FSD Beta 10.69

[AlanSubie4Life]

Musk must have known that this .1 release has lots of changes. Such a slow rollout.

As someone else suggested here, may be he is using this .1 release to compare the performance vs .0 release using significantly different parameters...

It’s really hard to know what is happening; we have basically zero information about how any of these releases are performing. There is just no way to know.

But as I have always expected, 10.69 will be an incremental change. Just another update, probably similar to the one other update that showed a decent number of behavior changes.

I expect prudent intervention rate to remain in the same range (every 1-5 miles).

Hopefully, for some people, specific issues may go away (for everyone, nearly all the same issues will remain, of course). That’s progress.

 

[momo3605]

It’s really hard to know what is happening; we have basically zero information about how any of these releases are performing. There is just no way to know.

But as I have always expected, 10.69 will be an incremental change. Just another update, probably similar to the one other update that showed a decent number of behavior changes.

I expect prudent intervention rate to remain in the same range (every 1-5 miles).

Hopefully, for some people, specific issues may go away (for everyone, nearly all the same issues will remain, of course). That’s progress.

I think it's pretty clear that just about every youtuber has had significantly worse performance on 10.69.1 compared to 10.69. Even Omar wrote a nasty tweet about the update. If he's tweeting that, it must be real bad lol

 

[AlanSubie4Life]

I think it's pretty clear that just about every youtuber has had significantly worse performance on 10.69.1 compared to 10.69. Even Omar wrote a nasty tweet about the update. If he's tweeting that, it must be real bad lol

It’s just such a minuscule sample size it is hard to say.

 

[Cheburashka]

Are they pushing this update just because Elon said so? Hard to imagine they didn't do any testing before sending this to the YouTubers.

 

[2101Guy]

Are they pushing this update just because Elon said so? Hard to imagine they didn't do any testing before sending this to the YouTubers.

Especially when he PROMISED it would be

 

[momo3605]

Are they pushing this update just because Elon said so? Hard to imagine they didn't do any testing before sending this to the YouTubers.

I think Elon got a little upset when he saw James's comments on 10.69 so in fury, he made up something about how early beta is expected to have issues and 10.69.1/2 will be perfect. This was not the original plan as .1/.2 releases are just meant for safety, so the FSD engineers tweaked some parameters they thought would make it more assertive and with only 9 days between releases probably didn't have enough time to test it out.

 

[mtndrew1]

Especially when he PROMISED it would be

Thank you for the correct number of the fire.

 

[AlanSubie4Life]

I think Elon got a little upset when he saw James's comments on 10.69 so in fury, he made up something about how early beta is expected to have issues and 10.69.1/2 will be perfect. This was not the original plan as .1/.2 releases are just meant for safety, so the FSD engineers tweaked some parameters they thought would make it more assertive and with only 9 days between releases probably didn't have enough time to test it out.

This is false; there were various ways he alluded to improvements to *decrease* caution, which is arguably the opposite of a safety improvement. He also said they were for polish.

These all preceded James’ comments.

It’s extremely clear that changes were expected that would attempt to improve behavior, not just for safety. And it also seems clear that he expected 10.69 to have problems.

Before any release:

https://twitter.com/x/status/1560503326105649153

Post 10.69 release:

https://twitter.com/x/status/1561361064163516417

https://twitter.com/x/status/1561332565960806400

https://twitter.com/x/status/1561429523899482112

 

[momo3605]

This is false; there were various ways he alluded to improvements to *decrease* caution, which is arguably the opposite of a safety improvement. He also said they were for polish.

These all preceded James’ comments.

It’s extremely clear that changes were expected that would attempt to improve behavior, not just for safety. And it also seems clear that he expected 10.69 to have problems.

Before any release:

https://twitter.com/x/status/1560503326105649153

Post 10.69 release:

https://twitter.com/x/status/1561361064163516417

https://twitter.com/x/status/1561332565960806400

https://twitter.com/x/status/1561429523899482112

First 2 just say .1 and .2 are coming later (expected), and the response to chuck says "upcoming releases will do better" doesn't say anything about 10.69.1 or .2. Third quote is the only one that says 10.69 will be "polished", but that doesn't imply anything specific either. What does that even mean?

These quotes are all non-specific for a reason!

 

[AlanSubie4Life]

First 2 just say .1 and .2 are coming later (expected), and the response to chuck says "upcoming releases will do better" doesn't say anything about 10.69.1 or .2

I’m not sure what your point is. I was providing evidence that your claims were wrong. I think it is reasonable in context to think that .1 and .2 are included in “upcoming releases” in that Tweet, but of course there will ALSO be other releases that do even better.

[in response to James in anger] he made up something about how early beta is expected to have issues and 10.69.1/2 will be perfect.

He said prior to James’ comments:
* extra cautious
* .1 release will “accommodate feedback” (and it seems reasonable to presume that .2 would as well even though he did not explicitly say that)

This seems like he expected the first releases to have issues and he said nothing about it being perfect. (Said .2 should “really shine”)

This was not the original plan as .1/.2 releases are just meant for safety,

He said point releases were needed for extra polish, and the first release was being extra cautious. Clearly not “just for safety.”

