-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Speculation: configuration queue for non owner line waiters
- Thread starter McHoffa
- Start date
This is true. But say out of the first 200 people, I don't see a reason that one store would have significantly more or less owners than another.
Seems to me that this one went out to something like first 100 in line in CA, and first 25 or so in other states, but not everywhere.
Just an observation from reading this and other threads...
The west-to-east coast delivery assumption for non-owners doesn't appear to be holding up. It would seem that deliveries may be spread across the country by Fremont production, transportation considerations, and each delivery centers' ability to deliver cars.
Within that structure, it appears that place in line at each individual store has some bearing as early line waiters seem to be getting invites before us that slept in a little too long on 3/31. It's doesn't appear to be absolute time-based (e.g., GMT) since the guy reserving in Pacific time zone would have to have reserved at 7am to match a 10am ET reservationist. Sound reasonable? Does it appear to hold up?
The west-to-east coast delivery assumption for non-owners doesn't appear to be holding up. It would seem that deliveries may be spread across the country by Fremont production, transportation considerations, and each delivery centers' ability to deliver cars.
Within that structure, it appears that place in line at each individual store has some bearing as early line waiters seem to be getting invites before us that slept in a little too long on 3/31. It's doesn't appear to be absolute time-based (e.g., GMT) since the guy reserving in Pacific time zone would have to have reserved at 7am to match a 10am ET reservationist. Sound reasonable? Does it appear to hold up?
I reserved at 10 am in Seaside CA got invited 2/22/18. I have heard others in NJ that got invited too on the say day as myself. I think your logic makes sense. No west-to-east cost delivery priorities. That's good news for you folks out back east.
Hi
Long time lurker, first time poster.
My theory, invites are based on % of people reserving from each location. So it would make sense that more non owners from CA would get invited since there are probably more reservations from each site. Let me explain my reasoning.
I reserved from Newport Beach, CA. Reached the location around 11:30 am PST, thinking that most of the early morning rush would be done and I could breeze through with my reservation in my lunch hour. No such luck, there were at least 100 people still in line. Took me about 2.5 hours to get my reservation done. When I left, the line was still almost the same as when I started.
I don’t know how bad the lines were in the rest of the country, but probably not as long. So if the invites were based on % of people reserving from that site, it would make sense that some sites got more invites than others. For e.g. If a site got 1000 reservations on first day in CA, and another got 250 in another state, then first 10% getting the invites would translate to 100 in CA vs 25 in other state. All speculation of course.
By this math, I would be in invitee batch #8 or beyond
Long time lurker, first time poster.
My theory, invites are based on % of people reserving from each location. So it would make sense that more non owners from CA would get invited since there are probably more reservations from each site. Let me explain my reasoning.
I reserved from Newport Beach, CA. Reached the location around 11:30 am PST, thinking that most of the early morning rush would be done and I could breeze through with my reservation in my lunch hour. No such luck, there were at least 100 people still in line. Took me about 2.5 hours to get my reservation done. When I left, the line was still almost the same as when I started.
I don’t know how bad the lines were in the rest of the country, but probably not as long. So if the invites were based on % of people reserving from that site, it would make sense that some sites got more invites than others. For e.g. If a site got 1000 reservations on first day in CA, and another got 250 in another state, then first 10% getting the invites would translate to 100 in CA vs 25 in other state. All speculation of course.
By this math, I would be in invitee batch #8 or beyond
The only problem with that is that each location has a maximum number of deliveries they can make in a given period. Some more than others but there is still a limit and that limit would probably not coincide with their percentage of reservations. It seems that recent invites have been pretty evenly scattered across the country implying that most delivery locations are getting some love evenly.
Also, keep in mind that they're still doing S and X deliveries which have to be counted towards the number of deliveries each location can handle.
mwu
Member
There are many theories floating around, but without an official word, everyone is guessing. As such, I'll to add my own to the mix.
1. Tesla is a tech company, and as such, will tackle the problem with technology. The problem has multiple dimensions. First, production is going to roll forward very quickly which makes the method of building cars to order as they did for the S (and probably the X, but I haven't paid attention to that one) not feasible. So cars will need to be built predictively.
