Tesla Model 3 revealed
Discussion
Heres Johnny said:
For their sake I hope it’s sooner as all the indications are they have stopped MS and MX production for about a month so far
Retailers are selling off their in-stock and demonstrator cars. I was told I could get a virtually new Model-S for around 60k, which is well under what the "new" stock is listed at on the Tesla website. GFVs have also dramatically dipped.It all points to a refresh, and the expectation is that values and prices on the current car will take a big hit.
RobDickinson said:
This is over the raven refresh of the drivetrain etc, nothing to do with any Sept on update, which will cause even more chaos even if it is fake news..
Erm...They're not making any cars at the moment not even Raven ones. Everything points to that.
Unlike you to be less enthusiastic about what Tesla may be doing.
Witchfinder said:
I was told I could get a virtually new Model-S for around 60k, which is well under what the "new" stock is listed at on the Tesla website. GFVs have also dramatically dipped.
A new 100D for £60k?? Where?? If so get one, at £60k its a much better car than a £50k+ AWD Model 3 running uprated software.Change in GFV is due to parliamentary stuff, pretty much all PCP deals have dropped GFV for all makes/Models.
gangzoom said:
Witchfinder said:
I was told I could get a virtually new Model-S for around 60k, which is well under what the "new" stock is listed at on the Tesla website. GFVs have also dramatically dipped.
A new 100D for £60k?? Where?? If so get one, at £60k its a much better car than a £50k+ AWD Model 3 running uprated software.Change in GFV is due to parliamentary stuff, pretty much all PCP deals have dropped GFV for all makes/Models.
Tuna said:
RobDickinson said:
Because simulated miles are worth the same as actual miles.
On the whole they're better, because you can accurately measure the actual behaviour of the vehicle and compare it against the intent.If I take a Model 3 and it drives erratically along the road, nobody will know except me. The only time a Tesla knows something is going really wrong is if the driver has to break hard suddenly. Up until that point, it will be blissfully unaware of having misread a junction, or cutting up a driver.
Simulation isn't worthless, and Tesla are simulating. But simulation can only simulate the things that are imagined by the programmer or (if we're getting to that point) by the code.
Tesla are being quite smart. Their AutoPilot code seems to be running *the whole time* in "shadow mode." Whenever the driver does something the code wasn't expecting, it reports an exception. That's much more valuable than the exception during AutoPilot use (which is the much smaller proportion of the time for most drivers). The details of the exception go back to Tesla, where they are classified and used in refining the AP code.
The aim is of course to reduce the number of exceptions over time. This is a smart way to do it.
Simulation alone can't generate anything like enough edge cases because the real world is so full of random stuff. Rather than imagining the world, far better to capture the real world. That is then used as a *part* of the simulation.
Nobody else but Tesla is catching so many edge cases AFAIK. That's what is so great about this approach. Despite the confusion of at least one analyst in the room for the autonomy presentation, Tesla aren't training their neural network on the road in the cars but are instead capturing exponentially more examples of the real world to use to train the latest iteration of the NN.
gangzoom said:
A new 100D for £60k?? Where?? If so get one, at £60k its a much better car than a £50k+ AWD Model 3 running uprated software.
The Tesla salesman pointed at a showroom 75D and told me it has just been sold for 60k because it was "an older spec with less range". These cars are apparently listed on an internal sales system, and aren't shown on the Tesla website.60k is still too much for me. Doubly so now that the GFV after four years means a 10k deposit gets you monthly payments of over a grand on a base spec Model-S. A few weeks ago, it would have been £650 per month.
Witchfinder said:
The Tesla salesman pointed at a showroom 75D and told me it has just been sold for 60k because it was "an older spec with less range". These cars are apparently listed on an internal sales system, and aren't shown on the Tesla website.
60k is still too much for me. Doubly so now that the GFV after four years means a 10k deposit gets you monthly payments of over a grand on a base spec Model-S. A few weeks ago, it would have been £650 per month.
They'll be the ones I list quite a lot of - I've about double the inventory in the UK compared to what Tesla publicly declare - all they do is nobble the top level listing but the individual page for the car still exists.60k is still too much for me. Doubly so now that the GFV after four years means a 10k deposit gets you monthly payments of over a grand on a base spec Model-S. A few weeks ago, it would have been £650 per month.
And the other other thing they do is sell service cars with quite a lot of miles on as new inventory with massive discounts, if you want an example...
https://tesla-info.com/detail.php?ref=62e2be8084cd...
skwdenyer said:
Simulation isn't worthless, and Tesla are simulating. But simulation can only simulate the things that are imagined by the programmer or (if we're getting to that point) by the code.
For sure. But driving in real world conditions, you can't do the same junction but 3 inches to the left, then 2 miles an hour faster, then with three more cars crossing lanes etc. etc. without an infinitely larger number of cars on the road.Just as with the dashboard in the model 3, Tesla make a virtue out of a necessity - the reports are clear that they do not have the capacity to simulate at the scale of Google. Meanwhile Autopilot at full chat is processing gigabytes of information every minute - no-one is pretending that your car is uploading more than a few megabytes of that per day for training.
Heres Johnny said:
And the other other thing they do is sell service cars with quite a lot of miles on as new inventory with massive discounts, if you want an example...
https://tesla-info.com/detail.php?ref=62e2be8084cd...
