Announcement

Collapse
No announcement yet.

AMD Wowed Linux Users In 2020 With Their Fantastic Zen 3 CPUs, Timely New Open-Source GPU Support

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #21
    Originally posted by onlyLinuxLuvUBack View Post
    I am wowed by the paper launches.

    I would like amd nics and would like amd to figure out how to catch up to intel sata controllers.

    There is no paper launch



    Amazon: Všechny Ryzeny 5000 v top 12, nejprodávanější Intel je Core i7-9700K
    18. 12. 2020
    Ryzen 7 5800X - 1. příčka
    Ryzen 5 5600X - 8. příčka
    Ryzen 9 5900X - 10. příčka
    Ryzen 5 5950X - 12. příčka
    Ryzeny 5000 tehdy (začátkem listopadu) skutečně spadaly mezi nejprodávanější. Všechny čtyři modely se dostaly do top 16, což potvrzovalo reálné vysoké prodeje (jež následně potvrdily i statistiky německé Mindfactory).
    https://diit.cz/clanek/amazon-vsechn...-5000-v-top-12

    Trasnlated quote
    Amazon: All AMD Ryzen 5000 series CPUs n TOP12 COre i7 9700K as best selling Intel
    as of December 18th 2020

    Ryzen 7 5800X - 1st place
    Ryzen 5 5600X - 8th place
    Ryzen 9 5900X - 10th place
    Ryzen 5 5950X - 12th place

    At the beginning of november there were all 4 models (of Ryzen 5000) in TOP16. Proven by Mind Factory statistics
    and
    Mindfactory: Ryzen 9 5900X se prodává ve stejných objemech jako všechna Core i5

    4. 12. 2020



    Jen high-endový Ryzen 9 5900X se prodává ve stejném množství, jako všechny modely Core i5 této generace, tedy jako celý aktuální mainstream Intelu. Ryzen 5 5600X pak zhruba ve stejném množství, jako všechna Core i5 a i7 současné generace.
    https://diit.cz/clanek/ryzen-9-5900x...sechna-core-i5

    Translated quote
    Ryzen 9 5900Xsold at same level as all Core i5

    Ryten 9 5900X was sold at same unit level as all Core I5 (of this generation) so as all current Intel mainstram CPUs. Ryzen 5600X alone sold as many unit as all curent generation ofCore i5 and Core i7 combied
    So it cannot been paper launch.



    Edit: Jus Remeber year of 2017



    https://diit.cz/clanek/v-nemecku-se-...yzenu-nez-core
    Last edited by Peter Fodrek; 30 December 2020, 12:26 PM.

    Comment


    • #22
      Originally posted by EvilHowl View Post
      Why would AMD, which is not a networking company, make networking products?
      You act as if I asked them to make patch cables, operating systems for managed switches, and so on. I've asked that a microprocessor firm make microprocessors for those purposes as well, duh. Realtek and Broadcom are utter garbage under Linux. Intel products are good, AMD products would be good as well.

      Originally posted by Schugy View Post
      My new 292€-Notebook is equipped with rtl8821CE - BT and WiFi works (245Mbit@5GHz). Not super fast but much better than the old rtl8723be. I hope rtw88 kernel module will work in Leap15.3.
      That's still Realtek and precisely the WiFi chip that gave me so much headache only a few months ago (so not ancient past). I've had way too much trouble with Realtek drivers for Linux (broken suspend, botched performance,...) to trust them in the immediate future. I don't see why support by a dedicated driver recently shifting to it being supported by rtw88 would make any difference. The same people wrote both drivers.

      Comment


      • #23
        Originally posted by Awesomeness View Post
        You act as if I asked them to make patch cables, operating systems for managed switches, and so on. I've asked that a microprocessor firm make microprocessors for those purposes as well, duh. Realtek and Broadcom are utter garbage under Linux. Intel products are good, AMD products would be good as well.


