interesting fact, by waiting a second, you arent letting the ball spawn, you actually spawn slightly in the air, most time based training packs can be broken because you can wavedash before you hit the ground
this is the exact reason why the first level of musty's speedflip pack is impossible for some people even with a good speedflip, can't build momentum properly within the first fractions of the second you press boost and accelerate since falling down works against you
I realised the difference with eversax olympics air dribble, if you hold boost and throttle as you spawn with both hitboxes, fennec hits the ball perfectly to roll flat and octane makes it bounce
also im trying out the pack, and for the first shot I hit the ball but it went over the net ;-; update: doing them all in order and the catch and dribble is so stressful...
My best guess is that while the hotbox is the same, the spawn “center” is different for both models. Meaning a slight difference in spawn location, making them just a touch off
I could see the cars using the same hitbox, but the hitboxes will be aligned to the car's visual graphics differently. When the game spawns a car, the spawn point may be anchored to the visual representation of the car, which would lead to the cars spawning with different hitbox positions.
Yeah I also think it’s something to do with one hit box being more to the front or back compared to the other witch is one of the reasons pros use the fennec more than the octane because the fennec hit box is more aligned to its visuals and further forward of spawns maybe
if they have part of their hitbox more forward or back even with the same hitboxes, do you think one might have like a nanosecond of an advantage on kickoffs? we need rocket sledge for this one i fear
I did a full experiment years ago (2021, video on my yt homepage) and determined that fennec and octane are not the same. I have full videos, custom training pack which proves it with a breakdown, and even submitted a ticket to psyonix back in the day. Nobody seemed to take me seriously. I found that all vehicles sharing a hitbox fall into 2 categories when it comes to actual performance. This was a couple years ago but the data is probably still accurate
@@danielebrambilla2290 im pretty sure 90% of my videos are said to be pointlessly existing regardless of length. Also, I hadn't considered the screams. It's hard to keep up with the algorithm these days.
That's my guess. The cars spawn in the center of their model, but the hitbox is not centered over the center of the model to more accurately match the geometry of the car. It would be a simple fix to spawn cars based on the hitbox and not the model. It could be more complex than that, but that makes the most sense and is most plausible. Now to find the car with an Octane hitbox that's the furthest forward on the model and test them all out on 1v1 straight kickoffs to see which one hits earliest and how the ball bounces.
@@redd_rlthat’s actually false. In training, I made screenshots for every kickoff spawn, and Fennec is aligned ever so slightly different and the camera is looking in a slightly different direction too
do other cars with octane hitbox like scarab do different things too? i remember hearing a while ago with rl bot development that the bots just played worse if using scarab despite having “identical” hitbox to octane
I beleive the wheels placement does have an impact on how your car will turn. Anyway, there‘s some spreadsheet on reddit about it, even if most cars share a common hitbox, every car in RL got his how turn radius, like scarab is about 2.30 while octane got 3.32, making the scarab turn slightly slower than the octane (or octane does tighter turns if you look at it the other way) Edit : With how the car is placed at the beginning of shot 15, I‘m pretty sure it's about a wheeltouch. Since wheels are placed differently, you got another outcome
@@IfeelKindaSick The wheels are set different depths in a lot of hitboxes too. Hit a reset between octane and dingo. It's a significantly different feeling in spite of the same hitbox, whereas cars with similar visuals (like venom) feel very similar to reset with because it feels like the wheels are in a similar position.
Sometimes I think that Rocket League was a fluke. Did Psyonix design the mechanics that have been discovered over the last 10 years or are they just happy accidents?
@RandomGuy-qc8ml makes me wonder if there are more mechanics yet to be discovered. Also, if Psyonix updated Rocket League to Unreal 5 or made Rocket Legaue 2, how would they carry over these mechanics, that seem to be consequences of the physics, without breaking them.
@@panzer00 I worry about that scenario tbh. I have to imagine they would preserve things like flip resets (an example of something unintended that the community discovered), as they have been adopted into the community as a whole. However, something like a wall dash, zap dash, ceiling shuffle etc, I could see them not caring about as much because they aren't as popular and might be difficult to code in if they don't work by default.
@idemus168 I feel that this is why they haven't "upgraded" Rocket League because it might be a downgrade in terms of mechanics. Of course, that's just conjecture.
If you have access to the map asset, could you not maybe split the field in half overlap both asset and somehow highlight the area where it is not a 1:1 or make it list it in a text file or something? (I know nothing of how that works, might not be possible)
@@Lethamyr If the cars and the field is physically identical then all I could see is that physics is the cause of the difference in that corner. So it would be even harder to know what happens there unless psyonix/epic can tell us themselves.
@@Lethamyr From my completely inexperienced eye, it looks like a collision thing then. The transition from wall to ceiling seems to be the cause of the desync, so I wonder if something funky is happening where those mirrored quarters of the field mesh together that causes the collision to behave differently in the physics engine during the transition. I know from livery editors in racing games that simply mirroring something in software sometimes doesn't work 100% as intended, and can need manual touch up to get a perfectly symmetrical result.
