The physics are way off on this game

[HornetMaX]

Out of curiosity, would it be possible to take one MXB track (the default one would be OK) and port it to MXS ?

That's something Rafagas (known track creator from Mx Simulator) did before the game was released. He ported the example track.

The jumps in this game are smaller, it feels like everything floats in Mx Simulator, there's almost no weight to the bikes. I'm not sure if I still have the track, but if I have I might be able to make a quick helmet cam video for you.

That's my impression too: as I said commenting the video posted by al167, a bit of gas in MXS seems to send the bike 10m high.
But I'm more interested in seeing how the two handle tracks that are a bit bumpy.

Did anyone of you see my post about physics? It was in another topic, and I didn't get a single reply to it.

To me, most of what you reported is tied to what MXB does whenever at least one wheel is not in contact with the ground.

In GPB there's a similar issue: control while wheeling (even slightly) is definitely needed and not particularly good right now even if I think something improved a bit in the last GPB beta (6).

To me MXB (and GPB, to a lesser extent) should have, roughly speaking, 2 or 3 different virtual riders: one is the usual one, to be used when both wheels are on the ground, one to be used when wheeling and one when mid-air. A target lean angle may still be meaningful when wheeling but when mid-air ...

In general, this kind of switching from one control method to another (triggered by the fact the underlying physical model under control changes) is fairly hard to implement properly and probably even harder in our case, as we want the same stick input to be used in the different situations (likely).

From what I've seen between MXB beta1 and 2, I'd say that Piboso/Snappe are heading in the right direction.

Steer twitchiness is also something we notice in GPB: many think this should be solved by more steering damping. I think that maybe the virtual rider stabilizing PID control needs some attention, especially it's D term: no real rider can intentionally shake the bars that quickly.
The max torque the virtual rider can generate on the bars is limited (as it should be), but it seems as if there was no limit on the derivative of that (i.e. on how quickly you can go from lock to lock). What could generate that ? Can't say for sure: D gain too high, or maybe D term implementation with no roll-off ... or maybe something as simple as considering that on top of the inertia of the steering head, you also have the inertia of the rider's arms.

MaX.

[𝖙𝖋𝖈]

I'd be interested in seeing a video if you fancy making one. I don't agree with all of your points (note the 'all' please!) - Think I was going to reply but thought I'd wait to see what PiBoSo / others said.

Sometimes I feel like one of the only ones who doesn't suffer all the problems people constantly bring up. I think firstly a lot of issues people have are repeated and already mentioned, and also think that seeing as PiBoSo is aware of not only all genuine problems but also what the community think then you guys should wait and see what happens in beta 3 - As MaX said above, the difference between b1 and b2 is huge and heading in the right direction nicely (remember GP bikes is in beta 6c - That's a fair amount) 

Personally I'm starting to think that something is wrong with the weight of things. Note here that PiBoSo has already made posts regarding this and although we don't know if it's fixed or not, it's being looked at which is good enough for me. Sometimes it feels like the bike is too front heavy - Nosedives still happen in unexplained situations although not nearly as often as beta 1, and it could also be a factor in the front wash out and rear slide. Try getting the bike a bit sideways going up a hill and it behaves far more like you'd expect.

[GDUBMX]

I have to agree Fat's. Al also made some very interesting points about steering which I didn't actually notice until he mentioned it. Personally I'm buzzing about seeing this forum pick up again and feel there is a general buzz about the forthcoming b3. Piboso has been quiet lately so hopefully he decides to drop beta 3 so we can all get stuck in to the next stage of development. There really is some great discussion here and most of the time is out of my realm lol, not the sharpest knife in the draw! Its good tho because I learn so much from these types of threads. btw I think we need to sort a race out for all forum members this weekend,would be cool having a 15 man moto on Paleta raceway! Anyone fancy it?

[Phathry25]

Well, you may want to re-read my post then.

Point 1 was definitely not about graphics (albeit point 2 was).

I said "in random order", implying point x was not to be considered more important than point x+n (with n positive).

And finally (and most important), all the points past point 2 (and including point 1) were about physics.

So, when you're ready to discuss a post after having read all of it, let me know

MaX.

I read the whole thing. I was proving a point. Your judging a game off of one short video. It's laughable.

You're critical of an "automagic" save in the video but yet are singing the praises of a game that uses an automagic rider to control the steering. How can MX Bikes be considered a simulator using the lean angle control scheme?  That's like controlling turning radius and letting an automatic driver decide what to do with the wheel in a car simulator. Sounds like an arcade game to me.

IMO More time needs to be spent developing the "hardcore steering" mode. Gyroscopic stabilizer, or just more control options like al167 mentioned. Until then it's all a guessing game at what the virtual rider is doing, questioning the simulation is silly when you have no control over the handlebars anyways.

In one point you stated you said the track was too flat. Later on you said the bike was too permissive over bumps.  Which is it?  Either way the tracks aren't flat, and it's a motocross bike, it's going to be super permissive over bumps. They are designed that way.

Yes the gyroscopic stabilizer is pretty automagic when it goes to saving big slides, it's a little too effective IMO, but it is what it is. You can't be critical of the simulation merits of that and act like virtual rider and automagic leg push is any better.

[Mace-x]

Well, im sorry man but at least this auto rider works way better than mx simulator's brick rider, that rider doesnt react to anything, it's a dead boey except for leaning front and back, tell me how that can be called simulation pffft.

I understand, both games has flaws, each one has different strong points but your critics are biased towards mx sim, i respect if you like it more but looking it from outside mx sim is not far from mxb, and it had waay more development time.

[𝖙𝖋𝖈]

btw I think we need to sort a race out for all forum members this weekend,would be cool having a 15 man moto on Paleta raceway! Anyone fancy it?

I'm up for that, on any track! 

Lets keep this thread OT tho - Start a topic in the races forum.

[PiBoSo]

Yes the gyroscopic stabilizer is pretty automagic when it goes to saving big slides, it's a little too effective IMO, but it is what it is. You can't be critical of the simulation merits of that and act like virtual rider and automagic leg push is any better.

Please note that it will be possible to disable automatic dabbing.
Also, with hardcore steer mode the virtual rider is completely off.

[HornetMaX]

I read the whole thing. I was proving a point. Your judging a game off of one short video. It's laughable.

Oh come on man, I said *very explicitly* that I was judging from that one video (and the little I recall of when I tried MXS), so my "judgement" (better, impressions) was to be taken with care. Please.

Notice however that plenty of things can be deduced from a video. Not everything of course, but still plenty.

You're critical of an "automagic" save in the video but yet are singing the praises of a game that uses an automagic rider to control the steering. How can MX Bikes be considered a simulator using the lean angle control scheme?  That's like controlling turning radius and letting an automatic driver decide what to do with the wheel in a car simulator. Sounds like an arcade game to me.

That's a fair point, but there's an explanation for it.

You drive a car with 2/3 pedals and a wheel (with FFB), and you can have all this on a PC. For a car sim, you have a near perfect situation in terms of input device.

For a bike, it's a totally different story. As you're limited in terms of input device, you've to come up with something smart, within the boundaries of what you have. I think for Piboso (or MXS) it would be fairly simple to give you control on steering torque, full body weight shifting and plenty of things, but then what ? One would never be able to use it.

IMO More time needs to be spent developing the "hardcore steering" mode. Gyroscopic stabilizer, or just more control options like al167 mentioned. Until then it's all a guessing game at what the virtual rider is doing, questioning the simulation is silly when you have no control over the handlebars anyways.

But the hardcore steering mode is already complete. As soon as you add a "gyroscopic stabilizer" (whatever you mean by that), it's no longer hardcore. So non-hardcore for non-hardcore, I think (and that's just a personal opinion) that Piboso's virtual rider is a better control approach compared to MXS bunch of parameters.

The reasons for that are a bit technical and come from years of experience in designing control systems. In a nutshell: I think all the params exposed by MXS in its adavanced stability menu should not be exposed to the player. To the modders eventually (and GPB does that too, in a different manner), but not to the players. It makes no sense to expose them to the players. It's like having an oven that instead of having a tempreature set-point (which anybody can understand), gives you a temperature set-point, the parameters of the regulator in charge of achieving the desired temperature, a knob to set the sampling frequence of the control system and who know what else. It makes no sense. Again, that's just my opinion, anybody can disagree.

In one point you stated you said the track was too flat. Later on you said the bike was too permissive over bumps.  Which is it?  Either way the tracks aren't flat, and it's a motocross bike, it's going to be super permissive over bumps. They are designed that way.

I should have been more specific, sorry: flat "laterally", across a track section (from left to right) and in terms of small bumps (e.g. 10-15cm high).

The MXS tracks I've seen do have big bumps and big slopes/jumps, but overall the track seems very smooth, not rough enough. Maybe it's just any impression, maybe it's the 3rd person view that somehow "hides" that, don't know ...

Watching the slo-mo on MXB web-site you can easily see how the rear responds to this kind of rough track, and it looks pretty real. I haven't yet seen that in MXS (but again, maybe it's my fault).

Yes the gyroscopic stabilizer is pretty automagic when it goes to saving big slides, it's a little too effective IMO, but it is what it is. You can't be critical of the simulation merits of that and act like virtual rider and automagic leg push is any better.

I haven't yet tried the leg-push (not out yet), so I cannot have commented on it. It is my understanding that it will be optional (i.e. auto or manual), so I don't see any problem with it up to now.

Anyway, if it makes you feel any better, I can equally list what I think is off in MXB at the moment: it would be kind of redundant however, as it has already been reported by many others.

MaX.

[arnzzz]

I've heard people talking about "hardcore steering mode". What is this?

Is it actually an option called "hardcore steering mode"? or is it just a combination of settings that you guys are referring to AS "hardcore steering mode"?

