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

[Skilly]

Just because it is software, doesn't mean its free - Porsche have to spend engineering euros on developing new features and it should not be unreasonable for them to expect a return on that investment in the form of people upgrading their vehicle or winning sales from non-Porsche customers.

This is the wrong mindset, unless you are saying we are purchasing the PCM and getting the car for free. Makes zero sense in the car world...the PCM is essential for the car's operation. If the steering wheel only turned right reliably, or the brakes were occasionally buggy, this wouldn't even be a discussion.

Back to software, I would agree with you under a normal formfactor lifecycle management, that advancements in later versions wouldn't be inherent to the original. I don't think anyone is saying that. That isn't what this is at the core of the concern though.

We are talking about a completely new PCM that is unique to the Taycan (that itself seems just dumb) that had chips switched mid cycle. And, an announcement that shows a new OS while most of the fleet is hemorrhaging from a terrible experience in the current OS with ZERO acknowledgement of this CSAT issue.

At the very least, its tone deaf (and creating a lot of emotional responses) but past that, all of this change management, screams a design flaw; especially when hardware changes are down within the original release (more than once) without any OS changes.

It seems clear that engineering and hardware software design was a mismatch. CSAT issues aren't loud enough and marketing isn't aware of any of this.

Porsche doesnt have example of this kind of major change without a refresh to the platform, like 991.1 to 991.2.

This is pure damage control.

Sponsored

 

[schad]

Appreciate you posting this as I am a long term non-advocate of OTA updates for anything that has a potential safety impact - what you have described is precisely the type of incident we must avoid at all costs.

This is a really complicated topic and I'm not sure I can do it justice here.

But basically, OTA updates are a way of reducing the overhead associated with deploying software changes to cars after they've been sold. Think about it like this. If it takes 6 hours of tech time to deploy an update to your car, nobody's going to do it unless it's to fix a critical problem. It costs the automaker 6 hours of tech time, it costs the customer a day (or more) without their car, etc. And even when the decision is made to do an update, it'll tend to be as many updates as possible all at once, to hopefully get some economies of scale.

This is what we in the software engineering world call an antipattern. It's the opposite of a best practice; it's something that you specifically should not do. The safest kind of software update is a small one, one that makes only one or two changes. This is why the modern software development lifecycle is focused around making very frequent, yet very small, changes.

If OTA updates are used in this way, they can not only be safe, but safer than the current "bring your car in for a recall" approach. How long is the typical gap between the moment a defect is corrected by the manufacturer, and the moment the correction is actually applied to the median car in the wild? I'd bet it's at least a month. With OTA updates, that gap could be only a week.

The problem is that the traditional automakers aren't doing OTA updates this way. They're treating OTA updates as a way to do recalls, but without having to pay for tech time. That's... awful. And it reflects a very fundamental misunderstanding of the way software development works.

I understand why the automakers are getting this so wrong. Software development looks sort of like auto development, so it's natural to treat it the same way. But it's not the same at all. You have so much more freedom when it comes to software than you do with products that are designed to exist in the physical world. I understand why they're making the mistake, but I still don't forgive them for it. Software engineering is very much an unsolved problem, but we know a lot of things that don't work. This is one of them.

 

[raharris]

This is the wrong mindset, unless you are saying we are purchasing the PCM and getting the car for free. Makes zero sense in the car world...the PCM is essential for the car's operation. If the steering wheel only turned right reliably, or the brakes were occasionally buggy, this wouldn't even be a discussion.

Back to software, I would agree with you under a normal formfactor lifecycle management, that advancements in later versions wouldn't be inherent to the original. I don't think anyone is saying that. That isn't what this is at the core of the concern though.

We are talking about a completely new PCM that is unique to the Taycan (that itself seems just dumb) that had chips switched mid cycle. And, an announcement that shows a new OS while most of the fleet is hemorrhaging from a terrible experience in the current OS with ZERO acknowledgement of this CSAT issue.

At the very least, its tone deaf (and creating a lot of emotional responses) but past that, all of this change management, screams a design flaw; especially when hardware changes are down within the original release (more than once) without any OS changes.

It seems clear that engineering and hardware software design was a mismatch. CSAT issues aren't loud enough and marketing isn't aware of any of this.