@@Lethamyr Maybe float position having different rounded values since different original x values? like -100 rounding to -100.000184 while 100 rounds to 100.000217 (examples), hence a 0.000033 difference
yeah i also think this basically boils down to precision issues regarding float numbers. i guess if they used fixed point numbers for the physics it could be completely simmetrical, but unfortunately now it is not...
It's simply floating point precision issues. For example, there is no accurate floating point representation of numbers as simple as 0.3 or 0.4, and those errors accumulate reasonably fast. Keep in mind that every physics tick involves multiple floating point calculations just on the orientation and position of your car. Edit: The arena "not being symmetrical", that is. As for the hitbox, yeah, they must have messed something up.
They are not simple values, floats are 32 bits and have 6 additional floating points of precision in c++. The hitbox for both octane and fennec are exactly "X: 59.003689, Y: 42.099705, Z: 18.079536". And thats for the hitbox itself, quats are used for rotating in 6 degrees of freedom.
In theory you could fix any errors by placing the field so that the ball spawns at origin, right? Also something might have to be done in the baking process of the field's geometry. Still this kind of thing is unacceptable to me.
@@Anteksanteri Hmm, I'm not sure that would solve it. For simple translation, yes, floating point representation is "symmetrical" in that sense (because of the sign bit), but I have honestly no idea if quaternions and rotation matrices wouldn't mess this up as soon as rotation gets introduced (which is already the case to even place the cars). If I don't forget, I'll test it.
The octane and fennec hitboxes are the same, but I'm pretty sure the octane's hitbox is rotated slightly, making it lower at the front and higher at the back compared to the fennec. I'm also pretty sure that almost every car in the game is unique in this way. Some cars share the same hitbox, but with different rotations/inclines.
the placement of the hitbox, and the hitbox sizes are completely the same, but the wheel's hitbox on the Fennec are slightly offset, the touch you get while boosting for the Fennec is offset slightly lower, making it less of a wheel touch, the octane's hitbox however has the wheels positioned slightly higher, so when you hold boost on the shot, you get a "pinch" with the car's hitbox, and the wheel's, rather than just one or the other like most other hitboxes, oh and I tried it with the merc, and it actually worked on the merc's hitbox a couple times, though that might just be my stick drift tilting the car a little on spawn.
This has been a known phenomena for quite some time. All cars are in some hitbox category, however the alignment of the hitbox (rarely), and more importantly, the wheel locations are still independent for each car. The wheel location might be the main culprit here
@@nedasbolevicius8676 it appears lethamyr autofilters links, but rocketsim by zealanL on github has a carconfig file that showcases the objective values for every hitbox, with no difference per car. only per hitbox.
Seems like maybe a positioning issue of the hitbox. Fennac's hitbox must be positioned slighted behind the Octane's relative to the visual frame of the car. This has the effect of the hitbox starting at a different coordinate within the map. That would explain why shot #4 was so much harder on the Fennac because the hitbox needs to travel ever so slightly further than the Octane.
They are the same hotbox, but relatively placed differently when the game spawns them in, i think they are shifted differently relative to the wheels and the game also tracks the center of the car slightly different when it is spawned in
you know what it probably is on the mirror octane desync? i bet it's floating point errors accumulating until one car's angle rounds up while the other rounds down, thus the cars are basically the same until a very tiny change
Idk if this matters much cuz idk how the hitboxes are built but it would make sense that fennwcs result is different because the wheels on the fennec are surrounded by the frame of the car while octanes wheels are free and has nothing around the except ontop.
it's because of collision geometry. Even with the same hitbox, if the models contact points or physics meshes are handled slightly differently by the game engine, it could cause them to interact with surfaces like ramps or walls in unique ways. This can result in one car touching down slightly sooner or having a different trajectory after drops.
Only thing I can think of is I remember different cars used to have very marginally different stats like turn radius being slightly wider / sharper. The differences were always very small so it would never make a real difference, and to my knowledge this was patched out years ago when they standardised all the hitboxes. But this is the only thing I can think of, that maybe the Octane / Fennec have some slight stat difference which makes this happen.
the only thing i can think off Fennec vs Octane, is the following: have a look at what Dominus does. it grips the pillar and drives while holding boost only. Fennec seems to almost always do the same. could it be that the Fennec box is placed slightly lower, making it grip the pillar/wall more often then not? and by lower i mean, this could be 1mm and we just can't notice when comparing it to Octane
A lower hitbox would also be evident in how it hits the ball on the ground. The pop would be higher with a lower hitbox. It could also be a slightly taller hitbox where the top of the car is at the same height but the bottom is slightly lower and that would be much harder to test accurately.
I had the same thought. and it could be such a small difference that it isn't noticeable to most people. You do hear pros sometimes say that fennec feels different to octane. if its like .01% closer to the ground or something that could be why.
@@BenRHarsh exactly, and one thing i noticed with Fennec is that i can pop the ball easier even when driving slower then i would with Octane. Octane on the other hand i can get way more power out of barely any touch, its possible with Fennec but less likely somehow.
The hitbox on the fennec will have to be lower than the octane, potentially similar to the marauder (as the marauder is the octane hitbox but sits lower), which would make the marauder a RLCS level car lmao
The only thing I can think of is maybe the wheels or something are misaligned, or the the way the boost is put on the octane vs the fennec is different. That is very weird, besides that, I'm just as stumped as you!
