? PCM 6 Update Announcement [Update: Retrofit Being Explored For Existing Taycans]

[Scandinavian]

I bet Porsche will make the update available to all existing Taycan model as when they launched the car they have committed that the car PCM 6.0 will stay up to date over the time. EVs are like gadgets and no one buys a gadget which will be out of date in just months. I bought mine back in Sep 2021 (MY 2021) based on Porsche’s commitment that the car will receive future update. Now if they do not stick to that, I will be one of them to sue Porsche for miss selling a product as I wouldn’t buy an EV knowing in few months the software would be outdated. I know that Porsche already failed to keep their commitment about OTA update as the wired android auto update for MY 2022 did not get to the MY 2021 or older models but this new update is crucial as the updated route planner sounds very accurate along with the capability of filtering chargers based on output and see their availability in real time. Wireless android auto is also a great addition and btw the hardware it requires is already in the existing head unit because all Taycan support wireless CarPlay.

I fully agree with you. There is no way that Porsche can avoid to bring some of these new features to all Taycans. I can understand that Android Auto might not be available if it tequires any HW. But again I question that since as you also state, wireless CarPlay and WiFi is already there. I know nothing about requirements for Android Auto, but there are cars available that have both options as standard. So it can not be that difficult.

I just can not understand why Porsche have not responded to the questions asked. It seems there are different answers from dealers and countries.

This is a fine mess they got themselves into.

Sponsored

 

[atebit]

My MY21 PCM software levels feel positively ancient!

A proper OTA capability would make things so much easier for the development teams by allowing them to aggressively sunset older version levels.

You mean foist half-tested (at best) code onto unsuspecting “beta” users?

100% agree it appears that Porsche has missed the mark on the value of OTA. But maybe this is all part of their cunning plan to only use it when they are sure they have code worthy of OTAing vs just shipping whatever.

Can you guess how I really feel about Agile?

 

[stirthepot]

i wish they were using an agile development process, the reason we are all frustrated is because they are not

 

[Perry]

That is a software version, but I thought the PCM 5 vs PCM 6 was a hardware difference not a software version. Am I incorrect in that? Thought I read it somewhere.

As I understand it, it's a bit messy in the Taycan case because all Taycans shipped with something called PCM 6. The difference is that with MY22 they released a new version of PCM 6 for all models, including new hardware.

So MY20/21 Taycans are running the Taycan exclusive PCM 6 while MY22 is running the new fleet-wide PCM 6. The new version they released now is thus an update to the new PCM 6 which the old Taycans don't have and might not even get without a physical retrofit because of hardware differences.

 

[mlambert890]

As I understand it, it's a bit messy in the Taycan case because all Taycans shipped with something called PCM 6. The difference is that with MY22 they released a new version of PCM 6 for all models, including new hardware.

So MY20/21 Taycans are running the Taycan exclusive PCM 6 while MY22 is running the new fleet-wide PCM 6. The new version they released now is thus an update to the new PCM 6 which the old Taycans don't have and might not even get without a physical retrofit because of hardware differences.

People with 2022's on this thread have already shared that they have the exact same hardware. There is a new part number (PAD) that folks have found, but so far no one has reported actually having that. So if there really is such a significant hardware difference, then they've actually also divided a single model year. I can't believe even Porsche would allow such a train wreck to just play out that way.

 

[Fletch]

The cynic in me thinks the only thing they’ll learn from this is to not put out PR’s about new software versions and hope people don’t notice

 

[schad]

The new version they released now is thus an update to the new PCM 6 which the old Taycans don't have and might not even get without a physical retrofit because of hardware differences.

There is absolutely no excuse -- none whatsoever -- for not outfitting the Taycan with sufficient processing power to accept full-feature upgrades for at least five years. The first-gen iPhone SE, release date March 2016 (almost 6 years ago), can run the latest iOS. And that's a phone! Not only did it cost less than 1% of a Taycan, it also has all kinds of physical constraints that the car doesn't have.

With that said, it's quite conceivable that the PCM's hardware is in fact too old and slow to support any upgrades. That's exactly the sort of nickel-and-dime BS the auto makers just love to do. Why put in a $5 part that's future-proof, when you can put in a $4.98 part that's obsolete before the car even enters production? Not only do you save a whopping $0.02 per unit -- on a car that probably has an average sales price of over $130k -- it's a way to force people to buy a brand new car every two years... without having to expend any actual effort to make the product better!