Porsche doesnt have example of this kind of major change without a refresh to the platform, like 991.1 to 991.2.

This is pure damage control.

I think we are saying the same things about expectations of the system working as intended; my comment was more in reference to new capabilities being offered in potentially later vehicles.

As someone that runs hardware and software engineering organizations, I can tell you were are making changes to hardware components almost on a monthly basis due to ongoing supply challenges. Our hardware is not changing to accommodate a capability we want to introduce, but to ensure we can continue to ship hardware that fulfills on capabilities of previous systems.

Not saying that is what is happening right now at Porsche; but I also don't believe they are immune to supply challenges in current climate.

 

[W1NGE]

This is a really complicated topic and I'm not sure I can do it justice here.

But basically, OTA updates are a way of reducing the overhead associated with deploying software changes to cars after they've been sold. Think about it like this. If it takes 6 hours of tech time to deploy an update to your car, nobody's going to do it unless it's to fix a critical problem. It costs the automaker 6 hours of tech time, it costs the customer a day (or more) without their car, etc. And even when the decision is made to do an update, it'll tend to be as many updates as possible all at once, to hopefully get some economies of scale.

This is what we in the software engineering world call an antipattern. It's the opposite of a best practice; it's something that you specifically should not do. The safest kind of software update is a small one, one that makes only one or two changes. This is why the modern software development lifecycle is focused around making very frequent, yet very small, changes.

If OTA updates are used in this way, they can not only be safe, but safer than the current "bring your car in for a recall" approach. How long is the typical gap between the moment a defect is corrected by the manufacturer, and the moment the correction is actually applied to the median car in the wild? I'd bet it's at least a month. With OTA updates, that gap could be only a week.

The problem is that the traditional automakers aren't doing OTA updates this way. They're treating OTA updates as a way to do recalls, but without having to pay for tech time. That's... awful. And it reflects a very fundamental misunderstanding of the way software development works.

I understand why the automakers are getting this so wrong. Software development looks sort of like auto development, so it's natural to treat it the same way. But it's not the same at all. You have so much more freedom when it comes to software than you do with products that are designed to exist in the physical world. I understand why they're making the mistake, but I still don't forgive them for it. Software engineering is very much an unsolved problem, but we know a lot of things that don't work. This is one of them.

Thanks and know the rationale well - I understand (I am from a software background).

My position still holds true - OTA is not a safe medium to get safety critical updates posted and applied nor is it (likely) 100% secure. Lack of standards and cohesion. Porsche are particularly poor at it.

 

[stirthepot]

Thanks and know the rationale well - I understand (I am from a software background).

My position still holds true - OTA is not a safe medium to get safety critical updates posted and applied nor is it (likely) 100% secure. Lack of standards and cohesion. Porsche are particularly poor at it.

To each his own opinion, but I disagree. OTA can be completely safe, whether a tech loads the update or it happens OTA the net result is the same. Updated Car.

If you are questioning the testing methods prior to deployment that a different point altogether, but this notion that OTA isnt' safe has to stop, its something that the legacy auto world wants you to believe because they are behind. Heck, they said the same about electric cars, look where they are now and what they are saying.

You get updates for your computer, your phone all types of devices OTA, that doesn't make them unsafe or insecure by the fact that they are OTA.

I love the fact that these updates are happening more quickly and continually, its the way it should be IMO. It can be done safely.

 

[@lpher]

What code/letter in the VIN tells you that your car has the PAD PCM? Thx.

Sorry...it's not visible in the VIN itself... I use partslink24.com

Which is a subscription service, but i suppose there are free alternatives out there as well?

 

[Skilly]

Thanks and know the rationale well - I understand (I am from a software background).

My position still holds true - OTA is not a safe medium to get safety critical updates posted and applied nor is it (likely) 100% secure. Lack of standards and cohesion. Porsche are particularly poor at it.

I dont think both things are true.

Its true that Porsche is particularly poor at it. Lots of evidence to support that. To link it to a general statement that safety critical updates shouldn't be applied to OTA isn't true. Perhaps for Porsche, due to their current abilities; however, not generally true at all.

