Real speedruns don't necessarily adhere to the same rules that TASes use; in many cases it is done to ensure authenticity, which is mostly inherent to our regular process of making/submitting/publishing movies.
That being said, some unassisted runs do skip normally unskippable FMVs, see Diablo for a recent example.
Oh yeah, I still can't forgive MAME coders for adding artificial input lag in order to simulate arcade cab button strokes more accurately. So now we have your control device of choice input speed summed with that artificial lag to produce something nonsensical as a result. You know, fuck this idea.
Well, Zurreco had started an any% even before you joined, and quit after something like 200k rerecords due to luck manipulation failure, marketed under hard mode obsoletion precedent [1]. :)
Even more disheartening was the fact that it got obsoleted by using a single new trick… twice.
Do you mean "hardest to perfect" when you're saying "hardest to TAS"? There are many games like that, and most of them have something to do with thinking. Arkanoid comes to mind, simply because it's near-impossible to tell what positions and angles will result in an optimal solution. Scripts, bruteforcing, and memory addresses aren't of much help here. Yoshi's Island, on the other hand, can benefit from them.
PSX version could still use warrior, and not sorcerer. It's obvious there's not much to expect from the sorcerer run except some very minor optimization.
Not to mention that they are mostly irrelevant as the player skips plot elements serving as keys to those FMVs. I, for one, would rather play the game to see it all, not watch a speedrun.
I don't really like how you decide to spoil plot twists in the second post about the game. Seriously. RPGs are about the most plot-heavy genre of games period, and it's not good to do such a thing without any notification whatsoever.
Wow, this quiz is awesome. I recognized some of the tunes but didn't manage to identify most of them. Too hard for me overall.
Tracks like #4 and #8 are pretty cool by themselves, btw.
What non-bullet hell shooters have you beaten?
MAME: Armed Police Batrider, Battle Bakraid, Battle Garegga, DonPachi, DoDonPachi, Dragon Blaze, ESPRa.De, GigaWing, Guwange, Mars Matrix, Progear, Raiden Fighters Jet.
PS2: DoDonPachi DaiOuJou, Espgaluda, Ibara, Homura, Mushihime-sama, Psyvariar Complete Collection, Psyvariar Revision.
DS: Ketsui Death Label.
PC freeware: Blue Wish Resurrection, rRootage, Noiz2sa, Titanion, YGS2000 (dodgesim), nomltest, sdmkun (dodgesim), Hollow World of God (Touhou rip-off), XOP Black.
PC commercialware: Touhou games (especially Shoot the Bullet), Shikigami no Shiro 2 & 3, eXceed 3rd -JADE PENETRATE-.
There are some others as well, but I don't remember everything.
Yeah, I understand it's what it's for. But in practice it just doesn't serve the purpose. The reason you're getting 200k on Beginner and I got over 1.1 million on my second complete playthrough (that's not playing for score, paradoxically) is a clear indicator. I'm reasonably sure I could do over 2 million if I survived just a while longer.
First of all, reduce the inflector's bullet sucking strength. It will make it actually useful in a way that doesn't kill you most of the time. Next, make the multiplier penalty for a hit a percentage of the current multiplier. Say, 25% for a shield hit and 50% for a life hit. This will make the shields useful. You can also assign static amounts of points for passive grazing. Hey, there are lots of ideas to toy around with. :)
An average shooter gives the player three lives (one active and two in reserve) with one or two life items and two or more score extends. It's very rare when a game gives more than six lives in total in one playthough without being too easy.
Bear in mind that getting hit inadvertantly breaks your concentration for a short while — unless you totally saw that coming and knew what to expect. Cut the player some slack, you don't have to abuse them.
I am absolutely advising you to play more games like that. I can give you a list of them to check out if you want.
Lots of criticism incoming, I figure it's better than leaving it unaddressed. Hope you won't hate me for this.
First of all, my main issue with the game. With the recent (~1.5 decade old, actually) tendency of drastically increasing bullet count in most shooters and the beginning of bullet hell era started by developers like Cave, Raizing and Takumi, many people started getting wrong ideas about this subgenre. The nature of bullet hell games (and this is taken from the words of many Japanese developers) is to give the player a false sense of great difficulty — most bullet hell games are actually easier than your average Gradius III era snipefest — hence the large amount of gratification when they endure and overcome. And these games are carefully crafted around the idea of this balance of difficulty and gratification, with none of them being unable to go through with practice. There are thought-out scoring systems which are meant to make the game fun and keep the challenge for advanced players who don't play for pure survival anymore.
Western people tend to think it's mainly about bullet count and small hitboxes, and that such games are not really expected to be won (at least without using continues). This game is full of such wrong ideas, unfortunately. I'll elaborate now.
The scoring system is bad. Why? Because it revolves around one basic idea: do not get hit once. Once you get hit for the first time, you may as well restart, because you won't be able to make up for the loss when your multiplier goes over 70x. About 65% to 90% of your score comes from this initial no-miss period. Then the kicker: eventually bullets become unavoidable. Scoring system just dies there because it isn't needed anymore.
Now on getting hit. I know it's cool to have some story in a shmup (wait, story in a shmup?) providing the rationale for its design choices, like with that glass cockpit and stuff, but the hitbox position is skewed from the geometric center of the sprite, making it counterintuitive. Pre-emptive counterargument: when you're playing a shmup, you're not looking at your hitbox. That is, unless you are a beginner, in which case your stare is concentrated on the hitbox. I don't like to have to look at the hitbox just because it's so hard to pinpoint it without looking directly.
The hitbox problem doesn't end there, though, because the bullets are even worse. Basically every bullet hell game — from DonPachi to rRootage to Gradius V (not much of a bullet hell, I know) has bullet hitboxes smaller than their sprites, akin to the player's craft's hitbox. There is a sane ratio of bullet hitbox to player's hitbox, and keeping them close is a very good idea. Here you're having a two-pixel wide hitbox that even seems to connect with bullets that haven't hit it, and humongous bullets in comparison. At loops 6+, they tend to form impenetrable walls, adding further insult to the injury.
Next, the amount of lives. You're basically giving the player 9 (!!!) lives initially with the ability to get more. Considering the scoring system that urges you to restart upon the first miss, and the eventual undodgeability of the patterns, it seems not only needless, but also silly.
But what's more silly is the invulnerability period… or lack thereof. Which is why it's possible to lose several lives in the span of one second. I wonder where you got that idea, because all developing institutions (period) consider it a bad game design.
The inflector. A neat idea in concept, unbalanced in realization. The hitbox for grazing the bullets is small, while the bullets are sucked in from half a screen away at a huge speed. Considering their size, every use of the inflector is a potential invitation to ruin your score, rather than raise it. Who needs a 2x multiplier increasing speed when it's safer to just keep the multiplier?
Screen orientation. It is a vertically oriented game where the enemy spends most of the time at the top, and the player at the bottom. However, the distance between the top and the bottom is smaller than between the sides, making it harder to dodge shit at the bottom, compared to the sides. Not really clever.
The first 2-3 loops are dead tedious. I get tired of doing nothing before something challenging appears and start hitting random bullets I could have easily avoided otherwise due to misconcentration.
The patterns. They are too geometrical for their own good. I realize this is easier to code, but it's basically either random hailstorm, perfectly symmetrical streams, or aimed sprays. See games like DoDonPachi DaiOuJou, Mushihimesama Futari, Shoot the Bullet, rRootage, and Ketsui for examples of clever and challenging patterns. Or dodgesims like YGS2000 or sdmkun, which incorporate many of those.
The last but not the least. There's currently no incentive to purchase the game. The beginning is relatively small and fairly lackluster. People can literally get much more from a free dodgesim, let alone the multitude of good free Japanese doujins. There is no official online scoreboard or anything like that. There is no ability to save replays and put them online for others to see. The game is pretty forgettable in its current state. It's far from a commercial product.
Also, two questions: what are your scores on the first courses available in the beta, and what are your favorite shmups (and why)?
Even if you fall slightly short of beating the current run, there's always the possibility to add your submission to the published movie's description.