Slashdot Mirror

← Back to Stories (view on slashdot.org)

Mobile G-SYNC Confirmed and Tested With Leaked Driver

Posted by Soulskill on from the spilling-the-beans dept.

jones_supa writes: A few weeks ago, an ASUS Nordic support representative inadvertently made available an interim build of the NVIDIA graphics driver. This was a mobile driver build (version 346.87) focused at ASUS G751 line of laptops. The driver was pulled shortly, but PC Perspective managed to get their hands on a copy of it, and installed it on a ASUS G751 review unit. To everyone's surprise, a 'G-SYNC display connected' system tray notification appeared. It turned out to actually be a functional NVIDIA G-SYNC setup on a laptop. PC Perspective found a 100Hz LCD panel inside, ran some tests, and also noted that G-SYNC is picky about the Tcon implementation of the LCD, which can lead to some glitches if not implemented minutely. NVIDIA confirmed that G-SYNC on mobile is coming in the near future, but the company wasn't able to yet discuss an official arrival date or technology specifics.

42 comments

  1. No suprise by Anonymous Coward · · Score: 1

    A dubiously useful, expensive option marketed at people who are willing to pay through the nose to have their games run poorly on a laptop...

    Seems like the right audience to market it at to me.

  2. Wut? by TechyImmigrant · · Score: 2, Insightful

    I'm trying to interpret the summary, but I cannot.

    They found a 100Hz LCD inside of what? A graphics card? A laptop? The driver?

    --
    I should use this sig to advertise my book ISBN-13 : 978-1501515132.
    1. Re:Wut? by Anonymous Coward · · Score: 0

      The graphics card contains its own mini lcd panel that can display different colors so its easy for your fans to tell how well you are gaming !

      secret word : pedantry

    2. Re:Wut? by Luckyo · · Score: 5, Informative

      The original story goes like this:

      1. Nvidia claims that it needs the expensive FPGA chip to make variable refresh rate on current range of monitors. Calls it G-sync, tech adds significant costs and nvidia takes additinal licensing fee from monitors that include the said FPGA board.
      2. AMD finds the adaptivesync spec in current VESA spec for embedded displayport used in laptops. Gets VESA to add it to upcoming displayport 1.2a spec for desktop. This does mostly the same thing without needed FPGA board or additional costly licencing fee. Monitors with adaptivesync and same specs end up about 100USD cheaper than monitors with G-sync and same specs.
      3. Nvidia openly states that it cannot make G-sync cards compatible with adaptivesync any time soon and that it will continue supporting G-sync. Many pundits wonder how long Nvidia could keep attempting this kind of vendor lock-in on monitors before ceding its position due to rather extreme price differential between G-sync and adaptivesync monitors.
      4. Laptops use eDP (embedded displayport) to connect monitor to GPU card which already has adaptivesync in the spec.
      5, Alpha driver for nvidia mobile GPU sufraces which is made to work with adaptivesync over eDP, which driver itself calls "G-sync".

      Conclusion - Nvidia lied about its adaptation of adaptivesync and it now appears extremely likely that nvidia will be using adaptivesync in its future products and just call it "G-sync mobile" or something similar.

    3. Re:Wut? by Anonymous Coward · · Score: 0

      It's not just me, then? I thought I was having a stroke for sure. Why is such sloppy writing accepted?

    4. Re:Wut? by Anonymous Coward · · Score: 0

      I'm guessing the benefits of adaptive sync rates is reduced power usage? No need to quickly refresh a static image?

    5. Re:Wut? by Anonymous Coward · · Score: 2, Informative

      An animation running with fixed vsync on a 60 Hz display can only run at 60, 30, 15, ... Hz.
      Now if the game engine only manages to render 59 Hz, it will depict 30 Hz due to vsync.
      Adaptive vsync allows the system to run at 59 Hz without tearing.

    6. Re:Wut? by Anonymous Coward · · Score: 0

      "Conclusion - Nvidia lied about its adaptation of adaptivesync and it now appears extremely likely that nvidia will be using adaptivesync in its future products and just call it "G-sync mobile" or something similar."

      Yes and no. It is clear from testing with this laptop that it does not currently handle low frame rates as well as native G-sync, and can result in more flicker. It should also be noted that nVidia released working G-Sync before the Adaptive Sync standard was finalised (even for mobile).