It’s ok to be wrong because you forgot the exact Tweet sequence. Not a big deal. Everyone is wrong some of the time.

These quotes are all non-specific for a reason!

Yes of course. To the extent they are not specific, they are intentionally vague.

Elon never tells us what the actual timeline is (and never attempts to), so that is no surprise.

 

[momo3605]

He said point releases were needed for extra polish, and the first release was being extra cautious. Clearly not “just for safety.”

You're assuming polish means that the creep behavior would be improved...again he doesn't say what polish and shine mean. It could just be system crash bugs like dirtytesla noted in his first drive. Maybe you're right, maybe not

I’m not sure what your point is. I was providing evidence that your claims were wrong. I think it is reasonable in context to think that .1 and .2 are included in “upcoming releases” in that Tweet, but of course there will ALSO be other releases that do even better.

The point is that upcoming versions could refer to whatever comes after 10.69...10.420 probably...looking at previous history, whenever Elon says upcoming releases will improve ULTs or UFOs or xyz, it is the next point release, not the 0.0.x version. Again, you could be right...you could be wrong.

But anyway, my view based on an almost 2 year history of lackluster overhyped updates, I honestly don't think anything substantial was done in .1 or will be in .2 (if anything they'll just revert some stuff and call it improved). Any changes that WERE squeezed in to .1 were rushed out in the 9 days it took between the releases.

 

[EVNow]

Eh, I definitely think there should always be a perpetual license option. If they expect drivers to add to their data set/join FSD Beta for free then they should always have access to the software. $7k was the max for this to be reasonable, though. My license was basically free at $5k with my state’s EV rebate.

How much the market will accept depends on the take rate they want. It’s a sliding scale not one fixed amount.

As an investor I don’t want Tesla to offer perpetual option. As a customer I’m happy to have bought it.

 

[AlanSubie4Life]

You're assuming polish means that the creep behavior would be improved.

I assume that is one of the things they will attempt to improve. (But the most important thing they must do for improved success rates is to GO FASTER.)

As far as whether I am assuming it will be improved, you know where my money is.

The point is that upcoming versions could refer to whatever comes

Yeah of course, I made that extremely clear that I thought that was possible. But it doesn’t change the specifics of what you said being wrong:
- made up something after James’ comments about issues being expected.
- said that .1/.2 would be perfect
- .1/.2 were intended only for safety.

All three of these claims are contradicted by the record.

They have a lot more time for the .2 release (several weeks, per latest guidance) so we will see what happens.

Sounds like “next week” actually. We’ll see if that happens.

https://twitter.com/x/status/1564748039344082949

 

[boonedocks]

How many times did you ... actually buy ?

BTW, Rob of TeslaDaily thinks they want fewer people to buy and more people to switch to subscription. I think that makes sense - if Tesla achieves FSD, $15k is too less, and if they don't - its a needless headache to have to carry it on their books etc.

I think selling FSD made sense when they needed cash - now, not so much.

3 times. Currently have one of them, my wife’s ‘17 X, and 2 S’ that I sold and got no extra for the pre-purchased FSD package. (At least nothing worth paying $12k on my 2021 S as an “investment”).

 

[Oil Freedom]

Assuming the (non-Tesla) buyer of a used Tesla wants FSD, seems like that could be worth $12-15k for resale value. (With the minimum hardware, of course.)

 

[Supcom]

Assuming the (non-Tesla) buyer of a used Tesla wants FSD, seems like that could be worth $12-15k for resale value. (With the minimum hardware, of course.)

You have to discount the price of used FSD by the fraction of useful life remaining in the car. For example, if you are buying a 6-year old car with an expected 12-year lifespan, you should not pay more than 50% of the cost of new FSD.

 

[kabin]

You have to discount the price of used FSD by the fraction of useful life remaining in the car. For example, if you are buying a 6-year old car with an expected 12-year lifespan, you should not pay more than 50% of the cost of new FSD.

The way things are going with FSD progress, the economy, and a sky-high FSD asking price, Tesla might need to offer an FSD trade-in value similar to the recently reported program in China.

 

[tivoboy]

You have to discount the price of used FSD by the fraction of useful life remaining in the car. For example, if you are buying a 6-year old car with an expected 12-year lifespan, you should not pay more than 50% of the cost of new FSD.

It’s a good idea, but doesn’t seem to play out in practice at least not with the prices Tesla charges for say 2-3 year old M3, with or without FSD.. usually it’s about a. 1-2K pop at best.

 

[Supcom]

It’s a good idea, but doesn’t seem to play out in practice at least not with the prices Tesla charges for say 2-3 year old M3, with or without FSD.. usually it’s about a. 1-2K pop at best.

Sounds like a good deal for the consumer then.

 

[FSDtester#1]

So what are the latest bets on when anyone on this thread gets 10.69.xxxx???
Today? Tomorrow? This weekend? October?