Suggestion to make it easier to deal with variable effective range

[geosynch]

Hmmm. So, starting at say 9900' current elevation, highest point 10000', and destination elevation 0' (you didn't mention destination), and the reverse trip 0->10000->9900, you'd use the same energy? I'd be surprised, but then again you know rocket launches. I'd probably gone for an integral calculation over the road elevation changes for the trip. And with the weight of the car being roughly 6000 lbs, wouldn't four adults plus luggage make for more than a 10% difference in mass? I was thinking, though, that in many of the previous discussions on this topic, people have reported that wind speed and direction, road conditions, use of HVAC played a significant factor. Oh, and tires and tire air pressure and effect on rolling resistance. Seemed to be more of a multivariate problem to me.
As previously mentioned, the ABetterRoutePlanner site seems to do its calculations on range based on many of these factors and include battery degradation. As previously discussed, it would be great to have more accurate and dynamic range information available for road trips. But for me, the more significant concern is around being able to find working and available charging infrastructure along my long distance route such that I can travel more miles in a day, spend less time cursing about people who need to charge to 100% or leave their cars fully charged and taking up a charging spot or finding out of service units. Now if you have a nice algorithm in mind to handle that aspect of range planning, send me the code and I'll do a code review.

If you want to be pedantic, I can do that.

Wind speed, tire pressure, roof racks, if the car has been washed, if the widows are rolled down, if I'm listening the The Wall at 11, whatever you want to throw at the problem, save change in elevation, is going to be resolved by the current efficiency calculation. is it 3 mi/kWh? or 4? Don't care. The car already resolves this and it has been shown to be pretty accurate. Because energy is a scalar, I can simply add the quantities for electrical, potential and kinetic energy together. Kinetic energy is by virtue of the car moving through a fluid. The car's estimate is presently summing the kinetic and electrical energy in its efficiency calculations, but what it lacks is an estimate of the potential energy.

Estimating mass is possible, but probably little more than a noise generator. But the car definitely knows if seats are occupied, etc. Are there sensors on the shocks to estimate the load? I confess I don't know. I could find out, but you know what, this has already gotten silly.

Before ABRP was ever a thing I made a spreadsheet to perform these calculations and using the origin, peak and destination elevations got me within a mile for the drive from SF to Tahoe, assuming 80% recaptured on the downhill portions. As you suggest, a piecewise integration is an option and I'm sure that's what ABRP uses, but it may be driving in a thumbtack with a sledgehammer, too.

Honestly, I wouldn't care if all EV manufacturers would give up on range estimates and just use ABRP if I had access to it in the cockpit while driving. But you know.... ABRP doesn't include the Touring or the Pure as selections, just the GT. And Lucid doesn't support Android Auto, so, I'm SOL unless I get a new phone.

As for coding, I confess I haven't actually coded since FORTRAN 77 was the preferred solution for such problems. I forgot to use the sarcasm font in my original post.

This forum is starting to feel al ot like the Tesla forums. So many fanbois that are unable to critical look at the car they are driving and create a cult of personality.

I'm frankly disappointed in the direction this took, but I stand by my current statement. A range estimate the is void of the topography is useless for any driving scenario with significant altitude changes. Ya'll defending otherwise are kool-aid drinkers.

 

[Amster]

I have a basic question on this. How do you switch

Yes, I mean the native Lucid nav. That is accurate for arrival SOC%.

The display of percentage / miles left *at present* when not using native nav is still based on EPA.

I also use Waze for its warnings, but switch to the native nav if I am running low and really need to know if I’ll make it somewhere, or if I’m going to a charger, since it will now automatically precondition.

I have a question on this. How do you switch the display from miles left to SOC%?

 

[MHDave]

So many fanbois that are unable to critical look at the car they are driving and create a cult of personality.

Oh, I fully agree as I have stated in the past that the calculations could certainly use some sophistication. I just didn't agree with the suggestion that the calculation be simplified to m*g*h, and your elaborated response seems to agree.
As for the critical look, I think if you were to look at the list of suggested improvements that were assembled from a poll led by one of the moderators, you'd see lots of critical looks at features people on the forum think are important, many of them expressed as clear misses in the car's design. Some of these have been actively addressed by Lucid, others are patiently or impatiently being waited on. Including better range calculations. But Joe's point is valid -- not knowing the priorities and constraints and costs and components and restrictions that go into Lucid's feature development, we're just armchair quarterbacks to say they're all wrong and just don't get the basics. Perhaps we're right on some points, likely we're dummies ourselves on many others. I give them and you the benefit of doubt. The car's not perfect, but I'm happy that it's evolving and taking into consideration things that you and I and others observe to be potential improvements.
Cheers

 

[Bobby]

I have a basic question on this. How do you switch

I have a question on this. How do you switch the display from miles left to SOC%?