2. Limit the number of options initially to reduce the number of possible combinations (this first part is known -- the next part is my guess). Then predictively build cars in batches of similar cars to reduce overhead (swapping paints for example).
3. Configurations will help with knowing where to ship cars, but at least for the initial production volumes, a car transporter may not be full with the configurations being produced and distributed across the country, so add a predictive algorithm for sending cars to delivery centers ahead of time.
4. All of that combined is going to be a very difficult algorithm (likely NP complete -- has no fast solution / algorithm, but I haven't broken the problem down to confirm -- I'd think the problem is related to the traveling salesperson problem). We have seen that some cars have been sitting in lots for a bit and people are still being delivered cars with low VINS, but that has been reducing the further we go with production and deliveries. This is my wildest guess... Tesla has been investing in AI / learning algorithms for their self-driving cars, but that doesn't mean that is the only area they will be deploying that tech. They may be using it to help predictively configure, produce, distribute, and deliver cars. It would start off less accurate and as it gets more datapoints, improve significantly.
1. Tesla is a tech company, and as such, will tackle the problem with technology. The problem has multiple dimensions. First, production is going to roll forward very quickly which makes the method of building cars to order as they did for the S (and probably the X, but I haven't paid attention to that one) not feasible. So cars will need to be built predictively.
2. Limit the number of options initially to reduce the number of possible combinations (this first part is known -- the next part is my guess). Then predictively build cars in batches of similar cars to reduce overhead (swapping paints for example).
3. Configurations will help with knowing where to ship cars, but at least for the initial production volumes, a car transporter may not be full with the configurations being produced and distributed across the country, so add a predictive algorithm for sending cars to delivery centers ahead of time.
4. All of that combined is going to be a very difficult algorithm (likely NP complete -- has no fast solution / algorithm, but I haven't broken the problem down to confirm -- I'd think the problem is related to the traveling salesperson problem). We have seen that some cars have been sitting in lots for a bit and people are still being delivered cars with low VINS, but that has been reducing the further we go with production and deliveries. This is my wildest guess... Tesla has been investing in AI / learning algorithms for their self-driving cars, but that doesn't mean that is the only area they will be deploying that tech. They may be using it to help predictively configure, produce, distribute, and deliver cars. It would start off less accurate and as it gets more datapoints, improve significantly.
I'm a little confused by their delineation of order location and delivery location. I made 2 reservations at Boston Prudential and was ~75th in line. One is for me, delivery nearby in Dedham MA, and the other is for a relative who lives in Oklahoma City (which I think ends up being Dallas delivery center?). Both of our reservations were avail to configure 2/22.
mwu
Member
Yea, I'm pleasantly surprised I still had access to those text messages (Yay, Google Fi). I do wish I had taken more photos of the line before they opened the doors, though. From a photo I got of the line on one side of the building, I'd guess there were about 60 people ahead of me.
Last edited:
IslandGirl
Member
Yeah, I was going nuts looking for my exact time last week. I looked at the photos I took of the line, plus a photo that someone posted to Reddit that I was in to get an idea of my # in line. The photos information was an hour later that I thought I did my reservation. I then went and requested my credit card statements. No luck with time on that.
Then I just (about an hour ago) remembered good ol' Google timeline in Google Maps. I did reserve around 10:30 because I left the store at 10:34.
Then I just (about an hour ago) remembered good ol' Google timeline in Google Maps. I did reserve around 10:30 because I left the store at 10:34.
IslandGirl
Member
Dublin, CA here as well. But I ordered online at night during the announcement. I currently have a window of May-July. What about you?
Chris350
Member
As I am looking for a M3 - LR - RWD car and currently sit with a July - Sept 18 window. Troy's Model 3 Delivery Estimator shows a possible config date of 7/19/18 and deliver of a possible 8/26/18. I fear the opening of the AWD config.... I sure when that happens, I will get bumped back into late 18... Order date 4/28/16 - 3:40pm - Non-Owner - Online Order.
Similar threads
- Article