Jump to content
Phishing Emails & YouTube Messages - Fake Giveaway Read more... ×
Search In
  • More options...
Find results that contain...
Find results in...
  • entries
    5
  • comments
    2
  • views
    895

About this blog

Miscellaneous test results and information about displays and related technology.

Entries in this blog

 

V2 G-Sync Module Analysis

This post contains an analysis of the v2 G-Sync module's features and behavior. This analysis was performed with a Dell S2417DG, but is not intended to be a review of the monitor itself.   Input / Output This G-Sync module supports a single DisplayPort 1.2 input and a single HDMI 1.4 input. G-Sync is only supported over DisplayPort.   This monitor supports up to 165 Hz at 2560×1440 through DisplayPort. The timing parameters used by this monitor can be viewed here: https://linustechtips.com/main/gallery/album/4127-dell-s2417dg-edid-and-timing-parameters/   The DisplayPort EDID on this monitor reports a vertical frequency range of 30–165 Hz and a maximum bandwidth of 19.2 Gbit/s (640 Mpx/s with 8 bpc RGB color), just enough for the maximum format (2560×1440 @ 165 Hz 8 bpc RGB), which operates at a pixel rate of 635 Mpx/s, requiring 19.07 Gbit/s of bandwidth (about 88% of the 21.6 Gbit/s provided by the DisplayPort 1.2 interface).   The HDMI EDID reports a vertical frequency range of 24–60 Hz, and a maximum bandwidth of 9.0 Gbit/s (300 Mpx/s with 8 bpc RGB color). This indicates support for around 83% of the 10.2 Gbit/s limit specified by the HDMI 1.4 standard. The maximum format (2560×1440 @ 60 Hz 8 bpc RGB) uses standard CVT-RB timings by default, for a pixel rate of 241.5 Mpx/s and 7.24 Gbit/s bandwidth consumed, about 80% of the monitor's reported maximum.   DisplayPort Behavior Unfortunately, the G-Sync module carries the same behavioral flaws that other DisplayPort monitors have. When the monitor is powered down, the operating system considers the display disconnected, and will re-shuffle application windows and icons to the remaining screens. However, this particular monitor has a "Power Saving" option which, when disabled, prevents this behavior. When "Power Saving" is off, the monitor can be powered down without disconnecting from the operating system, and applications will not be moved around. I don't know whether other G-Sync monitors have a similar menu option.   This behavior does not occur with DVI or HDMI (in general, but also including the HDMI port on the G-Sync module), since DVI and HDMI supply a small amount of power from the source to read the sink EDID of the connected device even when it is powered down, which allows the operating system to still recognize the display. DisplayPort does not allow power to be transmitted from source to sink, as the DP_PWR pin is only intended for use by attached devices (such as adapters). Presumably, when the "Power Saving" option on this monitor is disabled, the monitor keeps its internal control chip powered up to some extent even when the monitor is off.   HDMI Limitations The HDMI port on the v2 G-Sync module has a flat 60 Hz limit regardless of resolution or bandwidth. While many 144 Hz monitors (particularly older ones) are limited to 60 Hz at full resolution over HDMI, this is usually due to a simple bandwidth limit of the hardware.   For example, in older 1080p 144 Hz monitors like the ASUS VG248QE, the manufacturers chose to implement HDMI controllers that were only capable of up to ≈210 Mpx/s, enough for only 60 Hz at 1080p (with an absolute upper limit of around 85 Hz at 1080p if the user sets custom resolutions). 100+ Hz at 1080p over HDMI was simply not possible on these monitors. However, since it was only a matter of limited bandwidth, higher refresh rates over HDMI could at least be achieved at lower resolutions. Usually something like 720p (which has less than half as many pixels as 1080p) would be enough to get 120 Hz.   However, the G-Sync module seems to have a software restriction which actually enforces a strict 60 Hz limit over HDMI at all resolutions, regardless of bandwidth. The monitor does work at up to 60 Hz at 2560×1440 over HDMI, so it supports at least that much bandwidth, but when attempting higher refresh rates at a lower resolution such as 1080p 120 Hz, 100 Hz, and even 75 Hz, it only results in a black screen despite the fact that 1080p 100 Hz and 75 Hz use less bandwidth than 1440p 60 Hz.   This limitation is not due to GPU scaling as one might suspect (which would scale the image to 1440p prior to transmitting it across the cable, which would mean the transmitted video is always 1440p no matter what resolution is selected, and would therefore be subject to the monitor's maximum refresh frequency for 1440p video, which is 60 Hz when connected via HDMI). Although display-side scaling is not supported over DisplayPort for some mysterious reason, display scaling is supported over HDMI and I made sure it was set when I tested >60 Hz formats.   