Pilot panel, displays, units (I think)

 

[joec]

This forum is starting to feel al ot like the Tesla forums. So many fanbois that are unable to critical look at the car they are driving and create a cult of personality.

I'm frankly disappointed in the direction this took, but I stand by my current statement. A range estimate the is void of the topography is useless for any driving scenario with significant altitude changes. Ya'll defending otherwise are kool-aid drinkers.

Don't call people names or make personal accusations about others. Read the guidelines. Final warning.

There's no reason not to be civl.

To my earlier point (and later @MHDave's), you are making giant assumptions with little knowledge to back them up. Another thing we try to discourage here. This—not the fact that you are criticizing the car—is what's causing the push back.

Criticism itself is certainly welcome. As is clearly proven by the content of this site. Do your own analysis of the vast majority of threads here, and you'll find far more negative comments about Lucid than positive. Your assertion that this place is full of "kool-aid drinkers" has no basis in fact.

Trust me. I have to read every post.

As someone who lives amongst the mountains, I have as much vested interest in your "feature request" as anyone. I'm just trying to be realistic about when that particular request might get addressed, as opposed to the hundreds of other things customers want at the moment. Also, I can do quick math in my head and figure out when I need to charge fairly easily. So I don't personally consider it a big deal. You likely disagree, and that's okay.

 

[Bunnylebowski]

If you want to be pedantic, I can do that.

Wind speed, tire pressure, roof racks, if the car has been washed, if the widows are rolled down, if I'm listening the The Wall at 11, whatever you want to throw at the problem, save change in elevation, is going to be resolved by the current efficiency calculation. is it 3 mi/kWh? or 4? Don't care. The car already resolves this and it has been shown to be pretty accurate. Because energy is a scalar, I can simply add the quantities for electrical, potential and kinetic energy together. Kinetic energy is by virtue of the car moving through a fluid. The car's estimate is presently summing the kinetic and electrical energy in its efficiency calculations, but what it lacks is an estimate of the potential energy.

Estimating mass is possible, but probably little more than a noise generator. But the car definitely knows if seats are occupied, etc. Are there sensors on the shocks to estimate the load? I confess I don't know. I could find out, but you know what, this has already gotten silly.

Before ABRP was ever a thing I made a spreadsheet to perform these calculations and using the origin, peak and destination elevations got me within a mile for the drive from SF to Tahoe, assuming 80% recaptured on the downhill portions. As you suggest, a piecewise integration is an option and I'm sure that's what ABRP uses, but it may be driving in a thumbtack with a sledgehammer, too.

Honestly, I wouldn't care if all EV manufacturers would give up on range estimates and just use ABRP if I had access to it in the cockpit while driving. But you know.... ABRP doesn't include the Touring or the Pure as selections, just the GT. And Lucid doesn't support Android Auto, so, I'm SOL unless I get a new phone.

As for coding, I confess I haven't actually coded since FORTRAN 77 was the preferred solution for such problems. I forgot to use the sarcasm font in my original post.

This forum is starting to feel al ot like the Tesla forums. So many fanbois that are unable to critical look at the car they are driving and create a cult of personality.

I'm frankly disappointed in the direction this took, but I stand by my current statement. A range estimate the is void of the topography is useless for any driving scenario with significant altitude changes. Ya'll defending otherwise are kool-aid drinkers.

There’s a difference between being a FanBoi and giving benefit of the doubt, that’s all we are asking here. They’re not morons, they likely know the intricacies involved, and so do you. Given the state of the software from 1 year ago as compared to now, I’m happy with it as is, expectations are managed, but am looking forward to what they come up with next as there is still, as everyone is aware, room for improvement.

In my limited experience, the power required to get this very heavy vehicle up to speed on an uphill on-ramp heavily impacts efficiency, like you’ll see that one on-ramp still having affected the miles/kWh after 100 miles of driving. So micro-elevation changes affect it just as much as what ABRP calculates. How come ABRP can’t figure out the on-ramps, what are they idiots with the software? No of course not, I’ll give them the benefit of the doubt that there’s a lot of variables to take into account and I don’t expect to have them all mastered, and Lucid is in the process of implementing more things with the software every day, and that’s just fine.

 

[Bobby]

There’s a difference between being a FanBoi and giving benefit of the doubt, that’s all we are asking here. They’re not morons, they likely know the intricacies involved, and so do you. Given the state of the software from 1 year ago as compared to now, I’m happy with it as is, expectations are managed, but am looking forward to what they come up with next as there is still, as everyone is aware, room for improvement.

