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

Firmware 7.0 Beta Discussion

This site may earn commission on affiliate links.
I'm going to point out again, since my other post probably got buried, that I'm pretty sure the UI overhaul is more of a performance improvement requirement vs a visual overhaul. The removal of all of the unnecessary brushed metal texturing probably drops CPU/GPU usage down significantly, which would improve performance significantly. The car is doing a lot more behind the scenes these days. Not just autopilot cars. I'm betting the new "range assurance" stuff is eating up quite a bit of background CPU time, along with the added overhead of everything the nav system does for trips now with the superchargers and supposedly polling them frequently. I've noticed a significant decrease in responsiveness of both the 17" and the IC since 6.x. The scroll wheel menus are so laggy nowadays it's ridiculous, and many touch responses on the 17" are delayed by at least 750ms or more.

Basically I believe the UI overhaul was to soften the load on the processors. The textured old look was probably fine when the IC was lower resolution and basically did nothing special and the 17" just had some basic functionality. But they've added so much they're going to have to cut back somewhere eventually to preserve responsiveness.
 
Hi, I wrote the article in question. Can you point me to anywhere "here or elsewhere" where it confirms the autopilot features enabled by v7.0? Because, when I have a publicly available source, I link it...

Who cares about linking to publicly available sources? The interesting news comes from obtaining, and if appropriate naming, information from sources that are previously not publicly available. In this regard, you state: "Sources familiar with Tesla’s new firmware...". I guess that means you won't list your source? So that really doesn't mean anything to me because I have heard that phrase used many times and unnamed sources have often proved to be wrong (and I often wonder if they even exist). The pictures you used are not new and thus not from your "source". In fact, there is no information in your article that is new. You claim that you "confirm" that autopilot will be part of 7.0 but, again, you don't name the source. We all know 7.0 will be autopilot so if it turns out to be true, that doesn't mean anything to me about your "source".
 
@wk057 - I have not noticed any performance or responsiveness decline on my "classic" MS. But if the flatter UI really cuts down on CPU/GPU usage then I'm all for it. Although if it is accompanied by a messy IC interface that removes a bunch of functionality then that's a whole different ball game.

To Tesla software devs:

Put the power and range displays back where they belong - front and center. These I consider very key and non-negotiable especially for an EV.

Get rid of that god awful clock that takes up an entire half of the IC and reduces what useful data can be displayed.

Put the ambient temp displays back - both on the IC and MCU.
 
Who cares about linking to publicly available sources? The interesting news comes from obtaining, and if appropriate naming, information from sources that are previously not publicly available. In this regard, you state: "Sources familiar with Tesla’s new firmware...". I guess that means you won't list your source? So that really doesn't mean anything to me because I have heard that phrase used many times and unnamed sources have often proved to be wrong (and I often wonder if they even exist). The pictures you used are not new and thus not from your "source". In fact, there is no information in your article that is new. You claim that you "confirm" that autopilot will be part of 7.0 but, again, you don't name the source. We all know 7.0 will be autopilot so if it turns out to be true, that doesn't mean anything to me about your "source".

Maybe I didn't phrase this right.

Which particular autopilot features are going to be enable by v.7.0?

Are you saying that you knew exactly what autopilot features are going to be in v7.0 and which ones will not be? Like lane changing through tapping the turn signal, the level of self-parking, etc. I mean lane keeping/auto-steer was a given, but the rest?

- - - Updated - - -

Who cares about linking to publicly available sources? The interesting news comes from obtaining, and if appropriate naming, information from sources that are previously not publicly available. In this regard, you state: "Sources familiar with Tesla’s new firmware...". I guess that means you won't list your source? So that really doesn't mean anything to me because I have heard that phrase used many times and unnamed sources have often proved to be wrong (and I often wonder if they even exist). The pictures you used are not new and thus not from your "source". In fact, there is no information in your article that is new. You claim that you "confirm" that autopilot will be part of 7.0 but, again, you don't name the source. We all know 7.0 will be autopilot so if it turns out to be true, that doesn't mean anything to me about your "source".

Oh and also.

Who cares about linking to publicly available sources?

I do. It's just a simple courtesy.
 
Also @wk057 -- you mention all the texturing and other "brushed metal" elements -- most all of those are likely static bitmaps that are displayed once and not redrawn unless the screen changes state. The dynamic things that actually change (on 6.2) are the speed and power lines -- those are just gradient color arcs. Also the text based numbers change (speed, trip odos, battery, time, temp, etc). I haven't noticed any laggyness on my car, everything is pretty snappy. Also, I think there are different processors handling the IC display and the nav/range assurance stuff on the main screen. So I really don't think performance is the leading issue causing the re-design.