Ive done significant testing on this recently. Do resets between the octane and dingo and youll realize you have to hit a different part of the bottom of the hitbox to get it because of the wheel position. I think that and the wheel height affect the hitboxes as well, but its all anecdotal, as im a console player and cant scientifically test it from specific physics positions with commands.
Same hit boxes but different center of rotation/pitch points? Or different amounts of steering input accelerations or aerial depending on how close the hit box is to the nearest surface or wall. (idk if that sentence is legible) Does the Octane hitbox ride higher than the Fennec hitbox when idle? Octane feels like a taller car with squishy-ier suspension travel to me than what the fennec feels like. Ride height wise.
Programmed environments like the rocket league arena are chaotic due to rounding error in the stored data. Error as small as machine precision will compound when both cars are set off on the same path. But hey maybe the arenas are actually not symmetric
If it was just a rounding error then it wouldn't happen on the same aide each time right? If you're taling about the thing at the start of the video, rapid and leth both tested it and the cars always took the same path on orange side and always the same path on blue side, it just wasn't symmetrical
I switched to Fennec a while ago from Octane , what i noticed is that now when i switch back to octane i just can't get as good of a first touch for example when going for air dribbles , but it is also easier to do stall resets with octane , octane is also better for general shots , more power but then again if you do speed flip shots then it doesn't really matter , i think something has to do with fennec's visual , i think it is closer to the hitbox than Octane visually so you can be more precise with fennec , at least that is how i feel , but i only have 1400 hours so more qualified people prob know a lot more about this
Truly makes you wonder about all the times Leth has specifically said that it feels weird to play as the Fennec compared to Octane, maybe there was no placebo after all? It really has a different hitbox! 🤔
I heard on a video that they are changing the hitboxes, not by the car but by just shifting and adjusting their hitboxes to match the car. For example the Octanes hitbox is the same as the Fennec’s except the hitbox is shifted towards the center point of the car. Hopefully that helped
oh and the fact the server physics on match making seem to be in the 60fps range now... which is why so many ghost touches and cars not in position in replays etc... this one is hard to prove but been slowly getting evidence
What if you were to replay a game but switch the blue and orange sides with the cars having the exact same inputs? Would the slight changes due to the field being unsymmetric effect the game drastically after 5 minutes of gameplay? Wonder if any historical matches could have been different if the teams were just on opposite sides of the pitch.
I attempted this map and successfully completed it within the 30 minutes! However, I was able to do every other shot except for shot #1 using the Fennec. I switched to Octane and tried again and got it on my first attempt. Not sure if that was pure coincidence or not, but the cars definitely play differently. I didn't finish the video so as not to spoil the map, but I figured that's what you were talking about.
So could the answer just be that the hitbox is in a different spot on the car? Like maybe the hit box is positioned further back on one of the cars, but the game positions you based on the entire model rather than the hitbox? Just a thought.
Okay, hear me out. Maybe its not the hitbox but the part of the car that the boost comes out, which is different on the octane compared to the fennec. Both cars have 2 parts where the boost comes out, but on the octane they are a little more towards the middle and a little bigger, while on the fennec they are a little more towards the outside and a little smaller.
Well. I always felt it, always doubted those are not the same hitbox. The feeling was too different when dribbling a ball with Octane or Fennec, I have realized it because of the Dribbling Challenge 2 Overhauyl map on the Steam Workshop : I mained Fennec ever since I had it, and so I trained with it a lot on that map. And sometimes I go back to the good old Octane, and when I go on the dribbling challenge, well.. Those are not the same hitboxes, that is for sure : the ball ain't react the same. Ok now I will watch your video, cause I wonder if you speak of the wheel gap too, cause I feel like one of those cars have a higher turn radius... :p
something else crazy if you do the same thing that they did in the first video where they boosted off of kick off the fennec travels a different path than the octane even if they are on the same side.
same hit box but they might have different center points so the spawn is tiny bit different maybe? like if the phennac and octane are same hit box, but the phennac might spawn like a milimeter back or something?
Lethamyr, i know you probably wont see this but i main fennec when i play and yesterday i saw your video on how the map is uneven as you put it, however whilst trying it in the fennec you get a completely different route crossing the goal. I then tried in the octane and got the same u turn path you did showing the octane and fennec have a minutely different hitbox
I like these kind of silly training packs so I tried it for myself. And idk what it is but some of the shots behave completely different on Playstation. Shot 8 is impossible because the ball doesn't bounce up, it bounces straight out of map. Shot 12 behaves completly differently, instead of bouncing of the pillar back to the blue side it goes towards the right side of the orange goal. And on Shot 15 when you do nothing and let your car just fall, with the Octane the ball will pop out the pillar at the end, regardless if you boost or not, while that won't happen with the Fennec. Also while doing nothing the Octane will end up on its roof while the Fennec will land on it's wheels. No idea why it's like that but it's somewhat interesting and I wonder if anyone has a explanation for it. I uploaded the shots in question to my channel just in case anyone wants to see it.
