1. This section is for official content brought to you by BeamNG. Please keep the discussion civil and posts constructive to avoid spam. Posts disregarding these guidelines may be removed without any notice.
    Dismiss Notice

Gavril T75

Discussion in 'Official Content' started by gabester, Sep 11, 2013.

  1. mtchll3

    mtchll3
    Expand Collapse

    Joined:
    Mar 31, 2014
    Messages:
    294
  2. JeremyP147

    JeremyP147
    Expand Collapse

    Joined:
    Aug 7, 2013
    Messages:
    59
    I have given this one a try just recently and I am quite impressed, even though it is not finished. I did notice that it is missing an interaxle lock, but I guess that will be added later.

    I think it would be a good idea to add an option for range change/splitter. It is difficult to get 13 buttons for gears on a keyboard or controller, and the gears are not supposed to be changed up in sequence. It would especially benefit those with a gear lever controller.
     
  3. SixSixSevenSeven

    SixSixSevenSeven
    Expand Collapse

    Joined:
    Sep 13, 2013
    Messages:
    6,960
    I was actually looking at the code for the DCT transmission and wondering if that could be adapted as a range selector
     
  4. Sebulba

    Sebulba
    Expand Collapse

    Joined:
    Jun 17, 2014
    Messages:
    90
    i tested the new experimental build and there is some strange cpu usage with the gavril T-75.
    only 1 out of my 4 cores is properly used when the T-75 is loaded.
    i tested 4 cars (G-15, T-75, Grang Marshal, Moonhawk) on the small pure grid map
    i loaded the vehicles via the standard vehicle selector and did then a 0-100-0 kph cycle.
    check the charts for cpu load / gou load / gpu memory usage.

    beamng_test.png

    system specs:
    Operating System: Windows 7 Professional 64-bit (6.1, Build 7601) Service Pack 1
    Processor: Intel(R) Core(TM) i5-3470 CPU @ 3.20GHz (4 CPUs), ~3.2GHz
    Memory: 8192MB RAM
    DirectX Version: DirectX 11
    Card name: NVIDIA GeForce GTX 660
    Display Memory: 2048 MB
    Driver Version: 9.18.13.4411
     
  5. Dummiesman

    Dummiesman
    Expand Collapse

    Joined:
    Sep 17, 2013
    Messages:
    4,685
    Its JBeam is much more complicated than other any vehicle in the game. So more CPU usage is expected.
     
  6. logoster

    logoster
    Expand Collapse

    Joined:
    Sep 5, 2012
    Messages:
    2,083

    i think his point is is that it's NOT using more cpu, and that it's actually using less
     
  7. Sebulba

    Sebulba
    Expand Collapse

    Joined:
    Jun 17, 2014
    Messages:
    90
    so the jbeam is only handled by one core? (cpu hasn't hyper threading)
     
  8. iheartmods

    iheartmods
    Expand Collapse

    Joined:
    Aug 8, 2012
    Messages:
    1,482
    I don't know why calculations aren't being done over all CPU cores yet; sticking to one core is not ok IMO. I can run Next Car Game astronomically smoother because it Hyper-Threads
     
  9. logoster

    logoster
    Expand Collapse

    Joined:
    Sep 5, 2012
    Messages:
    2,083

    really, are people STILL comparing NCG to beamng :/

    they aren't even going for the same thing, NCG is going for arcade racer, beamng is going for realistic soft-body physics simulation

    and the reason beamng doesn't use more then one core/thread is because of syncinc issues, NCG doesn't have to worry about that as it uses a far different system, so they can run on multiple threads without too many problems (and thats IF NCG runs on multiple threads, because NCG isn't going for soft-body physics simulation, it's no where near as intensive, because it DOESN'T use soft-body physics (or uses a really basic version of them)

    but seriously, don't compare beamng and NCG, they are both going for 2 different end goals :/

    now to get back to topic:

    was browsing the facepunch thread, and saw that gabester said that there's a bug in the code causing problems with the simulation of the tires on the t75.

    which is to be expected, it's called experimental for a reason after all
     
  10. SixSixSevenSeven

    SixSixSevenSeven
    Expand Collapse

    Joined:
    Sep 13, 2013
    Messages:
    6,960
    Each vehicle is simulated in its own thread. 2 vehicles = 2 threads, 3 vehicles = 3 threads etc. Hell its weirder that you have 4 cores utilised with just 1 vehicle, sure nothing else is hogging CPU time in the background?

