High-End Wheel Owners (CSW V2 and Above) Please Chime In - Experiences, Info, Reviews

[mark7]

I agree, just set up to what feels right.

Adrian, going to have a chat with John Beeson at Leo's tommorow regarding a purchase but after spending the afternoon looking at Simsteering forum posts have you heard anything about new hardware to go with the latest firmware updates?

 

[Lazza]

Ok, let's be specific here and try to keep it simple. Note I'm replying in this thread only just to keep things centred somewhat.

1. Why does the modder set a number for the torque? I thought all this stuff was calculated dynamically and "truly" from the physics engine (all sorts of complex calculations regarding suspension geometry, tyre forces, aero, etc.)?

Second sentence is absolutely correct. Don't doubt yourself there. There's a whole dynamic virtual world interacting with virtual tyres and wheels and suspension and steering geometry to produce dynamic forces at the virtual steering wheel. You can't fudge any of that as a modder.

But the game has to know how much steering wheel column torque is 100% FFB output. Maybe it could do some calculations based on the car parameters to work out a maximum torque figure and it would be close enough, but instead it's a figure the modder can set. This is probably a fair method because the modder knows more about how the car will be used than the game can guess at (a super speedway configuration of the DW12, for example, you could expect to encounter higher forces than the road course configuration; the game can't really know the difference [and the fact it doesn't is why the current single max nominal torque is leading to even lower forces in the road course version. A later update will have different values for the different configs, somewhat alleviating the problem]).

So something has to decide how much steering column torque is 'full force', and that's the modder's job. Design choice. Done.

2. i'm trying to determine how the car's "true" (as in no scaling or anything) FFB power, my real life wheel's power, and the game's STC all affect each-other...

The default STC is very low, so let's start there. Basically every car made will have a nominal max torque higher than STC, so the game scales all virtual steering column torque (0 -> nominal max) to your wheel at (0 -> 100%).

Example: The skip barber has a nominal max torque of 9.5Nm. Let's say at 3 different points around a track the steering column is currently experiencing (due to all the physics calculations) torque of: 4.25, 9.5, 13.75 at those points.

Since 9.5 is the 'full output' point, the game will attempt to send FFB to your wheel at: 50%, 100%, 150%. Of course you can't send more than 100%, so it's 50%, 100%, 100%.

If your wheel can produce 5Nm, you feel 2.5, 5.0, 5.0

If your wheel can produce 25Nm, you feel 12.5, 25.0, 25.0

Note that the powerful wheel is producing 12.5Nm when the virtual steering column is experiencing torque of only 4.25Nm. This is why it's a good idea to not ignore these figures and set the STC if you have a powerful wheel...

2a) How does the FFB scaling work if the car's true FFB can, let's say, hit 20 Nm, but my real wheel is only capable of 15 Nm, and my STC is set to 10 Nm?

Just like above, even though you've set a higher STC of 10Nm the game can see that the car has a nominal max of 20Nm, so it scales it just like before. When the virtual car is producing 20Nm the game says that's full force, so it sends 100% to your wheel which produces its 15Nm. Everything lower is scaled down, anything higher is lost (clipped).

2b). What about if we were to switch the real wheel and STC figures around in the example above (real wheel capable of 10 Nm, STC set to 15 Nm [same 20 Nm car])?

The car's max is still higher than your STC, so the game scales (0 -> 20Nm) in the car to (0 -> 100%) at your wheel. Car at 20Nm is sent to your wheel as 100% which produces its 10Nm. Lower forces are scaled, higher are clipped.


Switch those figures around again: car produces 10Nm, wheel does 15Nm, and STC = 20Nm

Now the game thinks your wheel can do 20Nm, so a car producing 10Nm will be simulated properly if it sends 50% FFB to your wheel; that would result in 10Nm at your 20Nm wheel. But because your wheel is actually only capable of 15Nm, sending it 50% FFB results in you feeling 7.5Nm. Setting STC higher than your wheel is giving you lower than real forces.

Finally, let's do it properly: car produces 10Nm, wheel does 15Nm, you set STC = 15Nm

Game thinks/knows your wheel can do 15Nm, so a car producing 10Nm will make the game send 66% FFB to your wheel, and you feel 10Nm. You feel the correct forces as calculated inside the game, and because your wheel can still produce more force you'll avoid nearly all clipping because your wheel can take up to 15Nm of simulated force before running out of room - which should be rare for a car set at 10Nm.


You can see in those last two examples that STC can help avoid clipping, but setting it higher than your wheel is capable of makes all forces weaker than they should be. It's much better to set STC correctly and use the FFB Mult if you want to reduce clipping in a specific car, so that only that car is made weaker.