In my limited experience, the power required to get this very heavy vehicle up to speed on an uphill on-ramp heavily impacts efficiency, like you’ll see that one on-ramp still having affected the miles/kWh after 100 miles of driving. So micro-elevation changes affect it just as much as what ABRP calculates. How come ABRP can’t figure out the on-ramps, what are they idiots with the software? No of course not, I’ll give them the benefit of the doubt that there’s a lot of variables to take into account and I don’t expect to have them all mastered, and Lucid is in the process of implementing more things with the software every day, and that’s just fine.

I still think that we are over complicating a simple calculation. Of course elevation, ambient temperature, accessory usage, etc. all come in to play. But I think the last 15 (or whatever) miles of driving history are the best prognosticator of range because it takes all that into account, unless you are in an area with significant elevation and temperature changes.… in that case, the car should provide the ability to reset it if the driver feels that the recent XX miles are not representative of what is upcoming.

 

[Spin Doctor]

I confess I did really like the energy / range graph in my previous vehicle. But the superior range of the Lucid makes it less of an issue I guess.

 

[Sandvinsd]

I still think that we are over complicating a simple calculation. Of course elevation, ambient temperature, accessory usage, etc. all come in to play. But I think the last 15 (or whatever) miles of driving history are the best prognosticator of range because it takes all that into account, unless you are in an area with significant elevation and temperature changes.… in that case, the car should provide the ability to reset it if the driver feels that the recent XX miles are not representative of what is upcoming.

Yes. A simple owner inputted toggle to.provide range estimates for the last so many miles, since last charge, lifetime, etc would provide much better estimates than using the current EPA efficiency.

 

[Gil]

Yes, I mean the native Lucid nav. That is accurate for arrival SOC%.

The display of percentage / miles left *at present* when not using native nav is still based on EPA.

I also use Waze for its warnings, but switch to the native nav if I am running low and really need to know if I’ll make it somewhere, or if I’m going to a charger, since it will now automatically precondition.

I agree that “expected miles on arrival” is amazingly accurate.

How can we explain the difference between this excellent computed number and the overall number (outside of the navigation system) for miles remaining—which we all know is just an optimistic starting point. I hope it’s not connected to this rather savage exposé of Tesla: https://www.reuters.com/investigates/special-report/tesla-batteries-range/

 

[borski]

I agree that “expected miles on arrival” is amazingly accurate.

How can we explain the difference between this excellent computed number and the overall number (outside of the navigation system) for miles remaining—which we all know is just an optimistic starting point. I hope it’s not connected to this rather savage exposé of Tesla: https://www.reuters.com/investigates/special-report/tesla-batteries-range/

Not connected in any way; the Lucid’s range calculation doesn’t change based on anything. It is literally just EPA times current battery percentage. I suspect they see it as a representation of the current state of the battery, rather then as a predictive tool, and that’s why it’s not using the predictive algorithm. Also, without navigation in place, it has no idea where you’re trying to go, and that may greatly affect range.

 

[Amster]

I agree that “expected miles on arrival” is amazingly accurate.

How can we explain the difference between this excellent computed number and the overall number (outside of the navigation system) for miles remaining—which we all know is just an optimistic starting point. I hope it’s not connected to this rather savage exposé of Tesla: https://www.reuters.com/investigates/special-report/tesla-batteries-range/

I haven't really paid a lot of attention to this, because for local driving the car has such a huge range. But tonight, I noticed that the expected miles on arrival was dead on accurate for a short trip. I had 295 in the parking lot at Vons. I put in Home as the destination. This is only 1 3/4 miles in actual distance, but it's mostly uphill, some of it pretty steep. It said I'd have 289 when I got home, a 6 mile loss, and that is exactly what I had. So the computation had to be based on the elevation change between the two locations.

 

[Genghis]

This forum is starting to feel al ot like the Tesla forums. So many fanbois that are unable to critical look at the car they are driving and create a cult of personality.

I'm frankly disappointed in the direction this took, but I stand by my current statement. A range estimate the is void of the topography is useless for any driving scenario with significant altitude changes. Ya'll defending otherwise are kool-aid drinkers.

Many of the owners here believe that the Air is absolutely the best car they have owned and/or driven so it may seem very fan(boy/girl)ish. However, we all have our gripes no matter how much we love the car, as it's nowhere near perfect, especially the software. Not much more you can do on the hardware based on the trim and options you bought, but the software will improve to make the owning experience better.

Regarding the topic on hand, it's the there are many ways to skin a cat problem, and everyone is going to have different thoughts on how to calculate the actual mileage but it is a complex problem like skinning a cat and some manufacturers have done a better job. I just don't trust any software calculated method right now as there are so many variables that affect energy consumption and battery drain, so I go with just SOC% keeping it 20-80 as most suggest and at extremes 10-100 when I need. I think that's the safest way for now so you never get caught in a bad situation.

