• Welcome to PiBoSo Official Forum. Please login or sign up.
 
March 28, 2024, 11:43:43 PM

News:

GP Bikes beta21c available! :)


MaxHUD plugin

Started by HornetMaX, September 26, 2013, 04:34:50 PM

Previous topic - Next topic

Vini

Any news?
Sorry but I didn't quite get if or what kind of data you need from me.

HornetMaX

I said I don't need more data in fact.

I added the values you requested and I thought the ball was in your camp, that you were thinking hard about how to use them ... I still don't see a lot that can be extracted to drive the rumble in a useful manner.

Maybe it could be useful to have 2 clean laps: one on a pretty fast pace and one at a more moderate pace.
Looking at the differences between the two may highlight something (still a long shot if you ask me).


Vini

September 21, 2021, 09:15:20 AM #947 Last Edit: September 21, 2021, 09:17:00 AM by Vini
Looking at the (R-CoG)/CoG values for my Silverstone lap, I don't see why you can't just cut it off at "0" (using the backstraight as the definition for the 0-point) and scale it to the rumble motor. Maybe do a log scale for smoothing out the peaks but I don't think that's even necessary.

I remember the corners where I pushed the rear harder on that track and the graph perfectly reflects that.
You can even tell where the intentional slides are and where it broke loose by accident.
And yes, it was a hotlap (record I think), so even in the "clean" corners the rear was very close to the limit.
Both the wheelie and braking problem is irrelevant when using the CoG speed for calculation.
At this point, I really struggle to see what you are looking for here.
Where exactly do you see the problem?

The only thing I see are the peaks when going over curbs but this is perfectly fine since the bike actually is at the limit at this point and would crash out if you stayed on them for longer (with the Silverstone curb size).

Of course you will never get a perfect representation of the true slip angle but in terms of usefulness for feeling the limit it seems miles ahead of the current chassis-accel-based rumble that really doesn't do that much most of the time until it's too late.
You seem to be looking for the perfect solution (which never exists), while I, as a player, would already be happy with one that is an improvement over the current one.

Vini

Sorry if that last part sounded a bit negative. The rumble is great for immersion but on some bikes (that lack realistic chattering under braking) it does little for predicting the limit.

HornetMaX

Current rumble was never intended to "predict the limit", so yeah that's not a problem.

Can you highlight on the plot of (R-CoG)/CoG where exactly you'd expect the rumble and where you would not expect it to rumble ? And above/below which level ?
Could you do that on two laps, one fast and one slower ?


Also, what about negative value of the above quantity ? Should it rumble too or not ?
Because negative values are essentially when you're "locking" the rear (or rear is hopping).
As I said, this would work likely pretty well (detecting longitudinal slip of front or rear under braking).

Vini

September 22, 2021, 04:08:07 PM #950 Last Edit: September 22, 2021, 05:04:41 PM by Vini
Like I said, just cut off at 0 and scale the graph above that line to the rumble intensity. No need to explain any further. It would be as simple as that.
Negative values are not necessary because the engine sound already tells you when the rear is locking and the whole process is much less critical in general (can't recall ever crashing because the rear slid out under braking).

Here is the difference between a slow and a somewhat decent lap with MotoGP bike at Valencia (nowhere near record pace though):

I think this speaks for itself.

HornetMaX


Vini


HornetMaX


TFC

Just wondering MaX, how would what Vini is talking about translate to MXB? I'm all for more 'feeling' or feedback from rumble, so from that point of view it makes sense, and as for MX where the rear end loses traction in a more consistent and controlled way I imagine it could be a very useful addition..

HornetMaX

Quote from: TFC on September 24, 2021, 12:50:33 PMJust wondering MaX, how would what Vini is talking about translate to MXB? I'm all for more 'feeling' or feedback from rumble, so from that point of view it makes sense, and as for MX where the rear end loses traction in a more consistent and controlled way I imagine it could be a very useful addition..
Hard to say. I'm tempted to think that in MXB the rear is slipping much more than in GPB, both laterally and, in particular, longitudinally: if that's true, then it may be difficult to convey something useful.

Also unsure if it is common practice to spin the rear during big jumps: it wouldn't make sense to rumble there, but that can probably be filtered out somehow.

I'll push out a new version of MaxTM that will provide the relevant telemetry shown in the graphs above even for MXB: you will be able to do what Vini has done and have a look at the graphs to try to figure out if it has a chance to work or not.

HornetMaX

V2.2.4 out (2021/09/27)
  • .ini file version change: will lose all settings. You'll have to manually redo them. Sorry. That's life.
  • Minor cosmetic changes: widgets options with fixed font and better aligned.
  • HUDRumble rework:
    • 2 possible sources: CoG acceleration (norm filtered by highpass) or rear tyre overrun (how much faster is the rear tyre spinning compared to CoG velocity norm). Overrun is only for GPB and MXB. In MXB it may not make sense at all.
    • Each source has 2 extreme points between which the values are interpolated.
    • Each source can be associated with each rumble motor.

If you test the two rumbles, I'm interested in your feedback, including what default settings could make sense.

Vini

Thanks a lot Max!
This is amazing.

Just did a quick test and the default settings already feel very nice.
I will play around with the settings to see how much it can be improved.

Vini

September 28, 2021, 11:17:44 AM #958 Last Edit: September 28, 2021, 12:02:42 PM by Vini
Only thing is that the braking phase now feels a bit "empty" relatively speaking.

Switching to a "front underrun" mode when front brake is applied would be the icing on the cake I think.

HornetMaX

What settings exactly are you using ? (all rumble settings, or just settings screenshot).