Modifying 991.1 steering software
alt theory, Adam, is that the default _is_the GT3 and all of the other maps are artificially adding something that some marketing person wanted. Porsche cars are race cars first so that makes sense.
see: what they did with the temperature readings
I understand the logic you’re basing the argument on, ie software is in the component, and PIWIS just ticks the software options on, off or changes settings. That is solid.
I’m not a software engineer or Porsche product planner, but I’m unsure if bc the GT3 wasn’t released to the public means / doess not mean the component supplier was / was not including the software already in builds of the hardware (ie the EPS rack/motor/etc). Product development takes awhile and launch dates move up and back regularly.
Let’s look at dates. This is not meant to be argumentative, rather trying to break this mystery down. I too am puzzled by this. It would seem that my car would not be able to even be coded to “GT3” EPS software.
- Aug 2012 - Spy photo of GT3 at Nurburgring etc - also spy photos of GT3’s around Europe published without camo and real dashboards. If it’s on that track, GT3 software (even if beta) exists.
- Late 2012 - My C2S was built
- March 2013 - 991 GT3 was introduced March 2013 at Geneva.
- May/June 2013 - Chris Harris drove and other UK/EU reviewers drove - and published in June 2013 all published around June 2013. Software deffo done at this time.
- This means the components for these cars were built early 2013. .... Which means the software was probably finalized sometime before that.
- The earliest 2014 GT3’s to hit the US were built in maybe October- November 2013. Ish?
A few other points:
- Why would PIWIS not be intelligent enough to lock us out of coding options that simply do not work? Seems pretty obvious. IDK though. Not a PIWIS Jedi here.
- I've not yet run into a failsafe mode for an engine, steering, transmission or other component on a car that didn’t completely neuter the component or driving experience. Engines lose all power, transmissions start in 3rd gear or don’t shift, that sort of thing. When something is in limp mode, it limps. It’s obvious. This is not that. At all. I’m making vast and dumb assumptions, so this is probably a garbage point to make. But - it does seem to make some anecdotal sense based on my automotive history.
- EPS was controversial on the 911 when 991 came out. In the launch technical documentation, I can find no mention of failsafe mode in any launch or subsequent docs. Maybe I don’t know where to look. (again, making lots of assumptions)
- The improvements that owners of, for example, 2015 GTS models document here and elsewhere when adding GT3 steering coding seem to be exactly the same improvements I note on mine with the yellow error message, when I select GT3 steering. Characteristic for characteristic, it matches. This is not seat of the pants imagination as can happen with performance modifications... or group think. It’s a real difference and pretty simple to tell the differences.
After 200-210 miles of driving, the GT3 software is a notable and welcomed improvement to my C2S. I’m going to keep watching it and seeing if I notice anything else. The one guy @vincenz0 who noted a domino of other negative effects is surprising and he’s quite smart. So -- yeah. Mystery. Not solved.
The PIWIS, launch etc do not have the map details within the software, that means it must be on the rack firmware . Which means older racks cannot have the software of a car that didnt exist when the rack was built
The default 991.1 steering map is so bad that some failsafe map is an improvement.
Id guess, though, that it was refined and validated by the time the 991.1 GT3s went into production in 2013. The validation process may also explain why later cars dont throw an errorperhaps theres something in the code that confirms its completion (like a certificate or similar mechanism).
Regardless, I sourced a 2016 rack and plan to install it in my 02/2012 car. Ill share the details eventuallyjust need to find the time to do it.
Last edited by aarodynamics; Mar 5, 2025 at 10:56 PM.
I dont like the idea of suppressing warnings so I switched back to stock C2/S config until we know more. I think the hypothesis that the software program exists on the EPS module and the PIWIS simply tells the module which config to use makes the most sense. If thats the case, Im guessing that the GT3 software on the early EPS modules were either beta or worse, incomplete. Seems like the only way forward for folks like me would be an updated firmware for my EPS module. Subscribed to this thread in case anyone figures it out.
Anyways, heres the info for my 2014 C2S which throws the EPS faults when GT3 mode is set:
Part number: 99161814706
DSN: 000004
Software version: 1330
Hardware part number: 99161814706
Hardware version: 011
Steering Porsche PN: 99134700513
My experience was a bit different from the common consensus here—I found the change in steering input and feedback to be subtle, not the dramatic "night and day" difference often mentioned. Interestingly, switching to GT3 Steering also seemed to blur the difference in suspension feel between Sport and Sport Plus modes. It's worth noting that my car had the stock PASM controller replaced with a DSC controller, making that observation even more perplexing.