For the beginning of the video were you borh using keyboard? If not there may have been a tiny difference in the value of one of you pressing down the trigger
im gonna take a wild guess and say that floating point accuracy is the issue for the map not being even. im assuming that the "origin" (0,0,0 in world space) is on the blue net, while the orange net is 100,0,0 (not actual values, just an example). as numbers get bigger and bigger, computers cant store them correctly which means sometimes movement, physics or other bugs might occur. 0.1+0.2=0.300000004 not 0.3 for example.
Aren't they the same hit boxes, but positioned differently on the car graphics? So maybe this has a bigger impact in terms of spawning in the air on a wall vs spawning on the ground w/gravity holding you down. Probably wouldn't make a huge difference, but I wonder if these hit box placements could affect your ability to reach the ball sooner in those speed ones for example. That's what I was thinking watching him go through the ball over and over. maybe that hitbox is slightly farther back on the car, even though its the same hit box.
I’ve always said they are not the same hitbox. If you use both cars it’s so clear. The ball interacts with the cars so much differently. I’ve also always said that I feel like the ball just glues to the octane more than the fennec. Maybe that’s why lol, the octane could literally have some sort of pull on the ball for some reason 😂 it’s why the octane is so good lmao
i think the octane does that because the fennec is actually moved in the bgack by a pixel or 2 so it makes it go ro the wall instead of doing that the octane did
Hey, Leth! I just installed UDK 2015 and am planning on making custom RL maps, including 'totally normal Rocket League maps'. Would you be interested in giving them a try when they're finished? I'd put links in the RLMM discord.
I always had a feel the sides of the field weren’t symmetrical and that the fennec was different than the octane. What other weird quirks are there that we thought were placebo just to be true 😂
I thought we knew this,if u play with them they both feel different.Some players are better in octane and some are better in fennec,it's hard to find someone who doesn't prefer one over the other.
Wait till you start investigating the balls and game speed and car touches on a day to day basis. The game is filled with inconsistencies. Thats why if you dont play for 4 days your cold. Cuz your coming back to a diff game every time and gotta adapt to that new game.
Even if cars share hitboxes, every car got their wheels and center mass at different places. Result in the same hitbox but different turn radius and also maybe why you get a different outcome with the shot 15
interesting fact, by waiting a second, you arent letting the ball spawn, you actually spawn slightly in the air, most time based training packs can be broken because you can wavedash before you hit the ground
next training pack from leth incoming
this is the exact reason why the first level of musty's speedflip pack is impossible for some people even with a good speedflip, can't build momentum properly within the first fractions of the second you press boost and accelerate since falling down works against you
Didn't they dub that the Spawn Dash?
I believe Twinser is also kinda weird
@@Zel-Veraanwait can u do that on demo spawns?
24:35 for why the Fennec isn’t like the Octane
24 minutes in should be illegal
Real
No fucking way it's 20 PLUS MINUTES in... Holy shit. Thanks dog
thanks for letting me skip 20 minutes of yap
@@redd_rlit wasn’t yap it was rapid completing a training pack that leth made
I realised the difference with eversax olympics air dribble, if you hold boost and throttle as you spawn with both hitboxes, fennec hits the ball perfectly to roll flat and octane makes it bounce
also im trying out the pack, and for the first shot I hit the ball but it went over the net ;-; update: doing them all in order and the catch and dribble is so stressful...
update 2: holy I just did it (kbm gc1, definitely does not take half an hour to do for me xD) can't wait for 'normal' and 'hard'
My best guess is that while the hotbox is the same, the spawn “center” is different for both models. Meaning a slight difference in spawn location, making them just a touch off
OK DUDE
@@benjikrafter like it is with the lightning mc queen car it hast the dominus hotbox but it's further back
I could see the cars using the same hitbox, but the hitboxes will be aligned to the car's visual graphics differently. When the game spawns a car, the spawn point may be anchored to the visual representation of the car, which would lead to the cars spawning with different hitbox positions.
Yeah I also think it’s something to do with one hit box being more to the front or back compared to the other witch is one of the reasons pros use the fennec more than the octane because the fennec hit box is more aligned to its visuals and further forward of spawns maybe
if they have part of their hitbox more forward or back even with the same hitboxes, do you think one might have like a nanosecond of an advantage on kickoffs? we need rocket sledge for this one i fear
@@JermstainFor sure yes, but which one ? 😅
Not sure why, from a programming perspective, anything about the visuals would be used for spawn. Kickoffs wouldn’t be even.
this is false, they're in the same location.
I did a full experiment years ago (2021, video on my yt homepage) and determined that fennec and octane are not the same. I have full videos, custom training pack which proves it with a breakdown, and even submitted a ticket to psyonix back in the day. Nobody seemed to take me seriously. I found that all vehicles sharing a hitbox fall into 2 categories when it comes to actual performance. This was a couple years ago but the data is probably still accurate
hey hats off for your research and efforts
Yeah but was your video pointlessly long and full of annoying screams like this one? It's basic TH-cam, man.
@@danielebrambilla2290 im pretty sure 90% of my videos are said to be pointlessly existing regardless of length. Also, I hadn't considered the screams. It's hard to keep up with the algorithm these days.
it's false, and data has changed.
Maybe like the centering (or like "center of spawn" of the octane and fennec is different?