This is an unfortunate and seemingly needless software restriction.   Can AMD graphics cards run a G-Sync monitor at full refresh rate? There has been some concern in the past as to whether G-Sync monitors will be limited to 60 Hz when using AMD graphics cards. Unsurprisingly, there are not very many people with the means to test this, as most people with G-Sync monitors don't have AMD graphics cards laying around or vice versa, and there don't seem to be any reviewers who have seen any reason to test it either (at least to my knowledge).   Fortunately, I have an AMD RX 480 on hand, so I have tested it and found that this monitor (the Dell S2417DG) works perfectly fine up to its maximum overclock of 1440p 165 Hz on AMD cards. G-Sync, of course, is not supported, but there does not appear to be any restriction requiring you to have an NVIDIA graphics card to achieve the full resolution and refresh rate of a G-Sync monitor. https://i.imgur.com/EIrj9jN.png   G-Sync Behavior   G-Sync behavior at low frame rates G-Sync operates from 0 Hz to the maximum refresh rate of the monitor (in this case, 0–165 Hz). Some people are under the impression that G-Sync has a "minimum range" below which it does not operate, such as 30–165 Hz. This is untrue, and comes from people incorrectly assuming that G-Sync stops operating once the framerate drops below the monitor's physical operating limits. Although monitors do have a minimum refresh frequency, usually around 24–30 Hz, G-Sync does continue to operate below the monitor's physical limit by duplicating frames. This technique is visually indistinguishable from single long frames, so there are no disadvantages caused by this behavior. Using this technique, G-Sync can operate at any framerate below the monitor's maximum refresh frequency, even at extremely low framerates.   I demonstrate this on the S2417DG here, where you can see G-Sync continuing to operate at around 18.5 FPS:   Does G-Sync work through a DisplayPort daisy-chain? No. I tested this monitor daisy-chained from a Dell U2414H. The S2417DG was recognized, and worked at up to 1440p 120 Hz (higher refresh rates are not available since it exceeds the bandwidth limitations of DP 1.2 when combined with the 1080p 60 Hz U2414H). However, it was not recognized as a G-Sync monitor, and the G-Sync (and ULMB) options were missing from the NVIDIA control panel.   Does G-Sync work through a DisplayPort MST hub? No. I tested this monitor through an Accell K088B-004B two-port DisplayPort 1.2 MST hub. The S2417DG was recognized, and worked at up to 1440p 165 Hz. However, it was not recognized as a G-Sync monitor, and the G-Sync (and ULMB) options were missing from the NVIDIA control panel.   ULMB Behavior   ULMB Overview ULMB (Ultra-Low Motion Blur) is NVIDIA's implementation of backlight strobing built in to G-Sync monitors. Backlight strobing is a form of reducing perceived motion blur by eliminating the "sample-and-hold" behavior of LCDs. It makes the screen behave in a manner more similar to CRTs, where the image fades to black shortly after it is drawn. This changes the way that the human eye tracks motion, resulting in less perceived motion blur. Backlight strobing does reduce the maximum brightness of the monitor significantly, since the monitor only spends a fraction of the time illuminated, which reduces the total light output of the monitor.   Similar to PWM brightness control, backlight strobing can cause noticeable flickering if the strobing is done at low frequencies. Usually 85 Hz is the recommended minimum for strobing, which is why 85 Hz was a standard refresh frequency in the days of CRTs, where it seems most people stop noticing flickering at or above that level.   PWM brightness control does not achieve the same effect as backlight strobing because the pulses are not synchronized with the monitor's refresh operations, and PWM brightness control usually operates at a much higher frequency than backlight strobing does.   NVIDIA's backlight strobing implementation, ULMB, is only available at 85 Hz, 100 Hz, and 120 Hz. It cannot be activated at other refresh frequenies. ULMB uses single strobes, so at 85 Hz refresh rate, the backlight strobes at 85 Hz, and so forth.   For technical reasons, ULMB is not compatible with variable refresh technologies like G-Sync. The user must choose between either ULMB or G-Sync, they cannot be used at the same time.   Relationship between ULMB Pulse Width setting and actual pulse width Monitors often give settings in unitless quantities. The most universal example of this is the "brightness" setting, which most monitors allow you to adjust between "0" and "100", but with no indication of what these numbers actually represent, other than arbitrary relative values.   