The Metronome Glitch Nobody Found After 15 Years

[@BigNation1]

To my knowledge this is the first documented glitch for PBR regarding Metronome, enjoy! This is my final video of the year. Thank you for your continuous support! Feel free to check out the archives channel where we continue to document PBR like highlighting tournament matches. These take place in the Discord server. Let’s revive the game 🥲

[@jacobkerner2593]

Am I questioning why a pokemon battle revolution video popped into my feed? Yes. Am I now remembering my favorite 3 pokemon games (this, soul silver, and platinum) yes. Time to try and find a Nintendo wii and a copy of the game to play with my pokemon platinum on the Nintendo DSI

[@BigNation1]

Now that’s gaming!

[@Blazingflare2000]

It's not so much a glitch as just a small oversight that only comes into play because of the extreme lack of variation in moves and teams. PBR just either uses the same starting rng seed on game boot, or sets the same starting rng seed at the start of vs matches, and then seeds all rng rolls off the previous rng roll. Either way, it really isn't an uncommon thing for most games to do. PBR probably doesn't have much or any elements that roll rng in the menus, and players that were joining the tournament likely quickly mashed through the menus as fast as they could, and as a result a few of the tournament rounds started off with the rng seed on the same step. And since everyone only used the same exact move, it had the same exact outcome. And the reason why the match rng would slowly diverge away from each other is because the pokemon doing their alternate idle animation is based off of rng, so how fast the players selected their moves ends up affecting the metronome rng. If you decide to do another metronome tournament in the future and don't want this to happen again, you will just need to ask each player to do a single offline random battle before they join the tournament, or tell everyone that they need to randomly wait 30 seconds to a minute before they select their first attack in battle. That will guarantee that the rng seeds should diverge far enough that you don't see the same moves again.

[@KrakatonMain]

This seems like a good time to finally mention something I've noticed in my journey back through the games. I think Pokémon's hidden personal IDs are factored into RNG for moves, abilities, and items used by that INDIVIDUAL Pokémon. And I have a few Pokémon who support this theory. First off, meet ShakeNBake the Wigglytuff, who I caught in my playthrough of Gold on the Virtual Console. I caught her as a level 3 Jigglypuff, and in GS, that meant she started with 1 move: Sing. 55% accuracy, right? Welp, ShakeNBake' hit RATE with Sing is more like...90-95%. She almost never misses with that move. Heck, it's nearly as reliable as Spore. And yes, I discovered this during my many balls thrown whilst catching her back then. Secondly, we have Jell-bro, my East Sea Gastrodon from Diamond. It took me a while to decide on a held item for him, but eventually I settled on a Quick Claw. This item is meant to have a 10% chance to activate, but Jell-bro gets it about 50-60% of the time, and it's always when I need it most. The amount of times I've said "... but it's okay bc Jell-bro outspeeds" followed immediately by the Quick Claw animation is HILARIOUS. In my exact words (as said animation plays): "he 👏 wuvs 👏 his 👏 Quick 👏 Claw! 👏" He is so ridiculously reliable with it. Why do I think this RNG is tied to them? Well, I transferred them into Home, along with my other team members. So these are Pokémon from official copies of the games (not ROM files, which can have occasional glitches even in normal playthroughs for any game), and they have been ported directly into both Brilliant Diamond for the Battle Tower and Scarlet for the Blueberry Academy E4 challenge. In both of those games, these same two Pokémon have had EXTREMELY consistent results that mirror that of their original games. Now, this is not to say that some Pokémon always get better dice rolls. After all, my experience with Focus Blast has never been good. And most of that largely comes from my first playthrough of Black Version, where I believe I used it with the same one Pokémon the whole time. Across a few dozen attempts, I think I landed it once, MAYBE twice. I think each Pokémon uses its personal ID as a factor in RNG, and each thing handles that differently. But for Battle Revolution, just like Stadium 1 and 2, you do not "move" your Pokémon into the game; you "copy" them. Why is that important? Well... what if it's only copying "surface" data? Species, stat values, level, moves, ability, item, and nature. At that point, the game may not be copying the personal ID value. And if that's the case, then it needs a filler number for the math to run on. So what if it just generates A number upon startup and uses that for ALL Pokémon in all fights until you restart or cycle the power? Then, you would see EVERYONE'S Pokémon getting similar results, like omni-boosts activating frequently or scarcely, or Focus Bands constantly triggering. Besides the two Pokémon I mentioned, I have plenty of other examples, but those two made for the best demonstration due to just how consistent they are. Hopefully this can be enough to point an investigation in the right direction. But this tournament may very well be the puzzle we needed to really confirm my theory!

