What CAVE did.

[Rob]

The stage order in Psikyo games makes very little difference in score or difficulty level.

[Obiwanshinobi]

I for one, with IL-102 (not even fully powered up), found the New York stage boss in Sonic Wings Special a breeze even on Hard, no matter the order of stages, whereas Paris as the fourth stage was significantly more menacing in my book (on Hard, that is).
Parhaps for you it makes little difference, but can you PROVE your best scores aren't as high as they are partially because stage order turned out favourable for you during these particular runs? Only by bettering them under different conditions methinks, but beyond some point this might be hardly possible.
As I understand it, the beef some people have with randomisation and playing for the score is that different runs are only 100% comparable score-wise if the conditions were identical. To meet these conditions, games like the aforementioned should have a separate scoreboard for each and every one route possible.

[RNGmaster]

The stage order in Psikyo games makes very little difference in score or difficulty level.

Unless you really don't want to see Sand as Stage 4 in Dragon Blaze - that shit is much harder than any other version of Stage 4.

[Obiwanshinobi]

Dudes, random elements in games doesn't mean they don't require memorisation. Muscle memory, memory of control dynamics, different patterns and attacks in later levels, how far to fire ahead to catch that blasted space invaders ufo etc etc.

Aye, many cool dramatic situations may be happening in arena shooters or the likes of GTA thanks to their non-scripted nature, but to fare well at those you still need to know the intricacies of weapons, vehicles, maps layout... You can win by luck a deathmach of Quake against people more familiar with the game than you, but how many times in a row?

[BPzeBanshee]

When I think of randomness I think on two different levels:

1. Random where the entire game is based on randomness and nothing else
2. Controlled random figures on a small level to maintain consistency while keeping things interesting.

In case of the latter level the first thing that comes to mind is ESPRade's stage selection. Pressing the A key as an unsuspecting gamer when picking your character will get you a random order for the first three levels, while pressing the other two buttons will have different presets.
Personally when I code bullet patterns I like to use the irandom(x) function for things like spray attacks, but then that spray attack happens at the same time within the same attacking phase of the enemy or boss, and they're all relative to the player. Keeps my eyes interested but is still predictable. That does however, kill off exact playback recording support in Game Maker as far as I know, but I can't find any GM game that actually does playback recording properly anyway.

[BIL]

The one thing score is absolutely not supposed to indicate is good fortune.

Variable order of stages in Psikyo shmups and Sonic Wings series makes fortune quite a factor in those games, and yet nobody in their right mind says "your run was better than mine because you had a better luck". Mutable conditions are something you must take into account if you want to get better at those games.

I don't play the Sonic Wings games, but for stuff like Dragon Blaze and Gunbird 2, if you're better at some stage orders than others you just need to practice. A player's most hated stage will always be the same the next time it's encountered. And as Rob said, the impact of stage variation on scoring potential is negligible.

[TrevHead (TVR)]

Aye, many cool dramatic situations may be happening in arena shooters or the likes of GTA thanks to their non-scripted nature, but to fare well at those you still need to know the intricacies of weapons, vehicles, maps layout... You can win by luck a deathmach of Quake against people more familiar with the game than you, but how many times in a row?

Randomness might be fine in GTA, TF2 and maybe arena shooters where competion is more about winging it. Its doenst work as well in shmups due to the way players analyse stages and attempt to maximise their scores (perfectionist play). The main reason is that for many shmuppers playing for that 1CC or top score on the leaderboards only need to do so just that one time and its set in stone.

[Obiwanshinobi]

I don't play the Sonic Wings games, but for stuff like Dragon Blaze and Gunbird 2, if you're better at some stage orders than others you just need to practice. A player's most hated stage will always be the same the next time it's encountered. And as Rob said, the impact of stage variation on scoring potential is negligible.

I don't disagree, by my point is - the game imposes random conditions on you in a way that might affect difficulty of the run to an extent. If people who actually play these games competitively don't mind, it's just a proof it can be done well. By no means trivial achievement as you wouldn't want first few stages of any arcade game to be equally difficult, would you? Those games mix stuff randomly, but still have something of a difficulty curve akin to more conventional shmups.

[rockaroller]

you guys means people who dont like the 1cc ( I suppose this means 1 credit coin or credit continue? im wrong? ) or people who don't like after beat the game try to play again and again and again...watching the same everytime............. and again and again....just to improve, just for the scoring !!.....that kind of gamers are casual???
I think people who like many shmups since a longgg time shouldn't be considered casual gamers.

I don't think it's really needed to label shoot'em'up players, but if we really had to, I think we'd be, IMHO, the most hardcore among the hardcore type of gamers.

[BIL]

If people who actually play these games competitively don't mind, it's just a proof it can be done well. By no means trivial achievement as you wouldn't want first few stages of any arcade game to be equally difficult, would you? Those games mix stuff randomly, but still have something of a difficulty curve akin to more conventional shmups.

Sure, I agree it's possible to include random elements with some sensible limitation, and Psikyo's use of more rank-dependent opening stage difficulty is certainly one example. Another would be Garegga bosses tending to use randomly-selected attack patterns, ensuring the player can't reliably hide behind panic-bombing.

Then again, as I've said before in this thread, it's never been the principle of more free-form shooters I find distasteful, so much as the misguided if not moronic assumptions people tend to make about high-level play of the more static ones.

[BER]

I don't play the Sonic Wings games, but for stuff like Dragon Blaze and Gunbird 2, if you're better at some stage orders than others you just need to practice. A player's most hated stage will always be the same the next time it's encountered. And as Rob said, the impact of stage variation on scoring potential is negligible.