Since these settings usually go between 0 and 100, many people use the term "percent" when discussing these settings (i.e. "I set the monitor to 50% brightness"). However, some people will recognize that these numbers do not actually represent a percentage of the maximum setting, otherwise a brightness setting of "0" would leave the monitor completely dark. This being the case, a brightness setting of 50 is not actually half as bright as the 100 setting and so forth; in reality, the setting follows an arbitrary (and in some cases, non-linear) scale which differs from display to display, so it can be informative to measure the actual values of these types of settings.   In this case, the subject of discussion is the ULMB pulse width setting. Naturally, the "100" setting does not equate to a 100% pulse width (which would mean no strobing at all), so I decided to measure the strobe at various settings to determine the actual pulse width, and to see how it reacts when the setting is adjusted. Since ULMB is available at three different refresh frequencies, I performed the tests on all three to see if that affected the behavior too.   The ULMB Pulse Width setting does behave differently at different refresh rates; neither the pulse width nor the duty cycle remains the same. The setting is variable between "10" and "100", in increments of 1. The pulse width responds linearly to the setting, meaning that each decrease of 1 in the setting decreases the pulse width by the same amount every time. When set to 100, the pulse width is twice as long as when set to 50, and ten times as long as when set to 10.   Pulse width is often represented in terms of the duty cycle, which is the pulse width as a percentage of the total period. For example, at 100 Hz a single period would be ¹⁄₁₀₀ seconds or 10 ms. A duty cycle of 20% would mean 20% of that period (2 ms) would be spent with the backlight active, and the remaining 80% (8 ms) would be spent off.   Results: At 120 Hz, the pulse width was configurable between 2.22% (185 µs) at pulse width setting "10", and 22.1% (1.84 ms) at pulse width setting "100". At 100 Hz, the pulse width was configurable between 2.44% (244 µs) at "10", and 24.1% (2.41 ms) at "100". At 85 Hz, the pulse width was configurable between 3.03% (356 µs) at "10" and 30.1% (3.55 ms) at "100". Actual measurements at every interval of 10 may be viewed here: https://linustechtips.com/main/gallery/album/4129-dell-s2417dg-ulmb-pulse-width-measurements/   Brightness reduction when using ULMB Lowering the strobe duty cycle will reduce the total light output of the monitor, which reduces the overall brightness. Brightness is directly proportional to strobe duty cycle; cutting the duty cycle in half will cut the brightness in half. Since the duty cycle scales linearly with the monitor's ULMB Pulse Width setting, the brightness will also scale linearly with it.   Since the monitor uses DC brightness control, it has a "100% duty cycle" when not in ULMB mode. Activating ULMB will reduce the brightness significantly from the monitor's maximum, since the duty cycle will drop to 30% or less. This is not as much of a problem as it might sound, since the monitor has a powerful backlight capable of excessively high brightness (well over 400 cd/m2), presumably for this exact reason. Even 20% of maximum brightness will be enough for most users, and most people will not have the brightness set anywhere near maximum in normal mode. The monitor also keeps separate brightness settings when switching between normal and ULMB mode.   Can ULMB be used with AMD graphics cards? No. The ULMB settings in the monitor's internal menu are greyed out in any situation where the monitor isn't recognized as a G-Sync monitor, including when the monitor is attached to an AMD graphics card. ULMB must be enabled through the NVIDIA control panel, and the monitor will not show up in the NVIDIA control panel unless the monitor is plugged into an NVIDIA graphics card.   Does ULMB work through a DisplayPort daisy-chain? No. I tested this monitor daisy-chained from a Dell U2414H. The S2417DG was recognized, and worked at up to 1440p 120 Hz (higher refresh rates are not available since it exceeds the bandwidth limitations of DP 1.2 when combined with the 1080p 60 Hz U2414H). However, it was not recognized as a G-Sync monitor, and the G-Sync (and ULMB) options were missing from the NVIDIA control panel. The ULMB settings were also greyed out in the monitor's internal menu.   Does ULMB work through a DisplayPort MST hub? No. I tested this monitor through an Accell K088B-004B two-port DisplayPort 1.2 MST hub. The S2417DG was recognized, and worked at up to 1440p 165 Hz. However, it was not recognized as a G-Sync monitor, and the G-Sync (and ULMB) options were missing from the NVIDIA control panel. The ULMB settings were also greyed out in the monitor's internal menu.  

