayed and when I click on it, I am redirected to a "login to google account"...

[john4p]

Originally Posted by malko

...AGA is a less well programmed game and may quickly become a frustrating game for players not used to such kind of game.

Easy definitely is too hard for me, I'll switch to Medium tonight. Hope that's easier.

[lifeschool]

Shatterhand - yes if the big scores only come from the fruit machine, the round is slight broken. It is possible to time the rolling tumblers, especially on a slow stock A1200, and this means people who can get far have a greater chance of huge score. It would be handy if someone can list the prizes in the bonus round, so we know how they did it. I know two or three cherries increases the score.

John, the game comes with its own AI, and this is rather dumb when played on a stock A1200, as it can only calculate maybe 10 moves in advance. If you play it on an expanded A4000 like me, the game thinks 40 moves in advance and can block you at every attempt. With the game in ECS mode, I suspect it can only think 5 moves in advance and seems easier?

Maybe we were wrong by trying to always go for the AGA versions this year, and with Exile AGA etc. The rules say that if a game is available as OCS and AGA, OCS will be chosen to make sure it is available to the most players, and also the game plays identically on all systems. I say lets go back to this rule next year, and then OCS/ECS is the standard.

[Shatterhand]

Ok, I made a few tests with Winuae.

A600 config, CPU "Approximate A500", chipset ECS Cycle Exact...

The game is as slow as on my A600. If I turn off Cycle Exact it gets a little faster, but not by much.

If I turn on "CPU fastest possible" the game runs at a speed that actually makes it feel more like Puyo Puyo and I guess that's the speed the game should be running.

With an A1200 setup at bare minimum for a A1200 (020 and "Approximate CPU speed") the game seems to be running at full speed, with either cycle exact on or off. So I guess any AGA setup will play the game at full speed, but a 7 mhz cpu isn't fast enought.

So I think my guess was right. The game was made with the mininum A1200 config as a base and probably just "hacked" to work on ECS machines, because it's slower than normal on a bare minimum A600. (The game is a lot more enjoyable at full speed... on the normal A600 speed, when I do a chain bigger than 2x it takes like ages before I the next piece falls, as I have to slowly wait for my side of screen to "drop" the pieces, make the animation for them dto dissapear, then drop them again, etc etc. It really takes a while and its kinda annoying. On the A1200 setup is as fast as a normal Puyo Puyo game and thats ok )



With that being said, if you are able to actually "see" the slots on the slot machine and get 3 grapes all the time you either have some amazing nearly inhuman eye-hand coordination or you are playing the game slower than you should.

[john4p]

^ Interesting, thanks for the tests.

Originally Posted by lifeschool

John, the game comes with its own AI, and this is rather dumb when played on a stock A1200, as it can only calculate maybe 10 moves in advance. If you play it on an expanded A4000 like me, the game thinks 40 moves in advance and can block you at every attempt. With the game in ECS mode, I suspect it can only think 5 moves in advance and seems easier?

Ah okay, that explains everything. I've played on an emulated 68040 "Fastest Possible" with JIT on my i5.
Because of that my "EASY" opponents were probably on Kasparov-level.

I'll change to cycle-exact 14MHz 68020-emulation from now on.

Thanks Dan.

[malko]

Originally Posted by Shatterhand

Ok, I made a few tests with Winuae.
A600 config, CPU "Approximate A500", chipset ECS Cycle Exact...
The game is as slow as on my A600. If I turn off Cycle Exact it gets a little faster, but not by much.
If I turn on "CPU fastest possible" the game runs at a speed that actually makes it feel more like Puyo Puyo and I guess that's the speed the game should be running.
With an A1200 setup at bare minimum for a A1200 (020 and "Approximate CPU speed") the game seems to be running at full speed, with either cycle exact on or off. So I guess any AGA setup will play the game at full speed, but a 7 mhz cpu isn't fast enought.[...]

Thanks for your tests Shatterhand !
For having played both till the end, I still think the ECS version has a better progressing curve regarding the difficulty (speed of the falling cats ).
As there is three modes available : easy, medium & hard, there shouldn't be any difference if you use ECS or AGA. Difference should be in the difficulty mode.

On the other hand, emulation speaking, that this game is coded in this way also brings a "good" side as there are at least 6 levels of difficulty :
ECS easy, ECS medium, ECS hard, AGA easy, AGA medium & AGA hard .

Originally Posted by lifeschool

Shatterhand - yes if the big scores only come from the fruit machine, the round is slight broken. It is possible to time the rolling tumblers, especially on a slow stock A1200, and this means people who can get far have a greater chance of huge score. It would be handy if someone can list the prizes in the bonus round, so we know how they did it. I know two or three cherries increases the score. [...]

Originally Posted by Shatterhand

[...] With that being said, if you are able to actually "see" the slots on the slot machine and get 3 grapes all the time you either have some amazing nearly inhuman eye-hand coordination or you are playing the game slower than you should.

The linked manual says nothing regarding the scoring system, so if I have observed correctly the score while playing I would say that big scores is the addition of several factors :
- the ability of the player to line up 4 cats of the same colour ;
- the ability to line up more than 4 cats ;
- the ability to line up the cats in a row (x2, x3, x4, etc.) ;
- and at the end, the slot machine.

I don't know if the speed at which you are able to fill up the opponent's left colum has in incidence on the score (?)

Regarding the slot machine, I am not Steve Austin ( [ Show youtube player ]) but it's feasible to get three grapes (4'000 points) or two grapes (2'000 points). But I never said it's easy AS it's not (I putted it in the last place in the above list on purpose). You have to get the fruit pattern for each row and "just" press (not too hard in ECS and with some luck in AGA) accordingly

Originally Posted by lifeschool

[...] John, the game comes with its own AI, and this is rather dumb when played on a stock A1200, as it can only calculate maybe 10 moves in advance. If you play it on an expanded A4000 like me, the game thinks 40 moves in advance and can block you at every attempt. With the game in ECS mode, I suspect it can only think 5 moves in advance and seems easier?[...]