    7. Re:Wut? by fuzzyfuzzyfungus · · Score: 4, Insightful

      Displayport is trying to be several different things to several different people; but in this case 'adaptive sync' serves the purposes of both:

      For power-sensitive situations, being able to modify the frame rate can reduce power consumption(especially if combined with 'panel self refresh', also part of the embedded displayport standard, which calls for the panel driver to have enough RAM to store the entire frame so that the display driver and DP link can be shut down entirely if a static image is being displayed, only needing to wake back up when the image needs to be changed.)

      The other advantage(and the one that Nvidia would be shooting for) is that it allows you to avoid the 'tearing' you get if your GPU's frame rate differs from your panel's refresh rate and you end up with part of one frame and part of the next frame drawn on the panel at the same time. If you can change the panel refresh rate, you can ensure that it refreshes when, and only when, the GPU has a new frame ready(obviously, you can't push the panel above a certain refresh rate, even if the GPU is doing something simple and could spit out hundreds or thousands of FPS; but it's a lot easier to cap at X FPS than it is to ensure that the framerate never dips under heavy load.)

    8. Re:Wut? by fuzzyfuzzyfungus · · Score: 1

      On the plus side, with 'DisplayID' replacing legacy EDID, it should be possible to artificially disable this 'adaptive sync' on any monitor that doesn't have an Nvidia-signed 'G-Sync Supported!' flag purchased by the vendor and flashed into the firmware, thus allowing Nvidia to charge more without the need to include the FPGA!

      This clearly, um, incentivises innovation and, er, stuff. It's too bad, really. I could really use a GPU with better thermal efficiency; but Nvidia are being such a bunch of dicks that I feel bad considering them. Surely they could leave the dirty tricks for a generation where they are behind on pure technical aptitude?

    9. Re:Wut? by Anonymous Coward · · Score: 0

      Thank you very much for this excellent summary. If I had mod points, I'd give them.

    10. Re:Wut? by drinkypoo · · Score: 1

      On the plus side, with 'DisplayID' replacing legacy EDID,

      Well, that could be good, EDID is fucking awful, and has been a sticking point for Linux since forever. Is DisplayID also fucking awful?

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    11. Re:Wut? by Anonymous Coward · · Score: 0

      Inside an ASUS G751 review unit.

    12. Re:Wut? by gl4ss · · Score: 2

      well there's some other hacked drivers for desktop however.

      which provide the benefits of gsync without gsync and indeed it seems gsync just provides a) a certification that the monitor can handle the used displayport extensions and b) the monitor manufacturer paid for it.

      and the choice of the specific fgpa chip was for security. the chip doing shit all _nothing_ to enable "gsync" on the monitor except provide authentication to the drivers that it's there and it's safe to use the extensions, basically.

      what it means that freesync(amd version of this) and gsync work pretty much exactly the same - only that gsync has a security chip in the loop(bypassed in the hacked drivers).

      me, I don't care too much about tearing in 30-50fps range so tough nutters.

      --
      world was created 5 seconds before this post as it is.
    13. Re:Wut? by cheater512 · · Score: 1

      Isn't double buffering supposed to make it run at 60hz with a single frame doubled?
      Yep single buffering would cause a progressive tear moving down the screen constantly.

    14. Re: Wut? by Anonymous Coward · · Score: 0

      It is weird, but double and triple buffering start to fail at refresh rates like 45.
      Adaptive syncing really is a panacea for laptop gaming. The sync it'd matched perfectly to rendered frames. Forget 59hz, forget double buffering, just site a frame when you can, not just on whichever boundary of 1/60th of a second it is closest to.

    15. Re:Wut? by AllynM · · Score: 1

      1. The FPGA *was* required for the tech to work on the desktop panels it was installed in.
      2. FreeSync (as I've witnessed so far) as well as the most recent adaptive sync can not achieve the same result across as wide of a refresh rate range that G-Sync currently can.
      3. Nvidia could 'make it work', but it would not be the same experience as can be had with a G-Sync module, even with an adaptive sync panel (as evidenced by how this adaptive sync panel in this laptop intermittently blanks out at 30 FPS or when a game hitches.
      4. ...
      5. The driver was not a release driver, and was not meant to call the experience it gives 'G-Sync'. It was meant to be internal.

