Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Model S/X deliveries with Intel-based MCU

This site may earn commission on affiliate links.
Hope MCU2-Intel does not use too much battery, otherwise folks will complain and want to replace it back with MCU1-Tegra.
The battery impact of the computers — all of them — is negligible compared to the environmental controls and the motors. You could drive the car with the MCU and Autopilot computers turned off entirely and you'd barely notice a difference in your Wh/mile consumption.

Both the Tegra 3 and Intel Atom we designed as tablet processors (perfect for use in a big touch screen interface), where power consumption is a key design factor. I'm not sure exactly which Atom processor is being used in the MCU2, but there's a good chance it's actually lower wattage than the seven-year-old Tegra 3, even though it performs far better.
 
  • Like
Reactions: HankLloydRight
For those who are lucky enough and have the MCU2, are you still running on 2018.6.50 or have you received and update to 2018.10.4? If I have understood it correctly guys who have MCU2 are running Vector-maps and are able to load web pages like teslainventory.com. I am curious if this is because the are running on 2018.6.50 and if they upgrade to 2018.10.4 they will not be able to load those webpages and will loose vector maps...

I cannot shake my suspicion that MCU1 will not get vector maps or new browser.... because why have those with MCU2 vector maps and not those with MCU1? if the argument regarding the same code can be used on Tegra, why is vector maps not a part of the lates release if they are willing to ship new cars with vector maps?

Am I missing something?
 
I cannot shake my suspicion that MCU1 will not get vector maps or new browser.... because why have those with MCU2 vector maps and not those with MCU1? if the argument regarding the same code can be used on Tegra, why is vector maps not a part of the lates release if they are willing to ship new cars with vector maps?

Am I missing something?
@verygreen discovered that vector maps could be enabled and used on the Model S/X with the old MCU, before the new MCU appeared. This demonstrates that there is nothing about the old MCU which would prevent the use of vector maps.
 
  • Like
Reactions: jrad6515
For those who are lucky enough and have the MCU2, are you still running on 2018.6.50 or have you received and update to 2018.10.4? If I have understood it correctly guys who have MCU2 are running Vector-maps and are able to load web pages like teslainventory.com. I am curious if this is because the are running on 2018.6.50 and if they upgrade to 2018.10.4 they will not be able to load those webpages and will loose vector maps...

I cannot shake my suspicion that MCU1 will not get vector maps or new browser.... because why have those with MCU2 vector maps and not those with MCU1? if the argument regarding the same code can be used on Tegra, why is vector maps not a part of the lates release if they are willing to ship new cars with vector maps?

Am I missing something?

MCU1 will get higher definition vector maps and an updated navigation system with better route planning, etc. Period.
MCU1 will probably also get an updated browser and MCU1/MCU2 will follow the exact same feature path for a long time. Only difference will be MCU2 being a bit snappier and faster.

I posted this link and mentioned the same thing a few pages back in this thread. But here is the link again:

Vector Maps incoming?
 
At least with the model S, I’m now speculating that the MCU roll out has to do with geography (and thus production dates). I have yet to see anyone east of Texas with the new MCU. Not sure where the line was going north and south, but WI and MN seem to be all old MCU. Will be interesting to see if any IL or other states south have new or old ones.

Also might explain how Dec orders from Cali are now being produced
 
At least with the model S, I’m now speculating that the MCU roll out has to do with geography (and thus production dates). I have yet to see anyone east of Texas with the new MCU. Not sure where the line was going north and south, but WI and MN seem to be all old MCU. Will be interesting to see if any IL or other states south have new or old ones.

Also might explain how Dec orders from Cali are now being produced


Seems like it’s production date and, since the East Coast people have to wait for delivery, their production date was earlier. More timing than geography.

I’m San Diego and ordered on 1/28, production 3/14 and delivery will be 3/22 with the new MCU.
 
  • Like
Reactions: jrad6515
Supposedly, my Model S went into production the around March 5th or so on a canceled preproduction inventory model I ordered the end of February. It finished manufacturing by the 12th and will be delivered tomorrow in NJ, and I receive it Wednesday, the 21st. Vin is 247XXX. Am I out of luck since it was pre-production inventory or will I have a shot at the new MCU?
 
AP1 improvements didn’t stop with Ap2 roll out.
All the software updates for pre-AP and AP1 cars isn’t the latest hardware when they got furthe software update support.

And the hardware improvements upgrades are a very obvious kind of support to the legacy hardware since you can only do so much with software support.

Again you couldn’t be wrong to claim that Tesla doesn’t support legacy hardware when they do more hardware and software support of legacy hardware than any car mfr, and really any other mfr of any kind.

Again, “Simply: I think Tesla will spend a lot less time on the Tegra MCU now.” You’re free to disagree.
 
  • Like
Reactions: Matias
So, at this point, we have at least three areas where MCU falls behind MCU2, software-wise:

- spatial enhancement of audio (aka iTunes "Sound Enhancer")
- vector maps?
- navigation flyover view
- web browser (unusable)

And hardware wise:
- 5 Ghz WiFi

Can any owners of the new MCU add to this list? Can we maintain this on a wiki?

Can you elaborate on the 'Navigation Fly over view' ?

Is that something new or you're saying it's just much faster on the new MCU?

Can someone post a video?
 
Simply: I think Tesla will spend a lot less time on the Tegra MCU now.

Sure, but that's a far cry from the straw man arguments in your prior posts

"They tend to only include improvements that will also benefit the latest hardware." I’m talking about continued software support for previous hardware versions"

They may not so it quickly but they haven't stopped supporting ANY hardware yet. Not even the roadster.

