Fanatec Announcements: CSW V2 Reviews Out

  • Thread starter Crispy
  • 13,388 comments
  • 1,320,608 views
It hasn't been confirmed, but coyly hinted at several times. I do believe they are at least seriously considering it, and possibly have already started designing, or possibly have tried a prototype or two.

I would definitely want to test and review one if the chance comes about. Although its hard to imagine how much better a directdrive would feel over the v2.
My biggest point of interest would be to discover what types of sim racers would benefit enough from this type of wheel to justify the price.
 
I would definitely want to test and review one if the chance comes about. Although its hard to imagine how much better a directdrive would feel over the v2.
My biggest point of interest would be to discover what types of sim racers would benefit enough from this type of wheel to justify the price.
"some says" that it is all about feeling before seeing (so reducing the input lag under the rendering and display input lag which by the way change a lot my driving performance). But the are no "scientific" ore benched proof that a belt driven wheel is dramatically slower than a direct drive wheel. I have never seen a wheel check graph comparison between a CSW and a Leo Bodnar wheel until now.
 
I have seen some partial graph comparisons, the direct drive wheels out now are relatively much quicker to get moving. Whether it would make a very big difference to one's driving ability in-sim is another story. I'm positive it wouldn't hurt, but don't know how much it would help without driving with both wheels back to back.
 
In the end the only discussion is whether any wheel is worth the money and whether it makes more expensive wheels effectively redundant.

I would assume any direct drive wheel can achieve the following:
1) lower latency and very low internal resistance due to no cog or belt mechanism
2) due to having vastly more motor power and no gearing, it should be able to work against wheel rim inertia much more effectively to produce the tiniest subtle effects and quicker changes of direction (but only if sims are built to make use of those capabilities)
3) it should have enough torque to allow the wheel to spin as fast as needed so that the simulated wheel position will always follow the front tyres position

These attributes will obviously help most effectively in oversteer situations. Drifting in particular. Would it make a smooth driver any faster over a single lap using a direct drive wheel over a Logitech DFGT? I doubt it. But identifying and reacting to any unexpected situations should be a lot easier.
 
  • Spotted this on VirtualR


  • Thomas Jackermeier13 hours ago
    In order to add to the Servo vs. Direct Drive discussion I might want to add a few words.

    A Direct Drive wheel is much easier to develop and therefore the choice of small companies with limited R&D budget. The huge size of the motor makes it very expensive as the cost is also raised by the bigger motor driver and power supply.

    And in theory we could develop a belt driven wheel with 20 NM or more so a lot of things can be done with a belt drive.

    But there is no doubt it has some technical advantages over a belt drive.

    However the really interesting question is: How big are these advantages in practice and how much money is it worth?

    If a belt wears out after 10 - 15 years of daily use, does that really matter?

    The biggest difference between current Direct Drive wheels on the market and the V2 is the maximum torque. If you crank up the FF forces you will be shocked how brutal those DD wheels are and it certainly is impressive if you try it on a show or at a friend who wants to impress you. But will you run at that force level on a 4 hour race?

    The cogging (notchiness) of a V2 cannot be felt by humans. You need a machiene to measure it. So if a DD wheel is better on that, how much does it really matter if only test instruments can feel the difference?

    There is also a big difference between belt drive and belt drive. On our Porsche wheels for example we are using GT belts on plastic pulleys and with the need of belt tensioners. We use some glide bearings instead of ball bearings. (btw the competition does it the similar). In such a system there is some noticable loss.

    On the new CSW B V2 however, we are using Poly V belts on metal pulleys with much higher precision and there is no need for belt tensioners. Other than in the CSW V1, the use of big ball bearings on the pulley and in the motor make sure that we can use a lot of belt tension in the V2. The higher the belt tension, the lower the negative effects of dampening subtle force effects. And there is no belt slip at all.

    This system is so effective that we even had to add artificial dampening (which can be de-activated) by default to the motor just like direct drive wheels do.

    So if our belt drive is already with such a low drag that we had to add artificial dampeneing in order to avoid problems on the center what does that tell you about how close a belt drive can get to a direct drive?
 
This is what I left in reply:

"The plus side of this Thomas is that new games will be able to add drag/dampening themselves or let that fall-out of their Physics simulation, and do this on a per car basis. This way you can have anything ranging from a Fiat Grande-Punto that has speed sensitive powersteering which you can pinky park, up to a more sports oriented heavier resistance powersteering rack.

So with no more mechanical drag you can live that control up to the game developers, instead of forcing them to try to work around hardware limitations.
This should technically lead to a jump in conveying the personalities of cars much better."
 
Random question. When you are using the Formula Rim for example and in game you switch to the BMW rim; can the button assignments be done individually to each rim and the game automatically can detect which rim is which or do you have to go through and readdress your button assignments?
 
