GP Bikes beta5b

[Jose Reina]

I think it makes too Wheelis, too unreal.
And on .... Piboso wobbles, you have not thought about putting a steering damper? in which one could control the hardness as is done in reality. That would make depending on how you configure tubieses + wobbles but more agility on the bike, or - wobbles but slower bike (in terms of direction).

[Arvoss]

PiBoSo, was alby's crash picture 'luck' or are you working on a simple crash damage model like KRP? Losing a front wheel or a side panel would be awesome!

[PiBoSo]

The rpm changes at full angle only depends of the tyres radius, if the 500 had the same tyre as the norton commando, we quasi can't hear the difference but it's not the case. If you want less rpm changes, take less angle or build new thin tyres.

Indeed. The RPM change with lean is not due to a fake "effect" that can be tuned. It is the end result of torus radius and lean angle.

[HornetMaX]

EDIT: Just thought of a good description for the feeling. It's exactly like really heavy engine braking on a four stroke bike, that does not happen on a 2 stroke engine(well practically negligible).

Except it doesn't brake at all, it's just the RPMs going down, speed stays the same (it was probably already clear to you Hawk, just to be sure it is clear for all).

I haven't tried the 500 enough, on the 990 it seemed OK: we need to get accustomed to it and take it into account when shifting gears entering/exiting turns.

MaX.

[janaucarre]

but if you have well done the calculs the rpm change is good, and all here want a hyper realistic game, so if it's well calculated it significate that this is the reality, and so, why change it?

[LOOPATELI]

https://www.youtube.com/watch?v=MEfpOXCWr5c

https://www.youtube.com/watch?v=MEfpOXCWr5c

an other weird thing that happened right now online
first of all when i crash rider disappeared and only his shadow was visible. Is not the first time that this happens but is not an important thing, no problem.
The important thing is that the front suspension of the bike is like a spring! with default setting this doesn't happen, it happened with a 4b setting (The setting was good I was qite fast with it)
Maybe all the front tires problems (wobbles, waves,wheelies..) is because of this. maybe this movements happened in a minor way, and my setting just exagerate it. no idea
Someone with the same problem??

[Alby46]

PiBoSo, was alby's crash picture 'luck' or are you working on a simple crash damage model like KRP? Losing a front wheel or a side panel would be awesome!

no just a crash, don't know if you ever played beta 1b but there was something like that also on that beta, and it was all about crashes

[Stout Johnson]

For the developer browsing this forum, it might sometimes be a bit hard to get a quality feedback from what is posted here. For the sake of criticism/praise balance MaX thankfully already put this up

http://forum.piboso.com/index.php?topic=1241.msg15832#new.

But for the sake of development, I think it might be helpful, to find sort of a short resumé:
(I tried to find a general resumé based on what I read here numerous times; please note that not every single aspect that was mentioned by only one or few can be taken into a resumé - I tried to find a resumé with the right balance between overview and detail. In terms of brake feeling the majority found it to be better compared to beta4, while some seemed to prefer beta4 braking)

First off, it seems to be the vast majority's consensus that - wobble/weave issue aside - beta5(b/c/...) is definitely a huge step in the right direction. It feels much more like really riding a motorcycle - especially in terms of engine brake, leaning and braking. I would like give an overview of positive and negative aspects compared to beta4b, also indicating importance and probable root causes for the negative aspects. I sorted aspects by observed frequency (descending).
Note: aspects which in general are positive, but still need adjustment are listed positive and negative and are marked with *.

positive

• much improved leaning/turning (importance: high)
• much improved brake feeling/braking feeling more like a motorcycle brake(importance: high)
• engine RPM's depending on lean angle/tyre radius(importance: high)
• much improved feeling for rear wheel, power weaving not exaggerated anymore (importance: high)
• improved engine brake(importance: high)*
• tyre pressure has more realistic effects (importance: normal)
• new rider animations (importance: normal)

negative / still needs to be worked on

• wobble/weave (importance: very high, possible root cause:virtual rider, new tyre simulation? causing left/right/left/right oscillation of front wheel. )
• bike wheeling too much (importance: normal, possible root cause:new chain simulation, new tyre simulation, anti-wheelie electronics not adapted to new engine power? )
• engine brake only useable on settings 2 and 3 (importance: normal, possible root cause: Engine brake settings not adapted to new engine power? )*

Feel free to discuss my list or make other proposals.[/list]

[Stout Johnson]

EDIT: Just thought of a good description for the feeling. It's exactly like really heavy engine braking on a four stroke bike, that does not happen on a 2 stroke engine(well practically negligible).

Except it doesn't brake at all, it's just the RPMs going down, speed stays the same (it was probably already clear to you Hawk, just to be sure it is clear for all).

I haven't tried the 500 enough, on the 990 it seemed OK: we need to get accustomed to it and take it into account when shifting gears entering/exiting turns.

MaX.

Judging from the rear tyre of a race-bike http://www.bridgestone.de/motorrad-reifen/~/media/Images/Motorcycle/Central/Products/Tyres/battlax%20racing/v01/closer%20look/battlax%20racing_v01_closerlook1.ashx?h=379&mh=379&mw=379&w=379 it would be coherent to have a RPM decrease of about 15-18% when going from zero-lean to full lean, for a wider MotoGP tyre this could possibly mean about 20%.