But if what you propose is true, they could have just removed all the texturing and other elements WITHOUT also removing or relocating major portions of critical functions.


 
Well thank you. That's what I was looking for in regards to the original claim. Although your are assuming that I've read this which I haven't.

and I maintain that there's new information in my article.

Sorry if I sound skeptical of your "source". Maybe it has to do with another thread here when I repeatedly pressed someone for his Tesla source and it ended up being that he lived close to the factory...

Dual chargers - Page 5

I'm sure your source is better. I wonder why anyone at Tesla with inside information would risk their job, and serious legal ramifications, for leaking information to you, but that's another issue I guess.
 
It's amazing to see just how far we've coming in 6 years.

Searching around I found this image from this thread from 2009: Model S Dashboard and console controls

3433445696_6e29d357ba_b.jpg
 
I haven't either.
@wk057, what are you referring to when you mention superchargers and say "polling them frequently"?

Elon's comments for the release of the range assurance feature indicated that the car would be polling for the status of nearby superchargers constantly.

I've definitely noticed decreased performance on the P85D coming from my P85 and since I took delivery. The IC side panels/scroll wheels are mostly unusable they're so laggy. The 17" is also pretty slow most of the time.

My wife's P85 isn't nearly as bad (non-autopilot and has the lower resolution IC), but we've both noticed it isn't nearly as snappy as it was on day 1.
 
I think this is another instance of Elon being far ahead of the curve. I have seen no evidence that the Nav polls SpC status in real time to factor into the est. charge time.

I've definitely seen increased data activity since 6.2 over Tesla's VPN to the car. Pre-LTE update I was occasionally using an LTE wifi tether, and logging pcap data, when I would be in an area with spotty coverage. Definitely more activity on the VPN after 6.2, and given the frequency of some of the bursts of data I'd be willing to guess there is some polling going on, likely the car just pulling a status file from Tesla or something, but no way to know for sure unless I hack up my car for VPN keys to MITM the info... which I don't plan on doing for at least another 7 years and change. lol.
 
@wk057 - I have not noticed any performance or responsiveness decline on my "classic" MS. But if the flatter UI really cuts down on CPU/GPU usage then I'm all for it. Although if it is accompanied by a messy IC interface that removes a bunch of functionality then that's a whole different ball game.

For what it's worth, I have. I've noticed a bit of "schizophrenic" behavior on the energy graph during longer trips. Take a look at this video:


Start watching the energy meter at 0:20. Also note that despite me driving at 75 mph, the "since last charge" doesn't change from 103.3 miles. In fact, it didn't change for well over 15 minutes, then started jumping by a few minutes until it (mostly) caught up, although it still lost some mileage.

When it does this, responsiveness would go away.

(Video taken by my wife, in car with me. Sorry for the shakiness, she's balancing the phone on the steering wheel.)
 
Last edited by a moderator:
For what it's worth, I have. I've noticed a bit of "schizophrenic" behavior on the energy graph during longer trips. Take a look at this video:


Start watching the energy meter at 0:20. Also note that despite me driving at 75 mph, the "since last charge" doesn't change from 103.3 miles. In fact, it didn't change for well over 15 minutes, then started jumping by a few minutes until it (mostly) caught up, although it still lost some mileage.

When it does this, responsiveness would go away.

(Video taken by my wife, in car with me. Sorry for the shakiness, she's balancing the phone on the steering wheel.)

Yep, noticed this as well. On the P85D it's been worse though. I've seen the speedo and energy bar "stick" for several seconds at times, then quickly play catch up. It's pretty bad sometimes. :(
 
Last edited by a moderator:
Yep, noticed this as well. On the P85D it's been worse though. I've seen the speedo and energy bar "stick" for several seconds at times, then quickly play catch up. It's pretty bad sometimes. :(

I've found that it only tends to happen when miles since last charge is > 100 or so, and I'm on a trip on the Interstate. I don't know if it's the navigation system sucking up all the CPU, or something else. During this time, the odometer still works perfectly and the rated range decreases. When the "since last charge" does catch up, it's clear that it's missing some data, as the Wh/mile can be 280 and yet rated range + miles since last charge is significantly reduced.