[@JG0NE]

Reminds me of early stages of SV competitive where someone discovered that Sheer Cold would always hit if used after Fake Out/Air Slash/Icy Wind 😂

[@BigNation1]

Wait what there’s no way 😂😂

[@Magikarpador]

online battles were all in the same seed for a period.

[@UncalibratedAimbot]

Like the other guy said, it wasn’t about using a specific move first. It was that online battles always used the same rng seed so sheer cold would always connect on the second turn no matter what. Thankfully they patched that after a few weeks, but the incompetence was jarring.

[@TwiliPaladin]

​@@UncalibratedAimbotActually, it was both. Sheer Cold would always land on the second turn if Fake Out was used on the first.

[@slimestudios9748]

wasn't it actually if Sheer Cold & co. were the third action taken on the first turn?

[@mnm1273]

RNG is best understood by the speed running scene. I think some games always start with the same seed, and certain player inputs inputs modify the RNG over time

[@BigNation1]

Which is super interesting to think about when talking about the inputs that’s possibly effecting it too. Perhaps players just made it all the same as each other somehow without realizing?

[@mnm1273]

@@BigNation1 It seems likely, the input windows are generous enough that speedrunners can force an RNG window and if you're doing a pure metronome battle they might be simple enough that people do the same. After all you'd just start the battle and the select the only move as fast as possible most of the time. As battles drag on the odds of different input timing increases and so it slowly diverges. I don't know which inputs impact RNG and which don't so I can't suggest a test. But I think RNG in the mainline games is well documented (it changes from gen to gen and sometimes from game to game) so if you're interested I'm sure you could set up a test

[@GeorgeDCowley]

@mnm1273 I guess any switches would affect it.

[@nicholasapodaca9886]

Fire Emblem (GBA) is always the same seed. You have to burn RN's to get random outcomes.

[@pheeeshy]

@mnm1273 this seems reasonable

[@SethAbercromby]

Seems that while a bit cumbersome, for future PBR tournaments you could make a rule that each player needs to do at least one offline battle before moving to the online mode. The differences in player inputs and game decisions should diverge the seeds far enough where battles are sufficiently random, but that might warrant some further investigation.

[@BigNation1]

Yeah this would be a good idea. Though from what we’ve seen so far in general battles that weren’t metronome it all played out just as fine as you can expect it to.

[@CytricAcid]

reminds me of the glitch in scar/vio launch that caused all online battles to use the same seed, meaning that correctly timed usage of a ohko move would always hit, etc etc would be interesting to see if that's the case here as well. I wonder why this glitch was not found in the same amount of time it was in Gen 9? I don't think many official tournaments were hosted in PBR were they? Probably why it wasn't as well researched until now. (Man, I miss these stadium games...) To be specific-- is the battle seed the same every time, or just the first game after booting up? Does this sequence of events change if you reboot the Wii? Whose Wii is used to determine the battle seed? Some can be answered easily, others may require digging into the code, but I think these answers can be found empirically and could shed some light onto how PBR handles random number generation.

[@BigNation1]

I had no idea about the Gen 9 glitch, that’s interesting! Correct not much was played on PBR back in the days for official VGC tournaments. So the technical side of things weren’t dived into. Seed is new after every reboot of the game though who’s seed is used, not exactly too sure when it comes to online.

