Hey everyone,

I'm MarthSR, and I've just been appointed as a new moderator for this game.

There were a few issues when submitting runs on certain boards, namely because the board would always expect 2 players even in the 1-player subcategory. So we've gone ahead with a complete overhaul of the board layout. While this effectively changes nothing in how runs are categorized, it did help with resolving the aforementioned issue and reorganizing the layout.

We've also clarified a few rules:

• In 2-player mode, the Player vs Player fight after defeating Willy does NOT count towards the run's total time. So no need to manage extra lives in order to "save time".
• All runs we are unable to properly identify difficulty-wise (either via showing the difficulty, or by determining how much damage the first few enemies take before dying) will be classified in the "Easy" leaderboards.
• MiSTer FPGA is now its own platform and has been added to the game's available platforms.

Should anyone have any questions, feel free to ask! I will be glad to assist however I can! :)

They really oughta look up the internet's history, and see how the approach they're taking "worked" for ebaumsworld...

Just use Brave as your web browser. It'll solve that specific problem.

No response, huh? Guess this should come as no surprise to anyone, honestly. The closure of transparent communication lanes, their rather cavalier handling of the site since they took over... this is sad, really.

Let's look at the NES part.

As per https://docs.google.com/document/d/1apP69VDIzUx-iVO_sj5YjTxBAE2ukHPa6kl33AOW9hA/edit , "Mesen, Bizhawk/NEShawk, puNES, Nestopia and FCEUX are the most acceptable emulators, listed in order of accuracy, most other emulators are not accepted. Retroarch should not be used."

Exhibit A: https://www.engadget.com/2014-09-28-retroarch-authors-retron-5s-emulators-code-violate-licenses.html

It states that RetroN5 uses FCEUmm. NOT to be confused with FCEUX. It doesn't support all mappers the other emus do, and its audio emulation leaves to be desired.

Let's look at the GBA part.

As per Exhibit A, the RetroN 5 uses VBA Next, which is banned as it is known to be inaccurate. This is common knowledge for all GBA speedrunners.

Let's look at the SNES part.

As per Exhibit A, the RetroN 5 uses SNES9X Next, an "optimized version with speed hacks". Not what I'd call accurate by any stretch of the imagination.

While the Genesis emulator used by the RetroN 5 seems to be accurate, that's one amongst a whole bunch.

Plus, the latency on all platforms is horrible. Plus it plays at 60Hz as opposed to the various systems' native frequency (ie: 60.0988 for NES/SNES, 59.94 for the Genesis, 59.7275 for GB/GBC/GBA, just to name a few), which results in either faster or slower gameplay when compared to the original consoles. Sure, it may be "trivial" when playing a 10, or even a 20-minute game, but when you play RPGs, those "trivial" milliseconds turn into seconds, which can ultimately make or break a WR.

And this has been known since 2014. So yeah, it fits the "well-known inaccuracy" label pretty well.

Given their well-known inaccuracy, they definitely are NOT original hardware. If anything, they're bad emulators.

https://www.speedrun.com/biomiracle

It's an ungrateful job you want no part of if it's just to show off!

Sure, the feeling of "being in a position of power" may seem appealing, but the appeal goes away once you understand your responsibilities, and the number of ways ish can hit the fan.

But at the same time, it's empowering when you actually have something to bring to the table when it comes to improving your community.

So overall, it depends on what your true intentions are:

• You want "ULTIMATE PAHWA"? This is gonna become a headache real quick!
• You genuinely want to help? It'll be fun, barring those few times you'll come across an unpleasant runner who wants to mess with you "because they can". You'll learn to make abstraction of those soon enough.

Might I add that the 4-mob lag from the arcade archives version also applies to the original arcade version (whether that be on PCB, MAME, or on FPGA techs such as MiSTer). The PC version doesn't have this as it really is a redux moreso than a port (see: sprite limitations that were commonplace for 80's games to avoid taxing RAM, the 16:9 aspect ratio, the transitions that are different from the arcade version, the HUD, etc.), which also explains the faster times on that version as opposed to anyone who'd play on an original PCB / MAME / FPGA.

For those reasons, I'd argue that the original arcade version should not be paired with the PC version as it would be unfair to pair a version that visibly runs faster than the other on the same leaderboard. From what I saw, the Arcade Archives version seems to be running at around the same speed, although AA tends to run games at 60KHz cycles (which is essentially the FPS the game runs at internally) whereas 80's arcade games always had slightly slower cycles by a few KHz (for instance, Double Dragon originally runs at 57.4KHz cycles). I had to split categories on Ninja Gaiden arcade between the original and the AA version due to the cycle frequency differences as it made AA run faster.

I wanted to run this game originally, but the pairing with the PC version made me hesitant as I'd rather play on either the original systems, or FPGA hardware replications of said hardware via the MiSTer FPGA (which is the closest thing to playing on an actual PCB as this platform is 100% cycle-accurate to the original material, with no speedup functionalities). There is simply no way anyone playing on the arcade version can compete with the PC times on the board, and it's a bummer that the OG version doesn't have a board separated from the PC version.

