Announcement

Collapse
No announcement yet.

AMD Radeon RX 5500 XT Linux Performance

Collapse
This topic is closed.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by schmidtbag View Post
    I really don't think anybody would have a problem with a 5500XT 4GB model not existing. But, a 5500 with 8GB might be unnecessary.
    but an 8GB 5500XT is pretty much exactly what is recommended for HL Alyx...

    Comment


    • #32
      I know that most would not care - but it is striking that Navi which was said to be optimized for 4k and 8k is in mid range (i.e. Navi 14) with 120 W just a FullHD card. This doesn't sound right to me!

      Haswell iGPU is capable of running STK in FullHD, two player mode and maximum karts - and is suitable for 4k on the desktop.
      Concerning the review of Icelake (only extremely low power up to now) it is much slower concerning its graphics performance.
      After more than 6 years no one can provide HW which will be usable for 8k when screens get available, or do I miss something?

      bridgmanWhat is the maximum resolution of the Navi 14?
      Can we expect good support for 4k and 5k/8k for Navi in place for Linux graphics stack or at least near the finishing line?
      Is there hope for a Navi based APU (mobile Navi 14 was already announced - but likely for laptops only?) ... is there anything one can say right now about the next 6 months?

      I have set a deadline for my purchase of June 2019 the latest, as my Haswell got really hit by mitigation and is now hard to stand concerning special use cases (not only games, but also work - and I have seen that Skylake suffered severely, too - it really got ridiculous).

      MichaelHave I missed benchmarks about the current situation of Intel mitigations - in the desktop it feels in several cases that the performance has halved or even worse than that in the last three weeks of Linux patches and several firmware downgrades. It seems that the system is even repeatedly stuck for a second in some cases - for different systems running different distros and kernel versions.

      Intel seems to have nothing to offer giving April 2020 for Comet Lake start ... and Icelake given up for the desktop - otherwise Comet Lake would not make any sense. So it is a pity that AMD seems to be not willing or capable to use this situation.

      Comment


      • #33
        Hmm, yeah a comparison of the 4GB and 8GB models would be welcome in this scenario where people claim the 4GB has a bandwidth problem.

        Comment


        • #34
          Originally posted by JMB9 View Post
          Have I missed benchmarks about the current situation of Intel mitigations - in the desktop it feels in several cases that the performance has halved or even worse than that in the last three weeks of Linux patches and several firmware downgrades. It seems that the system is even repeatedly stuck for a second in some cases - for different systems running different distros and kernel versions.
          Your Haswell system getting stuck for a second on routine tasks actually points to a software problem...

          Just to give you a reference point:
          My Intel Core 2 Duo E8400 (45nm Wolfsdale) that I still use regularly never "gets stuck" on common tasks (e.g. trading while browsing pr0n sites on Firefox with heavy pop-up activity), no matter the load. (No pun intended, really!)

          So, you might be running different distros & kernel versions, but you are most probably always using Intel's 'pstate' driver, which is the default and at the same time the actual culprit of those "getting stuck" situations. (Besides running a standard "generic" kernel without full preemption, of course!)
          [Trust me; been there, done that...]

          Here's how you can easily get rid of your problems (also verified on my i5-3350P & i5-4430):

          - Put this to your Linux kernel command-line: --> intel_pstate=disable
          - Switch over to the "performance" governor
          - If you are on Ubuntu or its derivatives (i.e. Linux Mint), install the so-called "lowlatency" Linux option/flavo[u]r

          I really don't know why Intel loves to shoot themselves in their feet so much, but their "P-State" CPU driver only worsens their already miserable situation!

          Also, be sure to report back if these solved your issues, so others can get enlightened, too!

          Comment


          • #35
            Why is beta software used for testing? Why not use Radeon™ Software for Linux® on longterm/stable O.S.???

            Comment


            • #36
              Originally posted by JMB9 View Post
              I know that most would not care - but it is striking that Navi which was said to be optimized for 4k and 8k is in mid range (i.e. Navi 14) with 120 W just a FullHD card. This doesn't sound right to me!
              What we said was that the display engine was optimized for 4k and 8k, not that every GPU in the Navi family would be suitable for 4k/8k gaming. See first slide:

              https://www.techpowerup.com/256481/a...ngine-detailed

              We typically have a wide range of GPU "sizes" in each family - over 20x scaling for the last couple of generations - and the (complexity x pixel count) that each GPU can support scales with the size. In general display and video capabilities stay relatively constant within a family while drawing capabilities scale ~20:1.
              Last edited by bridgman; 13 December 2019, 01:02 AM.
              Test signature

              Comment


              • #37
                Originally posted by JMB9 View Post
                bridgman What is the maximum resolution of the Navi 14?
                Can we expect good support for 4k and 5k/8k for Navi in place for Linux graphics stack or at least near the finishing line?
                I think we are saying 8K HDR 60 Hz on a single DP 1.4 link with data compression for Navi10, and I expect the same would hold for other dGPUs in the Navi family.