      Conclusion - Adaptive sync alone is not the same experience you can currently get with a real G-Sync panel, which is why any possible future G-Sync that does not need a module it's not yet a real thing.

      --
      this sig was brought to you by the letter /.
    16. Re:Wut? by Luckyo · · Score: 1

      1. Actually the current claim is that FPGA is mostly for DRM, as it's basically a DRM wrapper around activesync, which is what G-sync appears to be. According to the guy behind this discovery at least.
      2. Factually incorrect. G-sync is in fact inferior to Adaptivesync in refresh rate range in the current implementation. G-sync range is 30-144Hz, where Adaptivesync can handle (depending on the scaler) 36-240Hz, 21-144Hz, 17-120Hz and 9-60Hz.
      Source: http://www.geforce.com/hardwar...
      http://www.anandtech.com/show/...
      3. In general, when someone uses a word "experience" to describe something as simple as syncing refresh rate, you know you're about to read some marketing bullshit. You certainly delivered, by arguing that alpha leaked driver delivering worse results than finished product is evidence of superiority of finished product.
      5. Correct. However this is not about the driver, but what finding the fact that G-sync "mobile" is actually Activesync in a wrapper means for previous statements of Nvidia on the topic. Specifically that it means that a lot of those statements have been lies.
      And then there's the issue of the claim that G-sync is actually activesync in a DRM wrapper, presented by the original source that got its hands on the driver.

    17. Re:Wut? by aliquis · · Score: 2

      I thought vertical sync just removed the tearing and would show 59 Hz on the 60 Hz display with one frame being twice as long.

      Anyway Adaptive VSync isn't Adaptive-Sync.

      Adaptive VSync is an Nvidia thing which switches between vertical sync on and off depending on the frame rate.

      Adaptive-Sync is the the VESA standard which let the graphics hardware decide when the monitor should refresh.

    18. Re:Wut? by aliquis · · Score: 1

      Or well, forget the first line.

      I'm tired and didn't really think it through but just didn't thought it limited to 30 Hz.

      I expected it to vary with the in-between frame-rates where the monitor updated at its 60 Hz but the graphics card never delivering any in-between frames content but only full frames and just drop frames in-between or provide no new content if that was the case depending on the actual in-between frames wait period.

    19. Re:Wut? by citizenr · · Score: 1

      Not only is it awful, but no one uses it :) DP monitors still use EDID :)

      --
      Who logs in to gdm? Not I, said the duck.
    20. Re: Wut? by Anonymous Coward · · Score: 0

      Correctly implemented double buffering doesn't fail at any refresh rate. The problem is dickhead developers switching out the displayed buffer outside of the VBLANK interval.

    21. Re:Wut? by Anonymous Coward · · Score: 0

      An animation running with fixed vsync on a 60 Hz display can only run at 60, 30, 15, ... Hz.

      Nope.
      It can run at 60, 30, 20, 15, 12, ... Hz (any integer divider) and look ok.
      Slightly below one of those and it's still pretty decent (the odd frame shown N+1 times is hard to catch unless you're explicitly watching for them).
      Half-way between those are the worst case spots, every other frame is shown N+1 times leading to blatantly obvious stuttering in what should be smooth motion.

    22. Re:Wut? by Anonymous Coward · · Score: 0

      Dumb v-sync would drop to 30fps.
      Double and triple buffering would do 60fps with the occasional frame being shown twice = dip to 30.
      Double buffering introduces extra delay if you can produce frames significantly faster than display refresh rate (you end up blocking on the buffer swap).
      Triple buffering fixes that.

    23. Re:Wut? by AllynM · · Score: 1

      1. That is a false claim - Gamenab didn't even cite the correct FPGA model when he made that DRM claim.
      2. G-Sync is actually good down to 1 FPS - it adaptively inserts additional redraws in between frames at rates below 30, as to minimize the possibility of judder (incoming frame during an already started panel refresh pass). FreeSync (it its most recently demoed form) reverts back to the VSYNC setting at the low end. Further, you are basing the high end of G-Sync only on the currently released panels. Nothing states the G-Sync FPGA tops out at 144.
      3. I use the word 'experience' because it is 'my experience' - I have personally witnessed most currently shipping G-Sync panels as well as the FreeSync demo at this past CES. I have also performed many tests with G-Sync. Source: I have written several articles about this, including the one linked in this post.
      5. I believe the reason it is not yet released is because Nvidia wants to be able to correctly cover more of the range (including the low range / what happens when the game engine hitches).