[@renakunisaki]

Were these being played on real consoles, and did those consoles have correct clocks? The current time is often used for random seeds. If the consoles' clock batteries had died, they might have been frozen at some default time.

[@BigNation1]

Some participants used real consoles and some used emulators during the tournament which produced these results.

[@GabeLucario]

I've seen this type of thing before in some Gen 6 battles by maybe HoodlumCallum and the gang a good half decade ago. They got a few moves like 3-5 times in a single battle multiple times - more than enough to be statistically significant. Metronome should pick a move out of all the moves available, but it simply doesn't. Odd.

[@BigNation1]

I’m hearing a lot about the Gen 6 time period regarding Metronome. Interesting to hear the consistency with how this happens.

[@hexbug101]

Man this reminded me of something that happened ages ago in the gen 6 days, would do metronome battles with my friends and one of their mons had a bizzare affinity to roll imprison, which obviously just wins them the match since everyone had metronome

[@BigNation1]

Imprison is just diabolical 😭

[@Rainchus]

What I believe the issue to likely be is that the game only advances the rng seed when something random happens. This would mean that the metronome moves are not predetermined, but since the seed doesnt advance unless a random event occurs, the same outcomes happen. Most games advance rng at least once a frame to prevent this type of problem

[@grahamtek]

Nice

[@Chowder908]

My guess is the first cold boot of the console has everything cleared from ram, but a warm boot from a simple restart is able to tell the game not to load the same previous seed. Since there is no previous seed in memory maybe that's the default one and it just so happens to work well with metronome?

[@GeneralKatarn]

Love that there's someone still passionate about PBR to this day. Keep the flame alive. ❤

[@BigNation1]

Trying my best! 😁

[@GeneralKatarn]

@BigNation1 Got access to my Wii again. If there's any PBR event in the future, I'd love to join

[@BigNation1]

Feel free to join the Discord, we currently have tournaments that go on monthly!

[@Auke1993]

I noticed this same glitch during my Tournaments as well. Earlier in the year, I hosted a PBR 1v1 Metronome Tournament. I recorded many matches per day, but for a certain amount of time I noticed that the first battle of a recording session had the exact same moves for some time, similarly to how you described it. Since then, I've tried to make sure that whenever I record/stream Metronome Battles on PBR, I do a short battle first with any team I feel like to 'reset the RNG'. It'd be interesting for me to look back at the videos for the Tournament to see if the moves that kept repeating were the same as the ones you showed in the video.

[@BigNation1]

Yep that just confirms it here. Seems like this whole thing wasn’t just a fluke!

[@Tribow]

Someone probably said this, but this is definitely an oversight and not really a glitch. Basically, the game doesn't randomize a seed until after a battle happens so the first battle will always use the same seed.

[@plentyofpaper]

Generation 9 had a near identical glitch on wifi battles at launch. Due to the same seed being generated every time, a number of players figured out specific setups that would basically guarantee OHKO moves hit. As for why this glitch is here, I'd speculate it's a left over from bug testing. Always starting with the same seed makes it so running code is completely reproduceable. For the finished product, random variables should really be linked to a timer, but a developer just forgetting to switch the random variable to timer mode was likely the culprit. In my humble opinion as somebody with only a basic understanding of coding. It's likely it went uncaught as the random number chain probably takes player choices as an input. A game that starts off with player 1 using Hydro Pump and player 2 using Ember may result in a different set of random numbers than if player 1 used Hydro Pump and player 2 used Hydro Pump. But since both players are selecting the same move every time without switching, you get to see the problem that would normally go unnoticed.

[@keirii6003]

Really interesting to finally see someone document this. I used to do metronome battles with my friends waaayyy back in the day and we noticed this and would just start a new battle whenever it occurred. I didn't know this was such an unknown thing! There are a few other common strings too but this is the most noticable since it seems to happen somewhat frequently on the first battle after booting the game. Happy PBR battling!

