NXP 4330 Xplorer dev board

Post here questions, problems and suggestions in English language
  • Mario_r4 Kernel Developer
    Posts 2328
    Joined: Sun Feb 10, 2013 12:37 pm

    blue4meridian wrote:Is a Kolibrios for ARM embedded (M series RISC) possible?
    No.
    Spoiler:Image
    Всем чмоки в этом проекте! Засуньте эти 11 лет себе в жопу!
  • User avatar
    blue4meridian
    Posts 16
    Joined: Fri May 09, 2014 6:39 pm

    What is the greatest obstacle which makes it impossible?
  • Mario_r4 Kernel Developer
    Posts 2328
    Joined: Sun Feb 10, 2013 12:37 pm

    1) 7,519,268 reasons
    2) Assembler basically in source code (kernel, main applications)
    x86 - monogenic platform
    ARM - heterogeneous (multi segment) platform
    3) Very good for the hobby already have a PC and bad for a hobby to not have a lot of ARM devices (see p.2)
    Всем чмоки в этом проекте! Засуньте эти 11 лет себе в жопу!
  • Serge Kernel Developer
    Posts 3952
    Joined: Wed Mar 08, 2006 6:25 pm

    Mario_r4 wrote:1) 7,519,268 reasons
    Ты продал за год дистрибутивов на 55000$ ?
    Spoiler:
    ohloh.png
    ohloh.png (15.89 KiB)
    Viewed 6378 times
  • User avatar
    blue4meridian
    Posts 16
    Joined: Fri May 09, 2014 6:39 pm

    @ Mario_r4... I totally understand the obstacles you identified. But theoretically... (as a proof of concept) would using FASMARM eliminate one of the obstacles?
  • Mario_r4 Kernel Developer
    Posts 2328
    Joined: Sun Feb 10, 2013 12:37 pm

    blue4meridian wrote:@ Mario_r4... I totally understand the obstacles you identified. But theoretically... (as a proof of concept) would using FASMARM eliminate one of the obstacles?
    It's completely different hardware platforms - different command syntax, different organization of many things. x86 and ARM as two different kinds of animals. FASM for ARM compiler completely different than FASM for x86 - they have only the common name of FASM. You in their views assume portability of code written in C and C++, as is done in Linux. With assembler and KolibriOS it won't work. It will take another 4-5 years to start anything run on ARM. Kernel, drivers, and most applications we will need to write from scratch. What you ask is impossible. ARM as the platform itself is not unified as x86, it is very heterogeneous and vendors hide any information for competitive advantage.

    One does not simply walk into Mordor!
    Spoiler:Image
    Всем чмоки в этом проекте! Засуньте эти 11 лет себе в жопу!
  • Mario_r4 Kernel Developer
    Posts 2328
    Joined: Sun Feb 10, 2013 12:37 pm

    Serge wrote:
    Mario_r4 wrote:1) 7,519,268 reasons
    Ты продал за год дистрибутивов на 55000$ ?
    Spoiler:
    ohloh.png
    Я пытаюсь объяснить человеку затраты на "работу из под палки", поскольку вероятность того что присутствующие на этом форуме личности будут разрабатывать на ARM (с учетом всех присутствующих к этому противопоказаний: ассемблер, другая аппаратная платформа, отсутствие документации, отсутствие самого железа в наличии для тестирования) на основе хобби - стремится к нулю.
    Spoiler:З.Ы. Вообще чувствую, что под конец в этой теме появится кто-то знающий английский лучше меня и все объяснит, а я в очередной раз буду чувствовать себя круглым идиотом. Хотя сразу ответить этому человеку было лень и он посмеиваясь наблюдал над моими попытками. Зато после этого в очередной раз будет сказано, что мы поворачиваемся к новичкам жопой. Такие вот козлы здесь обитают. Всем спасибо!
    Всем чмоки в этом проекте! Засуньте эти 11 лет себе в жопу!
  • User avatar
    blue4meridian
    Posts 16
    Joined: Fri May 09, 2014 6:39 pm

    Indeed... Mordor is a very scary place. However... it's dark influence is now ubiquitous. Mobile is everywhere (even in the rainforest and the desert). Because of ARM the future of computing looks darker than the command line.

    I doubt if Android can save the world from bloated O$ on the ARM platform. But KolibriOS/ARM (if it existed) could. I could only image how my nVIDIA Jetson supercomputer (low power cpu + quad core + 192 gpu cuda cores which can be used as cpu's) would perform with your creation. All for $192 USD.

    Only in a hypervisior on a bloated O$ (Android?) is this possible which defeats the purpose. Better to run Android on KolibriOs with a virtual machine (Qemu?) than KolibriOS on Android with a hypervisor (VirtualBox/Docker?).
    Last edited by blue4meridian on Sun Jun 29, 2014 8:02 pm, edited 6 times in total.
  • Mario_r4 Kernel Developer
    Posts 2328
    Joined: Sun Feb 10, 2013 12:37 pm

    blue4meridian wrote:Indeed... Mordor is a very scary place. However... it's dark influence is now ubiquitous. Mobile is everywhere (even in the rainforest and the desert). Because of ARM the future of computing looks darker than the command line. I doubt if Android can save the world from bloated O$ on the ARM platform. But KolibriOS/ARM (if it existed) could.
    OK. But KolibriOS project - just for fun. If you find people to just for fun for ARM, then no one will not interfere with doing this. But here in the forum you probably will not find these people or their strength is small.
    Всем чмоки в этом проекте! Засуньте эти 11 лет себе в жопу!
  • User avatar
    blue4meridian
    Posts 16
    Joined: Fri May 09, 2014 6:39 pm

    Alas... you are absolutely correct. Maybe... if there is a small group, it could be the beginning of something truly game changing.
    • Post Reply

    Return to “English”

    Who is online

    Users browsing this forum: No registered users and 15 guests