Jump to content

stormrider

Members
  • Posts

    876
  • Joined

  • Last visited

Posts posted by stormrider

  1. 1 hour ago, SoW Reddog said:

    I can absolutely see why ED are defensive about this. They have updated a lot of the visual aspects of the core over the years and what they've done generally deserves praise (love the clouds). But I suspect that's not what the OP, and others commenting elsewhere are talking about. The gameplay loops within DCS haven't really changed very much, the AI is still somewhat dodgy (being generous) in many different scenarios and as a result, while the IN AIRCRAFT experience is great, the external experience can get old quite quickly. Once you've mastered a given aircraft's start up, which buttons to press in which sequence etc, and systems, there's not a whole lot to do with it that isn't frustrating to do. 

    BN, could you maybe give an example of what a week in the shoes of the core team looks like at some point. How much of their time is spent reviewing bugs, how much developing new engine capabilities, how much is spent on new models or whatever. Doesn't have to be what they are doing next week, I mean just a representative "average" week to understand better what they're up against?

    That's about it, you explained it quite well and with respect.

    I could add that it is just as frustrating when you use the mission editor and is forced to bug hunt the scenario for days, sometimes weeks, simply because most things just don't work like they should. It completely kills the immersion that one could have had if the mission had been put together and worked out of the box on first try. Every scenario would be something new even for the scenario creator.

     

    • Like 10
    • Thanks 2
  2. How about a hyperconverged computing? Has anybody thought of offloading AI calculations over the network into another node? If you think about it, AI calculations are not affect and can tolerate a slight increase in delay.

    This means one could use a dedicated server connected to a 10gb network, to calculate all the AI in the session.

    • Thanks 1
  3. 10 hours ago, TAW_Impalor said:

    They just keep telling us to wait patiently (and pay for new unreleased modules). I too would like to see a detailed report with specific milestones. My fear is that lua scripts cannot be multithreaded.

    Where that fear came from, any source?

    • Like 1
  4. 9 hours ago, okopanja said:

    Hi @stormrider,

    I wish to apologize and to thank you.

    I was very skeptical (as software developer working actually in IT) about disabling Hyperthreading and at the same time was a lazy to go to BIOS multiple times and make measurements.

    Meanwhile I have extended my grafana dashboard to also collect nvidia internal performance metrics (nvidia_exporter).

    With hyper-threading enabled:

    - GPU usage up to 50 %

    - Power usage 50-60W

    - CPU usage low (as before)

    - FPS in multiplayer 24-55 FPS

    With hyperthreading disabled:

    - GPU usage is above 90% (close to 100%)

    - Power usage 100W

    - CPU usage is higher than before (there is still juice left)

    - FPS 60-70, and much more stable!

    There must some conflict between:

    - DCS itself

    - WIndows scheduler

    - Implementation of hyperthreading

    Furthermore I believe this affects the server side as well.

    I think I will order Ultrawide, since my test show that RTX 2070 Super can drive WQHD easily.

    Thank you for sharing your results.

  5. On 2/2/2022 at 1:50 AM, Rick50 said:

    There used to be a company that was well known for making the "only"  simulations of car racing... Papyrus. They covered CART / Indycar, Nascar Winston Cup, Nascar Trucks and probably a few other genres I don't recall. Their last Nascar apparently still lives on today through mods, it's physics engine and car / suspension setup is second to none even today.

    Papyrus exists today in iRacing through Dave Kaemmer.

    https://www.iracing.com/papyrus/

  6. 2 hours ago, okopanja said:

    Your CPU should be faster than my mobile version. Yet, it looks like mine distributes the work across all cores with one being above average, while yours uses only 4 cores.

    Can you check if yo have Hardware-accelerated GPU scheduling enabled?

    (I will try to disable mine and see if it will still distribute the load)

    image.png

    Why would I want to do this? (aside from maybe not benefiting from this on DCS)

    I work with IT and recently we tried different settings to improve the performance of a large scale HPC cluster. By turning HT off, it turns out, performance increased on specific applications.

    Disable Hyper-Threading

    Quote

    Some MPI applications get better performance by disabling Hyper-Threading in the guest OS. Hyper-Threading—more generally known as simultaneous multi-threading (SMT)—allocates two virtual cores (vCPU) per physical core on the node. For many general computing tasks or tasks that require lots of I/O, Hyper-Threading can increase application throughput significantly. For compute-bound jobs in which both virtual cores are compute-bound, Hyper-Threading can hinder overall application performance and can add unpredictable variance to jobs. Turning off Hyper-Threading allows more predictable performance and can decrease job times.

    https://cloud.google.com/architecture/best-practices-for-using-mpi-on-compute-engine

  7. 9 hours ago, BIGNEWY said:

    I can read all text and the quality is good. 

    You may have a different preference, I have no issues. 

    Cmon, you have a 3090...

    Do you think, with the contacts you have with the team, that you could get 90 with vulkan?

     

    • Like 1
    • Thanks 1
  8. 5 minutes ago, MikkelCanNotFly said:

    they haven't mentioned anything about that, but I doubt it because who is going to fly the plane then?

    In Janes F-15E you could switch seats even in SP. It was good because you could setup your avionics differently front to back and then switch as you wished.

    • Thanks 1
×
×
  • Create New...