Have you check the code ?
Regarding your comment, I better understand why, while playing the AGA version, I often complained to myself that cyberfishes & blocks were covering the cats I wanted to use with my next falling ones.
We as human player can only see 2 pieces : the falling one & the next one. If the computer is able to compute the next 40, 10 or even 5 next moves, there is an imbalance.
For a chess game, I don't mind that the computer "think" the next 100 moves as the game area is in front of my eyes as well. It's up to me to also "think" about my next moves. Here, we only see two pieces (hope you get what I mean ).

Originally Posted by lifeschool

[...] Maybe we were wrong by trying to always go for the AGA versions this year, and with Exile AGA etc. The rules say that if a game is available as OCS and AGA, OCS will be chosen to make sure it is available to the most players, and also the game plays identically on all systems. I say lets go back to this rule next year, and then OCS/ECS is the standard.

+1 for OCS/ECS as standard (but it should not exclude AGA only games from the competition).

[Shatterhand]

We as human player can only see 2 pieces : the falling one & the next one. If the computer is able to compute the next 40, 10 or even 5 next moves, there is an imbalance.

Dude, how did you manage such a big score thinking like that? It's like classic Tetris, even if you can only see the next piece you KNOW at some point you'll get an Z or an I and you plan accordingly. When you play Puyo Puyo you are setting up chains all the time... I usually try to build my screen like "If I get a blue I can chain like that but if I get a yellow I can chain like that". On ECS I never saw the CPU doing anything beyond a 2x chain, while I was doing 4x or 5x constantly. The Fishes and blocks fall on your screen because your opponent made a chain. So you have to chain more and faster than your opponent.

Puyo Puyo series has a "Counter-Attack" feature that, if you are about to be attacked, your next chain will first ELIMINATE the enemy attack instead of actually attacking the enemy... If your attack is bigger than his, the difference will fall on his screen (And if yours is smaller, then the difference will fall on your screen). On Minskies Furballs that's not how it works unfortunately.

I didn't test the game on Winuae long enough to see if the AI actually changed, but I found the game really too easy (and slow) on my A600. I wonder if on an A1200 the CPU would do better chains.

But just the fact the game is slower should make it easier too. The slots machine was also a lot slower on my A600 (Also on emulated A600) than on emulated A1200, and on A1200 it was fast enough that I really don't think you could constantly get 3 grapes just by skill/reflexes. (On the A600 is doable for sure).

And while I'd be excluded to play this because of the AGA version, I honestly think this game should be played on a 14mhz CPU. The ECS version on a 7mhz CPU is too slow.

[john4p]

14,870 x 2 = 29,740

Don't feel much of a difference. Still lost to the 4th opponent on Easy and again on Medium which is what this score is from.

[malko]

@Shatterhand : A good score doesn't prevent me from being critical
Will try to free some time to test Puyo-Puyo on MAME.

[Shatterhand]

There's another Puyo Puyo clone for Amiga called Super Foul Egg, but I found that one to be really failed on many aspects. This game here is way better.

[john4p]

Also on Hard got up to the 4th stage:



7,948 x 4 = 31,792

[malko]

John, do you feel any difference between easy, medium & hard (I have not tested hard - but found easy & medium quite similar).

[john4p]

Not really felt a difference between easy and medium either, but on hard there's a speed increase in stage 4.

Maybe the only difference between difficulties is at which stage the speed is increased?

[malko]

I gave a go to the Hard mode... and reached level 4 : 82'456 (20'614 * 4)


By the way, I didn't try to get grapes from the slot machine as I only wanted to see how far I can go in hard mode.

Originally Posted by john4p

Not really felt a difference between easy and medium either, but on hard there's a speed increase in stage 4.
Maybe the only difference between difficulties is at which stage the speed is increased?

Instead of saying that the speed increase I prefer to say that more blocks & fishes are falling even if the opponent does not line up cats. Also I received a 'star' malus that makes a lot of cats falling faster than with a 'speed-up' malus . 3 cats of the same colours were also regularly blocked with blocks & fishes making the line up with a fourth really difficult or impossible.

Could be nice if we can get our hands on an original manual. Mabe the scoring/bonus/malus system is more detailed ?

[Shatterhand]

I am willing to give this a proper go on Winuae even though I really don't like using Winuae for gaming too much, just to see if I can clear up those doubts and actually make a proper score, just because I love Puyo Puyo.

This guy:

https://www.youtube.com/channel/UCW_...Jh4VLgS3CFXU3Q

Commented on a video about this game saying he is a designer of the game. Maybe theres a way to get in touch with him and ask him directly?

[malko]

Originally Posted by Shatterhand

I am willing to give this a proper go on Winuae even though I really don't like using Winuae for gaming too much, just to see if I can clear up those doubts and actually make a proper score, just because I love Puyo Puyo. [...]

Cool idea
Waiting your score

[rexsu]

436,166 (218083x2)
The last enemy, it looks like tetris speed after 200 lines

[john4p]

Hard is too hard. After the speed increase in stage 4 there's another in stage 5 which makes it nigh impossible. I had the exact same thought there as rexsu's meme pic says.



11,300 x 4 = 45,200

Won't try hard again, only medium.

[malko]

Well done John. Nice performance !

I have seen rexsu's score. Looks like I have to try another go...
F__k (wanted to turn the page of this game )
Anyway, well done rexsu !