[@BigNation1]

Yeah it’s so strange, glad to hear this wasn’t just a one off issue that was discovered!

[@Dw7freak]

Did you know that the Missingno encounter glitch can also happen at the Seafoam Islands? If you're coming from the Fuchsia side, the first column of Surf tiles left of the the entrance is also bugged the same way as the ones right off of Cinnabar.

[@GuckFoogle507]

Has anyone looked at the seed using a memory editor or something? I have a feeling it will be seed 0 in a similar fashion to how Emerald always starts its RNG at seed 0 lol

[@BigNation1]

Doesn’t seem to be as simple as that considering it depends on the game first and foremost.

[@Lagbeard]

Because of the sheer amount of different possible actions for players to take, they probably assumed it would be fine for the RNG seeds to not have that much variance.

[@donlyphans7801]

Hmm yes I would have reached out to a coder before making the video. You've discovered the glitch but I'd appreciate knowing more about whats going on behind the scenes or how to work the glitch in the user's favor. This video would be legendary if you uncovered how to abuse Metronome to call specific moves in say a normal HGSS playthough. Everytime I try PBR, it tastes watery. What am I doing wrong??

[@BigNation1]

Seems to be more complex than you think to have it potentially work in one’s favor since it’s all predetermined. Resources are scarce.

[@pi_xi]

Does this only occur on emulators or also on real consoles? I wonder if the seed is based on variables like current time, MAC address etc.

[@BigNation1]

I believe there were participants who were using consoles. So I doubt the platform effected what seed was generated tbh.

[@SoDaFiiZzGaming]

I think you guys should try and find more glitches and exploits in pbr, would be cool to see more

[@BigNation1]

If there’s anything else we encounter then for sure!

[@Sadet1991]

sounds like someone would need to go into the games code to see how many seeds related to metronome there are

[@PersonaSama]

That’s super interesting, thanks for making this video and raising awareness about this glitch! A game I worked on had the exact same issue of using the same initial RNG seed in online matches and it wasn’t found out until years later, so I totally understand how this flew under the radar until now. (In our game’s case, for some reason multiplayer battles would desync after an hour or so and for years we couldn’t figure out why until we discovered it was the RNG seed not randomizing properly between players.) Programming online multiplayer is so complex, and debugging under a strict release schedule is so hectic, that just making sure it mostly works is the most the devs can check. As for PBR, there’s so many conditions I’d wanna see tested to see if this is consistent! For example: if this also happens offline solo in single or double battles, in offline local multiplayer in single or double battles, and in online single battles. Other conditions: was this a problem only limited to the western release? Does the original Japanese version also reproduce the results? It raises so many questions! Beyond that, I’m also curious if PBR’s battle system is taken directly from the gen 4 games, and if so, could the same results be replicated on the DS in multiplayer double battles?? This is such a fascinating wormhole you uncovered and I would love to learn more! Anyway, thanks again for the video!!

[@BigNation1]

I really appreciate you for sharing your insight on this regarding the development aspects! Really interesting!

[@LordAziki]

When I played this game years ago, I did notice my snorlax's metronome seed to become fly and jump kick a lot, but I didn't think too much of it since pure randomness means something like that was bound to happen to at least one person. Guess there may have actually been something to it though. We can only know why this happens if somebody can peek into the code to see where the seed comes from. It's possible that it resorts to a default seed because it can't properly use what it's meant to use. A typo could potentially make it use the wrong thing to generate a seed. It could even just always use the same seed since it's a bit less effort to use a static than a variable and would only be noticed by people doing a metronome tournament. It's also somewhat sad that so few people know about Pokemon Battle Revolution. It's one of my favorite Pokemon games, up there with Pokepark Wii. So much love clearly went into it, especially the animations.

[@BigNation1]

Yeah we need to shed some more light on these type of games. People are missing out on some great quality titles.

[@KingNidoYT]

Best move in the game!

[@BigNation1]

Facts!!

