I’ve started doing back to back testing on some of the different Reconstruction filter settings and I am finding some interesting things so far… Basically I have been testing between the 5 and 7 filters currently as those seem to be the most used settings.
my Notes on the #5 filter is that it has consistency in feel, and by signal testing should be the base for natural feed back from iRacing.
However the interesting thing is the #7 - this filter is very similar to the #5 filter in many ways and i can see why many have migrated to it… The main difference is that it imparts a reasonable level of inertia “like” feel to the wheel which lightens up the feel of the wheel in corners that require more turn in the wheel (i.e. hairpin corners) but it does seem to weight the wheel a little more in loaded more consistent corners.
In my back to back A-B-A-B-A testing I found that I was actually faster with #7 but more consistent with #5… So I did some playing with settings…
My starting settings were #5 + Dampening @ 2.25%
next was #7 + Dampening @ 2.25%
I was about .2 seconds faster around the Kansas road Course with the Ferrari 488GTE @ night
My next was to change use #5 + Dampening @ 2.25% + Inertia @ $3%
Still a little faster with the #7 but it closed up a little
Then upped inertia to 5%
and the results were close about …075 difference.
The last test was Inertia @ 10% which seems a bit excessive but at this point I actually ended up faster than the #7 by about .05 but also was more consistent… This point however did show the inertia filters kind of ugly side as it causes a deadness in the wheel when the FFB changes direction as when you turn into inertia it makes the wheel heavier but turning away from the inertia signal makes the steering lighter. this however is not the case with the #7 filter so it is not actually putting inertia in the signal but whatever it is doing is imparting inertia like feel.
Anyway I thought I would just share some of my findings.
BTW 10K Large Mige, .6:1 (60%) specific output, Full Power.
Ive often thought with AC that the centre ffb could always be a bit firmer.
Upgrading from .4 to .6 I could feel some noise even at recon 10 on .6 that wasnt there in 0.4. It turned out the new firmware changed my mcc and mmc to 12.0 12.9 from 12.9 13.0.
Increasing difference between the 2 values to 12.8 and 13.0 by just .1 has firmed up the centre FFB a big effect for such a small change.
At 12.7 13 you can definitely feel the dampening effect but its a bit too much and felt horrible at recon 5.
Ill settled for the moment with it at 12.75 and tweek later and see if there are any down sides.
These changes improve the feel of the ffb with the ford gt no end Ill have to check others cars.
Changing settings, especially those that are only stored in Ioni, is strange and definitively should not happen. Are you sure you didn’t upload a drc file and used “just use my current Ioni settings”?
Also, the recon filter is indeed basicly only meant and optimized for 60 Hz updates with the setting at 5 being only one that we tested, and the other settings are just guesses.
We might optimize the filter and make the expected frequency parametrized too, to be optimal for other sims as well.
thats possible at some point, but maybe we won’t limit or optimize it “sim-by-sim”. We are thinking about the filter with setting called “input rate” which would be “low” “medium” and “high”, with low being iRacing 60Hz, Medium being something like 200 Hz and high being >500 Hz.
Mika and I talked a little about that, because really the recon filter in a away is not totally meant for tuning BUT at the moment there was only the 60Hz optimized but knowing other sims use other refresh rates it may end up being best to optimize for a few input levels I was personally thinking 60, 120, 240, 360, 480, and maybe 600 as it seems that output frequency seems to be many times 60Hz apart… Granted this might miss the exact frequency of some sims but hopefully will be close… It will also allow for people to use it a little bit for tuning… I am trying to figure out the fundamental feel differences between filters hence my post above just so when Mika and Tero revisit the filter there is some distinct information regarding what changes within the filter have what feel effects… as if all that is known there might be some other methods of implementation to make it a little bit of a tuning thing as well as reconstruction… This is why in the Guide we indicated that it was a work in progress… For others out there if you can please indicate which setting is being used and why you chose it… It will help me try out and reconcile the feeling changes at least in iRacing as that is what I mainly use… I have not done much testing at all in Assetto Corsa but I have found in the little testing tha I did that Assetto Corsa does impart quite a bit of the inertia type feeling within their base feedback (to me almost distractingly so). That feeling in its extreme sense is when the wheel feels too light but at the same time feels like the wheel is steering you rather than you steering the wheel… Problem is to little inertia and it seems that the wheel doesn’t return to center firmly and directly enough as this is where you would want the steering (or in reality the caster) to drive the wheel back to center.
Hello,
I don’t know how to crop pictures, but here are my big mige simucube settings for the Mx5, I suggest trying it on limerock and all stock. I get great response and feel and this is truly the 1st time it felt like a real wheel! Some feedback is missing ie curbs, textures but its still impressive with the simucube. I’m still working on fine tuning it with other carss, but give it a try and give me your feedback.
TBW 680
Power 25% of 24.5 max 6.13 A
Recon 3
Damp 2.38
Friction 1.2
Inertia 0.99
in game ffb 15nm
Snow
Simracingbay Big Mide, simucube, 10k encoder, Large wheel
Those that have driven it:
Is the spring effect strength slider supposed to work in-game? I can’t get it to set any different parameters to SimuCUBE. However, there is a new slider for spring effect strength in the UI now.
Also, first ever test with sine effect vibration. It works!
Solo looking forward to being able to play ETS2 and ATS with FFB again! And sorry, can’t answer the Spring question. Previously all my settings were right in the middle and didn’t fiddle around with anything.
I can however find my trusty old DFGT and test it for you if need be?
I I stalled the 2048 SinCos earlier today so MMOS is not an option as far as I know.
At first I thought the slider was in-game, but I’m unable to find a such. On the DFGT Profiler setting the Spring Effect slider has the same results as using the “Force feedback ce teeing” in-game - they seem interchangeable
