hate these ads?, log in or register to hide them
Page 45 of 48 FirstFirst ... 3542434445464748 LastLast
Results 881 to 900 of 943

Thread: Musk's statistical analysis appreciation thread. (Not Tesla related at all)

  1. #881

    Join Date
    April 10, 2011
    Posts
    10,036
    Quote Originally Posted by Baarhyn View Post
    All of the other auto companies are probably too busy trying to get Patrick Dempsey on the phone...
    Too bad Patrick Dempsey is busy to return those calls...

    Last edited by Don Rumata; October 10 2018 at 06:43:35 AM.

  2. #882
    Timaios's Avatar
    Join Date
    April 10, 2011
    Posts
    982
    Just to point out that the bar chart is meaningless because we have no idea what the standard errors are for each bar so we don't really know if there are meaningful differences between the car models in that chart. I'd assume that there aren't, and the classification-by-stars method actually is based on statistical differences between some of the measured variables.

    Le coeur a ses raisons que la raison ne connaît point. - Blaise Pascal, Pensées, 277

  3. #883

    Join Date
    April 10, 2011
    Posts
    10,036
    Quote Originally Posted by Timaios View Post
    Just to point out that the bar chart is meaningless because we have no idea what the standard errors are for each bar so we don't really know if there are meaningful differences between the car models in that chart. I'd assume that there aren't, and the classification-by-stars method actually is based on statistical differences between some of the measured variables.
    I think it is also meaningless since it only shows data for Tesla and lacks any references on how these numbers were calculated.

    It just shows desperation on Tesla's part to push some kind of positive "news" before the Q3 numbers

  4. #884

    Join Date
    April 10, 2011
    Posts
    10,036
    On a more serious note, this is just the latest indicator that Tesla struggles with finding demand. Every fanboy faggot who could afford Tesla already has one. Now Tesla tries to go after the mom-and-pop Toyota Camry crowd, who buy "safe" cars.

    The problem is that this market group does not buy sedans anymore. They buy crossovers and minivans, because they are taller and have more "utility". The only vehicle close to that is Model X, and it is way beyond their reach.

  5. #885
    Joe Appleby's Avatar
    Join Date
    April 9, 2011
    Location
    in front of the class
    Posts
    14,400
    Why are we discussing five year old news?

    https://eu.usatoday.com/story/money/...-test/2678557/

    Check that date:
    Published 6:53 PM EDT Aug 21, 2013
    Back then Tesla claimed a 5.4, NTHSA said it didn't rate beyond 5.0. And a Tesla broke a roof crush machine.

    According to an article I read back when those results, you can pile 4 fully loaded Tesla S on it's roof before it caves in. And that the NHTSA wasn't really capable of flipping it over in their standard tests.
    nevar forget

  6. #886

    Join Date
    April 11, 2011
    Location
    Canada
    Posts
    3,137
    Quote Originally Posted by Don Rumata View Post
    Quote Originally Posted by Timaios View Post
    Just to point out that the bar chart is meaningless because we have no idea what the standard errors are for each bar so we don't really know if there are meaningful differences between the car models in that chart. I'd assume that there aren't, and the classification-by-stars method actually is based on statistical differences between some of the measured variables.
    I think it is also meaningless since it only shows data for Tesla and lacks any references on how these numbers were calculated.

    It just shows desperation on Tesla's part to push some kind of positive "news" before the Q3 numbers
    well here are the data behind the charts:



    I've linked the article on jalopnik (not a usual tesla fansite) that goes into details about this. You should all go read it and come back afterward cause you all sound pretty uninformed about the whole situation.

    edit: Article, again.

    https://jalopnik.com/how-tesla-made-...afe-1829610576

  7. #887

    Join Date
    April 10, 2011
    Posts
    10,036
    Can't carefully process data over the phone, but it looks like Tesla used completely different testing lab than the rest. Even for standard crash tests, this makes the comparison at best suspect. This is just systematic lab-to-lab difference in measurements. I can get similar data variation by crashing two identical cars on two separate days of the week. Thus, any "statistical" comparison is absolutely meaningless.

    Also, Jalopnik is an internet news blog from the Gawker stable. Might as well link articles from Electrek or Teslarati... I am not reading click-bait articles entitled "how Tesla made its cars so totally safe and awesome".

    I'll wait for an ADAC-administered test, thank you very much.

    Furthermore, there's such thing as "active safety". A 911 GT2 RS with ceramic brakes and a roll cage is intrinsically safer than Tesla (or any other production car) simply because it can out-brake and out-turn all of them.

