Hands-on With Direct Axial Drive: Thrustmaster T598 First Impressions


The Thrustmaster T598 offers a low-end Direct Drive wheel base with console compatibility. We tried the new base - here are our first impressions.

Two years after unveiling its first foray into Direct Drive territory with the T818, Thrustmaster announced a lower-end DD wheel base that also features compatibility with PlayStation 4 and 5. The T598 boasts 5 Nm of constant torque, a Direct Axial Drive Motor that is intended to eliminate cogging, and it comes in a bundle with a new 300-mm Sportcar wheel rim and the Raceline LTE pedals - not a bad deal for €/$499.99.

How does it feel, though? At ADAC SimRacing Expo, our own @Michel Wolk had the opportunity to try the T598 both on PlayStation and on PC to gather his first impressions.

Thrustmaster-T598-First-Impressions-Sim-Racing_Expo-2024-GT7-PS5.jpg


No Cogging, Smooth Steering​

According to Thrustmaster, the T598 heralds nothing less than a new era in Direct Drive steering wheels, as it replaces the radial flux motor with an axial flux motor, which differs in the arrangement of the stator tubes and thus the magnetic flux generated parallel to the shaft. This completely eliminates the cogging that can be felt when you turn your Direct Drive steering wheel when it is switched off.

When driving, this is usually dealt with by software and damping. According to Thrusmaster, the whole Axial Flux design has only advantages - less material, lower cost, a pure signal for sharper and cleaner effects, plus the eliminating clipping, so the loss of FFB effects during torque peaks.

We can already see from the wheelbase layout that it is significantly shorter, as this type of motor is simply less long. There are some drawbacks to this design as well, however: It is said to be very difficult and expensive to produce - which Thrusmaster seems to have this under control if they are offering it at the price point they do. On the other hand, heat dissipation is said to be a problem, which explains the rather tall base and why they start with a constant torque of 5 Nm.

Thrustmaster had the T598 set up on desks in their booth, as many sim racers out there do not have a dedicated rig. Michel found the base to be rather stable and smooth in Gran Turismo 7, and the Direct Axial Drive technology meant that the wheel felt quick to react when steering from side to side quickly.

Thrustmaster-T598-First-Impressions-Sim-Racing_Expo-2024-ACC-PC.jpg


The Clipping Question​

And since the base is compatible with other Thrustmaster steering wheels in addition to the Sportcar one it comes with, Michel finds the bundle to be a good starting point for those who want to take their casual racing on the PlayStation to a new level.

On the other hand, running the PC version of Assetto Corsa Competizione on the T598 mounted to a rig and with the Ferrari 488 GT3 wheel, the base continued to feel clean and smooth, but not immune to clipping when running over curbs, for example - according to ACC, that is, and not too much, either.

Interestingly, clipping should be completely avoided by the tech used in the T598, according to Thrustmaster, so we are eager to find out more. We should have a review unit avavilable soon to take a closer and more in-depth lool.

If there are any specifics that you want us to look into, feel free to leave your remarks and questions in the comments below, and we will try and answer as many as we can. Feel free to also join the discussion in our hardware forum!
  • Like
Reactions: Wipeout999 and b)b
About author
Yannik Haustein
Lifelong motorsport enthusiast and sim racing aficionado, walking racing history encyclopedia.

Sim racing editor, streamer and one half of the SimRacing Buddies podcast (warning, German!).

Heel & Toe Gang 4 life :D

Comments

On the other hand, running the PC version of Assetto Corsa Competizione on the T598 mounted to a rig and with the Ferrari 488 GT3 wheel, the base continued to feel clean and smooth, but not immune to clipping when running over curbs, for example - according to ACC, that is, and not too much, either.

Interestingly, clipping should be completely avoided by the tech used in the T598, according to Thrustmaster, so we are eager to find out more.
Seriously?? The owners or employees of a simracing site who we're supposed to give credibility to their impressions, opinions, and reviews on sim racing hardware & software don't know the difference between software clipping and hardware clipping?? Really??

You can have a wheel capable of 9,000,000 Nm and have it clip like crazy in any game. You can have a wheel with 0.00001 Nm and have it never clip in any game.

Software clipping is when the game should be outputting higher levels of FFB but is already at the 100% point so can't tell your wheel to output any higher since 100% is the highest...this is what we can see in games since it's the signal/info coming from tbe game and is completely irrelevant to the real-life wheelbase being used.
 
