To try everything Brilliant has to offer-free-for a full 30 days, visit brilliant.org/fusarihan . You’ll also get 20% off an annual premium subscription.
The spine really needs at least one more point to bend. Preferably around the stomach since most of the bending happens around the lower half of the torso.
I remember playing with Pivot as a child and being mind blown at how much my animations improved once I started adding a middle point on the spine of my stickmans
Hey man, I've been following your development and its coming out amazing. I would love if you could have your videos go a bit more into depth technically as it would benefit both the algorithm (having 10 min + vids) and those like myself who are interested on the technical side of this amazing full body physics based dribbling system. Kind of like your older videos. Either way, good luck on your journey and can't wait to see more. Hope you blow up in popularity and btw your English has improved tremendously since the beginning that you now sound native.
I really admire how deep you go into your solutions. If you're going to add a head you might want to keep it from bouncing so much during movement. Usually our bodies try to keep the height of the eyes still while moving as to not get disoriented. I think you can get away with letting just the neck do some balancing as to not have to rewrite a bunch of stuff. Maybe you already knew this and didn't have a chance to add it in but thought it could be worth a mention.
As always, a great update Furkan! It's good that he objective scope is getting bigger, it means you actively try to implement new things and are constantly improving the overall quality of the simulation, it is a great info for us! Keep it steady, it already looks very promising, I will be patiently waiting for the new updates. It's a game/environment we all look forward to, a proper simulation of football. Take care, much health and prosperity for you brother!
Hey Furkan! Really thankful for this updates, keep them coming! Your game is evolving a lot each day! Keep up the good work and when you feel its ready for testing we will be here happy to help you improve the game!
This is the first attempt at making a proper physics based movement system that hasnt been done by EA and it's one thats slowly getting insanely good. Can't wait for the demo!
I understand that you want to show the game when its closer to your dreams, but sharing it on this stages could help you identify current issues & possible improvements that you didn’t think of. It can be tougher at your future versions since it would be complicated. But i still admire your improvements and passion. Keep up the great work ❤
This looks really great so far! The ragdoll-y movements are really charming, are you planning on keeping it that way or are you going to move towards more realism?
Super interesting man! I wonder how performant procedural character simulation like this is. I think it would be interesting to design an rts game someday with a lot of units. I know traditional bone animation systems can be slow with 1000s of animated characters, so rts developers often bake the animations into the mesh and upload an "animation sprite sheet" to the gpu instead. I wonder how efficient an approach like this would be in comparison? A system like this is so dynamic and could add a lot of depth to any game! Keep up the good work!
Thank you! Performance mostly depends on the specific problem. I’ll share more details about the limits of the system when I increase the character instances!
This looks amazing. The character system looks promising for all types of games, not only football, so you'll be able to use it in your future projects, which will make them unique. If you'll need money, maybe you'll be able to adapt it for one of the most popular game engines and sell it in specific asset store (Unreal or Unity, not sure if there's something like that for Godot). This project is a great investment, no matter if you'll make a succesful football game.