Probably the best explanation
i was thinking the same thing. maybe has to do with the spawning being different even if the hitbox is the same. idk
That's my guess. The cars spawn in the center of their model, but the hitbox is not centered over the center of the model to more accurately match the geometry of the car. It would be a simple fix to spawn cars based on the hitbox and not the model. It could be more complex than that, but that makes the most sense and is most plausible.
Now to find the car with an Octane hitbox that's the furthest forward on the model and test them all out on 1v1 straight kickoffs to see which one hits earliest and how the ball bounces.
this is false, use hitbox plugin and align them perfectly.
@@redd_rlthat’s actually false. In training, I made screenshots for every kickoff spawn, and Fennec is aligned ever so slightly different and the camera is looking in a slightly different direction too
do other cars with octane hitbox like scarab do different things too? i remember hearing a while ago with rl bot development that the bots just played worse if using scarab despite having “identical” hitbox to octane
I beleive the wheels placement does have an impact on how your car will turn. Anyway, there‘s some spreadsheet on reddit about it, even if most cars share a common hitbox, every car in RL got his how turn radius, like scarab is about 2.30 while octane got 3.32, making the scarab turn slightly slower than the octane (or octane does tighter turns if you look at it the other way)
Edit : With how the car is placed at the beginning of shot 15, I‘m pretty sure it's about a wheeltouch. Since wheels are placed differently, you got another outcome
Scarab makes you play worse confirmed
I think the positioning of the hitbox is what makes the difference. Slightly more forward/backwards.
@@IfeelKindaSick The wheels are set different depths in a lot of hitboxes too. Hit a reset between octane and dingo. It's a significantly different feeling in spite of the same hitbox, whereas cars with similar visuals (like venom) feel very similar to reset with because it feels like the wheels are in a similar position.
@@014Darknessit makes u play better*
Sometimes I think that Rocket League was a fluke. Did Psyonix design the mechanics that have been discovered over the last 10 years or are they just happy accidents?
I think it's both,
They definitely add some of the basic mechs intentionally but the pros pushed the game to its limits.
@RandomGuy-qc8ml makes me wonder if there are more mechanics yet to be discovered. Also, if Psyonix updated Rocket League to Unreal 5 or made Rocket Legaue 2, how would they carry over these mechanics, that seem to be consequences of the physics, without breaking them.
they didn’t design these intentionally
but whatever the case they made it easiest for all these mechanics to flourish bc they had already made SARPBC
@@panzer00 I worry about that scenario tbh. I have to imagine they would preserve things like flip resets (an example of something unintended that the community discovered), as they have been adopted into the community as a whole. However, something like a wall dash, zap dash, ceiling shuffle etc, I could see them not caring about as much because they aren't as popular and might be difficult to code in if they don't work by default.
@idemus168 I feel that this is why they haven't "upgraded" Rocket League because it might be a downgrade in terms of mechanics. Of course, that's just conjecture.
This is gonna change the ball inside pillar meta forever
If you have access to the map asset, could you not maybe split the field in half overlap both asset and somehow highlight the area where it is not a 1:1 or make it list it in a text file or something? (I know nothing of how that works, might not be possible)
The field is made up of quarters that are the exact same, just mirrored!
@@Lethamyr If the cars and the field is physically identical then all I could see is that physics is the cause of the difference in that corner. So it would be even harder to know what happens there unless psyonix/epic can tell us themselves.
@@Lethamyr From my completely inexperienced eye, it looks like a collision thing then. The transition from wall to ceiling seems to be the cause of the desync, so I wonder if something funky is happening where those mirrored quarters of the field mesh together that causes the collision to behave differently in the physics engine during the transition. I know from livery editors in racing games that simply mirroring something in software sometimes doesn't work 100% as intended, and can need manual touch up to get a perfectly symmetrical result.
@@Lethamyr Maybe float position having different rounded values since different original x values?
like -100 rounding to -100.000184 while 100 rounds to 100.000217 (examples), hence a 0.000033 difference
yeah i also think this basically boils down to precision issues regarding float numbers. i guess if they used fixed point numbers for the physics it could be completely simmetrical, but unfortunately now it is not...
It's simply floating point precision issues. For example, there is no accurate floating point representation of numbers as simple as 0.3 or 0.4, and those errors accumulate reasonably fast. Keep in mind that every physics tick involves multiple floating point calculations just on the orientation and position of your car.
Edit: The arena "not being symmetrical", that is. As for the hitbox, yeah, they must have messed something up.
They are not simple values, floats are 32 bits and have 6 additional floating points of precision in c++. The hitbox for both octane and fennec are exactly "X: 59.003689, Y: 42.099705, Z: 18.079536". And thats for the hitbox itself, quats are used for rotating in 6 degrees of freedom.
In theory you could fix any errors by placing the field so that the ball spawns at origin, right? Also something might have to be done in the baking process of the field's geometry. Still this kind of thing is unacceptable to me.
@@Anteksanteri Hmm, I'm not sure that would solve it. For simple translation, yes, floating point representation is "symmetrical" in that sense (because of the sign bit), but I have honestly no idea if quaternions and rotation matrices wouldn't mess this up as soon as rotation gets introduced (which is already the case to even place the cars). If I don't forget, I'll test it.
rapid sounds like he has a deep voice changer on
The octane and fennec hitboxes are the same, but I'm pretty sure the octane's hitbox is rotated slightly, making it lower at the front and higher at the back compared to the fennec. I'm also pretty sure that almost every car in the game is unique in this way. Some cars share the same hitbox, but with different rotations/inclines.