        That's still Realtek and precisely the WiFi chip that gave me so much headache only a few months ago (so not ancient past). I've had way too much trouble with Realtek drivers for Linux (broken suspend, botched performance,...) to trust them in the immediate future. I don't see why support by a dedicated driver recently shifting to it being supported by rtw88 would make any difference. The same people wrote both drivers.
        I'm not so sure AMD would be a good idea. They have their Open Source graphics division, but apparently they don't have much communication with AMD and they don't really care about linux. They believe it's sufficient that Linux can boot at all and they don't care that a lot of hardware that works flawless on windows has no interfaces or sometimes even drivers on linux.

        I'd call AMD's graphics support on linux fantastic, but their platform support is just barely minimal.

        Comment


        • #24
          Cool. Intel are up cos they get an advice letter from a fund dumb enough to have a big position in Intel.

          They certainly seemed to pay top dollar for their acquisitions - their sale prices could be very embarrassing.
          Last edited by msroadkill612; 30 December 2020, 01:57 PM.

          Comment


          • #25
            Now if they'd only wise up and get competitive security-wise by releasing their source and letting folks verify their secure boot mechanisms. Amazing how 95+% of "security experts" will fight to the end online to promote 'security through obscurity' in this case. "But you're just a lowly customer and they need to make money..." Ya, well maybe I see something you do not see? Apple is going to have better security, RISC-V is completely open, and it isn't like AMD wasn't affected by a bunch of the same stuff as Intel in the last few years - though not to the same degree.

            Comment


            • #26
              Originally posted by duby229 View Post
              I'd call AMD's graphics support on linux fantastic, but their platform support is just barely minimal.
              They'd need to support their own networking chipsets because that would be a requirement for ChromeOS.

              Comment


              • #27
                Originally posted by Awesomeness View Post
                They'd need to support their own networking chipsets because that would be a requirement for ChromeOS.
                And amd has some time too, as long as they deliver a 2GbE nic chip in the next 10 years they will have caught up with the switches and the networking industry.

                Comment


                • #28
                  Originally posted by Awesomeness View Post
                  That's still Realtek and precisely the WiFi chip that gave me so much headache only a few months ago (so not ancient past). I've had way too much trouble with Realtek drivers for Linux (broken suspend, botched performance,...) to trust them in the immediate future. I don't see why support by a dedicated driver recently shifting to it being supported by rtw88 would make any difference. The same people wrote both drivers.
                  Because the rtl8821ce driver isn't included in the default kernel I have to deactivate secure boot. I don't really care about the name whether it is 8821ce or rtw88, but I want secure boot to be enabled. I added the repo and the driver works well for me. Also suspend to RAM doesn't even take a second and the wifi connection is restored when the lid is opened. The rtl8723be on the other hand sometimes fails and I need to unload with rmmod rtl8723be rtl8723common btcoexist rtl-pci rtlwifi and load it with modprobe rtl8723be again (and also systemctl restart NetworkManager). A dmesg |grep Trace barely shows a result.

                  Comment


                  • #29
                    Originally posted by Schugy View Post
                    Because the rtl8821ce driver isn't included in the default kernel
                    Apparently it is now included in kernel 5.10: https://github.com/torvalds/linux/bl...88/Kconfig#L64

                    I added the repo and the driver works well for me. Also suspend to RAM doesn't even take a second and the wifi connection is restored when the lid is opened.
                    I've used the same. It actually forced me to use Plasma on Wayland because that way the driver had less negative effect on waking up from suspend, ie. opening the lid and then hammering the keyboard for a while (to give it time to fully wake up and not turn off the screen again in the meantime). Under X11 everything outright froze. I've initially blamed AMD's drivers but for whatever reason I've booted a live system without the Realtek driver and everything worked (except wireless networking, of course).

                    I'd rather use Intel CPUs and GPUs than to use anything with Realtek for the time being.

                    Comment

                    Working...
                    X