For Gunbird and Gunbird 2, this impact becomes significant once you approach a record-breaking level. In both games, you get to see all four random stages in 1-1, 1-2, 1-3, and 2-1. Then for 2-2 and 2-3, you'll draw two of the three stages you saw in the first loop. Since each stage's scoring potential is different, you'll want the two stages with the highest potential to show up in both the first loop and second loop. Otherwise, it'll be difficult to make up the difference from stage 2-3. It may be even mathematically impossible if the records are high enough—TAD-NETA, former record holder for Marion, said something like this in his site, which is gone (the URL was http://www5e.biglobe.ne.jp/~tad-neta).

Imagine spending 20 minutes having the perfect run through stage 2-1 in Gunbird 2, only to see a weak scoring stage like the English castle or the Japanese snowed-in town in 2-2 or 2-3.

I suppose this wouldn't impact many of us, since we have a long way to go to beat the records for the Gunbird games. But to the motivated Gunbird player who'll read this years from now, be prepared to restart your game deep into your run.

[esreveR]

I like CAVE games because TYPE A

Actually, were they the first to come up with Type-A ships?

[BIL]

Imagine spending 20 minutes having the perfect run through stage 2-1 in Gunbird 2, only to see a weak scoring stage like the English castle or the Japanese snowed-in town in 2-2 or 2-3.

I suppose this wouldn't impact many of us, since we have a long way to go to beat the records for the Gunbird games. But to the motivated Gunbird player who'll read this years from now, be prepared to restart your game deep into your run.

I forgot about GB1/2's odd stage distribution in the loop. I stand corrected BER!

Dragon Blaze's even distribution of its stages seems a lot less arbitrary a decision by Psikyo now.

[Acid King]

Changing speed instantly and switching between atack modes wasn't uncommon in console shmups from the nineties. Hyper Duel (1993) brought that to the arcades.

What games? I just loaded up Hyper Duel and that has a separate button for each shot and the attack modes are like separate settings. If you tap B, you stay in mech form and move more slowly even when you aren't firing.

[Obiwanshinobi]

Changing speed instantly and switching between atack modes wasn't uncommon in console shmups from the nineties. Hyper Duel (1993) brought that to the arcades.

What games? I just loaded up Hyper Duel and that has a separate button for each shot and the attack modes are like separate settings. If you tap B, you stay in mech form and move more slowly even when you aren't firing.

Hyper Duel has got two attack modes, two "bomb" modes, two speeds - all controlled with just two buttons. It's a neat simplification of control schemes Soldier Blade, Super Aleste, Thunder Forces, Gate of Thunder, Eliminate Down and the likes had. More buttons on console joypads encouraged development of nuanced control schemes in all sorts of games, including shmups (Einhänder and R-Type Delta being the pinnacle of the fad), but arcade games need immediate accessibility. Even the SNES joypad alienated quite a few gamers with its complexity and some people still wish games had just two buttons for attack and jump etc. Cave's solution was original as far as I can tell, but Hyper Duel executed the very idea earlier.

[Acid King]

Hyper Duel has got two attack modes, two "bomb" modes, two speeds - all controlled with just two buttons. It's a neat simplification of control schemes Soldier Blade, Super Aleste, Thunder Forces, Gate of Thunder, Eliminate Down and the likes had. More buttons on console joypads encouraged development of nuanced control schemes in all sorts of games, including shmups (Einhänder and R-Type Delta being the pinnacle of the fad), but arcade games need immediate accessibility. Even the SNES joypad alienated quite a few gamers with its complexity and some people still wish games had just two buttons for attack and jump etc. Cave's solution was original as far as I can tell, but Hyper Duel executed the very idea earlier.

It's similar but not really the same. The buttons in Hyper Duel effectively are mode selections that change the essentials of the ship. It changes not only the shot and speed but the shape of the player and the location of the hitbox. The other games used selectable static settings for speed and speed wasn't tied to shot type.

Cave's doesn't function as a switch that sets the speed, it only slows the ship while actively firing a specific shot. The ship stays the same but momentarily has tighter control and more fire power. That specific change in control I'd say is the difference maker that opened up design opportunities for both score systems and bullet patterns.

[Obiwanshinobi]

Er, "more firepower"? I thought it just switches between spread and focused shot, both being as powerful as many power-ups you've gathered.
I agree that no shooter before DP I know had a "score system" as, um, strict as DP (and there's no wonder you need to take attack modes into account if you play such a game for the score), but that does not impress me at all. Mars Matrix did that in a way more appealing to me.

Cave's doesn't function as a switch that sets the speed, it only slows the ship while actively firing a specific shot.

Does remind me of Hyper Duel nonetheless. The difference being hitbox size, but this alone adds arguably as much "depth" to the mix as chaining in DP.

[Acid King]

Er, "more firepower"? I thought it just switches between spread and focused shot, both being as powerful as many power-ups you've gathered.
I agree that no shooter before DP I know had a "score system" as, um, strict as DP (and there's no wonder you need to take attack modes into account if you play such a game for the score), but that does not impress me at all. Mars Matrix did that in a way more appealing to me.

I meant that it was usually the stronger shot type when held. And while you may not be impressed with the change it introduced to Donpachi, that they've used the shot/laser/slowdown control mechanic, what, more than a dozen times now and the varying ways it has played into the score systems, it's quite obviously an important part of what Cave did.

[Hagane]

The stage order in Psikyo games makes very little difference in score or difficulty level.

In Tengai if you get the sky stage you are pretty fucked score-wise, especially if you get it along with the temple stage. Stage order wouldn't be such an issue if you played all four of the initial stages instead of just three. Getting the wrong stages can hurt your score badly.

Also, scoring high is really hard in the woods stage if it comes in 1-3. Still one of my favourite games but I would take random stage order away if I could.