You spent several posts making it sound like they just completely drop support on old hardware by focusing on the lack of significant auotpilot improvement during a time where a vendor pulled supply and support without mentioning that vendor issue.

Tesla has little history of supporting previous hardware. HW1 was supposed to have on-ramp to off-ramp, traffic signal and stop sign recognition... it hasn’t seen a significant upgrade in about a year.

If you want to walk back from that line you can, but you went there.
 
Last edited:
Again, “Simply: I think Tesla will spend a lot less time on the Tegra MCU now.” You’re free to disagree.

Less time? I mean, the MCU will look the same and have the same features for MCU1/MCU2 for many years ahead. They will simply just not split them out and offer anything specific for MCU2. The only difference will be speed, that's it. The functionality/UX/UI team work independent of chipset.

Just out of curiosity, what type of features do you think they will develop specifically for the MCU2 and WHY? I just don't see it coming at all to be honest.... My strong prediction is that we won't see any differences in features until the next gen Model S/X.

Time will tell :)
 
  • Like
Reactions: techmaven
So, at this point, we have at least three areas where MCU falls behind MCU2, software-wise:

- spatial enhancement of audio (aka iTunes "Sound Enhancer")
- vector maps?
- navigation flyover view
- web browser (unusable)

And hardware wise:
- 5 Ghz WiFi

Can any owners of the new MCU add to this list? Can we maintain this on a wiki?
Aside from the 5Ghz Wi-Fi, none of this is necessarily dependent on the MCU2 processor. Vectors maps and the new instrument cluster map display have both been demonstrated on the Tegra 3-powered MCU1, the web browser sucks ON BOTH because it's very poorly coded, not the power of the CPU.
 
Supposedly, my Model S went into production the around March 5th or so on a canceled preproduction inventory model I ordered the end of February. It finished manufacturing by the 12th and will be delivered tomorrow in NJ, and I receive it Wednesday, the 21st. Vin is 247XXX. Am I out of luck since it was pre-production inventory or will I have a shot at the new MCU?

My VIN is the same and I take delivery the same day. I think my production date was the same or day or two earlier. I have been told that I will have the new MCU so your chances are good.
 
Again, “Simply: I think Tesla will spend a lot less time on the Tegra MCU now.” You’re free to disagree.

As @evster noted, plenty of Tesla programmers will be working on enhancements that are at levels well above the difference between the new MCU and the old one. It'll be high level code... for example routing and navigation enhancements that are processor independent. These days it is common across a number of platforms to write code that gracefully handles varying hardware capability and performance.

At this point, my P85 is old enough that firmware revisions are usually not particularly interesting. I don't have any of the new hardware to take advantage of most of the changes going on these days. However, I still got enhancements to the backup camera (the lines finally), Supercharging near real time usage and the easy entry-exit. The #1 issue now is still the navigation and route planning software and I expect the ongoing enhancements for AP are going to help in that regard. We should be able the share the same POIs, real time traffic, and eventually enhanced awareness of routes and individual lanes within a route (HOV versus not, toll versus not, etc).
 
  • Like
Reactions: MarkS22
Less time? I mean, the MCU will look the same and have the same features for MCU1/MCU2 for many years ahead. They will simply just not split them out and offer anything specific for MCU2. The only difference will be speed, that's it. The functionality/UX/UI team work independent of chipset.

Just out of curiosity, what type of features do you think they will develop specifically for the MCU2 and WHY? I just don't see it coming at all to be honest.... My strong prediction is that we won't see any differences in features until the next gen Model S/X.

Time will tell :)

One thing to keep in mind is that it isn't just a different processor but different wiring harness as well. This could be important because they could release a front camera, 360 cameras, etc which would be an easy retrofit for cars with the new MCU/wiring but not possible for old cars.
 
Aside from the 5Ghz Wi-Fi, none of this is necessarily dependent on the MCU2 processor. Vectors maps and the new instrument cluster map display have both been demonstrated on the Tegra 3-powered MCU1, the web browser sucks ON BOTH because it's very poorly coded, not the power of the CPU.

I agree, the browser is poor due to poor coding, not processor speed. It seems that it was a rush port job by a junior programmer, and maybe even not that. It looks like a “demo ware” browser that a chip manufacturer (tegra) might shove out the door to demonstrate that, yes, you can theoretically run a browser on the chip. For instance, why is there the ability to select text when you can’t copy and paste? It is too bad that Tesla doesn’t allow third party apps, because our community could code a much better browser.

Actually, has anyone run a browser capabilities script on the new MCU to see if has different functionality?
 
Less time? I mean, the MCU will look the same and have the same features for MCU1/MCU2 for many years ahead. They will simply just not split them out and offer anything specific for MCU2. The only difference will be speed, that's it. The functionality/UX/UI team work independent of chipset.

Just out of curiosity, what type of features do you think they will develop specifically for the MCU2 and WHY? I just don't see it coming at all to be honest.... My strong prediction is that we won't see any differences in features until the next gen Model S/X.

Time will tell :)

If we're speculating, I'd say new features might include: the ability to show more robust camera views, like a 180 degree stitched view of multiple cameras. Or it could mean a streaming video player when parked (using HVC/H.264). Or enhanced sound processing, like we see on the Model 3. Or screen mirroring of your phone. Or a larger assortment of apps, like weather, stocks, or voice-activated text messaging. It could mean safety features, like Pre-Safe Sound from Mercedes which plays an 85db white noise immediately before an accident to protect the ears. It could mean better situational awareness with more accurate depictions of the vehicles on Autopilot's display that are 3D models instead of bitmaps that only have two angles.

I mean, that's just me thinking of a few things in three minutes. There are teams of people thinking of new ways to use the processing power available to them. My question back would be... why wouldn't they use the extra processing power? :)