 

[Fletch]

Feels like the delay/uncertainty is less about hardware and more about testing surface area. there’s a lot of upgrade permutations to work through, and the risk of bricking a car is high. a bad upgrade means flatbed trucks taking cars to dealerships. That’s…pricey. I sure hope it happens, but it may happen slowly.

 

[vaelin]

i wish they were using an agile development process, the reason we are all frustrated is because they are not

Agile has nothing to do with it. And honestly, for something so closely tied to a mission-critical (2.5 ton land rocket) piece of equipment, you probably want waterfall-esque regression testing anyways, which clearly isn't happening with all the various regressive bugs that pop up after every release.

IMO the problem is whoever is driving the actual requirements aren't in touch with what people are expecting in modern UX. The iOS/Android era has completely changed perception of the consumer experience, but when auto manufacturers are still stuck trying to bridge the gap between old and new, they end up with a half-assed hybrid of both.

 

[atebit]

i wish they were using an agile development process, the reason we are all frustrated is because they are not

I think it’s more like updating the existing fleet in terms of features/functionally isn’t a priority, or even on the radar for Porsche. “Here, have map update” appears to be their horizon of what OTA means.

 

[stirthepot]

Agile has nothing to do with it. And honestly, for something so closely tied to a mission-critical (2.5 ton land rocket) piece of equipment, you probably want waterfall-esque regression testing anyways, which clearly isn't happening with all the various regressive bugs that pop up after every release.

IMO the problem is whoever is driving the actual requirements aren't in touch with what people are expecting in modern UX. The iOS/Android era has completely changed perception of the consumer experience, but when auto manufacturers are still stuck trying to bridge the gap between old and new, they end up with a half-assed hybrid of both.

I think it’s more like updating the existing fleet in terms of features/functionally isn’t a priority, or even on the radar for Porsche. “Here, have map update” appears to be their horizon of what OTA means.

a rapid iterative approach where they release small updates vs a major release would definitely be better IMO. that is what 90% of the world has moved to - micro releases with small updates vs wait for a big one where you decided what to deploy 12 months ago and didn't have the requirements rights.

i think most would rather get a new update to the nav algo one week, and a couple weeks later get some colored icons, and then a week later get a bug fix, etc., etc., vs wait months in between updates and deal with no improvements or bug fixes during the wait.

regression testing on a micro release is always faster and easier than on a major update. too many variables all the update changes and all the hardware changes in one go is ALWAYS problematic. waterfall is a disaster, and a very dated methodology for a reason. show me a start up that uses waterfall, you won't find one. it's a legacy approach that only those that 'don't get it' use, and they continue to fall further and further behind.

porsche is struggling with just this problem. a legacy auto maker, trying to compete with new expectations of continual updates and being outpaced by start ups that are excelling in this regard.

 

[nickmdp]

a rapid iterative approach where they release small updates vs a major release would definitely be better IMO. that is what 90% of the world has moved to - micro releases with small updates vs wait for a big one where you decided what to deploy 12 months ago and didn't have the requirements rights.

i think most would rather get a new update to the nav algo one week, and a couple weeks later get some colored icons, and then a week later get a bug fix, etc., etc., vs wait months in between updates and deal with no improvements or bug fixes during the wait.

regression testing on a micro release is always faster and easier than on a major update. too many variables all the update changes and all the hardware changes in one go is ALWAYS problematic. waterfall is a disaster, and a very dated methodology for a reason. show me a start up that uses waterfall, you won't find one. it's a legacy approach that only those that 'don't get it' use, and they continue to fall further and further behind.

porsche is struggling with just this problem. a legacy auto maker, trying to compete with new expectations of continual updates and being outpaced by start ups that are excelling in this regard.

I realize I'm probably in the minority here, but the thought of weekly or even monthly updates for my car scares the crap out of me, and this is coming from a younger person in IT with a literal rack of computer equipment at home that I keep up to date on a regular basis. I want my car to be a static experience from a technology perspective though.

It sounds great in theory, but in practice I don't think it actually works out that great. You likely think of all the tiny inconveniences you have with the car and how great it would be to fix all of them, but you ignore all of the things that might get added that not only inconvenience you more, but also require you to learn a new process because there were some drivers that didn't like where the heating controls were.