If the mods feel there is any merit in addressing what I just brought forth, it would certainly be much appreciated.

Very well then!

I definitely agree that the arcade version should have its own LB. There are several key differences between the two that wouldn't apply to what you folks know of the NES version.

I do have knowledge on the more technical aspects of the arcade version, so I'll go ask on the general forums how we could go about getting the arcade game its own LB. I'll link the thread here when done so you can add in your point of view on the matter. Perhaps it's going to hold weight in favor of having this done.

Meanwhile, if you agree, I'd say we can leave the current board as is, with any board revision being eligible. It would then be the player's prerogative to choose which version they'll run it on. Although I doubt there would be a whole lot of new GnG Arcade runners in the near future.

Thanks for your response on this, by the way! Much appreciated.

[EDIT: Good call on asking to state which rev the run is played on. As stated before, judging on the playstyle and what I see from the playthrough, the WR run is played on "Japan A", which seems to be the revision used on the PS2 Capcom Classics Collection.]

Hey all,

There is something I'd like to bring to the attention of the mods here concerning the arcade version.

https://tcrf.net/Ghosts%27n_Goblins_(Arcade)#Revisional_Differences

To sum it all up, there are gameplay-impacting differences between revisions of the board, which could potentially lead to runs not being on equal footing. Indeed, the japanese version of the game is different from the US/world (save for one revision), especially on later levels. For example, Satan bosses from stage 5-6 taking 4 hits to kill on the JP version, whereas on the US/world version, they take 8 hits. The number of unicorns (ie: level 1-2 bosses) changes depending on the revision. Furthermore, some versions allow for a glitch to despawn bosses from stage 4 and 5 by taking the screen as far as possible from their spawn point. And finally, some mob spawns in stage 6 are changed from one revision type to another.

While the other glitches mentioned on the link above are not relevant to speedrunning, the ones that are can make quite the difference.

Which brings me to my question: how should this be handled? Do we leave the runner the option to use whichever revision they like? Do we split the LB between 8-hp and 4-hp Satans (as the 2-unicorn variation really is a non-issue when you can easily d-boost to the platform above) in the event that some people would run it under the "harder" versions? It appears that the boss despawn can be done on at least one of each X-HP Satans variations, so I'd argue that's also a non-issue.

For reference, the WR seems to have been done on a "Japan A" revision. I personally plan on playing on Japan C to take advantage of the spawn glitch, as it seems to work far more consistently on Japan C for some reason.

Just thought I'd bring this up for the sake of transparency, and in the event that the arcade port ends up picking up steam at some point.

Thanks in advance! :D

Hey all,

Just wondering if the upcoming Jotego CPS2 core for the MiSTer FPGA platform would be allowed as a platform eventually. It is currently in beta, so obviously that wouldn't be an option right now, but given how generally accurate arcade FPGA cores tend to be even over MAME, I'd say it should be an option to investigate alongside the others currently under investigation.

Jotego has made various arcade cores, namely the Ghosts n Goblins and CPS1 cores, with the latter being accepted as a viable platform on the SF2 leaderboards.

Just thought I'd bring that option up. Thanks! :)

Hello,

Just wondering if the CPS1 MiSTer FPGA core is acceptable as a platform to play on. The main difference between standard emulators and an FPGA core is that the FPGA core is a hardware replication of the original hardware, whereas standard emulation attempts to emulate a platform through code.

The two main advantages are in terms of accuracy compared to other emulators, and no extra latency that standard emulators would incur.

Thanks for your response.

Hello all,

The MiSTer FPGA core for Alien vs Predator will be released shortly, and will be approved for future runs. They will still have to be classified as "Emulator", and will have to be mentioned in the run's description.

MiSTer FPGA is a system that replicates the hardware components of systems rather than emulate them through software, and is known for its accuracy and lagless gameplay. Ergo: it'll be the closest thing you can get to an actual CPS2 board of AvP. Therefore, as it will be far more accurate than MAME and/or FinalBurn, runs on MiSTer will be allowed.

Further information on MiSTer: https://emulation.gametechwiki.com/index.php/MiSTer

@Neural89 Except not all emulators have run-ahead.

What this essentially does to put it in laymen's terms is reduce input lag to a point that is even lower than with original hardware. This is why so many communities are banning emulators that use that feature as it effectively reduces the "skill gap" between players, making some of the harder strats easier to perform. RetroArch also has that feature, but there is a notification that appears when you enable it at least. Mesen has no indicators, which would make any sort of cheating using the feature untrackable.

@AkibaMomoro Have you come across the Twins glitch yet? Seems like for every twins fight I've had thus far, I was able to move before the fight started, and thus was able to kill them the second the fight "officially" started.

I basically use the "Noble Sacrifice" cheat to farm money.

The cheat has changed since the PS3/X360 days.

[While holding jump] up up down down left right left right block

@JSR_ From what I've heard, the Complete Edition board has been discussed for a while with speedrun.com staff. So it seems to be a matter of "when" rather than "if". :)

Thanks for the heads-up, WB! Much appreciated.

Meanwhile, I'll get my runs ready! :D