[@Alegriacaptions]

that's not on the bulbapedia's page on metronome, it would be interesting to write on it

[@pi_xi]

Technically this has nothing to do with Metronome, but with the PRNG.

[@Alegriacaptions]

@pi_xi Yeah but that's at least trivia worthy

[@BigNation1]

I’m not sure what the prerequisite/requirements are to post on that site. 🤔

[@Alegriacaptions]

@BigNation1 technically it's a wiki, anyone can butt in and write. Well if you don't respect the guidelines, your addition might be reverted, so maybe go to the "talk:metronome" page to mention it

[@jacksonbranby]

3 Wood Hammers in the thumbnail Pog

[@BigNation1]

Bonk

[@cookedrice6318]

I gotta say, looking back I remember using my Munchlax against my Buddy and almost always landing a roar of time 3rd or 4th metronome. Mayhaps this is why

[@BigNation1]

Yeah that sounds about right haha

[@Draco_WarriorEX]

It is at best an "exploit", since it is "working as intended", and therefor not a "glitch". What I like about this video is that it is basically saying "we've found something interesting, see? And would like more information about it if anyone can explain how this works". I'd say go ask the speedrunning community since we know a lot about game seeds. They're usually connected to the in game timer, and therefore chosen at a specific time in the game. So if everyone is just mashing buttons trying to get their game to load as fast as possible, assuming everyone is running similar hardware, it should load at a similar speed, and therefore a similar time and end up on the same Seed each time. If only a small number of seeds are available then that increases the likelyhood that you'll get the same one over and over, or if it is randomized less frequently. Sometimes game seeds can be connected to inputs, so if you use the exact same number of inputs each time you get the same seeds, but this is not likely the case here.

[@deltaflamestorm7400]

fuck now im sad because you made me remember playing pokemon battle revolution so much as a kid getting to see my pokemon from platinum and soul silver come to life was so exciting

[@BigNation1]

Such good times 😭

[@SoDaFiiZzGaming]

Very cool glitch I must say

[@648]

Cool video but I really don't like the letterbox blocking off content

[@SeaSlug]

Do the same moves happened in ds Battle or local battles ?

[@BigNation1]

From what I gathered yeah because of the seed

[@EulerAlert]

Nice touch with the TCGP music!

[@BigNation1]

Thanks! Great OST in that game!

[@bomberhood7234]

Never knew about this glitch. It’s always nice that even though a game is old, it still has some interesting secrets hidden that are yet to be discovered.

[@BigNation1]

That’s what makes the game so fun. Some things will always be hidden.

[@icculusburly]

i am also a PBR fanatic, love picking some up after work!

[@BigNation1]

Now that’s a vibe 💪

[@BaalFridge]

So does that mean we can use metronome for RNG manips?

[@daniellelondon439]

Oh I have this game! I played it a lot as a kid but never noticed this!

[@BigNation1]

The more you know!

[@novapulser9396]

Man I missed PBR. Used to have a friend when I was way younger that had it before I got my own copy, all I'd do when I went over was play PBR with them either online/offline cuase of the silly formats (that damn spinning wheel, even if it has a method to the madness)

[@BigNation1]

Neon Colosseums wheel is so aggravating if you don’t master the technique to get the pokemon you need haha

[@dd-di3mz]

This title is questionable. ...

[@BlueHairPronouns]

Cool glitch! Glad you documented it. But consider putting the thesaurus down and sticking to words you know well when writing a script.

[@ikeneats]

That and run-on sentences; I had to rewind the first 15 seconds of the video over and over since my small brain was immediately bombarded by information.

[@diegodankquixote-wry3242]

The battle revolution might have better animations, but it has the worst rentals in the history of the franchise. Big root gaming

[@BigNation1]

I disagree with you about the rentals. You can trade them away in game and get much better ones like Lucario, Garchomp, Charizard, etc. and they can all have good movesets and natures.

