While messing around with GCW attempts yesterday, I discovered a way to manipulate a perfect spawn (right next to a speeder bike) in Dune Sea. The downside is that this only works from a "clean slate" in the memory. In other words, it only works from a console reset.
A TL:DR can be found in the video I'll post in the message following this one.
This manip is rather simple to do and can be fun to do every now and then. First, it's important to understand how it seems the game stores your spawn locations. A note: this was ONLY tested on Dune Sea in GCW, but it should in principle hold for other planets in GCW (maybe in IA too?).
There are two components to a "spawn" - the first is where your cursor is on the map. For Dune Sea, there are two default locations: the Dune Sea command post and the Sandcrawler command post. The second component of the spawn is your location on the map when you spawn in at a command post. The game appears to store a giant list of these spawns (cursor location and position) in memory at some point during load up. I generated a table of the first 30 spawns in this list and will publish it once I verify I didn't make any mistakes.
So how does this list work? Basically, the game keeps a counter that starts at 1 (technically it probably starts at 0 because computers index things starting at 0, but for clarity's sake lets say it indexes starting at 1). The first spawn in the list is:
cursor: Sandcrawler position: to the right of the right speeder bike
You will ALWAYS get this spawn when picking Dune Sea GCW from a console reset. If you're curious, the next two spawns are:
cursor: Sandcrawler position: back left
cursor: Dune Sea position: back right
Now the next part is to talk about how the game increases this index that chooses your spawn. Currently, there are 3 things I know of that change this index:
- Choosing "Quit" and going back to planet select. This increases the index by 1.
- Your teammates movement on the map.
- Your ammo/the number of bullets you've shot. Not sure which value effects it and Idk if grenades impact it yet.
Capturing command posts and killing enemies are unknown to affect this because it's impossible to do either of these before your teammates spawn in and so it's not clear if either have a direct effect on it since your teammates's movement and their effect on it is so large/random. I'd assume capturing command posts messes with it since it has to update the master list with the new command post you control, same thing with losing a command post. The easiest way to influence this index is to do certain actions prior to your teammates spawning in. With the right actions, you can give yourself a perfect spawn next to a speeder bike.
From a console reset, select Dune Sea in GCW. Note, your cursor movements on any of the main menu screens have no effect on this, nor does the amount of time you spend on the main menu. Once you load into Dune Sea, spawn as a Stormtrooper. You'll notice you'll be at the first spawn I listed above. From here, shoot 8 bullets into the ground (your ammo left in your clip will now be 47). Now, quit out. When you select Dune sea again, you'll have your cursor on the Dune sea command post and when you select Sandcrawler you will spawn in next to a speeder bike for a perfect spawn.
For the record, it is NOT a 1:1 match for the number of bullets you shoot to your spawn in the master list. In other words, shooting 8 bullets and quitting out doesn't give you the 10th spawn (index incremented by 9). It gives you a spawn that is somewhere outside of the first 30. It's impossible to know for sure the spawn number it gives without compililng the entire master list to know the pattern in full (assuming when the game reaches the end of this master list it rolls back to the start of it). But that list could be 100s of spawns long.
Given that this only works from a console reset, this is absolutely not viable for grinding attempts as you'll go crazy after only a couple hours. Instead, it's mostly an interesting thing that shows that this game takes some shortcuts to determine some things and it isn't as random as may appear. It also shows what may be possible if more concrete methods of RNG manip are discovered. Watch the video in the next message to see it in action with an explanation similar to what I typed here.
Forgot to mention that Im on PS2 version. Can't imagine it would be any different for XBox/PC (unless the game messes with memory on those platforms).
dude, this is awesome. Not sure if you still run this game but I might make a master list documenting spawns for several maps. Last night I ran Rhen Var Harbor, spawning as a droideka at citadel probably 200 times and noting the randomness. In that case it'd be useful to know if the "Y" button press that puts you into ball mode affects the RNG, and if so, how. Would be nice to shoot a certain number of bullets to guarantee a god spawn right by the door.
Video is below, you can see what I'm talking about at 1:05:55
I played with this last night on PC. What I assume is the RNG value increments a somewhat inconsistent amount on Spawning, Respawning, shooting, and going into ball mode. I might be missing another factor to determine where to spawn because not everyone spawns in the same spot (and even I can't get it to stay completely consistent, which is why I think I'm missing something), but it seems to go in cycles rather than changing every value. If you knew where your spawn point was on the cycle, you could theoretically manipulate it but you would need access after spawning to alter values or be able to predict how much it is going to move when AI spawn (AI impact the cycles) I can create pictures of the key spawns (I have the PC mod tools and they didn't change the basis of the maps between systems) but it might be a little much. For example, I use the tactic of only running CW on a clean run of the game. This ensures that I get the same spawn next to the tank that I want. There are 12 spawn nodes at that CP and you always get node 10 on a clean boot if you spawn first. 12 nodes is going to be hard to predict and in this case you have to have Cheat Engine running to keep track of the values.
Hi, just checking out this thread now. I haven't run this game in almost 2 years (I have no desire to go back, not a fan of how XBox and PS2 are side-by-side since XBox can create unbeatable runs for PS2 b/c of loading), so I haven't looked into this any more since I first did it a couple years ago.
But, it's awesome to see other people taking in interest in this and how it works on other versions of the game besides PS2. I'd love to see anything documented on it, since it's a topic that still interests me greatly. I'd love to see this game broken down into some more manipulable states.