 

[thecodingart]

Before I retired, I was the program manager for a NASA spacecraft. And not only did I have an entire software team in my org, some of them work now at Lucid. I happen to know exactly what the situation is at Lucid.

Not once did I say I could do it in an afternoon. I think you need to take your condescending tone down a notch and give the guy on the street a bit of credit.

I’d reverse the comments here and ground yourself in a bit of reality. I have plenty of people I’ve mentored or had oversight on working for the biggest tech companies out there in high and low positions. Does that mean I can hop in their tech stack and effectively change whatever I want in a day, no.

Ignorant comments like this display a lack of understanding, they dont solidify a decorated resume. It’s very akin to what GeoHotz was claiming at Twitter (which is not something anyone should look to replicate).

As a software engineer myself reading comments from someone else who was supposedly in software, these comments are a bit insulting and absolutely rub me the wrong way.

With that being said, Lucid needs to fix their range estimates.

 

[borski]

It’s very akin to what GeoHotz was claiming at Twitter

LOL. I’ve spent a lot of time with geohot playing CTF. He’s a very very smart guy, just also incredibly full of himself and self-assured. He’s really fun to mess with, tho - if we were meet in person, ask me about how we stole his seekingarrangement date in Korea. Great times.

 

[ken43]

It does this extremely well when you are navigating somewhere. The “X miles on arrival” works very very well.

The issue is if you aren’t navigating it doesn’t use the same algorithm. I imagine that’s an oversight (or hope it is) and they’ll fix it.

I'm not sure I understand what you are looking at when you say the

That navigation predicted SOC% on arrival is what I'm talking about being within 1 mile. The general range is very off, and still based on EPA.

I'm not sure I understand what you are looking at when you say the SOC% is within 1 mile. I just finished a 1200-mile roundtrip between OH and NY, and really couldn't trust that the 'X miles on arrival' on the left upper panel was accurate. Here is a screenshot at the completion of one leg. On this leg I had an average usage of 3.8mi/KW, and the left display indicated 60 miles on arrival, while the SOC% was 25%. If I do the math: 25% x 112KW battery (AGT) = 28KW remaining. 28 x 3.8 mi/KW = 106 range remaining. Am I doing something wrong in deciphering the info? How do you interpret that SOC% is within 1 mile?

 

[Adnillien]

I'm not sure I understand what you are looking at when you say the

I'm not sure I understand what you are looking at when you say the SOC% is within 1 mile. I just finished a 1200-mile roundtrip between OH and NY, and really couldn't trust that the 'X miles on arrival' on the left upper panel was accurate. Here is a screenshot at the completion of one leg. On this leg I had an average usage of 3.8mi/KW, and the left display indicated 60 miles on arrival, while the SOC% was 25%. If I do the math: 25% x 112KW battery (AGT) = 28KW remaining. 28 x 3.8 mi/KW = 106 range remaining. Am I doing something wrong in deciphering the info? How do you interpret that SOC% is within 1 mile?

You are not at your destination. The car is showing miles remaining at your destination which is an hour and 56 minutes away. Your battery %SOC is showing the current state of your battery.

 

[ken43]

You are not at your destination. The car is showing miles remaining at your destination which is an hour and 56 minutes away. Your battery %SOC is showing the current state of your battery.

I am confused still. The one hour and 56 minutes is showing time to my FINAL destination. The 60 miles on arrival is what was displaying when I stopped to recharge. I interpreted the 60 miles to be reflective of distance remaining when I made the stop at my recharging point? How could my car possibly be displaying 60 miles at the final destination, before I recharge the battery, if the car doesn’t know how much juice I’m going to pump in at this recharging station?

 

[Adnillien]

I am confused still. The one hour and 56 minutes is showing time to my FINAL destination. The 60 miles on arrival is what was displaying when I stopped to recharge. I interpreted the 60 miles to be reflective of distance remaining when I made the stop at my recharging point? How could my car possibly be displaying 60 miles at the final destination, before I recharge the battery, if the car doesn’t know how much juice I’m going to pump in at this recharging station?

The number shown on the left is the miles remaining at your final destination, not an intermediate stop. I am not sure what navigation does with charging stops other than assume that you are going to add whatever it tells you to add at the stop.

 

[Bobby]

I am confused still. The one hour and 56 minutes is showing time to my FINAL destination. The 60 miles on arrival is what was displaying when I stopped to recharge. I interpreted the 60 miles to be reflective of distance remaining when I made the stop at my recharging point? How could my car possibly be displaying 60 miles at the final destination, before I recharge the battery, if the car doesn’t know how much juice I’m going to pump in at this recharging station?

After you recharged, did it change?