      --
      this sig was brought to you by the letter /.
    24. Re:Wut? by Anonymous Coward · · Score: 0

      1. That claim is pretty much bullshit. The FPGA is mostly there because it does double buffering. Panels don't like refresh rates below a certain threshold (and generally the faster the panel, the higher the minimum refresh rate), so if there's no new frame coming in from the GPU after X ms the g-sync module just scans out the stored last frame again.
      AMD does pretty much the same thing in the GPU/driver for FreeSync (which is also why it will only work for 3D on Hawaii/Tonga and future GCN1.2 cards, earlier GPUs lack the internal logic in the DCE to do that).
      2. Correct in theory, incorrect in practice. Those are the maximum ranges supported, monitors are free to implement narrower refresh ranges. The first models that should appear in the next months *will* do that. 30-120Hz will be common, with quite a few only doing 30-60.
      3. Agreed, likely the alpha driver simply assumes no lower refresh rate range limit, which that specific panel/chipset doesn't like (thus blanking out if refresh rate drops below the minimum, it thinks it lost signal).
      4. see 4.
      5. Exactly. This also means once they get it working properly it's likely only a matter of a driver or DPCD hack to have desktop FreeSync monitors "mobile G-Sync" capable.

      The claims by that guy seem rather bogus, as outlined above the FPGA *does* have actual functionality, his claimed "hacked" driver is 100% identical to that internal alpha driver, ...

      Now in theory a AdaptiveSync display could do the exact same thing as a g-sync module in the scaler/t-con.
      They need to store the last frame anyways for response rate compensation, so they could just scan out that one again if there's no frame in X time and have another frame worth of buffer for cases of a new frame coming in while scanning out a repeat frame, but to my knowledge none of the current or the soon-to-be-released AdaptiveSync chipsets actually do that.
      And why would they, they can just report what the panel actually supports and let the GPU handle sending duplicate frames. Adding more memory would drive up display cost. GPUs have GBs of memory and already do double/triple buffering, so they can do it with a tiny bit of extra logic in the display engine and driver.

  3. Not a huge surprise... by fuzzyfuzzyfungus · · Score: 1

    I'm not sure why 'mobile G-sync' would be a surprise. While laptops are a bit more of a niche(for gaming purposes, they might well have pulled ahead of desktops as a whole by now), and laptop GPUs are less likely than desktop ones to achieve truly heroic framerates; they still share most of their design and drivers with desktop parts and laptops have the advantage of having tight control over the display being used. Your laptop OEM isn't going to want to go too custom, for cost reasons; but most laptop models only ship with a few different panels across their entire production run, and the manufacturer isn't on the hook if swapping in a different panel doesn't work for you. If you are dealing with a desktop, or an external display on a laptop, you have to cope with anything that more or less complies with the spec for that video output, which doesn't make your life easier if you are trying to implement a nonstandard tweak on top of a standard interface without breaking compatibility(especially since EDID is notoriously accurate, well formed, and properly implemented).

    1. Re: Not a huge surprise... by Anonymous Coward · · Score: 0

      Also, G-Sync makes more sense for situations where the GPU can't keep up with 60fps. Laptops are much more likely to struggle with high frame rates so it just make sense.

    2. Re: Not a huge surprise... by Anonymous Coward · · Score: 0

      The surprise is the lack of dedicated hardware that nvidia claimed was necessary.

  4. its Nvidia FREESYNC by citizenr · · Score: 2

    http://gamenab.net/2015/01/26/...

    Sure, its asus releasing the driver, suure

    --
    Who logs in to gdm? Not I, said the duck.
    1. Re:its Nvidia FREESYNC by Anonymous Coward · · Score: 1

      Thanks, the link you provided is VERY interesting. Freesync on every Display Port 1.2 capable monitor? Wow... Just wow... I will try here and And hoping to work

    2. Re:its Nvidia FREESYNC by abies · · Score: 1

      Unfortunately, that site already got nvidiadotted....