Glenwing

Glenwing

AOC G2460PF 120 Hz over HDMI Testing

The AOC G2460PF supports HDMI 1.4. I will now demonstrate it operating at 1920 × 1080 @ 120 Hz over HDMI. These tests are performed with an NVIDIA GeForce GTX 780 Ti, which also only supports HDMI 1.4.   Display Settings Demonstration These settings show the G2460PF (EDID identifies itself as the "2460G4", Windows however does not read the name) connected via HDMI at 1920 × 1080 @ 120 Hz with full RGB color. A custom resolution was necessary to expose the 120 Hz option (CVT-RB timing was used, with a resulting pixel clock of 285 Mpx/s). Without custom resolutions, only options up to 60 Hz were available. Higher formats such as 144 Hz were also attempted, but failed. The monitor's HDMI port appears to support a maximum TMDS clock of approximately 300 MHz. Timing Parameters and EDID The EDID on this monitor reports a maximum of 170 Mpx/s, around the same as the maximum limit of SL-DVI or HDMI 1.2 (165 Mpx/s). However, in practice, the monitor's hardware works up to around 300 Mpx/s. Several custom resolutions were attempted. 1920 × 1080 @ 120 Hz worked with both CVT-RB timing (285 Mpx/s) and CTA-861 timing (297 Mpx/s), but anything above this point resulted in a black screen with a floating "Input Not Support" text. I attempted 1920 × 1080 @ 144 Hz at 317 Mpx/s without success, and even 138 Hz with a pixel rate of 304 Mpx/s (shown below) was rejected.

This monitor makes a good demonstration for two important points: The maximum limit of an HDMI device can be any arbitrary limit that the manufacturer decides, or that the hardware is capable of. It is not simply "a device can support either HDMI 1.4 speed (340 Mpx/s) or be limited to HDMI 1.2 speed (165 Mpx/s)", or anything like that. The limitations can be anything, and may differ on every individual model. The limits listed in the EDID are simply values typed in by the manufacturer. The EDID does not have some method of magically detecting the actual hardware capabilities of the display. The EDID limits therefore do not necessarily represent the capabilities of the actual hardware. Verification Of course, it is possible that the monitor is simply skipping frames, or failing to truly operate at 144 Hz in some other way. Some form of verification would be desirable. Verification By Oscilloscope This is measured using a Keysight EDUX1002A oscilloscope and a Texas Instruments TSL14S light-to-voltage converter. A pattern of alternating black and white frames was generated by the blurbusters flicker test (https://testufo.com/flicker). Since oscilloscopes are designed for measuring oscillating waveforms, a set of one white frame and one black frame is counted as a single "wave" (indicated by the two vertical orange lines marking the boundary of "one wave"). For this reason, the frequency displayed on the scope is half the actual refresh frequency, and the displayed period is twice the actual refresh period. In this case, 60.00 Hz indicates 60 sets of black-white transitions (2 frames) per second, for a total of 120.00 frames per second. This demonstrates flawless 120 Hz operation. Verification By High-Speed Camera This is a high-speed video of the blurbusters frame skipping test (https://testufo.com/frameskipping) shot with a Casio Exilim ZR100 at 1,000 FPS. Each frame of video represents 1 ms of real time. The video is played back at 30 FPS, meaning that every 1 second of video shows 30 ms of time. At 120 Hz, the display refreshes at intervals of 8.333 ms. This means that we should see slightly fewer than 4 refreshes per second of video, which the video does show. This can also be verified more precisely by examining the video frame by frame and counting 8–9 frames between each refresh. We can also observe from this video that the display is operating properly, without any frame skipping. High-Speed Camera Complete Demonstration Just for good measure, this video shows the display operating at 1920 × 1080 @ 120 Hz over HDMI with the frame skipping test in a single take at 1,000 FPS.