If this scenario, there are two ways to fix it. The smartest way in my opinion would be to correctly set the STC to what your wheel can do. So raising it from 10Nm to 15Nm. However a 20Nm virtual steering wheel torque will still be clipped because it is above the STC value of 15Nm.

If the car is given a nominal max torque of 20Nm, the game scales that to your wheel. So you don't get clipping until the virtual steering column exceeds 20Nm. Remember we all have a default STC of 2.5; if the game just blindly used that we'd all be clipping nearly the whole time.


Adrian: you might want to reconsider your ignoring of these figures

 

[Adrianstealth]

I agree, just set up to what feels right.

Adrian, going to have a chat with John Beeson at Leo's tommorow regarding a purchase but after spending the afternoon looking at Simsteering forum posts have you heard anything about new hardware to go with the latest firmware updates?

hiya mark

I updated today -but those may have been there a while and may be more new ones coming,
the reason why I was on the old ones for so long is that I don't really check as everything feels fine etc they don't need to make any changes IMO ( notice the Bodnar theads don't have countless posts about trying to get stuff right etc & fiddling etc )

Bodnar is a bit of a financial blow but it will last a lifetime had mine over a year now and it's still like totally bran new & I'm confident it will be the same case in 5 year times! a beautiful piece of kit

let us know how you get on

 

[DrR1pper]

Oh Jesus, got it wrong again. Thanks for correcting Lazza. Also, beautifully explained.

 

[Spinelli]

So, the nominal max torque is just whatever number the modder decides to use?

The STC setting only does anything if the game recognizes that the STC setting is higher than the particular car in use's nominal max torque setting?

The default STC is very low, so let's start there. Basically every car made will have a nominal max torque higher than STC, so the game scales all virtual steering column torque (0 -> nominal max) to your wheel at (0 -> 100%).

Example: The skip barber has a nominal max torque of 9.5Nm. Let's say at 3 different points around a track the steering column is currently experiencing (due to all the physics calculations) torque of: 4.25, 9.5, 13.75 at those points.

Since 9.5 is the 'full output' point, the game will attempt to send FFB to your wheel at: 50%, 100%, 150%. Of course you can't send more than 100%, so it's 50%, 100%, 100%.

If your wheel can produce 5Nm, you feel 2.5, 5.0, 5.0

If your wheel can produce 25Nm, you feel 12.5, 25.0, 25.0

Note that the powerful wheel is producing 12.5Nm when the virtual steering column is experiencing torque of only 4.25Nm. This is why it's a good idea to not ignore these figures and set the STC if you have a powerful wheel...

Was that an example of what would happen if you had a 25Nm wheel but left a very low STC? If not, then why is the wheel not outputting more power when the in-game virtual wheel increases from 9.5 to 13.75?


EDIT: One more thing... I forgot to mention...I want to try and have all the cars' FFB 1:1 in relation to eachother. For example, 12 Nm should always feel like 12 Nm no matter the car I'm in. Wouldn't the pre-set nominal max torque setting need to be ignored for this to work?

 

[DrR1pper]

So, the nominal max torque is just whatever number the modder decides to use?

Yes. They may or may not choose an optimal value for the nominal max torque setting in which case we can either change the nominal max torque setting for the specific car or a much more quicker and easier method is to simply change the ffb multiplier which has the same effect/result. ISI (at least in the past) set the nominal max torque lower than what we'd experience on the virtual steering wheel on track and so i would have to lower the ffb multiplier to (in effect do the same thing as) raise the nominal max torque per car. (I think the reason is/was that if you set the nominal max torque too high, the majority of consumer wheels would have weak feeling force back due to the small torque range of their ffb wheels.)

E.g. If some cars specific nominal max torque = 20Nm but you don't want a virtual steering torque of 25Nm to be clipped. Then we either raise the car-specific nominal max torque to 25Nm or the easier/better option is to leave it at 20Nm and simply lower the car-specific ffb multi to 0.8 (calculated by dividing the nom max torque / desired max torque = 20Nm / 25Nm = an ffb multi of 0.8) in game.

The STC setting only does anything if the game recognizes that the STC setting is higher than the particular car in use's nominal max torque setting?

Correct. So in my earlier post where i used the STC to calculate ffb output it should have been the car specific nonimal max torque setting instead (thanks again Lazza for correcting me ). Only when the STC value exceeds the car specific nominal max torque setting does the ffb calculation start to use the STC value in replacement of the nominal max torque value to compute the ffb output.

Was that an example of what would happen if you had a 25Nm wheel but left a very low STC? If not, then why is the wheel not outputting more power when the in-game virtual wheel increases from 9.5 to 13.75?