Random question. When you are using the Formula Rim for example and in game you switch to the BMW rim; can the button assignments be done individually to each rim and the game automatically can detect which rim is which or do you have to go through and readdress your button assignments?

Wasn't the case with the CSW v1. Games aren't yet supporting the CSW v2 because of embargo forced game devs to keep the existence of the wheel hidden, so I can't really tell if it is possible now.
For now to keep on the safe side, I don't think it's possible.



@Fanatec... Could you please implement the above suggestion in the driver and SDK for game devs, or implement a better solution with an equal result? With more and more wheels added, I think this piece of comfort is something that will be necessary.
On a personal note I know how bad resetting your button functions is. I had to do this for 3 years with Project CARS as you have to delete your profile with each new build you'd download (meaning all your settings would be gone).




As a (temporary) workaround for everyone:
What you could do though is to use X-Padder to make profiles for different wheels that are hooked to different key combinations used within the game for a certain function.
I am not sure if you can switch between X-Padder profiles with a key combination from memory, but if so than you could use a program called VoiceAttack to switch the profiles by voice command. Example 'Switch wheel profile' + 'Porsche'.

X-Padder: http://xpadder.com/
VoiceAttack: http://www.voiceattack.com/
 
Same here Mark,

I was happy to help and got the feeling of being part of something special :)

Here's another quote from Thomas about the CSWv2 from the Iracing Forum, which reveals a bit more of the CSWv2 for those who are not Iracing members:






Berney Villers wrote:^^ Just kidding
3b63d1616c5dfcf29f8a7a031aaa7cad.gif
I couldn't help myself.

Some time ago we did a survey in this thread and it was accuracy / resolution that this forum's readers prioritized over torque.

In the interest of setting hype aside and letting the products speak for themselves, I thought maybe we could compile an encoder resolution list here.

Leo Bodnar Sim Steering - 10,000 pulses per revolution
SimXperience AccuForce DIY - 16,000 pulses per revolution
SimXperience AccuForce Pro - 16,000 pulses per revolution
SimXperience AccuForce Team - at least 16,000


Does anyone have specifications for the consumer grade wheels?



Berney Villers wrote:It's a pretty complicated subject to be honest and worthy of a deeper dive.

There is positioning resolution and there is torque resolution. If you have no force feedback you still have positioning resolution. Positioning resolution is what David's test was measuring; the resolution of the physical encoder relative to the range of direct input.

...

By my definition, torque resolution (you could maybe call this force feedback resolution) accounts for all the variables in a motor control system and is a measure of how granularly a wheel can reproduce force. ...

As titled, the AccuForce aims to accurately reproduce the force and in great detail.​
.


I am replying to Berneys' post here in the CSW B V2 thread as courtesy to Berney because I want to stay out of his thread where he made the posts above.

He was asking for the pulses per revolution and here is some first hand information he can add to his list:

ClubSport Wheel Base V2 - 30,000 pulses per revolution (force feedback resolution)
ClubSport Wheel Base V2 - 4096 pulses per revolution (positioning resolution)

And I fully agree that the resolution makes a huges difference in the smoothness of a wheel. The more pulses the less you will feel cogging. The belt drive allows to multiply the resolution of the motor sensor and this is one of the advantages of a belt drive which should be noteworthy if you are making a serious and competent comparision between a belt drive and a direct drive without hyping one of those.
3b63d1616c5dfcf29f8a7a031aaa7cad.gif


If you will get your hands on a CSW B V2 you will feel why we described the feeling as "smoothastic". The use of two sensors (Dual Hall technology) allows both a precise input of the axis signals as well as a super high resolution of the Servo motor to avoid any cogging.

I hope this helps to set aside the hype
8a80c6485cd926be453217d59a84a888.gif

This message was edited 1 time(s). Last update was at 9/1/2014 2:09 p.m.
 
Last edited:
Does any beta tester want to sell me their V2 wheel? Top prices paid. :)

The wheels are still the full property of Endor A.G./Fanatec. So we can't sell them, but have to initially return them instead. Though this is up to courtesy of Endor A.G./Fanatec to change.

As you could read on the Fanatec blog there has been some changes to the motor. If those changes are highly critical it is best to ask the beta testers to return their wheels instead and if desired by Endor A.G./Fanatec to give them a different kind of compensation for their troubles. (E.g. Discount on a retail base)
 
Great reviews ;) Thank you guys.

Hope to see some more GT6 specific comments in the next ones.

GT6 is really a horrible horrible and extremely terrible game as far as Force Feedback goes. For all I care Midtown Madness might have better Force Feedback.

So yeah, ****** game. What do you want to know exactly?
 
Back