In my opinion, if you're getting error codes after changing to GT3 Steering, it might indicate the EPS module defaults to a generic configuration when it can't find the specific named configuration. Ironically, this default setup might actually feel better than the standard steering map.
Hank
has anyone ever tried coding the 918 spyder steering setting?
Last edited by mikescayenne; Apr 26, 2025 at 09:52 PM.
The Best Porsche Posts for Porsche Enthusiasts
EDIT: i went back in PIWIS, and noticed there is a GT3RS setting, i know that has RAS so i tried it and it worked! Immedietly took the car for a test drive, but didnt notice much difference. Nothing like what you guys have mentioned when coding carreras to GT3 steering. The steering just felt a little heavier at all speeds, not necessarily better or worse. Wasnt too exciting, but ill keep this setting for a while and see if i notice more differences or just go back to the normal turbo S steering setting. Im thinking with the RAS cars, software change cant really have a big affect on the steering.
Last edited by mikescayenne; Apr 27, 2025 at 01:10 AM.
Does anyone have access to SW 1880 version? If so, please DM me and share?
I'll try this firmware update and let everyone know how it goes.
I just ordered 991.347.005.18 (the latest Supersession EPS rack). My plan is to upgrade the firmware, if that bricks the rack, then just replace the rack with this latest version. Hopefully the rack doesn't brick though, and I can verify a working solution for everyone else here with a 2012-2013 991.1/981.
Here are all the relevant EPS rack part numbers in my research relevant to my early build (02/2012) car:
- 991.347.011.02 (991.1 C2S Original)
- 991.347.011.03 (Supersession)
- 991.347.011.04 (Supersession)
- 991.347.011.05 (Supersession)
- 991.347.011.06 (Supersession)
- 991.347.011.13 (Supersession)
- 991.347.011.81 (991.1 GT3 original)
- 991.347.011.14 (Supersession)
- 991.347.005.14 (Supersession)
- 991.347.005.15 (Supersession)
- 991.347.005.18 (Latest Supersession)
Would it be easier to just replace servo rather than the entire rack?
Does anyone know approximately much work it would be to replace just the servo or the entire rack?
...
Thanks!
From my research, the steering rack appears to be Bosch steering Gear KS00002888. FCP Euro made the connection, but verify before buying.
From my research, the steering rack appears to be Bosch steering Gear KS00002888. FCP Euro made the connection, but verify before buying.
Last edited by aarodynamics; Jun 26, 2025 at 08:28 PM.
These codes that we alter are by and large variant coding - they are pointers in the software that instruct the ECU to use different addresses for some parameters. E.g. the variant code for model designation may instruct the EPS controller to use different assistance maps or gains. It seems highly unlikely that Porsche / their suppliers would have their homework for the GT3 done so far in advance that the calibrations would be available on 2012, early 2013 cars. The most likely phenomenon here is that the enumeration for GT3 points to default calibrations, and falls outside the range of what an early EPS module would have been calibrated to accept as valid, hence the error. Ill be interested to see how the rack (and therefore EPS module?) replacement goes.
That said, if you like the uncalibrated SW better than the released cal, more power to you. I dont know the specifics for Porsche in this case, but Ive been doing embedded SW and calibration for OEMs and suppliers for a little over a decade.
Last edited by Sh0ty; Jun 27, 2025 at 07:45 AM.
and here's Rohrl talking about developing EPS for the Carrera S in a 2012 article.
https://www.carmagazine.co.uk/car-ne...ng-with-rohrl/
and here's Rohrl talking about developing EPS for the Carrera S in a 2012 article.
https://www.carmagazine.co.uk/car-ne...ng-with-rohrl/
i put the car into other modes and drove around.
-the 4S mode gave the error and steering feel was just like the 2S.
-there was another mode that wasn't the GT3 mode that threw the error and felt different from the 4S or 2S modes.
-the GT3 mode throws error and steering feels more like my 993.
so all modes besides the one for the car had the same error and those other calibrations were clearly ready for production.
anywho
got a new steering wheel to match my awesome, perfect, problem-free GT3 calibration. :P