Eline saglik guzel video olmus bu kol hareketinin awkward durmasini muhtemelen sunlar sebep oluyor omuzlarin da kosarken ileri geri hafif oynamasi gerekiyor bence bu ilk sebep ikincisi jogging temposunda kol hareketi normalde de komik durur bence cok iyi olmus bi diger arkadasin dedigi gibi mide kismina bi point daha iyi olabilir bu arada yuksek hizda on bacagi biraz daha kaldirmasi daha iyi olabilir mi genel hiza oranla bacak fazla mi hizli falan tam emin olamadim cok hafif bi oynama yapilabilir belki o konuda , max hizi uzun sureli inceleyemedim kesin bisey demek zor
Hey totally random, but could you add the french flag in the back with the other flags? ^^' Anyways i loved the video as always, i want to see how it progresses
your work has been excellent but i think you're disappointed because these results aren't at the same level. i don't pose the arms straight down but at a small angle out from the body. also, when a person walks or runs, they pivot on the Y axis, which is going to add a lot of more natural looking flailing to your arms. when a person wants to run fast, they pull their arms in tighter. the hips and shoulders are lifted somewhat as well, you can even animate these using sines (or anything derived from them) at different phases. i'm sure you can get believable, much better results. see if you can make it move like a sassy runway model, then tone it down some. take some time to do it silly, and learn from it :) and add a waist and another pivot around heart level :) and see my point to segment et c. codes! :)
hopefully y'all can figger out point to point yousef. and similarly, here's a constrainment method for skeletal points.. lw (left wrist) to le (left elbow), so eg. lwv is left wrist velocity. prev is a temporarily declared point used to store the preconstrained position until the end, where the difference is used as the velocity :) which makes it very stable. and hopefully you can discern td is a distance not a point, and my weights are arbitrary.. oh, and lwi is an ideal position the actual position is trying to reach prev = lw; lw += lwv; t = le - lw; td = (t.mag() - arm2length) * soft; t = t.norm() * td; lw = lw + t * 0.8; le = le - t * 0.2; // weight these, .7 .3? t = lwi - lw; lw = lw + t * strength2; lwv = lw - prev;
To try everything Brilliant has to offer-free-for a full 30 days, visit brilliant.org/fusarihan . You’ll also get 20% off an annual premium subscription.
The spine really needs at least one more point to bend. Preferably around the stomach since most of the bending happens around the lower half of the torso.
Yep
I remember playing with Pivot as a child and being mind blown at how much my animations improved once I started adding a middle point on the spine of my stickmans
I think one point that is more flexible then 2 other less flexible would be great
two.. unless teddy bears are ok :p
Or go crazy with 9 points
Hey man, I've been following your development and its coming out amazing. I would love if you could have your videos go a bit more into depth technically as it would benefit both the algorithm (having 10 min + vids) and those like myself who are interested on the technical side of this amazing full body physics based dribbling system. Kind of like your older videos. Either way, good luck on your journey and can't wait to see more. Hope you blow up in popularity and btw your English has improved tremendously since the beginning that you now sound native.
I really admire how deep you go into your solutions. If you're going to add a head you might want to keep it from bouncing so much during movement. Usually our bodies try to keep the height of the eyes still while moving as to not get disoriented. I think you can get away with letting just the neck do some balancing as to not have to rewrite a bunch of stuff. Maybe you already knew this and didn't have a chance to add it in but thought it could be worth a mention.
you're making me more excited for the next video each time you drop.
As always, a great update Furkan! It's good that he objective scope is getting bigger, it means you actively try to implement new things and are constantly improving the overall quality of the simulation, it is a great info for us! Keep it steady, it already looks very promising, I will be patiently waiting for the new updates. It's a game/environment we all look forward to, a proper simulation of football. Take care, much health and prosperity for you brother!
another banger, the movement is really coming together :D Keep it up man!
inanılmaz bir gelişim , bunu görmek çok heyecan verici gelecek güncellemeler için sabırsızım🤩
Hey Furkan!
Really thankful for this updates, keep them coming! Your game is evolving a lot each day! Keep up the good work and when you feel its ready for testing we will be here happy to help you improve the game!
The game would be pretty unique if chacters will be just pair or legs, nothing more nothing less
and creepy😅🙏
Bro we need tutorials of how we can make a game engine like this! Awesome vid btw❤
Looks amazing man!
Another amazing vid man keep it up😃🇧🇬
Brilliant progress on the body! Seeing it flop around reminded me of Chicken Run for some reason
This is the first attempt at making a proper physics based movement system that hasnt been done by EA and it's one thats slowly getting insanely good.
Can't wait for the demo!
This looks amazing!!!
I understand that you want to show the game when its closer to your dreams, but sharing it on this stages could help you identify current issues & possible improvements that you didn’t think of. It can be tougher at your future versions since it would be complicated. But i still admire your improvements and passion. Keep up the great work ❤
Thank you! You’re totally right! I hope I can share the game as soon as possible.
This looks really great so far! The ragdoll-y movements are really charming, are you planning on keeping it that way or are you going to move towards more realism?
I though the character will be only the legs
damn those stairs looked good
Yas an update! Cant wait to go my hands on this!
Tebrikler. Videoda görülen sadece kaygan zeminde ayakların kayabilme özelliği ile bile güzel bir oyun yapılabilir
nice job
am going to be here everytime u upload pls keep working on this project
Pelvis orientation calculation
it's hard to talk about stuff on the internet *issues ban* :)
Great job ! Keep up the good work.
başarılı
Really cool stuff!
şuraya artık bi antreman sahası yapmanın vakti geldi
i hope ur game gets published succesfully
Türkiyenin gururu❤
Super interesting man! I wonder how performant procedural character simulation like this is. I think it would be interesting to design an rts game someday with a lot of units. I know traditional bone animation systems can be slow with 1000s of animated characters, so rts developers often bake the animations into the mesh and upload an "animation sprite sheet" to the gpu instead. I wonder how efficient an approach like this would be in comparison? A system like this is so dynamic and could add a lot of depth to any game! Keep up the good work!
Thank you! Performance mostly depends on the specific problem. I’ll share more details about the limits of the system when I increase the character instances!
nice job!
This looks amazing. The character system looks promising for all types of games, not only football, so you'll be able to use it in your future projects, which will make them unique. If you'll need money, maybe you'll be able to adapt it for one of the most popular game engines and sell it in specific asset store (Unreal or Unity, not sure if there's something like that for Godot). This project is a great investment, no matter if you'll make a succesful football game.
Thank you! You described the path that I’m following!
TEACH ME MASTER!!!
Amazing stuff yo👏
cant wait to change the model with Ronaldo
Eline saglik guzel video olmus bu kol hareketinin awkward durmasini muhtemelen sunlar sebep oluyor omuzlarin da kosarken ileri geri hafif oynamasi gerekiyor bence bu ilk sebep ikincisi jogging temposunda kol hareketi normalde de komik durur bence cok iyi olmus bi diger arkadasin dedigi gibi mide kismina bi point daha iyi olabilir bu arada yuksek hizda on bacagi biraz daha kaldirmasi daha iyi olabilir mi genel hiza oranla bacak fazla mi hizli falan tam emin olamadim cok hafif bi oynama yapilabilir belki o konuda , max hizi uzun sureli inceleyemedim kesin bisey demek zor
Geri bildirim için teşekkürler!
Why do you need interactions with stairs, when working on a soccer game?
Nimsony leave that body 😂
Bro how do this 😮😅
Have you tried Jolt for the physics engine?
I saw it but never tried it.
@@fusarihan cool, I myself never used it either, but it seems to have positive feedback
@@MirralisDias I’ll check it out. It looks stable and performant with a modern api.
ellerine sağlık
Hey totally random, but could you add the french flag in the back with the other flags? ^^'
Anyways i loved the video as always, i want to see how it progresses
somehow you accidentally didn't type west papuan independence movement flag
@@atomictraveller surprising ikr
əlasan !
Are you planning to add networked multiplayer down the road?
Yes, I hope I can do it!
In the conclusion, where is the rest of the player's shadow?
Shapes are custom renderable and I needed to implement the renderDepth function for them. So, I just didn’t but I will 😅
your work has been excellent but i think you're disappointed because these results aren't at the same level. i don't pose the arms straight down but at a small angle out from the body. also, when a person walks or runs, they pivot on the Y axis, which is going to add a lot of more natural looking flailing to your arms. when a person wants to run fast, they pull their arms in tighter. the hips and shoulders are lifted somewhat as well, you can even animate these using sines (or anything derived from them) at different phases. i'm sure you can get believable, much better results. see if you can make it move like a sassy runway model, then tone it down some. take some time to do it silly, and learn from it :) and add a waist and another pivot around heart level :) and see my point to segment et c. codes! :)
It's incredible how you’ve almost completely lost that thick accent in just 7 months.
I’m sorry but this is a clone voice from ElevenLabs based on my thick accent 😅
@@fusarihan Oh I see 🤯 Clever solution, thank you for clarifying!
My boy's idea is so freaking unique, I expect for it to blow up and get everywhere🙏🙏❤🔥
Turkish John Carmack
Yok hocam o kadar değil 😅
I'm not sure you've ever seen a person move
I appreciated your observation but it’s a work in progress 😄
Doesn’t look as good, give it some time and think it will have potential.
That’s correct, it’s a work in progress!
@@fusarihan hey man, big inspiration for many people, keep it up!
bool pointtosegment(vec3& a0, vec3& b0, vec3& b1, double r0, double r1, double r2, double w0, double w1) {
vec3 v0 = b0 - b1; vec3 v1 = a0 - b1; double td = v0.dot(v1); bool colltrue = 0;
if (td > 0) {
double td1 = v0.dot(v0);
if (td1 < td) { // closest to b0 end
double ts = r0 + r1;
vec3 t = a0 - b0; td = t.mag();
if (td < ts) { colltrue = 1;
td = (td - ts) * segdamping; t = t.norm() * td; b0 = b0 + t * w1; a0 = a0 - t * w0;
} }
else {
double ts = r0 + (r1 + r2) * 0.5;
vec3 t = b1 + v0 * (td / td1); t = a0 - t; td = t.mag(); // segment
if (td < ts) { colltrue = 1;
t = t.norm() * (td - ts) * segdamping;
b0 = b0 + t * td * w1; b1 = b1 + t * (1 - td) * w1; a0 = a0 - t * w0;
} } }
else { // closest to b1
double ts = r0 + r2;
vec3 t = v1; td = t.mag(); if (td < ts) { colltrue = 1;
td = (td - ts) * segdamping; t = t.norm() * td; b1 = b1 + t * w1; a0 = a0 - t * w0;
} }
return colltrue;
}
bool segmenttosegment(vec3& a0, vec3& a1, vec3& b0, vec3& b1, double r0, double r1, double r2, double r3) {
vec3 sega = a1 - a0; vec3 segb = b1 - b0; vec3 r = a0 - b0;
double a = sega.dot(sega); double e = segb.dot(segb); double f = segb.dot(r);
double s, st;
double c = sega.dot(r); double b = sega.dot(segb); double denom = a * e - b * b;
if (denom != 0) {
s = (b * f - c * e) / denom; s = max(0, min(1, s));
}
else s = 0;
st = (b * s + f) / e;
if (st < 0) {
st = 0; s = max(0, min(1, -c / a));
}
else if (st > 1) {
st = 1; s = max(0, min(1, (b - c) / a));
}
vec3 closesta = a0 + sega * s;
vec3 closestb = b0 + segb * st;
double radab = r0 + s * (r1 - r0) + r2 + st * (r3 - r2); // interpolation of radii of both segments
vec3 diff = closesta - closestb;
double dsq = diff.magsq();
if (dsq < radab * radab) {
diff = diff.norm() * (radab - sqrt(dsq));
diff *= segdamping; // damping
a0 = a0 + diff * s;
a1 = a1 + diff * (1 - s);
b0 = b0 - diff * st;
b1 = b1 - diff * (1 - st);
return 1;
}
else return 0;
}
hopefully y'all can figger out point to point yousef.
and similarly, here's a constrainment method for skeletal points.. lw (left wrist) to le (left elbow), so eg. lwv is left wrist velocity. prev is a temporarily declared point used to store the preconstrained position until the end, where the difference is used as the velocity :) which makes it very stable. and hopefully you can discern td is a distance not a point, and my weights are arbitrary.. oh, and lwi is an ideal position the actual position is trying to reach
prev = lw; lw += lwv; t = le - lw; td = (t.mag() - arm2length) * soft;
t = t.norm() * td; lw = lw + t * 0.8; le = le - t * 0.2; // weight these, .7 .3?
t = lwi - lw; lw = lw + t * strength2; lwv = lw - prev;