    - - - Updated - - -

    "it hyperthreads", hyperthreading is an intel exclusive trademark. I think you mean multithreads. And BeamNG multithreads, each vehicle is in its own thread.
     
  11. gabester

    gabester
    Expand Collapse
    Vehicle Director
    BeamNG Team

    Joined:
    Jun 6, 2012
    Messages:
    2,653
    NCG simulates 5 data points per car. We simulate tens of thousands. That's why NCG is so much smoother.
     
  12. vladmir poopin

    vladmir poopin
    Expand Collapse
    Banned

    Joined:
    Mar 17, 2013
    Messages:
    844
    i get more fps on here than on ncg
     
  13. BaronRedStorm

    BaronRedStorm
    Expand Collapse

    Joined:
    Jul 12, 2014
    Messages:
    817
    Oh so is that why i can run NCG On my pc with high graphics so much smoother compared to BeamnNG, This will shut up the people complaining about this game being poorly optimized :p
     
  14. JDMClark

    JDMClark
    Expand Collapse

    Joined:
    Aug 10, 2013
    Messages:
    1,347
    The new official CopyPasta
     
  15. B25Mitch

    B25Mitch
    Expand Collapse
    Vehicle Designer
    BeamNG Team

    Joined:
    Aug 27, 2013
    Messages:
    274
    After some testing I have come to the conclusion that the T75's cab and front end are far too strong. It's a well-known phenomenon that heavier vehicles suffer more damage than light ones when colliding with an immovable barrier. The T75 however doesn't conform; it stays remarkably intact even in 130km/h front-on wall impacts - in the image below it hasn't even lost any parts.


    (imported from here)

    Contrast this with the crash test of a similar vehicle at much lower speed where we see the side steps and several other parts fall off as their momentum tears them from their mounts:



    Now let's look at a rollover event in BeamNG.drive - in this example the T-75 has flipped while skidding sideways, and the cab impacts the ground at 70km/h with 6 tonnes of truck chassis behind it.


    (imported from here)

    Yet even though the impact (taken solely by the cab) is strong enough to bend the frame of the truck, the cab remains completely intact with hardly any deformation.

    Contrast with the state of truck cabs after real-world rollovers, in which they are frequently crushed beyond recognition:



    It would seem that the non-structural components of the T75 should not only be less strong, but also more easily detached from the chassis.
     
    #775 B25Mitch, Oct 7, 2014
    Last edited by a moderator: Oct 21, 2015
  16. 14ramosr

    14ramosr
    Expand Collapse

    Joined:
    Sep 7, 2013
    Messages:
    979
    I agree with you, real life trucks are surprisingly..floppy, even in (relatively) calm crashes, as evidenced here [MEDIA=youtube]http://www.liveleak.com/view?i=2b9_1412376310[/MEDIA] But that isnt the reason im posting. I just want to thank you for presenting an argument, and backing it up with evidence and elaborating upon that. That stuff doesnt happen enough here :p Most people would say "the big truck is too weak" and end it there. Not you good sir :3
     
  17. gabester

    gabester
    Expand Collapse
    Vehicle Director
    BeamNG Team

    Joined:
    Jun 6, 2012
    Messages:
    2,653
    It probably is too strong, but keep in mind that all the footage you've shown have thousands of pounds more weight behind the impact. If you had a big trailer or sleeper cab on the T75 behind it in a high speed crash it'd crush a lot more. The T75 is around 14,000 pounds. Some fancy sleeper cab trucks with no trailer can weigh 25,000-30,000. Add in a 40,000 pound trailer like a lot are carrying and you're increasing the force by a factor of 5.

    As for roof strength: try a direct drop onto the roof, not just a gentle rollover, and it'll flatten like a pancake at low speeds.
     
  18. CaffeinatedPixels

    CaffeinatedPixels
    Expand Collapse

    Joined:
    Oct 2, 2012
    Messages:
    184
    AND don't forget:

    This is an alpha. It is not yet finished.
     
  19. Happyfaller23

    Happyfaller23
    Expand Collapse

    Joined:
    Dec 4, 2013
    Messages:
    21
    A highly overused phrase see's it's way in again...
     
  20. logoster

    logoster
    Expand Collapse

    Joined:
    Sep 5, 2012
    Messages:
    2,083
    Its true though, the game is still in an alpha state, and because of that many things are unfinished (there's even a bit of stalling code still left in the game, which im sure will be eventually finished)

    Sent from the 3rd galaxy via the talks of tapping
     
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice