This post is meant to list some of the current possible improvements to our understanding of the game. Much of this is easily testable; I just haven't had time to fire up an emulator.

Post proofs if you have info.

BP Mechanics
===========

Combo Kill Multiplier: We need to figure out how the multiplication applies. Getting a multiplier on boss fights could be an enormous boon to BP progression if it can be done in a time efficient manner. Does the boss need to be killed after the initial person is killed to get the multiplier or does it just need to be one kill in the chain?

Combo BP Bonus: Do kill combos increase the base amount of BP per kill before multiplication?

Enemy BP Values: Tracking each enemies' BP value on kill.

Combat
======

Damage Values: Tracking the damage values of each players moves both singularly and in a combo. This can be tested with controls in training mode.

Power Stat: Testing what the Power stat does exactly. Does it affect only basic moves? Does it affect throws? Does it affect learned moves? Does an increase to Power stat correspond to a direct increase in damage per hit by the raised amount or is there a proration / decimal multiplier on the Power Up stat per move?

Move Set: Refining and recording the effects of moves, particularly damage values, bouncing direction, interesting attributes, and potentially frame data to maximize DPS / safety.

Bouncing: Determining how bouncing damage is calculated. Is it based on the initial hit that caused the bounce? Is it a base value or base percentage? Is it modified by the Power stat?

Analog Button Moves: Basically need to find out how "hard button presses" work. If it is just holding the button a certain number of frames, then it should be fine. If it isn't, then we would need to figure out what we wanted to do about emulator since controlling which moves you execute is a skill linked to the combat system. This could be bypassed in emulator.

Kou
===

Stealth Section: Comparing time taken to complete a stealth section vs fighting. My early testing seemed to indicate fighting was faster or on par and led to an influx of needed BP. Needs verification.

In-Game Timer
============

Basically we just need to test thoroughly how it works exactly on emulator, and then try to duplicate results on actual hardware.

Consistency of shieldashing without stuttering, good transitional movement, and good jumps are the real key. There is more to SotN than the clock rush.

I am more interested with what we want to decide about end of game timing. If we want to change that, it seems fine. I can say personally i will still be tracking IGT in comments and probably will have additional splits for the old timing methods.

So if a person who has WR wants to compete/give more validity to their WR and skills as a runner, they would have to grind out runs on PS2, PSP, PSTV, and PSX. Can my time be the top time for PS2 77k Silver Colored systems since their is differences between that version and other PS2s?

Or I guess we can just give out gold stars to leaderboards with 2 runs on each.

We have had similar talks in the Symphony of the Night community regarding splitting versions. My question has always been "Who benefits from this?". The usual response is so people can compete against each other, but that is kinda bullshit if there are only 2-3 people running a version. The real question becomes "Are there runs on sub-optimal consoles that are strong enough to merit their acknowledgement on the leaderboard seperately?". So far, only 1 run has fit that criteria in SotN, and the runner agreed it was silly to split the boards at this time. I won't speak for Parasite Eve in this regard. The only other reason would be if there is a sizeable difference in the actual gameplay/routing. To my knowledge, the only thing that varies in the gameplay itself is analog access to save ~15 secs total.

The only way I support this would be to ease set up for the future, but let's be entirely realistic. Parasite Eve is NOT a Mario game. The probability that there will be even 100 runners is almost non-existent. I really don't even see the sorting mechanism as a great excuse since the community is still extremely small. Good runners will know what a good run is, so I don't see what being at the top of a limited list does for people other than stroke their ego.

Thoughts (literally a copy paste from Discord)

1. You aren't "saving time" splitting at the earlier point if you have to watch the ending as well. You are still sitting and mashing text to at least the credits.

2. The credits are reasonably easy to anticipate length and return with time to spare.

3. I have always run on an empty/1 safety save memory card. You should never lose appreciable time on the final save menu.

4. I get the feeling most runners are still probably going to track their final save split. Is there a point to this change then really? I suppose some were already tracking the boat escape.

5. It feels extremely silly to split a 12 run leaderboard to me, but if we maintain an overall board and it is considered a provision for the future, then I guess it is ok. I just really dislike creating leaderboards so that people can place higher on the list because the dominant players don't play on the suboptimal platform.

6. If we were to change the timing to reflect the last point of danger in the run, then escaping the battleship seems correct.

7. If we make this change to the boards, what would be suggested for NG+ Good Ending?

Summary: Although I think these changes are premature, I think the changes seem reasonable enough to the end game timing. We would need to consider our stance on Good Ending. I think play to the credits cutscene would be the next valid option for timing. Removal of IGT is a lateral change for me; it can always be put in the submission comment anyway.

There is no real accurate way to know. Even if we calculated the loading differences, we would have no great way of figuring out time loss due to graphical lag. Time loss would also depend on if you could sub 8. If you can sub 8, then that entire chunk of time is irrelevant. If not, then it can cost time.

Regardless, try not to worry about time loss and just play. You can always look to invest in better hardware later on.

Over the last few weeks, I have been reviewing and verifying runs for the boards. After conferring with another mod, we have come to an agreement that certain quality standards of footage are integral to the correct moderation of runs.

Due to this, footage that has significant drops in quality of video and/or audio may be rejected. This includes but is not limited to inconsistent FPS, screen freezes, and audio drops.

Note that this is not meant to make speedrunning Guacamelee! an exclusive club for the tech savvy or those with high end computers. It is meant to make the verification process easier, faster, and more effective at weeding out questionable runs for further review and critique.

Those experiencing problems are suggested to look into options for improving their video/audio quality including:

1. Making OBS minimized to reduce stress on your CPU and GPU.

2. Ensuring you have correct graphics settings in Guacamelee! itself with regards to your monitor and computer power.

3. Consider disabling V-Sync in Guacamelee! This will produce screen tearing, but it will often lead to improvements in FPS and input responsiveness.

4. Reducing the bitrate of your recording or downscaling the resolution to something more suitable for your internet connection (if streaming) or CPU/GPU (if local recording)

I am not a mod for this game. Typically, you can pause the game in an RTA run, but the timer will continue. You may not stop the timer until you finish. This usually means getting a separate timer program that tracks real time rather than the game's built in timer.

You usually do not want to pause the game in an RTA run unless you are trying to do something very specific like pause buffering to ensure something will work consistently.

Have not done speedruns of this game as of yet. In general, 100% tends to lean towards things that are actively tracked in the game as a starting point. The next points of contention would be:

- How much does picking up X positively and actively impact the speedrun?
- How much time is lost picking up X compared to passing over it?
- Does picking up X add anything interesting to the speedrun over Any% or is it just checking off a box?
- Does X event/item have rewards that are not duplicated elsewhere?
- Would anyone actively run the game on the definition presented or is picking up X thing de-incentivizing people to run it?
- Should NG+ items or events be considered?

Thinking about these things, I can give some general thoughts.

In

- Completing all chapters in the main game (definition for completing can be altered if something crazy is found later on)

- Collect all Keys (Very clearly tracked in two ways, key position and lockers opened. Also allows for diversifying builds, Could be redefined to All Lockers if key duplication is a consideration. All Keys will also provide if some sort of access to Lockers without keys is found.)

- All Map Fragments (again, tracked in two places, clearly the game's designated collectible feature, and easily identifiable)

Debatable

- All notes/data logs (content that is tracked in two places and easily identifiable, but does not actively impact the game world in a meaningful way.If this is in, I would debate All Key Items should be in as well since it is in the same menu list. This would again be a provision for possible new routing that might skip certain key items. That being said, the game does not actively have a XX/YY number for completion on this.)

-Akumu Difficulty (NG+ and would actively discourage running. Not a significant mechanical change. It does add tension and should change routing or certain areas.)

- Med Packs (tracked in two ways. Does this impact the run particularly? I have reservations just because many games do not include health increasing items in 100%, many do though)

-All Bosses (defining a boss could be difficult. Provides for boss skips later on.)

Out

- Sebastian's stats (I would argue that while these are tracked, they are more busy work than anything else. Did you collect enough goo? Did enemies happen to drop more random goo? If the improvement is good enough to merit its inclusion, then it will be obtained.)

- Rescue the Lady, kill the pig, do the crow time trial, insert Easter Egg here (busy work and is more to show something off than actively add something meaningful to the content. Strictly for showing gimmicks rather than core game mechanics. If pig is in, then Crow Time Trial in Chapter 3 should be in also. Why not look at the guy in the window in Chapter 9? Why not shoot the Fallout 3 poster 3 times tihe every gun and ammo?)

- Kill All Enemies (tracked in some places, not in all. Creates a scavenger hunt for ammo. Actively fights one of the core tenants/merits of the speedrun which is enemy avoidance.)

This post is meant to inspire discussion among active runners rather than cause an issue. Use it for discussion points on things up for debate.

Every RTA does not pause the timer ever, hence the name Real Time Attack. It seems like redundancy to say no pausing. The alternative would be an IGT, In Game Time(r).

For RTA, the only questions that matter are when does time officially start and end for any given game.

You should export to youtube since twitch has a nasty habit of changing VoD link formats every few months.

Briefly looking at the route, Incendiary Bolts make a huge difference in how you approach arenas. What would be the alternative? It seems like you would need to route in significantly more ammo to deal with these encounters.

if only I could do a frame perfect dupe 200 times.

I did not pre-order The Evil Within. I only have the PC version through Steam. Is there any way to gain access to the FCP without hoping I get a legit code from a 3rd party site?

Has there been any routing done on non-FCP at all?

The Death Skip mentioned in the tutorial works on XBLA, PSN, and PSX versions of the game. Saturn and PSP Drac X Chronicles are separate games.

2 options

1. enforce a rule that an FPS counter has to be onscreen (like overwolf)

2. Simply add a parameter for 30 FPS vs 60 FPS. This dodges the issue, but allows people to sort by one or the other without completely splitting the boards.

Also of note, you could always grandfather runs in, or store them for posterity in a locked/sticky thread.

I do not see this as an accessibility issue. Running a third party program should only be allowed when it is actually necessary to increase the reliability of game stability when that is a primary concern. SH2 PC to my knowledge has no enormous crashing issues that would need some patch to remedy them.

If the IGT is in fact affected by the 60 FPS hack, it should not be used. Imagine I ran a digitally altered copy of OoT that ran at 60 FPS. The community would lose their shit. While not on the same magnitude of change, 60 FPS PC still constitutes an actual change in what is possible compared to the base game.

tldr; ban 60 FPS if it is proven to affect IGT.

How could it be easily done? Not banning untransformed while jumping. You would have to accidentally hold jump and accidentally hit two buttons in near succession.

I'm very on the fence about swaphopping. Yes, it is basic movement tech, but the fact the game registers you in a standing state makes me iffy on this.

Shop/Checkpoint abuse seems fine to me.

So far, SDG and Bubble Warp seem universally banned. Wall Hops in.

Still in discussion: El Infierno and Swap Hops