Because in that example, 13.75Nm is above the skip-barbers nominal max torque of 9.5Nm. Also the default STC of 2.5Nm is way below that car-specific nominal max torque of 9.5Nm. If you changed the STC from 2.5Nm to your ffb wheels true 25Nm, then in this example the STC is greater than the default nominal max torque and thus the nominal max torque uses the STC value of 25Nm instead. Meaning a virtual steering wheel torque of 13.75Nm is no longer clipped and your 25Nm wheel will produce 1:1 matching torques for virtual steering wheel torques between 0-25Nm.

However if you had your STC set to say 15Nm (10Nm short of what you ffb wheel can really do), although 13.75Nm virtual steering torque would also no longer be clipped, it would not feel like 13.75Nm on your ffb wheel but instead 22.9Nm (because 13.75Nm/15Nm = 98.6% ffb output = 22.9Nm on a 25Nm ffb wheel), over-amplifying the virtual steering wheel torques nearly double what they should be. Thus highlighting the importance of using the right STC value if your ffb wheel is capable of significantly more torque than mainstream wheels (e.g. AF-Pro or above).

if i lost you, best wait for lazza.

 

[DrR1pper]

EDIT: One more thing... I forgot to mention...I want to try and have all the cars' FFB 1:1 in relation to eachother. For example, 12 Nm should always feel like 12 Nm no matter the car I'm in. Wouldn't the pre-set nominal max torque setting need to be ignored for this to work?

Yes, you'd need the pre-set nominal max torque setting per car to be the same or less than the STC value for it to be ignored. Then the torques will match 1:1. There is another way to achieve the same result but more confusing perhaps. This is the simplest methodical way.

The issue with this however is when driving cars with an expected virtual steering torque range greater than what your ffb wheel can output. For example, the indy car's which if i've heard correctly are in the 0-30/40Nm range? If so, the top half of the virtual steering wheel torque range would be completely clipped if you force the indy-cars nominal max torque of (e.g.) 30 or 40Nm value to your STC of 20Nm to get the STC to override the nominal max torque setting.

At the end of the day, 1:1 torque matching is limited by the range of torques your ffb wheel is capable of producing.

 

[Lazza]

Yep, if the car's nominal max torque is higher (right or wrong) than your STC it's like the STC setting doesn't exist. Before it was there all cars were scaled 0 -> 100% FFB based on the nominal max torque figure, so the game only uses it when it makes sense to do so.

Anyone can have all cars scaled in proportion to each other by setting the STC to a value higher than any nominal max torque they encounter. I could set mine to 25Nm and have the F1 cars peak 2.5x higher than the skip barber. Of course with my G27 the skip barber's going to feel very weak if I do that

 

[Slamfunk3]

Just wanted to add about nominal max torque; you can use Lazza's motec plugin to read out each car's "steering shaft torque". I believe the max value in motec should be what's put into the HDV file. Now of course even that leaves a lot of room for interpretation. Things such as hitting bumps, curbs or even too hard of a suspension setting can impact that value. So still a lot of room for the modder to use his creative licence as ISI did with the dw12. I imagine if you get a motec readout of the dallara at mountain peak raceway, at 300+ kph in turn entry, that would equal the cars nominal max torque set in the HDV (which is locked [emoji39]).

P.s. Just wanted to add another quick example. The ASR formula 643 has a nominal value of 19Nm in the HDV, using motec i get around 30Nm in a high df turn. Therefore I set the FFB multiplier to around 65% to avoid clipping.

 

[Spinelli]

Thanks guys, I really appreciate all the responses and info posted.

 

[DrR1pper]

sweet tip slam.

 

[Marc Collins]

This would all be a lot more fun and interesting to test if we knew what the correct Nm values for all the commonly available wheels were. Which ones do we know already?

 

[DrR1pper]

This would all be a lot more fun and interesting to test if we knew what the correct Nm values for all the commonly available wheels were. Which ones do we know already?

Off the top of my head and approximate values, i think are:

G25/27 - 4Nm
T500 - 5Nm
CSWv2 - 6Nm
AFPro - 13Nm
Bodnar - 16Nm

 

[Marc Collins]

If a CSW v2 is so much stronger than a CSW, which in turn is way stronger than a G25/27, something is amiss. If the CSWv2 is accurate, I would then estimate 4.5 for a CSW and about 3 or 2.5 for a G. I had a G (still have it, actually) and the CSW is in a different universe of force output.

But it doesn't really matter, because as I reported months ago, my STC can be set to 1 or 2.5 default, or 5 or 25 and it has no effect whatsoever on the FFB.

And my multiplier works as advertised. If I put it at 0.5, I get half the FFB, not more FFB as suggested above. Oh well.

 

[DrR1pper]

If a CSW v2 is so much stronger than a CSW, which in turn is way stronger than a G25/27, something is amiss. If the CSWv2 is accurate, I would then estimate 4.5 for a CSW and about 3 or 2.5 for a G. I had a G (still have it, actually) and the CSW is in a different universe of force output.