    3. Re:its Nvidia FREESYNC by TheDarkMaster · · Score: 1

      Dammit. Although is indeed possible to enable gsync on the hardware described (asus laptop), the author of the site is a liar. The driver is not his work, is merely a file for a beta version of the nVidia driver inadvertently leaked as described in the article. MAY work with just the right desktop monitor (able to use eDP and Display Port 1.2a), but is not his work as he claims.

      --
      Religion: The greatest weapon of mass destruction of all time
    4. Re:its Nvidia FREESYNC by valinor89 · · Score: 1

      The page is still on google cache. And there is a hidden youtube video that has a demo of the hacked drivers. https://www.youtube.com/watch?...

    5. Re:its Nvidia FREESYNC by AllynM · · Score: 1

      Gamenab stumbled across the leaked driver and tried to use it to spread a bunch of conspiracy theory FUD. I hope most people here can correctly apply Occam's razor as opposed to the alternative, which is that he supposedly designed those changes, those changes going into an internal driver build that was inadvertently leaked and happened to apply to the exact laptop he already owned.

      ExtremeTech picked apart his BS in more detail: http://www.extremetech.com/ext...

      --
      this sig was brought to you by the letter /.
  5. G-WHAT? by dohzer · · Score: 1

    Is G-SYNC some new kind of graphics card heat sink that runs at 100Hz?

    1. Re:G-WHAT? by Anonymous Coward · · Score: 0

      It's a new gay dating service.

    2. Re:G-WHAT? by Anonymous Coward · · Score: 0

      It's Nvidia-branded technology that turns around the old Vsync paradigm.

      Traditionally display devices have a Vsync line that tells the host computer when the display is painting something and when it is in a blanking interval, e.g.: a CRT/LCD might drive this line at a constant 60Hz for one resolution or 72Hz for a different resolution, but the frequency is always the same for a given resolution. The host computer would use this signal to dermine when it was safe to update/switch the display buffers - if done while displaying it would lead to tearing and all sorts of ugliness, if done while blanking it was update heaven. In double- and triple-buffering implementations, if the host could only draw frames as quickly as 59Hz on a 60Hz refresh rate it winds up only being able to update every other frame, so the effective throughput of the display updates would fall back to 30Hz (1/2). Likewise, if it could only draw frames at 29Hz, the effective throughput would fall back to 15Hz (1/4), etc.

      The idea behind G-Sync (and "Adaptive Refresh" being baked into DisplayPort) is that the host now tells the display device when it should be displaying something and when it should be blanking, so the host is now able to output at the fastest rate it can draw. i.e.: If the host can only draw frames at 59Hz then that will be the display refresh rate.

  6. Whaaat? Forced Slashdot Beta?! by matthias.loeffel · · Score: 0

    Since when? What did I miss?

    1. Re:Whaaat? Forced Slashdot Beta?! by matthias.loeffel · · Score: 1

      took a while until i found the link to ?nobeta=1 I'm relieved!

    2. Re:Whaaat? Forced Slashdot Beta?! by Anonymous Coward · · Score: 0

      Greasemonkey takes care of all that crap for me.

  7. Mod parent up by Anonymous Coward · · Score: 1

    You're completely correct, but I'd like to cite one situation where screen-tearing is a huge problem: The accurate preservation of old arcade games and computer systems.

    Due to the fact that arcade machines usually had wildly varying resolutions from game to game, it was actually pretty unusual for one to have a fixed refresh rate of 60Hz. Many of them had a vertical refresh rate in the neighborhood of 53-59Hz. Some of them, had refresh rates slightly above 60Hz, such as Pac-Man, which had a refresh rate of 60.606060Hz (repeating).

    When running these games in an emulator like MAME (Multiple Arcade Machine Emulator, http://www.mamedev.org ), screen tearing is just a given. With G-Sync or AdaptiveSync technology, the monitor can be run at the authentic refresh rate, eliminating screen tearing and making the overall experience much more accurate to how it was on the original arcade machine.

    NVidia and AMD can go on all day about how this technology benefits gamers, but from a historical preservation standpoint, it's an absolutely gargantuan windfall.