18:38 that "stop going" reminded me of Fire. I miss that mf🥲
the placement of the hitbox, and the hitbox sizes are completely the same, but the wheel's hitbox on the Fennec are slightly offset, the touch you get while boosting for the Fennec is offset slightly lower, making it less of a wheel touch, the octane's hitbox however has the wheels positioned slightly higher, so when you hold boost on the shot, you get a "pinch" with the car's hitbox, and the wheel's, rather than just one or the other like most other hitboxes, oh and I tried it with the merc, and it actually worked on the merc's hitbox a couple times, though that might just be my stick drift tilting the car a little on spawn.
this is also false.
23:08 thwomp 😂
I love the videos where you challenge rapid or compete together😂
This has been a known phenomena for quite some time. All cars are in some hitbox category, however the alignment of the hitbox (rarely), and more importantly, the wheel locations are still independent for each car. The wheel location might be the main culprit here
this is also false.
@@redd_rl it is true.
@@nedasbolevicius8676 it appears lethamyr autofilters links, but rocketsim by zealanL on github has a carconfig file that showcases the objective values for every hitbox, with no difference per car. only per hitbox.
Seems like maybe a positioning issue of the hitbox. Fennac's hitbox must be positioned slighted behind the Octane's relative to the visual frame of the car. This has the effect of the hitbox starting at a different coordinate within the map. That would explain why shot #4 was so much harder on the Fennac because the hitbox needs to travel ever so slightly further than the Octane.
They are the same hotbox, but relatively placed differently when the game spawns them in, i think they are shifted differently relative to the wheels and the game also tracks the center of the car slightly different when it is spawned in
RAPID GOT SCAMMED SO HARD. If you get the right frame at 21:43, you can see he hit the ball before it went green.
Ur right I got
It was probably somewhere around frame perfect and so the game just was like nah I'm good
you know what it probably is on the mirror octane desync? i bet it's floating point errors accumulating until one car's angle rounds up while the other rounds down, thus the cars are basically the same until a very tiny change
Yea I think so. I did see someone else in the comments say the same thing.
the angle conversion in rocket league uses a lookup table.
I appear in yet another Leth video, mwahaha!
bro got him a warm up for speed flips 😂
This officially brings back the question; is the fennec better than the octane? Or is octane better than fennec?
Neither. Always depends on the person playing as there are no boosts to stats per car in this game. The car you use more you’ll play better on
Shivaxi and Leth being friends isn't something I knew, but is somehow perfectly on brand :D
24:28 rookie training 💀
FRfr
Idk if this matters much cuz idk how the hitboxes are built but it would make sense that fennwcs result is different because the wheels on the fennec are surrounded by the frame of the car while octanes wheels are free and has nothing around the except ontop.
it's because of collision geometry. Even with the same hitbox, if the models contact points or physics meshes are handled slightly differently by the game engine, it could cause them to interact with surfaces like ramps or walls in unique ways. This can result in one car touching down slightly sooner or having a different trajectory after drops.
Only thing I can think of is I remember different cars used to have very marginally different stats like turn radius being slightly wider / sharper. The differences were always very small so it would never make a real difference, and to my knowledge this was patched out years ago when they standardised all the hitboxes.
But this is the only thing I can think of, that maybe the Octane / Fennec have some slight stat difference which makes this happen.
potentially, but they're visually identical, so only remaining candidate is suspension wackiness.
24:00 slow as diddy got me😂😂
the only thing i can think off Fennec vs Octane, is the following:
have a look at what Dominus does. it grips the pillar and drives while holding boost only.
Fennec seems to almost always do the same.
could it be that the Fennec box is placed slightly lower, making it grip the pillar/wall more often then not?
and by lower i mean, this could be 1mm and we just can't notice when comparing it to Octane
A lower hitbox would also be evident in how it hits the ball on the ground. The pop would be higher with a lower hitbox. It could also be a slightly taller hitbox where the top of the car is at the same height but the bottom is slightly lower and that would be much harder to test accurately.
I had the same thought. and it could be such a small difference that it isn't noticeable to most people. You do hear pros sometimes say that fennec feels different to octane. if its like .01% closer to the ground or something that could be why.
@@BenRHarsh exactly, and one thing i noticed with Fennec is that i can pop the ball easier even when driving slower then i would with Octane.
Octane on the other hand i can get way more power out of barely any touch, its possible with Fennec but less likely somehow.
It's probably positioning of the hitboxes, they must be a few pixels one way or another to match the car better.
The hitbox on the fennec will have to be lower than the octane, potentially similar to the marauder (as the marauder is the octane hitbox but sits lower), which would make the marauder a RLCS level car lmao
14:55 "Just need to line up my squishy correctly" Pause 🤣
He means lining up his squishy save correctly
@@eranfrith5924 I know. But out of context that's funny.
The only thing I can think of is maybe the wheels or something are misaligned, or the the way the boost is put on the octane vs the fennec is different. That is very weird, besides that, I'm just as stumped as you!
wheels aren't aligned.