    Crash safety is an exercise in engineering cost mananagement. All you want is to get 5-star rating by spending as little as possible.

  8. #888
    מלך יהודים Zeekar's Avatar
    Join Date
    April 10, 2011
    Posts
    15,158
    Quote Originally Posted by Don Rumata View Post
    Can't carefully process data over the phone, but it looks like Tesla used completely different testing lab than the rest. Even for standard crash tests, this makes the comparison at best suspect. This is just systematic lab-to-lab difference in measurements. I can get similar data variation by crashing two identical cars on two separate days of the week. Thus, any "statistical" comparison is absolutely meaningless.

    Also, Jalopnik is an internet news blog from the Gawker stable. Might as well link articles from Electrek or Teslarati... I am not reading click-bait articles entitled "how Tesla made its cars so totally safe and awesome".

    I'll wait for an ADAC-administered test, thank you very much.

    Furthermore, there's such thing as "active safety". A 911 GT2 RS with ceramic brakes and a roll cage is intrinsically safer than Tesla (or any other production car) simply because it can out-brake and out-turn all of them.

    Cars are an exercise in engineering cost mananagement. All you want is to get as much as possible by spending as little as possible.
    FYP


    

  9. #889

    Join Date
    April 10, 2011
    Posts
    10,036
    Yes, but you still need to hit a certain benchmark. Hence, the "star" rating. You don't want manufacturers chasing meaningless numbers just to get the "best" rating, simply because it is meaningless, given the nature of the testing. This what makes Tesla claim so retarded and desperate.

    What's next? "Rated by JDPower as #1 in the initial owner faggotry?"
    Last edited by Don Rumata; October 10 2018 at 11:05:15 PM.

  10. #890
    Lana Torrin's Avatar
    Join Date
    April 13, 2011
    Location
    Bonding around
    Posts
    18,929
    Quote Originally Posted by Don Rumata View Post
    Yes, but you still need to hit a certain benchmark. Hence, the "star" rating. You don't want manufacturers chasing meaningless numbers just to get the "best" rating, simply because it is meaningless, given the nature of the testing. This what makes Tesla claim so retarded and desperate.

    What's next? "Rated by JDPower as #1 in the initial owner faggotry?"
    Jag already claims that spot.
    Quote Originally Posted by lubica
    And her name was Limul Azgoden, a lowly peasant girl.

  11. #891

    Join Date
    April 10, 2011
    Posts
    10,036
    Quote Originally Posted by Baarhyn View Post

    well here are the data behind the charts:



    I've linked the article on jalopnik (not a usual tesla fansite) that goes into details about this. You should all go read it and come back afterward cause you all sound pretty uninformed about the whole situation.
    OK, now that I can see the charts properly. Lets look at the first three rows in table 1, comparing Model 3 RWD, Model 3 AWD and Camry. It appears that all three tests have been done at three different labs (there is empty cell for AWD model).

    Now, if you compare the "probabilities" for injuries of various body parts (neck, left femur, right femur), you can see these values vary drastically (sometimes by a factor of 3) between AWD and RWD Tesla models. The risk of injury to femur is apparently 3 times higher for the RWD version.

    What does it tell us? That the measurement error (for this lab, or labs, there is no way to tell) is higher than the differences between different cars. Now, compare data for Camry and Camry Hybrid. The numbers are identical, and one would expect that regular and hybrid Camry have more mechanical differences than AWD vs RWD Tesla. Furthermore, is we are to believe the numbers, Tesla RWD has actually worse probability numbers that VW Tiguan (two bottom rows). Note also that the numbers for VW Tiguan are identical for both AWD and RWD versions (as expected).

    What it means is that Tesla's claim is most likely based on the shitty data. It is tough to say more, in the absence of additional info. These injury probabilities (specifically neck ones) are calculated from the loads collected during all possible movements (tension or compression combined with either flexion (forward) or extension (rearward) bending moment). Those values are drastically different for each impact and vary as much as 4 times depending on the type of a dummy (male, female, child, infant, etc.). We don't even know whether the dummies were positioned in a manner reflecting typical Tesla driver (i.e. with its head up its ass).

    TL;DR: Tesla used a different lab and got shitty data. Quelle surprise.

  12. #892
    Lana Torrin's Avatar
    Join Date
    April 13, 2011
    Location
    Bonding around
    Posts
    18,929
    Quote Originally Posted by Don Rumata View Post
    Quote Originally Posted by Baarhyn View Post

    well here are the data behind the charts:



    I've linked the article on jalopnik (not a usual tesla fansite) that goes into details about this. You should all go read it and come back afterward cause you all sound pretty uninformed about the whole situation.
    OK, now that I can see the charts properly. Lets look at the first three rows in table 1, comparing Model 3 RWD, Model 3 AWD and Camry. It appears that all three tests have been done at three different labs (there is empty cell for AWD model).

