It seems the solution was quite simple.
If you get in this situation where iRacing is hanging in the loading screen leave the Simucube turned off until iRacing loads into the session and then turn on the control.
I have now upgraded to the lates Beta and Iracing launches into the session quite quickly.
Regarding the linearity of the scaling, it will depend on how the game scales the force as well as a lot of other factors.
In AC, the main gain, and the car gain are multiplied, and that needs to be taken into account when adjusting the gain. Consider the examples (gains are percentages, assumed approcimately small mige motor:
Main Gain 100
Car gain 100
MMC 12
Full scale signal is 20Nm
You want to reduce the gain 50%
Main gain 50
Car gain 100
MMC 12
You get 10Nm so this method is fine
Main gain 100
Car gain 100
MMC 6
You get 10Nm so this is fine
Now, if your main gain is 50 to avoid clipping
Main gain 50
Car gain 100
MMC 12
You get 20Nm (due to a car with stronger FFB from the game)
So, you want to reduce the torque for this car
Main gain 50
Car gain 50
MMC 12
You get 15Nm (Since you lowered the overall gain by 25%)
Doing the same with the MCM method
Main gain 50
Car gain 100
MMC6
You get 10Nm (Since you lowered the MMC and left the game gain the same)
So, I think the non-linearity might come from the scaling of gains in teh game, not from the hardware side of things. We know that the OSW system is very linear, so it āshouldā scale close to perfectly. Has anyone actually measured the torque at different MMC just to check? If not Iāll do that when I get home.
For me itās non issue I set the strength in the driver to 100 and forget about it.
Yes, for most people this is a non-issue for exacly that reason. I donāt see the benefit of scaling MMC instead of the torque requests anyway, unless teh torque requestst are really low resolution, which I doupt. Our tactile sense is not very sophisticated, and even with 10-bit resolution youād be hard pressed to feel the difference unless you go really low. And then you have negated one of the primary benefits of an OSW in teh first place, so why would you. Iād say remove the slider altogether to stop people from lowering their headroom, s some will invariably drive a game at default gains, and think āthis is too strongā, lower the strength in the driver and get loads of clipping instead. To me it doesnāt make sense to lower the output at the hardware level. Thatās what ingame gains are for!
I would like the option to lower the strength using the MMC scaling, the reason is my 11 year plays iracing and kart racing pro, itās good to be able to limit the torque across all games for him without having to mess about with my personal in game settings.
I totally see that point. And for that you have both MMC and the torque scaling, both are perfectly fine alternatives. I donāt see a reason to spend time reinventing the wheel here.
Having the force level configurable in some way via profiles is also good, so one can keep most of the settings in one place.
Yes. The MMC needs to e there to scale for different servos. I see no reason to combine the two sliders if it creates a problem. MMC is a jardware setting in my mind ,and the strength is a āsoftwareā setting. Liek I said, unless the resolution is really low thereās no way weāre able to detec a difference. I call placebo on those that do.
Thanks for your words.
The full resolution is -16384 - +16384 and if the resolution is set by software, then it would be proportional to that. At 50% power, that is +/-8192, which is already less than +/- 10000 that is the maximum in constantforce effect (as used by iRacing, for example).
We just want to experiment in making servo settings as transparent as possible to the end user, as the end goal is indeed to remove every need and even the possibility to configure via Granity.
So, Iāf weāre maximally ulucky with the values, the greatest difference in actual torque for iRacing is:
10.000 steps @20Nm means each step is 0,002Nm
8192 steps @20Nm is 0,00244Nm
So, maximal difference is 0,00044Nm. I sincerely doupt that you can feel that with your hands and brain. And this is a worst case scenario at 50%. Actual differences will be far smaller. Hardly anything to fuzz over.
Hi I had my simucube board wheel working in F12017 with MMOS but I cannot get it working in Simucube Firmware. Is there a reason do you know of for this? I makes the freeze and stutter really bad unplayable.