Glenwing

Glenwing

ViewSonic XG2401 144 Hz over HDMI Testing

The ViewSonic XG2401 supports HDMI 1.4. I will now demonstrate it operating at 1920 × 1080 @ 144 Hz over HDMI. These tests are performed with an NVIDIA GeForce GTX 780 Ti, which also only supports up to HDMI 1.4. Display Settings Demonstration These settings show the XG2401 connected via HDMI on both ends at 1920 × 1080 @ 144 Hz with full RGB color. These settings are available out of the box without requiring any overclocking/custom resolutions. 1080p 144 Hz was in fact selected by default when the monitor was connected over HDMI for the first time, I didn't even need to set it to 144 Hz manually. Timing Parameters and EDID For 1920 × 1080 @ 144 Hz, ViewSonic has decided to define a set of custom timing parameters, with an effective resolution of 2026 × 1157 or a pixel rate of 337.0 Mpx/s, just barely within the 340 Mpx/s maximum of HDMI 1.4. Curiously, when connected via DisplayPort, the monitor uses slightly different parameters defined by the standardized CVT-R2 formula, 2000 × 1157 or 333.2 Mpx/s, which would also fall within the 340 Mpx/s limit of HDMI 1.4. However, these timings are not used for the HDMI connections for some reason.

The EDID reports a maximum pixel clock of 340 Mpx/s, the highest allowed by HDMI 1.4. The 1080p 144 Hz format is defined within the CTA-861 extension block.

The EDID is the same on both of the XG2401's HDMI ports, and 1080p 144 Hz works on both ports. Verification Of course, it is possible that the monitor is simply skipping frames, or failing to truly operate at 144 Hz in some other way. Some form of verification would be desirable. Verification by Oscilloscope This is measured using a Keysight EDUX1002A oscilloscope and a Texas Instruments TSL14S light-to-voltage converter. A pattern of alternating black and white frames was generated by the blurbusters flicker test (https://testufo.com/flicker). Since oscilloscopes are designed for measuring oscillating waveforms, a set of one white frame and one black frame is counted as a single "wave" (indicated by the two vertical orange lines marking the boundary of "one wave"). For this reason, the frequency displayed on the scope is half the actual refresh frequency, and the displayed period is twice the actual refresh period. In this case, 71.79 Hz indicates 71.79 sets of black-white transitions (2 frames) per second, for a total of 143.58 frames per second. Verification by High-Speed Camera This is a high-speed video of the blurbusters frame skipping test (https://testufo.com/frameskipping) shot with a Casio Exilim ZR100 at 1,000 FPS. Each frame of video represents 1 ms of real time. The video is played back at 30 FPS, meaning that every 1 second of video shows 30 ms of time. At 144 Hz, the display refreshes at intervals of 6.9444 ms. This means that we should see slightly more than 4 refreshes per second of video, which the video does show. This can also be verified more precisely by examining the video frame by frame and counting 7 frames between each refresh. We can also observe that the display is operating properly, without any frame skipping.

Glenwing

Glenwing

 

How to Read the Product Description

"My laptop only has an HDMI port. I need to connect to the DisplayPort input on my new 144 Hz monitor. I found this adapter on Amazon, can I use it to connect my laptop's HDMI output port to my monitor's DisplayPort input port?"   To answer this question, we must apply some reading skills:                   No, you can't.

Glenwing

Glenwing

×