That’s not a massive discount compared to current price though, is it? It’s a massive discount from a price that isn’t charged any more https://tesla-info.com/detail.php?ref=62e2be8084cd...
Tuna said:
For sure. But driving in real world conditions, you can't do the same junction but 3 inches to the left, then 2 miles an hour faster, then with three more cars crossing lanes etc. etc. without an infinitely larger number of cars on the road.
Just as with the dashboard in the model 3, Tesla make a virtue out of a necessity - the reports are clear that they do not have the capacity to simulate at the scale of Google. Meanwhile Autopilot at full chat is processing gigabytes of information every minute - no-one is pretending that your car is uploading more than a few megabytes of that per day for training.
I would guess a lot of the value isn't the driving itself - that is probably pretty easy and done now? It's surely the real world driving which is the critical issue - either discovering issues with your sensors, or recording the unexpected and what do you see that could help predict it in future.Just as with the dashboard in the model 3, Tesla make a virtue out of a necessity - the reports are clear that they do not have the capacity to simulate at the scale of Google. Meanwhile Autopilot at full chat is processing gigabytes of information every minute - no-one is pretending that your car is uploading more than a few megabytes of that per day for training.
NRS said:
I would guess a lot of the value isn't the driving itself - that is probably pretty easy and done now? It's surely the real world driving which is the critical issue - either discovering issues with your sensors, or recording the unexpected and what do you see that could help predict it in future.
wasting your breath tbhHeres Johnny said:
They'll be the ones I list quite a lot of - I've about double the inventory in the UK compared to what Tesla publicly declare - all they do is nobble the top level listing but the individual page for the car still exists.
And the other other thing they do is sell service cars with quite a lot of miles on as new inventory with massive discounts, if you want an example...
https://tesla-info.com/detail.php?ref=62e2be8084cd...
How do they sell a 2 year old car with 30k+ as 'new'?And the other other thing they do is sell service cars with quite a lot of miles on as new inventory with massive discounts, if you want an example...
https://tesla-info.com/detail.php?ref=62e2be8084cd...
The inventory cars all have had a pretty abused life, even more than taxis ones I would say. All the ones I've had as loaners had knackered wheels, exteriors scuffs, and the interiors to match.
It also shows how massively overpriced the 'P' Models used to be, even with a £40k 'discount' its still more expensive than our 75D X new by more than £10k!!! Given the only difference is between the cars is the battery pack and a rear motor, thats a mighty profit margain!!
Tesla made a killing on the S/Xs before the 3 arrived, its probably unsurprising Tesla is now struggling to maintaining margins now the cash cows are no longer seen as good deals by the consumer.
Hopefully Tesla will drop S/X prices properly soon, they are great cars, but just way overpriced for the market at present.
NRS said:
I would guess a lot of the value isn't the driving itself - that is probably pretty easy and done now? It's surely the real world driving which is the critical issue - either discovering issues with your sensors, or recording the unexpected and what do you see that could help predict it in future.
Yes, it's all about correctly interpreting the environment around you and deriving an intent from that.It looks to an external developer a lot like the AI research in the 90's - initially there were great strides, and everyone got excited that it was a 'solved problem'. I remember a few companies that were spending millions on knowledge sets with the expectation that free inference would just drop out of the system if it was sufficiently primed. It never happened.
The reports that I'm seeing (and example here: https://forums.tesla.com/forum/forums/shadow-braki... ) seem to suggest two problems - first that the 'interpretation' phase is quite sensitive, leading to either phantom events or over-sensitivity to stimulus. The second is the problem of interpreting intent - if the brake lights go on, on the car in front, is it a prelude to braking hard, or just a tap? And these are exactly in the area that is hardest to solve - how do you tune a system to interpret its environment correctly when you only have partial information on what the correct interpretation is? Watching what the driver is doing doesn't help - we often react slowly, and let potentially dangerous situations build up because we can see that they will resolve themselves. We have no way of telling the car 'that wasn't meant to happen' apart from slamming on the brakes which isn't exactly a sophisticated feedback mechanism.
Rob doesn't like me being down on Tesla, but Musk appears to be betting the company on FSD, and I think that's a huge gamble - not helped by his habit of completely over-hyping capabilities. Anyone following the nonsense around Hyperloop and the Boring company will understand that what he says he is achieving, and what is actually happening tend to be two entirely different things. And you can't really apply 'common sense' to this sort of software research - it's not a problem that you just throw stuff at until it works, there are still some properly unsolved issues that appear to require a big breakthrough to fix. In the mean time, Tesla appear to be fine tuning things to make Autopilot as consistent as possible - but that's not the same as making the step up to FSD.
Have to agree with the above, there are some fundamental unsolved problems which mean a proper FSD as most people would define it is still a decade maybe more away, therefore it's a bit disingenuous to make claims that FSD is here when it really isn't.
For reference I'd define a complete FSD as a system that can complete 95-99% of normal, everyday journeys with the only necessary input from the driver being to programme in a destination.
For reference I'd define a complete FSD as a system that can complete 95-99% of normal, everyday journeys with the only necessary input from the driver being to programme in a destination.
Gassing Station | EV and Alternative Fuels | Top of Page | What's New | My Stuff