Results 1 to 8 of 8

Thread: RANCHCOMPUTING & Lightwave

  1. #1

    Angry RANCHCOMPUTING & Lightwave

    I just got a bad experience with RANCHCOMPUTING that i want to share to warn Lightwavers.

    I had an heavy scene to render. It need about 80Gb memory.
    The loading is very long and the rendering very fast, with a lot of frames to render.

    It render well on my small inhouse renderfarm with BNR and WIN7. But only one node can handle that memory. So i needed a renderfarm service.

    Estimation cost was 400-800€

    Unfortunately, the RANCHCOMPUTING was unable to render the scene without constant crashes that was forcing scene to be reloaded everytime.
    But there were no warning, so from the users point, everything seems to be ok, with frames outputed along the render range.


    In the middle of the night, i checkked again the render and discovered that it was only at 46% and coasting more that 3K€!... (because of the loading times)

    I stopped everything and started to search what was doing the crashes with guys at RANCHCOMPUTING, with no success. But we discovered that the render was running well on Windows7 and not on Windows10.

    From the RANCHCOMPUTING side, This is a LWSN.Exe problem or a scene problem. LWSN is unconstant with some scenes on Windows10.
    They claim to be not responsible and refund only 50% (in render points)


    It seems that if everything is good from your side but crash from there side: you have to pay half the cost price...
    I'm not very happy with that.... What do you think?

    Be warned.
    Last edited by papou; 06-03-2018 at 01:43 PM.

  2. #2
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,775
    Doesn't their system check scenes at submission? Did it report any problem(s) when the scene was submitted?

    Their just saying it's "a LWSN.exe problem or a scene problem" isn't really proof of anything. Definitely worth asking them for (or obtaining yourself, if possible) the crash reports from the machines, so you can see where exactly the crash was occurring.
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),64GB RAM, NV 980ti

  3. #3
    Scenes needs to be check by users.
    Then there is a automatic check, mostly for content directory and unsupported things by RANCHCOMPUTING.
    Nothing hurted it when submitting.

    When i discovered that something went wrong, they told me that LWSN.exe was crashing regulary on several nodes on loading (..."Preparing Objects/*.lwo"). With Memory error. (0X000...)
    And others nodes hadn't any trouble to load.

    (For sure, i havn't any problem to load the scenes on my side.)
    Last edited by papou; 06-03-2018 at 02:52 PM.

  4. #4
    Super Duper Member kopperdrake's Avatar
    Join Date
    Mar 2004
    Location
    Derbyshire, UK
    Posts
    3,147
    If *your* computer loads it okay and *theirs* doesn't, then I'd say it's a problem for *them*. At the very least they should have given you a headsup that things weren't okay with the scene loading, rather than rack up a bill like that. That's just lazy diligence at their end.

    In my book that's bad form. You have to trust your render farm, and if you've submitted all the information they've asked for, then it's in their hands once the project is being rendered.

    What they should have done, as a proactive service:

    1) An apology that your renders were taking much longer than estimated.
    2) Explain the cause of the extended render time.
    3) Explain how they could help such situations arising in the future with this particular render problem, or at the very least, how they could help mitigate such a massive difference in render costs, even if it meant halting the render at their end.

    What you don't want is a render farm to leave their PCs rendering, despite some crashing out mid-project, and them leaving *you* to pay for *their* problem.

    Garagefarm all the way.
    - web: http://www.albino-igil.co.uk - 2D/3D Design Studio -
    - PC Spec: Intel i9-7940X @3.1GHz | 64Gb | 2 x GeForce GTX 1080 Ti 11Gb | Windows 10 Pro -

  5. #5
    Quote Originally Posted by kopperdrake View Post
    What you don't want is a render farm to leave their PCs rendering, despite some crashing out mid-project, and them leaving *you* to pay for *their* problem..
    This is exactly what it happened... But, everythings ends well cause they accept, to refund me, exceptionally.
    They are not bad guys and they wanted to do an 100% automatic renderfarm. But somes of their automatic routine need to be optionable to prevent this case.
    I will discuss that with them and i'm sure they will integrate that.

  6. #6
    Super Duper Member kopperdrake's Avatar
    Join Date
    Mar 2004
    Location
    Derbyshire, UK
    Posts
    3,147
    That's great news papou - if they learn from it then everyone benefits
    - web: http://www.albino-igil.co.uk - 2D/3D Design Studio -
    - PC Spec: Intel i9-7940X @3.1GHz | 64Gb | 2 x GeForce GTX 1080 Ti 11Gb | Windows 10 Pro -

  7. #7
    Registered User ianr's Avatar
    Join Date
    Oct 2006
    Location
    Chiltern Riviera
    Posts
    1,458
    Quote Originally Posted by kopperdrake View Post
    That's great news papou - if they learn from it then everyone benefits


    I can't wait Papou to here the next narritive in this very important news

    Could it be at threnderfarm linked to this Bug sited here?


    Thread: Lightwave 2018.0.4 and memory issues
    Last edited by ianr; 06-05-2018 at 09:09 AM.

  8. #8
    Quote Originally Posted by ianr View Post
    I can't wait Papou to here the next narritive in this very important news

    Could it be at threnderfarm linked to this Bug sited here?

    Thread: Lightwave 2018.0.4 and memory issues
    I'm using the 2015.3 renderer so I don't think so.

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
  •