PHYSICS PROBLEMS IN POLYBRIDGE 2

[Antiriad]

1. Connection joints have an infinite stress limit, and apparently suffer no deformation (why?).
2. There seems to be no plastic deformation, at all.
3. Is the diameter & mass/momentum of the struts taken into account?
4. There seems only to be few physical laws being used which are: a) Hooke's law,b) Stress/Strain & young's modulus, c) the laws of acceleration, velocity, time and distance, d) the conservation of mass, momentum and angular momentum, and e) some kind of substitute law for gravity but I can't tell if its accurate or not. Oh, I forgot one! Friction is in there as well....
5. Why do the anchor points have no torsional stiffness but they have infinite strength (& cannot be deformed)?

please explain dry cactus. thanks!

[Gorby]

Publicado originalmente por Antiriad:

Publicado originalmente por Gorby:

Look, we get it; you think you're extra special so when you asked about some features you wanted in the game and were told you weren't getting them, you threw a little fit and claimed anyone who disagrees with you is childish. Mission accomplished! The fact that you had to go back 5 years to find a post from someone that agrees with you is hilarious, not to mention the fact that the replies on that thread prove my point - as you acknowledged.

Oh I'm on the floor laughing, that you think I even care. I think its clear you are a teenager or something, so anyway good luck with your gaming. I wish you no ill-will. Peace.

I can tell you don't care by the way you took the time to find and reply - multiple times in a row - to all my recent replies on other threads like some kind of stalker. I meant no ill will either. You were the one coming back with the condescending attitude to my genuine replies, which you apparently perceived as some sort of attack or challenge. Relax buddy, not everyone on the internet is trying to fight you.

[Garou]

Something that detailed wouldn't be sold as a game on Steam for $15.

[Technopiper]

1. The only instrument in the soundtrack is guitar (why?).
2. Coloration of the scales is lacking What about chromatic scale? Pentatonic? Dorian?
3. Please consider a symphonic approach: more instruments, give us an orchestra!

[Bamboofats]

If this game is going closer to FEM style of simulation with all the deformations, it's not going to run as well on many folks' hardware.
I rather have the simplified physics, with real time simulation, vs having to wait seconds to precompute the solution each time.

[Antiriad]

Publicado originalmente por VKNG:

Something that detailed wouldn't be sold as a game on Steam for $15.

Yeah. I think you might be right. A bit disappointing, but it is what it is.

Since I bought this game three days ago, I'm now finding out there are a lot of similar bridge building games that preceded it, almost to the point where I'm thinking this is just a straight rip-off of someone else's work....

[Doohicky]

I'm perfectly satisfied with the game. I did my research before buying so I knew what I was spending my money on.

[Antiriad]

Publicado originalmente por Harald Hardrada:

I'm perfectly satisfied with the game. I did my research before buying so I knew what I was spending my money on.

Did you know about all the other games out there? Like bridge builder? I just bought it on a whim thinking it was interesting but didn't know much about it.

[Bamboofats]

There is that bridge builder series, then went onto Pontifex and Bridge It. It had a more serious feel, but it seems polybridge was more well received on steam.
Polybridge, and a few other more "cartoony" games, like bridge constructor portal, and When Ski Lifts Go Wrong, allow more creative solutions and scenarios involving jumps and stunts. Different kind of fun for different tastes

Edit: if you want something a bit more serious. You can try the Bridge It demo on steam. I played it 6? years ago. There are tension and compression red blue colours, option for cross members, random vibration test, and hydraulics. But I think there were also serious physics shortcomings even with these sort of games. And generally not as polished as polybridge.
Also, it can get boring quickly for some of us, when the challenges eventually become building larger and larger spans, and taking on heavier loads. As compared to doing more unrealistic objectives in polybridge like jumps, loops, monster trucks

[Antiriad]

Publicado originalmente por Bamboofats:

There is that bridge builder series, then went onto Pontifex and Bridge It. It had a more serious feel, but it seems polybridge was more well received on steam.
Polybridge, and a few other more "cartoony" games, like bridge constructor portal, and When Ski Lifts Go Wrong, allow more creative solutions and scenarios involving jumps and stunts. Different kind of fun for different tastes

Edit: if you want something a bit more serious. You can try the Bridge It demo on steam. I played it 6? years ago. There are tension and compression red blue colours, option for cross members, random vibration test, and hydraulics. But I think there were also serious physics shortcomings even with these sort of games. And generally not as polished as polybridge.
Also, it can get boring quickly for some of us, when the challenges eventually become building larger and larger spans, and taking on heavier loads. As compared to doing more unrealistic objectives in polybridge like jumps, loops, monster trucks

Thanks. I'll check those out.

I really got into this at first, and then I started to feel like it's a bit pointless building bridges if the game physics are rigged in certain ways that don't make sense. Might as well play lemmings or something.

There is a review of Kerbal Space Program which says, "KSP's ultimate promise to the player is not that you'll crack a puzzle that has been set by a designer, but that you'll crack a puzzle set by reality" - maybe I'll give that a go as well.

[I|III||I|||II|IIII|]

Publicado originalmente por Gorby:

No, I don't... and judging by the lack of comments and complaints about this I'm guessing most players probably don't even know what torsional stiffness is, let alone care about it.

Complete all Main and Challenge Worlds levels under budget and unbreaking.

.1% of the people who bought this game actually bother trying to complete it. You can make an argument that people dont care because they don't complain or you could make an argument that people don't care, so they don't bother.

While I agree most people can't visually quantify torsional stiffness (myself included) doesn't mean they don't have a general understanding that things will twist and bend under load.

[Bandana Girl]

It's basically already doing the standard FEM that most engineers use for truss simulation. If you exceed the elastic limit in a structural element, that element might as well have failed (most engineers and FEM software will stop there for redesign, but PolyBridge just breaks it and continues on). Most engineers will ignore torsional stiffness for truss anchors (hell, most real anchors intentionally remove as much torsional stiffness as possible) and also usually assume that either the truss elements don't deform or the anchors don't deform, and if you're trying to solve for all of the forces in all of the truss elements, assuming that they're the ones deforming makes your system of equations much more consistent, especially if you're using Runge-Kutta, which works great for simulating on the fly.
Truss elements are also assumed to only act in tension or compression, which means other than for computing their elastic limit and linear weight (which are preprogrammed values) diameter means absolutely nothing. For roads rope and cable, their mass and momentum are taken in to account (which leads to the falling road exploits in the original; haven't checked to see if they're still there) while I'm pretty sure wood and steel are assumed to have negligible weight compared to their joints.
The only reasonable complaint you have is unlimited joint strength, but I can understand the programing constraints of deleting and creating new joints mostly arbitrarily while doing a live simulation. I can't imagine the headache getting split joints to merge without a noticeable stutter must have been, though I guess there are some optimizations you could make in preparing them as they approach

[Antiriad]

Publicado originalmente por Bandana Girl:

It's basically already doing the standard FEM that most engineers use for truss simulation. If you exceed the elastic limit in a structural element, that element might as well have failed (most engineers and FEM software will stop there for redesign, but PolyBridge just breaks it and continues on). Most engineers will ignore torsional stiffness for truss anchors (hell, most real anchors intentionally remove as much torsional stiffness as possible) and also usually assume that either the truss elements don't deform or the anchors don't deform, and if you're trying to solve for all of the forces in all of the truss elements, assuming that they're the ones deforming makes your system of equations much more consistent, especially if you're using Runge-Kutta, which works great for simulating on the fly.
Truss elements are also assumed to only act in tension or compression, which means other than for computing their elastic limit and linear weight (which are preprogrammed values) diameter means absolutely nothing. For roads rope and cable, their mass and momentum are taken in to account (which leads to the falling road exploits in the original; haven't checked to see if they're still there) while I'm pretty sure wood and steel are assumed to have negligible weight compared to their joints.
The only reasonable complaint you have is unlimited joint strength, but I can understand the programing constraints of deleting and creating new joints mostly arbitrarily while doing a live simulation. I can't imagine the headache getting split joints to merge without a noticeable stutter must have been, though I guess there are some optimizations you could make in preparing them as they approach

Thanks for the interesting comments! Some good points there. Glad to hear from an expert. Still have some questions tho about the weird world created by Dry cactus!

1) At an anchor point, why would you NOT constrain the rotational (or torsional) degree of freedom? Or at least give it some resistance? It supposed to be a structural safe-point on which to build everything else.....

That's the part I can't really understand. This means I can't build a single beam cantilever out from a fixing point because it just flops down, and a single beam vertical column simply falls over. This also means you have to design hack and build build a triangular support frame that rests on the cliff face! That's just weird. Although its been a long time since i used FEM, I don't remember that it was 'illegal' to control the rotational degree of freedom at the boundaries. Same applies to the connection joints that are just free to flop around, and its not to do with the underlying maths either.

I also get that you may allow torsional/rotational freedom on a real bridge to relieve stresses, but if its a cantilever then no way! You need that resistance otherwise the thing topples down man!

2) My whole point about plastic deformation is that the joint may well be considered 'failed' after reaching the elastic limit, but the bridge might still not fall down. Fair enough though on that one - its maybe a bit hard to model for a real-time simulation.Would be probably less fun as well if things were allowed to just stretch rather than go bang!

[XratedSippyCup]

this chat string is fun to watch. can anyone pick on me? and please assume i am whoever you want me to be in your rant

[Technopiper]

Things also don't move sideways in this game: it's a 2D world. Poly Bridge didn't set out to be an ultra realistic game so why should it improve in that direction? It's like asking for realistic blood loss in FPS.

[Bamboofats]

Publicado originalmente por Antiriad:

1) At an anchor point, why would you NOT constrain the rotational (or torsional) degree of freedom? Or at least give it some resistance? It supposed to be a structural safe-point on which to build everything else.....

The game's structural elements such as wood steel etc, are modelled as tension and compression trusses. In this case it requires the anchors to be pin joints.
If you have the anchors constrain the rotation as well, it would add bending moments to the elements.
I think this would also be an interesting option (anchors that can defined as pin/fixed/roller etc), but maybe it would add extra complexity the Devs didnt want to include for the style of gameplay.