Ive done significant testing on this recently. Do resets between the octane and dingo and youll realize you have to hit a different part of the bottom of the hitbox to get it because of the wheel position. I think that and the wheel height affect the hitboxes as well, but its all anecdotal, as im a console player and cant scientifically test it from specific physics positions with commands.
13:01 For those who wanted to see how rapid got it out with the fennec before Leth told him 😊
I always told everybody the fennec was really not the same as octane no one believed me here we are
Same hit boxes but different center of rotation/pitch points?
Or different amounts of steering input accelerations or aerial depending on how close the hit box is to the nearest surface or wall. (idk if that sentence is legible)
Does the Octane hitbox ride higher than the Fennec hitbox when idle? Octane feels like a taller car with squishy-ier suspension travel to me than what the fennec feels like. Ride height wise.
Programmed environments like the rocket league arena are chaotic due to rounding error in the stored data. Error as small as machine precision will compound when both cars are set off on the same path. But hey maybe the arenas are actually not symmetric
If it was just a rounding error then it wouldn't happen on the same aide each time right? If you're taling about the thing at the start of the video, rapid and leth both tested it and the cars always took the same path on orange side and always the same path on blue side, it just wasn't symmetrical
i think we desperately need a Rocket Science video on this topic and the other bugs mentioned in this video
Shivaxi is the GoaT and the creator of RLCraft fun guy. Love his streams.
9:36 chipmunks ahh sound 💀😭
Leth: let the ball spawn it’ll be easier (x1892884)
Rapid: okay *continues not waiting*
I always had a feeling that they weren't because I play better in the air with octane and I play better on the ground with fennec
The Fennec could be spawning in a slightly off-set position, relative to the designated point.
18:00 smooth cut ngl
Wtf
THE DUDE USING THE VOICE CHANGER IS HILARIOUS
I switched to Fennec a while ago from Octane , what i noticed is that now when i switch back to octane i just can't get as good of a first touch for example when going for air dribbles , but it is also easier to do stall resets with octane , octane is also better for general shots , more power but then again if you do speed flip shots then it doesn't really matter , i think something has to do with fennec's visual , i think it is closer to the hitbox than Octane visually so you can be more precise with fennec , at least that is how i feel , but i only have 1400 hours so more qualified people prob know a lot more about this
BRO THE MAP LETING ME ANGRY WHEN I WATCH YOU WHEN SO CLOSE LOSE😂
Iirc psyonix adjusted the hitbox *position* of every car to better align them with the visual model.
Goat I love these challenges & your funky findings
Truly makes you wonder about all the times Leth has specifically said that it feels weird to play as the Fennec compared to Octane, maybe there was no placebo after all? It really has a different hitbox! 🤔
I heard on a video that they are changing the hitboxes, not by the car but by just shifting and adjusting their hitboxes to match the car. For example the Octanes hitbox is the same as the Fennec’s except the hitbox is shifted towards the center point of the car. Hopefully that helped
You should prank JG7 by making his car slightly smaller
Quick explanation : Octane hitbow is lower than fennec
additional note : u might have trouble wavedashing when switching between the two
oh and the fact the server physics on match making seem to be in the 60fps range now... which is why so many ghost touches and cars not in position in replays etc... this one is hard to prove but been slowly getting evidence
You dont need evidence lol, just google it "rocket league server tick rate". Sixty is fine for an online game, I rarely have ghost touches.
Joe swanson mustve been expensive to hire for the voiceover
What if you were to replay a game but switch the blue and orange sides with the cars having the exact same inputs? Would the slight changes due to the field being unsymmetric effect the game drastically after 5 minutes of gameplay? Wonder if any historical matches could have been different if the teams were just on opposite sides of the pitch.
I hope to see a deep dive video on this. Imagine somebody finding / proofing the reason. I think this could be very interesting.
I attempted this map and successfully completed it within the 30 minutes! However, I was able to do every other shot except for shot #1 using the Fennec. I switched to Octane and tried again and got it on my first attempt. Not sure if that was pure coincidence or not, but the cars definitely play differently. I didn't finish the video so as not to spoil the map, but I figured that's what you were talking about.
So could the answer just be that the hitbox is in a different spot on the car? Like maybe the hit box is positioned further back on one of the cars, but the game positions you based on the entire model rather than the hitbox? Just a thought.
Okay, hear me out.
Maybe its not the hitbox but the part of the car that the boost comes out, which is different on the octane compared to the fennec.
Both cars have 2 parts where the boost comes out, but on the octane they are a little more towards the middle and a little bigger, while on the fennec they are a little more towards the outside and a little smaller.
Well. I always felt it, always doubted those are not the same hitbox. The feeling was too different when dribbling a ball with Octane or Fennec, I have realized it because of the Dribbling Challenge 2 Overhauyl map on the Steam Workshop : I mained Fennec ever since I had it, and so I trained with it a lot on that map. And sometimes I go back to the good old Octane, and when I go on the dribbling challenge, well.. Those are not the same hitboxes, that is for sure : the ball ain't react the same.