As an example, Tesla has been performing self driving OTA updates for years and there are watchdogs just dying to sue them for any miss steps - it hasn't happened, because the OTA updates haven't presented additional risk linked to the manner that they are delivered.

 

[@lpher]

Do you mind me asking where you got the VIN number? My car shows going into production yesterday but haven’t received a VIN. Thanks

Then you look to be in the same situation as myself... My car went into production on monday (24th) and i actually got an answer from my porsche salesrep today answering my question about the latest iteration of the PCM6.0, and more specificailly if he could confirm that my car would have it or not. He answered that he could confirm that it did not have this latest verison, and also that he could not state wheter it would be available as an update later on or not.

 

[Reg]

If nothing was promised, there is no legal basis. It’s just terrible customer service.

agree that not fixing bugs is terrible service

but these seems more like new features, like android auto and spotify integration and colored icons.

 

[Reg]

since this is not (yet, or at all) backward compatible, this was a very poorly thought out communication.

there was no reason for a release at all, since i doubt they are demand constrained, nor would spotify or android or some other things mention, be decidings factor in purchasing a car at this price.

all that it it did was alienate some existing owners.

 

[Jonmar]

Then you look to be in the same situation as myself... My car went into production on monday (24th) and i actually got an answer from my porshce salesrep today answering my question about the latest iteration of the PCM6.0, and more specificailly if he could confirm that my car would have it or not. He answered that he could confirm that it did not have this latest verison, and also that he could not state wheter it would be available as an update later on or not.

Thank you for letting me know. I’ll see what I can find out from my salesperson

 

[W1NGE]

To each his own opinion, but I disagree. OTA can be completely safe, whether a tech loads the update or it happens OTA the net result is the same. Updated Car.

If you are questioning the testing methods prior to deployment that a different point altogether, but this notion that OTA isnt' safe has to stop, its something that the legacy auto world wants you to believe because they are behind. Heck, they said the same about electric cars, look where they are now and what they are saying.

You get updates for your computer, your phone all types of devices OTA, that doesn't make them unsafe or insecure by the fact that they are OTA.

I love the fact that these updates are happening more quickly and continually, its the way it should be IMO. It can be done safely.

I would expect dealer / garage applied updates to be road tested rather than by the owner.

Case in point - Porsche has not pushed anything of substance OTA. I wonder what their rationale is.

 

[Scandinavian]

I would expect dealer / garage applied updates to be road tested rather than by the owner.

Case in point - Porsche has not pushed anything of substance OTA. I wonder what their rationale is.

Probably they never will. Having the architecture that Porsche has in their cars, it would be nearly impossible to do so. Over 100 different computers that all need to be addressed individually? How would you ever get that to sync properly?

 

[KensingtonPark]

While I agree that there is certainly no obligation on the part of Porsche to add additional functionality/features to existing Taycan PCMs, I will note that many "lesser" car manufacturers update both UI and functionality on their infotainment systems for much older model cars. My son's Hyundai Elantra GT had an update that I downloaded to an SD card, inserted in vehicle, ran an update process and - voila! - a new higher resolution and more modern interface that is vastly different, updated with additional (albeit minor) functionality. While not "OTA" (since the car has no way to do that), it was easy to do and quite impressive in its own way, particularly in light of the discussion that I'm seeing here.

 

[Visie]

soms info from pff.de forum:

Just got off the phone with "my" technology consultant. He has been trying to get some information about the PCM update in the last 24 hours, but the communication from Porsche seems to be a bit "difficult".

So what he found out:

• All vehicles that were built from week 3/22 (i.e. last week) already have the update on them
• Vehicles of the MY2022 will probably get the update (Porsche didn't want to officially confirm it, but probably let it be known). However, it is currently unknown when the update will be available. However, the hardware is probably identical in the "old" MJ22 and those built from last week.
• Whether vehicles from MY2020/21 will receive the update is currently being checked internally at Porsche. What is certain, however, is that if it comes, not all new functions will come (e.g. no Android Auto), since the hardware differs between MJ20/21 and MJ22. Overall, however, he was less optimistic about the update for MY20/21.

As soon as he has more information, he will get back to you. So the bottom line is that we have to keep waiting

Fingers crossed for MY22

Sponsored