The main problem with all of that is that 99% of the time I'm in my car, I'm driving it. I'm not going to get into my car and watch some sort of training video every month to tell me all about the new features, and even if I did, and played around with it while the car was parked, I'll still have all the muscle memory of how to do things while driving, and changes will be distracting to me. This is all of course ignoring any bugs that might slip through QA, which will absolutely happen in any development process.

To me, the goal of the tech in the car is to be completely transparent while driving, and even a consistently inconvenient interface would be better than something that's inconsistent.

 

[nafzal80]

The topic is getting carried away in a debate between RAPID iteration agile and long cycle waterfall, this group is likely way off topic (cause we are making assumptions, we are outsiders, not Porsche insider / developers / PCM team members) and exaggerating how it applies to Porsche PCM SW and HW development.

Taycans PCM isn’t horrible at the moment but it is pretty buggy. 2020 / 2021 / 2022 Taycans should have similar priorities from Porsche when it comes to stability / resiliency/ security and features. They are all CURRENT gen Taycans.

Pretty simple. Think like a customer. Porsche seems to be figuring out what there official answer will be, until then it would be wise to keep the pressure on so they do what is fair.

 

[stirthepot]

I realize I'm probably in the minority here, but the thought of weekly or even monthly updates for my car scares the crap out of me, and this is coming from a younger person in IT with a literal rack of computer equipment at home that I keep up to date on a regular basis. I want my car to be a static experience from a technology perspective though.

It sounds great in theory, but in practice I don't think it actually works out that great. You likely think of all the tiny inconveniences you have with the car and how great it would be to fix all of them, but you ignore all of the things that might get added that not only inconvenience you more, but also require you to learn a new process because there were some drivers that didn't like where the heating controls were.

The main problem with all of that is that 99% of the time I'm in my car, I'm driving it. I'm not going to get into my car and watch some sort of training video every month to tell me all about the new features, and even if I did, and played around with it while the car was parked, I'll still have all the muscle memory of how to do things while driving, and changes will be distracting to me.

To me, the goal of the tech in the car is to be completely transparent while driving, and even a consistently inconvenient interface would be better than something that's inconsistent.

i pretty much agree with you. i'm not suggesting massive UI changes every week. updating the nav algo to improve charging on trip wouldn't require learning anything new, nor would a bug fix, nor would adding color icons. changing out one component in your rack of equipment, one upgrading the software on one piece is much easier to do than buying all new gear at one and installing a new OS on the stack.

i too am an IT guy, spent many years on both the infrastructure and development side. AWS is a perfect example, they have thousands of releases a year to their platform, they dont try to do a massive udpate once a year. its this very reason, their agility, that they are the #1 in the space and have completely crushed the legacy installed base of Microsoft, IBM, Oracle etc. like them or not, Tesla is doing the same to the auto industry.

i think tesla has pretty much nailed it on the sofware side, latest UI update aside. they constantly release minor updates or add a piece of functionality and then once every 12-18 months a more significant update.

i enjoyed getting the bugs fixed of course, and getting into my car to find i now had access to spotify, or youtube, or a new game, or the cameras being able to be used for dashcam etc, allowed incremental changes to avoid the upheaval you indicate when a major update happens. agree, major changes can be problematic from a UI / end user experience.

take that major change and add in the complexity of testing across multiple versions of hardware and you have a mess - where porsche now finds itself with the subject of this thread.

 

[stirthepot]

The topic is getting carried away in a debate between HIGH iteration agile and long cycle waterfall. Getting way off topic and exaggerating how it applies to Porsche and PCM.

PCM isn’t horrible but it is buggy. 2020 / 2021 / 2022 Taycans should have similar priorities when it comes to stability / resiliency/ security and features. They are all CURRENT gen Taycans.

agreed,. but it is at the heart of the problem. updating the nav algo, or adding color icons should be able to be done to all cars / all years. 100% agree.

different hardware versions are apparently creating issues for the release of the update to all. since they haven't broken the update out into small components they can't provide an answer as to whether the update will work on old cars or not.

so, unfortunately, you can't divorce the two. agility is the answer to the problem and porshe isn't taking an agility driven approach. hence the lack of clarity and customer frustration.

Sponsored