Ok now I will watch your video, cause I wonder if you speak of the wheel gap too, cause I feel like one of those cars have a higher turn radius... :p
something else crazy if you do the same thing that they did in the first video where they boosted off of kick off the fennec travels a different path than the octane even if they are on the same side.
Bros content is amazing but he still only has 1mil subs he deserves more bro 💀
4:27 they both threw up at the same time haha😅
i already know this is gonna be a banger
same hit box but they might have different center points so the spawn is tiny bit different maybe? like if the phennac and octane are same hit box, but the phennac might spawn like a milimeter back or something?
Lethamyr, i know you probably wont see this but i main fennec when i play and yesterday i saw your video on how the map is uneven as you put it, however whilst trying it in the fennec you get a completely different route crossing the goal. I then tried in the octane and got the same u turn path you did showing the octane and fennec have a minutely different hitbox
fairly sure the whole uneven field thing is an optical illusion because of how the net shader moves separately from the actual wall hitbox
I’m wondering if the different boost thrusters have different effects.
Maybe with the octane the boosters are closer to each other compared to fennec it’s more spread out😅
13:30 the voices in my head on open nets
Hey Leth, are you planning anything for your second account? I miss the crab champions content! Thanks for your dedication, love your videos.
20:01 “I’m touching it” 💀
I like these kind of silly training packs so I tried it for myself. And idk what it is but some of the shots behave completely different on Playstation.
Shot 8 is impossible because the ball doesn't bounce up, it bounces straight out of map.
Shot 12 behaves completly differently, instead of bouncing of the pillar back to the blue side it goes towards the right side of the orange goal.
And on Shot 15 when you do nothing and let your car just fall, with the Octane the ball will pop out the pillar at the end, regardless if you boost or not, while that won't happen with the Fennec. Also while doing nothing the Octane will end up on its roof while the Fennec will land on it's wheels. No idea why it's like that but it's somewhat interesting and I wonder if anyone has a explanation for it.
I uploaded the shots in question to my channel just in case anyone wants to see it.
Makes it seem that the octane hitbox might sit a little lower than Fennec. Same hitbox but position/distance to the ground probably is different.
For the beginning of the video were you borh using keyboard? If not there may have been a tiny difference in the value of one of you pressing down the trigger
Maybe the center of mass is different so on spawn in ur occupying a slightly different space.
“You like that do you like it”
“I do like it”💀💀💀💀💀💀💀💀💀💀
im gonna take a wild guess and say that floating point accuracy is the issue for the map not being even. im assuming that the "origin" (0,0,0 in world space) is on the blue net, while the orange net is 100,0,0 (not actual values, just an example). as numbers get bigger and bigger, computers cant store them correctly which means sometimes movement, physics or other bugs might occur. 0.1+0.2=0.300000004 not 0.3 for example.
Aren't they the same hit boxes, but positioned differently on the car graphics? So maybe this has a bigger impact in terms of spawning in the air on a wall vs spawning on the ground w/gravity holding you down. Probably wouldn't make a huge difference, but I wonder if these hit box placements could affect your ability to reach the ball sooner in those speed ones for example. That's what I was thinking watching him go through the ball over and over. maybe that hitbox is slightly farther back on the car, even though its the same hit box.
The “IM HIM” followed immediately by dropping an easy dribble sent me
well time for another Rocket Science video
I’ve always said they are not the same hitbox. If you use both cars it’s so clear. The ball interacts with the cars so much differently. I’ve also always said that I feel like the ball just glues to the octane more than the fennec. Maybe that’s why lol, the octane could literally have some sort of pull on the ball for some reason 😂 it’s why the octane is so good lmao
DaddyLotus, you always have me laughing and farting with the sound effects and edits
i think the octane does that because the fennec is actually moved in the bgack by a pixel or 2 so it makes it go ro the wall instead of doing that the octane did
Hey, Leth! I just installed UDK 2015 and am planning on making custom RL maps, including 'totally normal Rocket League maps'. Would you be interested in giving them a try when they're finished? I'd put links in the RLMM discord.
Always better to have something ready before asking if the other party would be interested.
@@Mrlemar1 Sound advice! I already have started work on a couple of them and hope to have them ready soon.
I always had a feel the sides of the field weren’t symmetrical and that the fennec was different than the octane. What other weird quirks are there that we thought were placebo just to be true 😂
I thought we knew this,if u play with them they both feel different.Some players are better in octane and some are better in fennec,it's hard to find someone who doesn't prefer one over the other.
Wait till you start investigating the balls and game speed and car touches on a day to day basis. The game is filled with inconsistencies. Thats why if you dont play for 4 days your cold. Cuz your coming back to a diff game every time and gotta adapt to that new game.
Even if cars share hitboxes, every car got their wheels and center mass at different places. Result in the same hitbox but different turn radius and also maybe why you get a different outcome with the shot 15
To this day i still dont understand WHY CARS HAVE DIFFERENT HIT BOXES
Is the run with Rizzo available somewhere?
that'd be like torture 😭
They may have the same hitbox but they probably spawn in slightly differnet areas.
Nahhh we knew the hit box wasn’t the same already lol
i started dying on shot 13 where he could only shoot it not facing ball lmao
I think its because the fennec is very very slightly lower to the ground than octane. Just a guess, but I bet it is testable.