But it doesn't really matter, because as I reported months ago, my STC can be set to 1 or 2.5 default, or 5 or 25 and it has no effect whatsoever on the FFB.

And my multiplier works as advertised. If I put it at 0.5, I get half the FFB, not more FFB as suggested above. Oh well.

I think the CSW was around 4-4.5Nm. I know it appears inconsistent with what is perceived between the different wheels (as i also experienced what felt like a huge doubling to tripling of torque going from the CSR-E to the T500 only to find out later it was only a small increase in peak torque when measured directly but there is a good explanation for the misleading perception. Gear/transmission ratio.

So the smaller the gear/transmission ratio the more lively/responsive the motor feels which can easily be misinterpreted as having higher peak torque than actual when measured in steady-state. The reason being that if two wheels that can produce the same peak torque output but one is geared higher, the higher geared wheel (i.e. motor to steering wheel shaft output) will produce the same torque but at a slower rotational acceleration/velocity rate. The resulting effect is the same torque output but larger momentum change felt on your hands/arms in the lower geared ffb wheel. Which is why DD wheels feel so more responsive, it's not actually due to their increased torque range (if you hold that constant…e.g. compare a DD capped to 5Nm vs a t500's 5Nm).

It's analogous to like a small tracker vs a racing car, both with identical peak torque outputs but the racing car producing it at a higher speed thus having a significantly higher HP output.


Edit: Sorry, probably best i should have said is that you feel more power from the ffb wheel with the same final torque output bdue to the lower gearing and higher torque motor. The lower geared motor will do the same amount of work in less time despite the same final peak torque output which means higher power output per given length of time.

Here's a mock example of two hypothetical ffb wheels with the same final torque output at the steering wheel but having very different acceleration rates:



Same final torque output but a quarter the effective acceleration of the same load on the shaft (i.e. steering wheel mass/inertia + your hands/arms mass/inertia on it) when you use a motor with half the torque and scale it up with a 1:2 gear/transmission ratio. That significantly smaller acceleration will result is significantly less change in velocity per given length of time which will produce less power output and less change in momentum.

And interestingly enough the single motor of the T500 turned out to be something like almost double the total combined power of the two motors in the CSR-Elite, which could explain my perceived observation of the t500 feeling 2-3 times more powerful which at the time i mistakenly thought was 2-3 times higher peak torque where as it's only 25-30% higher when measured/compared at steady-state.

This is half the reason why DD wheels feel better, in particular why they feel so much more responsive, immediate and lag free in transmitting ffb information to you.

 

[Lazza]

When I googled earlier I saw 2.5Nm quoted for the G27, wasn't official but probably isn't a coincidence rF2 is there by default.

 

[DrR1pper]

Ooops, the CSWv1/CSR-E must have been 4-ish Nm then. Makes more sense in the gradual increase of torque of ffb wheels.

 

[vittorio]

Quoting eKretz who made the incredible CSW and CSR-E motor upgrade kit: "Stock motors @24VDC pulled 8.5 pounds@~5.38"R. So about 3.8 lbs./ft. (5.2 Nm) of torque."
(https://www.gtplanet.net/forum/thre...e-modders-thread-update-february-2014.274631/) I guess 5.38" is the radius of the CSR-E wheel.

With upgrade (which i have): "With the Buhler motors @24VDC pulled 14.5 pounds@~5.38"R. or 6.5lbs./ft. (8.8Nm)."

 

[DrR1pper]

Holy crap he did a big update to his post since the last time i saw it.

5.2Nm for the CSR-E/CSWv1? ….@%!$….can someone confirm i at least got the AFPro and Bodnar numbers right?

Found TechAde's old post: http://isiforums.net/f/showthread.php/6596-Torque-of-the-Town-G27-v-T500RS

According to his test, G27 was 3.0Nm and the T500 was 6.0Nm.

And according to fanatic, CSWv2 is over 7Nm.

So new list:

G27 - 2.5-3.0Nm
CSR-E/CSW v1 - 5.2Nm
T500RS - 6.0Nm
CSW v2 - (at least) 7.0Nm
AF-Pro - 13Nm
Bodnar - 16Nm

The upgrade must feel amazing Vittorio. That's a significant increase in torque (+79%) and power(+88%).

 

[vittorio]

The upgrade must feel amazing Vittorio. That's a significant increase in torque (+79%) and power(+88%).

Yes, it really does Most important the power doesn't fade anymore cause of the great cooling system. I guess the CSW/CSR-E are ~4Nm when hot and more worse feel "spongy" (dont know the english word) and they get hot very quick.

EDIT: But I'm already looking forward building an OSW next year. Never get enough. lol