                We share a lot of display code between OSes these days so I expect support for Linux should be pretty good. I don't know if there are any other obstacles though, eg something in X or elsewhere that might get in the way.

                Marketing focus seems to be more on single DP 1.4 than multiple slower links, so I imagine that would bubble through to development/test focus as well.

                Originally posted by JMB9 View Post
                bridgman Is there hope for a Navi based APU (mobile Navi 14 was already announced - but likely for laptops only?) ... is there anything one can say right now about the next 6 months?
                Sorry, as always "unable to comment on possible future products".
                Test signature

                Comment


                • #38
                  I'm curious to see how RDNA differs from GCN5 in terms of real world capabilities & performance. The 5700XT looks to perform almost identically to the Vega 56, on paper anyways. As a Vega 56 owner, I'd like to understand what compelling reason exists, if any, to upgrade from Vega to this new 5000 series. From what I can gather, the primary tangible advantage of RDNA today, is in 4k gaming, particularly at high refresh rates. Certainly we'll start to see games optimized for RDNA, as that's what the consoles will be running, but I'm guessing we'll have to wait a bit for the software to catch-up, in order to really see RDNA pull ahead of similar spec GCN parts.
                  Last edited by torsionbar28; 13 December 2019, 12:05 AM.

                  Comment


                  • #39
                    Originally posted by schmidtbag View Post
                    I really don't think anybody would have a problem with a 5500XT 4GB model not existing. But, a 5500 with 8GB might be unnecessary.
                    *I* would have a problem (in the abstract) with the 4GB version not existing, but I completely agree with your second opinion.

                    Just saw the GN review of these, and at +$40 for the 8GB version, with 0% uplift in *any* non-contrived scenario, that version definitely counts as "unnecessary" in my book.
                    The way I see it, the 4GB version serves a purpose. The 8GB version does not.

                    There's a VERY wide gap between the $160 5500 4GB and the $350 5700, in both $ and performance. The 1660 Super, landing almost exactly halfway in $ and (understandably) utterly destroying the 5500 in performance is completely uncontested in that range, and that's a huge range to not have a product in: a $200 span, right in "affordable" territory?!

                    AMD obviously has Reasons going on that I'm not privy to (whether it's tapeout / manufacturing capacity / whatever), but in an ideal world today's release would have been the 5500 4GB as the only card in that series, and a (much) wider 5600 XT that could compete with 1660 S, at a similar $250 price point. Since we know the extra 4GB on the second 5500 variant raises the price to $200, that leaves AMD with $50 to improve the actual GPU. Someone less tired than me can probably work out how roughly how many CUs that would translate to.

                    Right now though, AMD has a MASSIVE hole in the product lineup at what is historically a high-volume sweet spot for video cards, right before Xmas. The 5700XT / 2060S are going to be too big a reach for a LOT of buyers, and the 5500 / 1650 too weak for a smaller but still very large crowd. With only a single option (assuming you ignore the pointless 1660 non-S and 1660 Ti versions) in this $200 gap between low- and "high mid"- ends, the 1660S is going to clean up completely unopposed.

                    I'm sure AMD will sell enough at the edges to keep RTG funded for another year :P, but this looks like a huge missed opportunity to me. Like I say, it's probably the result of capacity etc rather than an outright error in judgement, but the absence of a 5600 is going to hurt them badly this season, and will continue to hurt them going forward until they produce something to fill that gap.

                    Comment


                    • #40
                      The RX 5500 XT seems like a nice replacement for my RX 560. In fact, I already ordered a Sapphire Pulse RX 5500 XT 4 GB card since I found it at the MSRP price (1990 SEK = €190 = £160 incl 25 % VAT, or $168 excluding VAT). Would I have liked it to be £20 cheaper? For sure, but at the moment this seems to be the only option for a small case without lots of fans, and focusing on a quiet system with open source drivers.

                      Now I only hope that the apparently necessary driver changes from linux-5.5-rc will be backported to stable (probably longterm?) linux-5.4. bridgman: Please?
                      Last edited by bitnick; 13 December 2019, 05:33 AM.

                      Comment

                      Working...
                      X