    Now, if you compare the "probabilities" for injuries of various body parts (neck, left femur, right femur), you can see these values vary drastically (sometimes by a factor of 3) between AWD and RWD Tesla models. The risk of injury to femur is apparently 3 times higher for the RWD version.

    What does it tell us? That the measurement error (for this lab, or labs, there is no way to tell) is higher than the differences between different cars. Now, compare data for Camry and Camry Hybrid. The numbers are identical, and one would expect that regular and hybrid Camry have more mechanical differences than AWD vs RWD Tesla. Furthermore, is we are to believe the numbers, Tesla RWD has actually worse probability numbers that VW Tiguan (two bottom rows). Note also that the numbers for VW Tiguan are identical for both AWD and RWD versions (as expected).

    What it means is that Tesla's claim is most likely based on the shitty data. It is tough to say more, in the absence of additional info. These injury probabilities (specifically neck ones) are calculated from the loads collected during all possible movements (tension or compression combined with either flexion (forward) or extension (rearward) bending moment). Those values are drastically different for each impact and vary as much as 4 times depending on the type of a dummy (male, female, child, infant, etc.). We don't even know whether the dummies were positioned in a manner reflecting typical Tesla driver (i.e. with its head up its ass).

    TL;DR: Tesla used a different lab and got shitty data. Quelle surprise.
    There is also a 0% chance of a chest injury in the AWD model 3..
    Quote Originally Posted by lubica
    And her name was Limul Azgoden, a lowly peasant girl.

  13. #893
    Approaching Walrus's Avatar
    Join Date
    March 8, 2013
    Posts
    7,515
    Quote Originally Posted by Lana Torrin View Post
    There is also a 0% chance of a chest injury in the AWD model 3..

  14. #894

    Join Date
    April 10, 2011
    Posts
    10,036
    Oh, and since Jalopnik article compared Tesla to Audi A4, here are the IIHS crash testing data, which shows Audi A4, BMW 3er and Lexus IS all to be superior to Tesla Model S.

    https://www.iihs.org/iihs/ratings/ve...door-hatchback

    https://www.iihs.org/iihs/ratings/ve...oor-sedan/2018
    https://www.iihs.org/iihs/ratings/ve...oor-sedan/2018
    https://www.iihs.org/iihs/ratings/ve...oor-sedan/2018

    Edit: Both ADAC and EuroNCAP has also tested Model S and found it to be worse than comparable BMW, Mercedes and Audi.

    https://www.adac.de/infotestrat/test...=Tesla+Model+S
    Last edited by Don Rumata; October 11 2018 at 03:43:24 AM.

  15. #895

    Join Date
    April 11, 2011
    Location
    Canada
    Posts
    3,137
    Quote Originally Posted by Don Rumata View Post
    Oh, and since Jalopnik article compared Tesla to Audi A4, here are the IIHS crash testing data, which shows Audi A4, BMW 3er and Lexus IS all to be superior to Tesla Model S.

    https://www.iihs.org/iihs/ratings/ve...door-hatchback

    https://www.iihs.org/iihs/ratings/ve...oor-sedan/2018
    https://www.iihs.org/iihs/ratings/ve...oor-sedan/2018
    https://www.iihs.org/iihs/ratings/ve...oor-sedan/2018

    Edit: ADAC has also tested Model S and found it to be worse than comparable BMW, Mercedes and Audi.

    https://www.adac.de/infotestrat/test...=Tesla+Model+S
    You really have a raging hate boner for the man lol.

    Go read the fucking jalopnik article and stop being an asshole. It's gonna take 10 minutes of your time and it has amazing gifs of sims and live crash test.

  16. #896

    Join Date
    April 10, 2011
    Posts
    10,036
    Quote Originally Posted by Baarhyn View Post
    Go read the fucking jalopnik article and stop being an asshole. It's gonna take 10 minutes of your time and it has amazing gifs of sims and live crash test.
    From the "article":

    "I’m admittedly no expert on interpreting this data..."
    "Tesla told me that..."
    "Tesla also talks about how..."
    "Tesla goes on..."
    in conclusion:
    "it’s clear that Tesla has this whole crash safety thing down pretty well."
    Why the fuck should I waste time on fanboys vomiting Tesla's talking points?