This is based on simple measurement and maths. I am not sure though, how the tyre deformation would change this calculation irl though. I would say that the tyre deformation would possibly change the 'perimeter' (is this the correct terminus technicus?) of a tyre under load. But probably not very much, so it could be estimated by measuring along the outer flank of the tyre. So roughly 15-20% of RPM increase should be the maximum while leaning for the big bikes if I estimated it correctly? For smaller tyres of other bike classes (esp. 125cc) the rpm change should be drastically lower.

Now it would be someone's turn to test it GPB. 

[HornetMaX]

I've made a simple spreadsheet before beta5 came out, when we started discussing this thing. Change the values in red (you have them in the tyre files of the bikes) and look at the table below them (the table on the right with the rim sizes is irrelevant).

https://docs.google.com/spreadsheets/d/1GvWPanOWV-jjMcInMpxvl5f2BC3_Ycg9Y5FG4MVwtpg/edit?usp=sharing

So for the 990 at 60deg, i have 20% of radius reduction (which means +25% in RPMs).

The numbers are not exact as they do not take into account the tire deformation due to the normal load: so that could justify why in GPB we see less RMP increase than that (the effective rolling radius under load being smaller than the free radius).

Notice also that GPB assumes (for good reasons, no critic here) a circular profile of the tire, while real tires are not really circular in profile.

MaX.

P.S.
Thanks Stout for the recap on wobble, weave and the other remarks.

[Vini]

I have already tried to put up a complete list of positive/negative points on page 8.
I think we can all agree that there needs to be a hotfix for the front wheel wobble (shattering) at low speed and high lean angle.
The weaving when accelerating out of corners (still at high lean angle) is good and realistic (you can see these little oscillations out of every corner if you watch a MotoGP onboard video). The problem is that it doesn't stop when the bike is stood up and that it quickly develops iinto a tankslapper.

[Furious]

Guys. If the bike behave is calculeted correclty with the dynamic physics it can't be wrong. I'm almost 90 % sure, that all that wobbles and stuff you are complaining about is caused by the virtual rider you ale "partly" controling. I assume, while you plug in a proper wheel and use direct steer, then all the negative effects might go away. I have no forcefeedback wheel right now to check it out, so i cant tell for sure.

I just want to point out that lack of true input to the bike you can't really tell if the bike behave is correct or not, because you are never sure what virtual rider is doing.

[C21]

Guys. If the bike behave is calculeted correclty with the dynamic physics it can't be wrong. I'm almost 90 % sure, that all that wobbles and stuff you are complaining about is caused by the virtual rider you ale "partly" controling. I assume, while you plug in a proper wheel and use direct steer, then all the negative effects might go away. I have no forcefeedback wheel right now to check it out, so i cant tell for sure.

Just read a few pages back....:

Okay..I admit it. I love it. On the strict caveat that somehow the front wheel oscillations are sorted out.. Braking is better. With a joypad and direct lean off it responds beautifully and turns in like it should do. Back end is much better in terms of encouraging a rear wheel slide.

Many more laps and I can fully see the potential now. My steering wheel and pedals are going back into their box in the cupboard.

See GRT on page 8 and 9 

[JamoZ]

Guys. If the bike behave is calculeted correclty with the dynamic physics it can't be wrong. I'm almost 90 % sure, that all that wobbles and stuff you are complaining about is caused by the virtual rider you ale "partly" controling. I assume, while you plug in a proper wheel and use direct steer, then all the negative effects might go away. I have no forcefeedback wheel right now to check it out, so i cant tell for sure.

I just want to point out that lack of true input to the bike you can't really tell if the bike behave is correct or not, because you are never sure what virtual rider is doing.

No physics engine is 100% accurate, of course there's a chance some calculations can be wrong.

[Stout Johnson]

I think we can all agree that there needs to be a hotfix for the front wheel wobble (shattering) at low speed and high lean angle.

Yes, pretty sure everyone seconds that.

The weaving when accelerating out of corners (still at high lean angle) is good and realistic (you can see these little oscillations out of every corner if you watch a MotoGP onboard video). The problem is that it doesn't stop when the bike is stood up and that it quickly develops iinto a tankslapper.

Exactly my opinion. The way the weaves start is always very realistic, many forget that. It also happens when you are accelerating on lean and suddenly close the throttle - it weaves --> very realistic. It can always be observed when the big load transfers happen - and that this brings the bike at its extremes is normal.

But, like you said, the weaves should not build up as much as they do. And they should happen only marginally with the 125cc. Maybe the 125cc should be used as a reference. If it is realistic there - the other bikes might also behave realistically.

But somehow I do think that the weaves, wobbles are not a problem with suspension and/or steering damper. I get the feeling that the virtual rider does play a role here. I also observed, sometimes when braking hard and already trying to lean a bit into corner - the rear sort of "overtakes" the front. It looks a bit like trail braking - but much, much more. In replay it looked like the virtual rider turned the steering too much inside  - the subsequent behaviour of the bike seemed very much realistic. It just looked like a problem with virtual rider. Maybe there is also something happening with the weave problems. Maybe the realistic behaviour of the bike and the reactions of the virtual rider add up to an un-wanted frequency?