Sucks that PS2 load times are stupid long, but to be honest, you could still do ILs. They're much more interesting anyway, since full-game requires playing so safely.
pls come back
At this point shaking up the leaderboards further (Splitscreen mode has been added to every category) to make separate categories for PS2 only (or PC only) would be a lot of work and a huge hassle to have everyone resubbmit their runs. The OOT community doesn't seem have a problem with this platform disparagement. I do not see what the need is to be hung up on a slower version of the game. You are the only person to have asked this of the community not even a PC runner has brought this up. We are in consensus on the topic, if you want to go fast, play Xbox.
Regards, Boston
PC has the same load time by default as PS2. Xbox is faster, but it's not that big of a deal. Other communities do "With/without" load timing, but I think this community is on the small side and probably shouldn't worry about it. This doesn't impact IL anyway, which I believe is being more heavily ran at the moment than full game runs.
Current times in campaign and full game runs aren't even close enough (at the top, anyway) to warrant different timing.
Full game runs are more fun because of the back-breaking RNG :)
At this point the game is still so fresh and unoptimized on lots of maps, so ILs are definitely where it's at. Just look at what happened with Rhen Var Harbor a few months ago, Boston's 'play normally' strat turned into Shade's out of bounds glitch turned into my tank blitz strat turned into BattlefrontRunner's insane Droideka moves.
As another example, I've been struggling to reduce the Naboo Theed time as droids, then someone completely fresh came in and beat the time by a solid minute playing as clones, of all things.
There's so much room to experiment with ILs and I feel like until strats get really consistent, full-game will always require a safer playstyle, especially since no one wants to reset a run once they're an hour in and their yolo strat didn't work
Just to clarify what I meant, I meant in context of full game stuff, not ILs since obviously loading doesn’t matter there.
@BostonBrew And I think you misunderstood my point. I’m not asking to change the LBs, I’m only stating why I stopped doing runs since Lemon mentioned it. LBs are meant for the masses, not for everyone. As you pointed out, I’m in the minority for wanting the consoles separated so it makes perfect sense to not do it and that’s 100% okay with me. Nothing wrong with saying “if you want to go fast, play on the fastest version” since that’s the name of the game so to speak.
(Sorry for the double post but this is kinda un-releated to what I said in my previous post a few hours ago).
Not to change the subject of the thread, but I just re-looked at the LBs after the addition of the split screen categories was mentioned, and have you all considered using sub categories for the difficulties? It would significantly cut down on the number of main categories you have since you'd only need say a single 100% category and you could just setup Easy, Medium and Hard as sub categories underneath it. You could also file sole categories like Splitscreen under Misc (since they’re essentially just extensions of Amy%) to further simplify the look for the boards
The lack of runs on most categories would make migrating them easier if you setup Easy as the default, for example, and then had to move a few Medium or Hard runs to their proper place. Idk, just a thought.
That's not a bad idea, the only gripe I can think of is that easy times would be displayed by default, which sorta unfairly prioritizes them over hard and medium runs. It'd clean up the category page, yes, but also might require another click to see what you're looking for.
Not that I mind, I'm mostly an easy IL runner anyway. I guess it'd be interesting to hear what other members of the community have to say about LB reorganization
In my opinion, the leaderboards have already been more cluttered with the addition of splitscreen, adding categories for each game would only make it even worse in that regard.
Not to mention, most runners would stick to Xbox anyways, there's far less of us who play on PC or PS2, so most people would just look to the Xbox section as the "Official" version, and just run that one; while the other two categories would sit empty or be very desolate
I'm a still newbie in the SWBF speedrunning community, but I also think that the page could profit from a rearrangement.
I agree with Timmiluvs about setting difficulties (easy, medium and hard) as subcategories (like the ones in the TMNT4 page, for example) for the 4 main categories - CW, GCW, 100% and All Maps. Considering that the splitscreen ones don't have any runs submitted atm, maybe they could also be set as a subcategory.
Regarding the "speedrunning standards": I, myself, don't care that much about the differences between the platforms for this game. Even though I can't state for a fact the console versions are easier/faster, at least from the VOD's I've seen, they apparently look a more "flexible" than the PC version, both in regards to loading times and to the AI in general. I don't see an urgent reason to differentiate all the platforms on the leaderboards - in my case, as a PC player, it only makes me more eager to try to beat those times :) Regarding the "easy" difficulty being shown as default: if the fastest time possible for a SWBF speedrun is achieved in easy mode, why not leaving it as the "standard"? To be fair, I also don't care that much about this aspect - but I'd love some sort of categorization of the difficulties on the leaderboard.
At least that's my opinion... I'd love to see more inputs from the community...
Actually, I have one more suggestion: what would you guys think of setting the GC campaigns as "IL's"? I was thinking about running campaigns like Revenge of the Sith and Dark Side Rising just to time them for a potential future 100% attempt. Considering that there are different strats for each map, different routes and different bonuses that can be chosen, I think it could be a fun addition and also make the GC more relevant to the leaderboards in general.
I'm definitely down to include GC campaigns as IL's, even though they'll likely end up being 30 minute "IL's" hahaha. I do think it's important to make clear though that ILs within GC runs are ineligible for the level leaderboard due to planetary bonuses.
Unless there's a good reason not to include each of the 8 GC scenarios on the IL leaderboard, I suppose I'll add those later this week.
Yes yes, I definitely agree with you: IL's are "Instant Action" levels, GC levels are way different due to the bonuses (even though some are pretty useless). I only meant to add those on the IL section so the main page doesn't get too messy.
Well, I already have a Revenge of the Sith run uploaded on youtube and ready to submission when the leaderboard updates. Just let me know when it does (I'll try giving the Empire's GC a look in the meanwhile)
Alright, the IL leaderboard has been updated and should be good to go. Please keep in mind that timing for galactic conquest scenarios begins on the first frame after selecting the name of the GC campaign and ends on the first frame of final "Victory" screen on the last map. Any and all planetary bonuses are allowed, just play through the scenario as fast as possible.