    Quote Originally Posted by Baarhyn View Post
    ...it has amazing gifs of sims and live crash test...
    The Finite-Element structural analysis is nothing new, there's absolutely nothing "amazing" about it, it has been the staple of automotive development process for the last 30 years.
    Last edited by Don Rumata; October 11 2018 at 04:14:25 AM.

  17. #897

    Join Date
    April 11, 2011
    Location
    Canada
    Posts
    3,137
    Quote Originally Posted by Don Rumata View Post
    Quote Originally Posted by Baarhyn View Post
    Go read the fucking jalopnik article and stop being an asshole. It's gonna take 10 minutes of your time and it has amazing gifs of sims and live crash test.
    From the "article":

    "I’m admittedly no expert on interpreting this data..."
    "Tesla told me that..."
    "Tesla also talks about how..."
    "Tesla goes on..."
    in conclusion:
    "it’s clear that Tesla has this whole crash safety thing down pretty well."
    Why the fuck should I waste time on fanboys vomiting Tesla's talking points?
    Since you won't read the fucking article detailing the WHOLE METHODOLOGY USED BY THE NHTSA followed by THE DATA THEY USE TO RATE CARS and how IT COMPARES BETWEEN VEHICULES because the writer doesn't pretend to know everything, i'll put your sentences into context:

    I’m admittedly no expert on interpreting this data, but a cursory look at how the 2011 and newer NHTSA crash ratings work shows that the “p(AIS3+)” figures above represents the probability of sustaining an injury that rates a score of three or above on the Abbreviated Injury Scale, with three being “serious.” These probability figures are obtained using established risk curves relating data measured in crash tests to probability of injury.
    Tesla told me that it dug into NHTSA’s data, and looked at overall probability of injury figures. The company sent me a link to some of NHTSA’s data; here’s a look at the injury probabilities to various parts of the body during a frontal crash test:
    Tesla also goes on to discuss how the car’s electric vehicle architecture—consisting of a “rigid” passenger compartment and a “fortified” battery pack—helps distribute crash forces away from the cabin, and optimizes the car’s crumple zone, with the company pointing out in a video how the Audi A4's motor gets shoved toward the passenger compartment.
    Tesla also talks about how the front of the car was also designed to handle the pole tests, which impart a narrow, focused load onto the vehicle between the main crash rails:
    Ho no, a dude writing about a story asks the company building the fucking car for details and convey them while backing it up with VIDEO EVIDENCE and full physics simulations to back the claim that the company is making, what a fucking tesla owned hack.

  18. #898
    מלך יהודים Zeekar's Avatar
    Join Date
    April 10, 2011
    Posts
    15,158
    Dude FEM is used in the auto industry for decades and people still test stuff and dont trust it blindly. Its an optimization tool and thats it.


    

  19. #899

    Join Date
    April 10, 2011
    Posts
    10,036
    Quote Originally Posted by Baarhyn View Post
    Ho no, a dude writing about a story asks the company building the fucking car for details and convey them while backing it up with VIDEO EVIDENCE and full physics simulations to back the claim that the company is making, what a fucking tesla owned hack.
    OMG, this awesome investigative journalist asked a company some questions and then posted their answers AND gifs that company PR drone has sent to him!!!!
    Give this man a fucking Pulitzer!

    What "video evidence"?!? That the Audi engine acts as a structural element and as an energy absorber, like its supposed to? Did you actually notice that Tesla's control arms and V-shaped brace bow out and collapse, while Audi X-frame absorbs the impact and stays intact? Did you read the motherfucking crash test reports that stated that there was an intrusion of the Tesla brake calipers into the passenger compartment? And did you notice that the video doesn't show that because fucking Tesla has cropped the gif to hide this?!? Because those calipers not only damaged the dummy's leg, but also perforated their shitty battery pack!?!

    Or you were too busy being distracted by the simulations that somehow "back the company's claim"?

  20. #900
    Movember 2011 RazoR's Avatar
    Join Date
    April 10, 2011
    Location
    The Motherland
    Posts
    29,731
    Quote Originally Posted by Lana Torrin View Post
    Quote Originally Posted by Don Rumata View Post
    Yes, but you still need to hit a certain benchmark. Hence, the "star" rating. You don't want manufacturers chasing meaningless numbers just to get the "best" rating, simply because it is meaningless, given the nature of the testing. This what makes Tesla claim so retarded and desperate.

    What's next? "Rated by JDPower as #1 in the initial owner faggotry?"
    Jag already claims that spot.
    Paging Mr. Dakflare.

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •