Announcement

Collapse
No announcement yet.

AMD Posts Navi Display Stream Compression Support For Linux

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

  • #21
    Originally posted by JMB9 View Post
    So if I am wrong just send the link and than it's all my fault - otherwise someone may explain why DP .. / HDMI ... / max resolution ... data can no longer been given. These are basic information and very relevant ... or is it really just me ???
    I'm not suggesting "it's all your fault" but I think the RDNA white paper should answer some of your questions:

    https://www.amd.com/system/files/doc...whitepaper.pdf

    Comment


    • #22
      Originally posted by bridgman View Post

      I'm not suggesting "it's all your fault" but I think the RDNA white paper should answer some of your questions:

      https://www.amd.com/system/files/doc...whitepaper.pdf
      Thanks for the heads up - but if a special white paper is where basic information is found than something is severly wrong.
      And as I am looking for small power footprint I would have aimed at APU as I have used currently iGPU only.
      From all we know now, there is no APU with Navi available in next months - isn't it.

      I had been more than 10 years as Unix consultant in industry wrestling for latest information - so please take a `thank you' for your info.
      But I am sure that you know that burying necessary basic information in a white paper is not good standard, is it.
      And `some of my questions' is not what I want to hear. Even Wikipedia had these pieces of information about Intel iGPUs before I bought my latest system.
      What has changed - is max res or port standards secret info today?
      Normally I would look at [email protected] and get dGPU, iGPU, APU versions which are capable of it (when screens become available).
      I don't blame ATI nor Intel for not giving such info before launch - but after launch - there must be something basically wrong.
      As my contact from a PC builder said Raven Ridge would be capable of much more and is limited by the chipset - and that AMD does not give that info.
      Just for the curious child in me - why? Or is this wrong?

      And as long as I don't want to program I would not consider a white paper - and I had already downloaded several AMD documents just in case it will be rejected. But not this white paper - may have missed it.
      Shouldn't open source feel different - don't you think?

      But you are right - there are answers to my questions:
      "8k textures / decode; optimized for 4k and 8k displays, HDMI 2.0b for [email protected], DP 1.4a [email protected] HDR".
      So basically the standards of 2013 and 2016, respectively (HDMI 2.1 was released 11/2017(?) - DP 2.0 is brand new: 06/2019).
      Vega was basically capable of 8k - and the performance of Raven Ridge should be similar on 8k than Haswell iGPU on 4k. Would you agree?
      And Intel Icelake may not reach the performance of Raven Ridge, right? And Intel started making Mesa capable of DSC for Icelake and 5k+ in 10/2018, true?
      Maybe you can see why I am disappointed. And maybe AMD will do with Navi-APU what they did with Vega/Raven - putting additional limits on their product by not-fitting chipsets.

      So in the end I am lost in the cold - [email protected] with low power system is not possible - and I understand the PC builder not getting into contact with me again ... as there is no current product - not to speak of an appropriate screen of about 40" to get work done - or the show going on.
      Research, Education, Creative Professionals are all longing for 8k ... what is holding it back?
      And it's just a pity that probably Intel will be the 1st to deliver a product of real [email protected] with a passively cooled system.
      Zen 2 {and even Zen+} is cool, Navi {and even Vega} is cool, but not wanting to support that resolution is a no-go for me.
      So I keep waiting ... technology is really slow these days - with 4k present in the analogue world in 1990 with 300 dpi Laser printer, transparencies and OHP (just A4 with 300 dpi = 3508*2480 pel = 8.7 Mpel, 4k = 3840*2160 pix = 8.3 Mpix).
      And we just reached that limit (diapositives anyone - much more than 8k real resolution).
      Even technical interested people here on this forum don't think of reasonable usage of 8k ... I worked on CCD images of 2000 x 2000 pix in mid 1990ies.
      Maybe a further negative effect of the use of smartphones ... where people claim to need res of at least FullHD.
      So the digital technology is kreepingly slow nearing analogue technology - maybe in 10 years. Ouch!

      Comment


      • #23
        Originally posted by Teggs View Post

        I hope 8K never becomes standard. I have nothing against its development for people who actually have a need for it, but for the majority of use cases a human would never be able to tell the difference between 8K and 4K at the screen sizes and distances involved. Even 4K itself runs into that issue. I see no need for pushing more expensive screens and technology, and requiring more power to drive them, when the difference to the human eye is... nothing.

        I'm not against technological improvement, but I hope the efforts are directed at something useful instead. OLED that doesn't suffer burn-in, usable holographic displays, flawless Adaptive Sync, etc.
        I disagree. Screen resolution evolution has been stalled for years, that is totally SHAMEFUL and SHOCKING.

        8K isn't enough. 16K isn't enough. 32K isn't enough. What the hell, even 640K isn't enough.

        If companies sell outdated and overpriced shitty GPU hardware to the mass market, it's not my fault.

        Antialising sucks and must die ASAP.

        Comment

        Working...
        X