Like 100% direct lean. Is that "hardcore steering mode"

I'm genuinely curious

[HornetMaX]

I've heard people talking about "hardcore steering mode". What is this?

Is it actually an option called "hardcore steering mode"? or is it just a combination of settings that you guys are referring to AS "hardcore steering mode"?

If it works like GPB, then you can read about it here (just the 1st and 2nd post): http://forum.piboso.com/index.php?topic=28.0

I don't know why it's not presented in MXB documentation session, I can only guess there's a good reason for that.
But as Snappe confirmed direct steer is there in MXB too, then ...

MaX.

[Phathry25]

I understand, both games has flaws, each one has different strong points but your critics are biased towards mx sim, i respect if you like it more but looking it from outside mx sim is not far from mxb, and it had waay more development time.

That's not true in the slightest bit. Get ready to learn.

I purchased GP Bikes in 2008. Same year I got MX Sim. Don't even try to pretend MX Bikes is wildly different from GP Bikes either. Terrain code and tire "digging" simulation that is identical to MXS's as far as I can tell from looking at the tire files, are the major differences. I'm sure there's more to it in the behind the scenes magic, but don't try and convince me they wrote the code for MX Bikes from scratch in the past year. As far as current development rate, I've learned over the years it's fueled by sales. They must have been good right away so now they're pouring time into this. If Beta 3 doesn't drive sales higher they'll focus more on another title. Likely KRP.  It will continue no doubt, but don't expect daily updates to last.

If I were to look at the development of GP Bikes and MX Sim across that same time period MX Sim wins hands down. Granted MXS started off way more featureless, but it has come a lot further in the same time period.

Cue PiBoSo to tell me I am wrong...  LOL

[Phathry25]

I should add I know I am coming off way harsher sounding than I intend to.  I own and play all PiBoSo's games, it's not like I hate the guy like some of you want to believe. I actually really enjoy his attitude and openness.

However it is my opinion that there needs to be more of a bridge between leaning the bike and steering the bike. Leaning feels very unnatural, steering is over difficult with common control systems. Controllers. Steering damping, speed sensitivity and some things I don't know about should be looked at in order to get most users playing via steering and not leaning. IMO of course.

[HornetMaX]

If I were to look at the development of GP Bikes and MX Sim across that same time period MX Sim wins hands down. Granted MXS started off way more featureless, but it has come a lot further in the same time period.

Not that I'm happy with GPB development speed (even if mostly I don't complain about the speed but about prioritization of what gets developed in GPB), but you may want to keep in mind that Piboso works on GPB (some), WRS (a little), KRP (a lot) and MXB (a lot) at the same time. Since the beginning of MXB, MXB has been the one with most dev time (makes sense), but even prior to MXB, KRP was getting more dev time than GPB by a fair margin.

The argument "MXB shares code with GPB" is valid up to a point: part of the code is totally reusable, part is totally not. And even if a majority was reusable, managing two projects in parallel is always bound to be less efficient than focusing on only one, especially for small dev teams.

As far as I know, KRP is the cash cow, followed by MXB. GPB is essentially eating up money with no return, WRS is free (well, you pay for GPB and you get WRS too). The crazy thing is that there would be plenty of GPB players ready to economically support GPB more (crowfunding, yearly fee/donation, whatever), but the idea is not welcome.

Anyway, to me the statement "MXS had way more development time than MXB" may well be true. Probably, even GPB 9or the sum of GPB and MXB) has had less development time than MXS.

Not that that makes a big difference anyway, as all we (should) care about is the final product we have at hand.

@Phatry25: I'm interested in your comments about the remarks I made about the physics in MXS, 3rd and 6th point (mass/traction and landings) in particular. Honestly, I am.

MaX.

[Phathry25]

MX Simulator is a spare time project for its developer as well.  It's hard to quantify time spent coding for either project.  The time I referenced was the 2008-2015 measure that we can guarantee. As far as I am concerned that other stuff doesn't really matter.

Max. I haven't watched the video, can't comment on its happenings. Almost always post from my phone. If I do hop on my PC I always try to work on a project instead of browse forums and waste time. I have a bad habit of hopping from one forum to the next for hours on end. So I don't.

I've never thought the bikes were too light in MX Simulator. I'm sure the exact figures could be obtained by searching the forum there though. I know it's been discussed. As far as physics environment it's all ODE so I'm sure gravity is accurate.

Traction is up to the track creator. Once again not sure of the quality of the track but you could be seeing a side effect of extreme traction values. You more than likely are.

Landings are hard to judge. There is a problem that needs to be corrected with the bikes collision shape, it's a little large as of now. Not sure if that's what you were seeing or not. Probably not since he didn't crash from the sounds of it.

[Phathry25]

On PC.  Watched video.  That view does make it look very floaty.  It was surreal watching it because the sounds sounded normal, but the bike looked to be in slow motion.

The landings you referenced were fine.  These are modern motocross bikes with 12" of travel and lots of bottoming resistance.  The 2nd one was close to being a crash, but he just caught the bottom of the landing with the rear wheel and made it work.