[@primechamp7841]

Well you’d have to ask someone to check the game’s coding but its possible that this game chooses a set number of moves at random at the start of the battle and then Metronome chooses randomly between those moves. Devs usually do that cuz the processing power couldnt keep up with loading and randomly selecting from all moves everytime Metronome is used which could lead to lagging and possibly crashing. If this is the case then this is not a glitch but a program design

[@BigNation1]

Hard to say because of the moves used was practically pre determined with same results that different users experienced. If it takes a pool of moves and the game chooses randomly between the ones selected then I believe we wouldn't have seen what we did during the tournament in that case. Such a mystery!

[@primechamp7841]

@@BigNation1If you’re all using the same server at the same time, then its possible that the chosen set of moves is chosen for that server and any user in it that uses Metronome will use a random move from that set.

[@tokugawashigeshige8862]

Processing power has nothing to do with it; randomizer functions are not computationally demanding. BigNation was correct about this being an RNG seed issue. Emerald Version actually has something similar- the initial seed when booting up the game is always 0. This makes it particularly easy to perform RNG manipulation to get shiny/good IV Pokemon in that game. Most games will randomize the initial seed based on something like the exact time that the game was launched.

[@primechamp7841]

⁠@@tokugawashigeshige8862Its not about the randomizing function. Its about loading the move list every time the metronome is used. Rather than loading the entire pokemon movelist every time, it loads from a smaller movelist so that the game runs quicker and smoother. They might’ve programmed it this way since this is online and the server has more functions to deal with.

[@Blazingflare2000]

​@@primechamp7841I really don't think that's the case. The attack graphical effects are simple enough that the wii is more than capable of loading them off the disk on the fly as needed. And the game doesn't need to use a turnicated move list. The full move list for every attack in the game, plus their battle effects, would only take up a couple kilobytes of memory at most and is almost certainly always kept loaded into memory for quick indexing. You vastly overestimate how demanding it is to load things that aren't textures or models, or to keep them loaded in memory.

[@StepBaum]

Til like 2:50 the video dragged out the intro until you actually came to the videos topic. For a 7 min video, thats a 1/3 which is pretty long and in this case kinda unnecessary imo. To top it of its not really a glitch as others mentioned in the comments. To me it seems (maybe I am wrong) like there hasn't been done either enough research or not smart enough research to find out the actual cause. Hope you take the as constructive criticism. The visuals and voice are nice to listen and watch to tough

[@BigNation1]

You will find that many videos always bring context and background information on the stuff that is being talked about before getting into the topic of the video which is normal. As mentioned in the video, information is scarce and my research was very limited. I have little to go off of and articulate theories that me and the community have made which was shared so it’s not really fair to criticize a lack of research. Thanks for the criticism overall but things are done for a reason when producing videos such as these or anything else that I’m trying to make you aware of.

[@StepBaum]

@@BigNation1 Thanks for not taking it the wrong way, not often do take criticism that well. I actually know the first one is common, doesn't mean its necessarily a good thing. Tbf maybe its a me thing but I always feel like its something way to constructed. I see the merits but personally its more holding back from the main point (generally speaking) than actually leading into it I feel. Figured I'd let you know despite that. Tbf in the end you can tell by the views what works best and makes the most sense, if that works the best to continue doing so. For the second thingy about research. I totally got that, and you got it across the video. I just feel like the reason came down to basic programming knowledge, and since I didn't actually know if you didn't take enough time, or actually did I figured maybe you just did it sort of the wrong way. So maybe I got it across the wrong way, and sorry for that, i can totally imagine, that you put in time and made the best out of what you could find. Generally I can recommend for these kinds of stuff to always go to Romhacking communities and/or speedrunning communities and sometimes even shiny hunt communities since they know they most niche stuff sometimes or even similar occurences. Usually their discords are pretty friendly too about that. That's what I meant by smarter research, should've put that in more direct words and be more clear mb. All in all still a solid video even tough I dislike some stuff and I will definitely check out the other stuff too. Also thanks for taking the time to listen to commenters, much appreciated :)

[@MannyMetronome]

👁️🫦👁️