Last edited:
If there are any specifics that you want us to look into, feel free to leave your remarks and questions in the comments below, and we will try and answer as many as we can.

I will be glad if you can test or provide information about Linux support for new wheels on the market. If someone know if already available Moza of Fanatec wheels works on Linux without issues then thanks in advance for information. T150 or T300 works well, G29 probably too but they are quite old and they works thanks to community. If some company provide Linux support out of the box it would be fantastic information.
 
Last edited:
OverTake
Premium
Seriously?? The owners or employees of a simracing site who we're supposed to give credibility to their impressions, opinions, and reviews on sim racing hardware & software don't know the difference between software clipping and hardware clipping?? Really??

You can have a wheel capable of 9,000,000 Nm and have it clip like crazy in any game. You can have a wheel with 0.00001 Nm and have it never clip in any game.

Software clipping is when the game should be outputting higher levels of FFB but is already at the 100% point so can't tell your wheel to output any higher since 100% is the highest...this is what we can see in games since it's the signal/info coming from tbe game and is completely irrelevant to the real-life wheelbase being used.
Hey @Spinelli,

Discussed with the team, just making sure we are on the same page. The clipping value displayed by ACC is reporting that the games output ffb signal can't be handled by the wheel base, nothing else. According to Thrusmaster the T598s motor design would deal with peak forces provided by the sim without any clipping.

this is what we can see in games since it's the signal/info coming from tbe game and is completely irrelevant to the real-life wheelbase being used.

That would mean, the graph would be the same, if you would use a 15+ Nm wheel base, when you hit certain kerbs and that is just not the case.
 
The clipping value displayed by ACC is reporting that the games output ffb signal can't be handled by the wheel base, nothing else.
I repeat: ACC has no way of knowing the Nm rating of the base, or if the base is clipping. What it's indicating is an overflow in the in-game FFB effects calculations. Why do you think iRacing asks for the Nm rating of the attached wheel base?

Your claim is akin to claiming that when a digital audio editor is indicating clipping its because the PC speakers are too small.

That would mean, the graph would be the same, if you would use a 15+ Nm wheel base, when you hit certain kerbs and that is just not the case.
It would be the same if you used the exact same FFB settings with both bases. And I'm sure you didn't.
 
OverTake
Premium
I repeat: ACC has no way of knowing the Nm rating of the base, or if the base is clipping. What it's indicating is an overflow in the in-game FFB effects calculations. Why do you think iRacing asks for the Nm rating of the attached wheel base?

Your claim is akin to claiming that when a digital audio editor is indicating clipping its because the PC speakers are too small.


It would be the same if you used the exact same FFB settings with both bases. And I'm sure you didn't.
Yes that is true I come from rF2 where the nm of each wheel is set in the controller JSON.

Side note: I'm sorry, only half looking at this today as I'm trying to build a new PC and it's taking forever
 
OverTake
Premium
Seriously?? The owners or employees of a simracing site who we're supposed to give credibility to their impressions, opinions, and reviews on sim racing hardware & software don't know the difference between software clipping and hardware clipping?? Really??

You can have a wheel capable of 9,000,000 Nm and have it clip like crazy in any game. You can have a wheel with 0.00001 Nm and have it never clip in any game.

Software clipping is when the game should be outputting higher levels of FFB but is already at the 100% point so can't tell your wheel to output any higher since 100% is the highest...this is what we can see in games since it's the signal/info coming from tbe game and is completely irrelevant to the real-life wheelbase being used.
Just sharing my experience and thoughts. I don't know exactly where the ACC FFB clipping graph gets its data from and how the API works for each game, we'd have to ask Kunos about that - but there is certainly a difference between running a 5 Nm wheelbase with 100% ingame / driver settings and a 15 Nm wheelbase with the same 100% settings. Just drive any car over the first kerb, with the 5 Nm wheelbase the graph will turn red. This will not happen with the 15 Nm. That was all I was referring to.

I repeat: ACC has no way of knowing the Nm rating of the base, or if the base is clipping. What it's indicating is an overflow in the in-game FFB effects calculations.
And that is why it makes no sense to me. If that were true, the ACC graph would be the same for any wheelbase and 100% FFB in-game.

I got some developer insights in the meanwhile, but as @Christopher E said, rF2 seems to handle it differently. What makes this even more complicated are filters the wheel bases add on top.
 
Last edited:
Just sharing my experience and thoughts. I don't know exactly where the ACC FFB clipping graph gets its data from and how the API works for each game, we'd have to ask Kunos about that - but there is certainly a difference between running a 5 Nm wheelbase with 100% ingame / driver settings and a 15 Nm wheelbase with the same 100% settings. Just drive any car over the first kerb, with the 5 Nm wheelbase the graph will turn red. This will not happen with the 15 Nm. That was all I was referring to.


And that is why it makes no sense to me. If that were true, the ACC graph would be the same for any wheelbase and 100% FFB in-game.

I got some developer insights in the meanwhile, but as @Christopher E said, rF2 seems to handle it differently. What makes this even more complicated are filters the wheel bases add on top.
You're overcomplicating this. Like others have also said, the real life wheel is irrelevant in this case (software clipping). It all has to do with the game's FFB, It doesn't matter what game it is - ACC, IR, RF2, or whatever. It doesn't matter what filters are added on top purely from the hardware side. It doesn't matter if the wheel's max Nm is 1, 10, or 100.

If there's software clipping (like in the ACC video) then the game has it's FFB settings set to a certain range that the game is surpassing at times. Simple as that.

Hey @Spinelli,

Discussed with the team, just making sure we are on the same page. The clipping value displayed by ACC is reporting that the games output ffb signal can't be handled by the wheel base, nothing else. According to Thrusmaster the T598s motor design would deal with peak forces provided by the sim without any clipping.

That would mean, the graph would be the same, if you would use a 15+ Nm wheel base, when you hit certain kerbs and that is just not the case.
There may have been a misunderstanding or miscommunication between you and the devs because that's not how it works. Software clipping is absolutely, purely, 100% hardware independent.

That clipping means ACC"s FFB settings (per-car multiplier or in-game setting or text file setting, etc.) was set to a certain FFB value that the game surpassed those times when it was in the red (software clipping).

A wheel with a max force of 1 Nm or 100 Nm could have been used in that situation and the software clipping results would have been 10000% identical if no game-FFB changes were made.
 
Last edited:
OverTake
Premium
You're overcomplicating this. Like others have also said, the real life wheel is irrelevant in this case (software clipping). It all has to do with the game's FFB, It doesn't matter what game it is - ACC, IR, RF2, or whatever. It doesn't matter what filters are added on top purely from the hardware side. It doesn't matter if the wheel's max Nm is 1, 10, or 100.

If there's software clipping (like in the ACC video) then the game has it's FFB settings set to a certain range that the game is surpassing at times. Simple as that.

There may have been a misunderstanding or miscommunication between you and the devs because that's not how it works. Software clipping is absolutely, purely, 100% hardware independent. That clipping means ACC"s FFB settings (per-car multiplier or in-game setting or text file setting, etc.) was set to a certain FFB value that the game surpassed those times when it was in the red (software clipping).

A wheel with a max force of 1 Nm or 100 Nm could have been used in that situation and the software clipping results would have been 10000% identical if no game-FFB changes were made.
Got your point now, thanks. I am still not sure, if the clipping graph of ACC is purely showing the software clipping, I will test that again with different wheel bases.
 
I am still not sure, if the clipping graph of ACC is purely showing the software clipping
It can't possibly show anything else than that as it doesn't know the NM force of the wheel, the gain the wheel is set to or anything similar.
ACC only knows the Direct Input API force value that it sends to the wheel and that's where the software clipping comes from, so ACC (like all other sims, including rF2) show software clipping on their FFB graphs (proportional value of the FFB force from 0 to max direct input allowed value).
 
I will be glad if you can test or provide information about Linux support for new wheels on the market. If someone know if already available Moza of Fanatec wheels works on Linux without issues then thanks in advance for information. T150 or T300 works well, G29 probably too but they are quite old and they works thanks to community. If some company provide Linux support out of the box it would be fantastic information.

Moza works well in wine/proton and native games, not out of the box but installation of universal-pidff driver with dkms took 2 commands, git clone and dkms install.

Then get boxflat which allows you to configure it.
 
Moza works well in wine/proton and native games, not out of the box but installation of universal-pidff driver with dkms took 2 commands, git clone and dkms install.

Then get boxflat which allows you to configure it.
Thank you so much for answer and direction to follow with further research! I really appreciate it!
 
Back
Top