Thanks
SimuCUBE firmware is beta, and the minimally supported games are listed in the User Guide.
Thanks for reporting this, though. Freezes and stutters are interesting, but I do have some type of an idea what could possibly cause that. But people wanted us to fix the USB crash issues firstā¦
Thanks Mika I just wished to make you aware of it as I could not see any other mention of it anywhere on the forums
Hi Mika,
looks like both F1 2016/17 have the same issue - 2016 has massive stuttering (1-3secs between frames) when starting a race and after 30secs the game crashes - as soon as SimuCUBE is disconnected the stuttering is gone.
Also when setting up the wheel you have to turn it forth and back quite fast, orherwise movement isnāt recognized.
I donāt quite understand your post here. I mean the first part is clear, while the torque with normal driving is of course reduced, in the first case, you still get 20 Nm peak torque.
So far, so good.
Then you add another reduction, 50% overall, 50% car, which means 25% total gain.
How on earth do you get to 15 Nm?
First, the peak torque would still be 20 Nm (if ever something would result in that output), second even if you would calculate with scaled forces, you would expect 1/4 of the normal scaled force, not 1-1/4.
Hence, if for the first case with 50% gain you would rate it ā10Nm, then in the second case you would rate it ā5 Nm.
And that would be approx. the same as the third case, which would be 10 Nm with 50% gain, thus ā 5Nm?
Apart from this, I think the maximum current/torque is more of a safety feature and should be labelled such.
If possible, it would be even good to state the approx. peak torque (it should be made clear that it is an approximation).
The force scale on the other hand is a different thing and should be there to scale the FFB down in case the game doesnāt support this.
This is not possible, as they required torque constant is not saved in the motor configuration. We are thinking of adding it, but not promising anythingā¦
@Berniyh I did that in my head at work, I may well have made a booboo. 
Iāll check my numbers and logic one more time. It made perfect sense when I thought about it, maybe it got lost in translation somehowā¦ 
OK I have actually done some of the testing on the 080 version MMC Scaling and I have sent a message to Mika about what I have foundā¦ There currently seems to be an issue with double scaling (old code, looping effects not sure) which is altering the way things feel and with that of course the overall strength becomes non-linear, I am sure Mika will get a handle on itā¦ as this is NOT the normal way that scaling MMC works (generally it is very linear) This is New information and I am not even sure if he has read my message yet regarding itā¦ It actually appears that the Prior version 07x of the firmware based on my feeling may have actually been reducing the real forces at the wheel by a small % when being run through the current Amplitude based strength slider as the New MMC based slider does feel a small bit heavier (not a ton but a little)
The MMC based slider is the way to go though in order to optimally judge and use the strength of the wheel it is just going to take a bit to make sure it is working fully correctly.
Having the Dual Strength modes, the MMC (drive Power) and Strength (signal Amplitude) makes it much harder for someone to understand what their changes are doing relative someone else who has the same equipment. In addition Signal Amplitude adjustment creates less fidelity than changing the power at the Drive level and could affect how filtering is applied to the signal which Mika explained above.
i have a lot more oscillation on straits testing 080
also iracing still crashes while launching it. but thats normall i guess since nothing has been changed for that yet.
Yes it appears that something may have been going on from the beginning with the Amplitude scaling (Mika is looking into both the Double Scaling and I assume the amplitude scaling)ā¦ as it appears from my testing that the 100% scale (for the amplitude) was not quite outputting at 100% or was filtering the signal a second time creating a more subdued feelā¦ What we are getting Feel wise with the 080 at 100% is the true power of the systemā¦ soā¦ We may all be altering settings again as all of this gets correctedā¦ If we figure out exactly what it was doing and is controllable we could always possibly have another filtering option (who Knows at this point)ā¦ These issues with the power scaling in the 07x releases and prior May have been the cause of a Major Oscillation issue that I ran across but we will see when everything gets straightened out.