OLED 4K 30" 60 Hz - Dell UP3017Q

Everything about displays and monitors. 120Hz, 144Hz, 240Hz, 4K, 1440p, input lag, display shopping, monitor purchase decisions, compare, versus, debate, and more. Questions? Just ask!
spacediver
Posts: 505
Joined: 18 Dec 2013, 23:51

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by spacediver » 23 Apr 2017, 01:05

thatoneguy wrote:
Ok let me explain it in words you might be able to understand
If you flash the same image twice or more times at a speed which the human brain/eyes cannot detect the double image effect goes away. The reason we see double images when we double strobe is because the persistence is within our viewing/detection threshold.

There, you happy?
You use the word "speed", which is not very well defined, but it sure sounds like you're talking about strobe frequency, not strobe duration (i.e. you seem to be referring to things like critical flicker fusion thresholds, or double pulse resolution thresholds, both of which involve the idea that when you temporally separate pulses of light with a small enough interval, you cannot distinguish them as individual pulses).


Yet in your post here:
thatoneguy wrote: I wonder how many lumens would be necessary to get something like micro/nanosecond persistence(potentially eliminating double/-image effect at any refresh rate entirely that way)? Maybe millions which I'm not sure is possible, but it's fascinating to think about

you are talking about strobe duration (persistence).

thatoneguy
Posts: 181
Joined: 06 Aug 2015, 17:16

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by thatoneguy » 23 Apr 2017, 01:31

The speed at which the frame is displayed. Or maybe I should say the time a frame is displayed on the screen for.
Say a frame is displayed for 1 nanosecond at 60fps@60hz, at 60@120 it would be slightly more than a nanosecond because it would have to display the same frame twice but at this point the frame is displayed for such little time either way that the human brain cannot detect the double-image because it is below their detection threshold.

spacediver
Posts: 505
Joined: 18 Dec 2013, 23:51

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by spacediver » 23 Apr 2017, 01:41

thatoneguy wrote:The speed at which the frame is displayed. Or maybe I should say the time a frame is displayed on the screen for.
Say a frame is displayed for 1 nanosecond at 60fps@60hz, at 60@120 it would be slightly more than a nanosecond because it would have to display the same frame twice but at this point the frame is displayed for such little time either way that the human brain cannot detect the double-image because it is below their detection threshold.
You're talking about persistence here (strobe duration).

And double pulse resolution thresholds (which is what I think you are referring to when you use the term "detection threshold") are primarily a function of frequency, not strobe duration (duty cycle).

In your example, each pulse is 1 nanosecond. The pulse by itself is detectable, since it's really fucking bright.

If you have two pulses that are separated in time by a very small amount, an observer will only see one pulse (that is twice as bright).

But the key thing here, which you seem to be confused about, is that it is the gap between these pulses that determines whether one pulse or two pulses are seen, not the length of the pulses.

Glide
Posts: 280
Joined: 24 Mar 2015, 20:33

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by Glide » 23 Apr 2017, 02:01

thatoneguy wrote:Ok let me explain it in words you might be able to understand
If you flash the same image twice or more times at a speed which the human brain/eyes cannot detect the double image effect goes away. The reason we see double images when we double strobe is because the persistence is within our viewing/detection threshold.
There, you happy?
There is no need to be so condescending.
When the strobe duration is reduced, the gap between strobes widens; this results in higher motion clarity, more flicker, and less brightness.
When the strobe frequency is increased, the gap between strobes is narrowed by repeating images; this causes double, triple, quadruple etc. images to be displayed in motion - depending on the source/display frequency.

If you continue to increase the frequency, eventually you will reach a point where you cannot see the gap between the strobes, and you will stop seeing multiple images.
But since we cannot see any gap between the strobes, this will just look like any other flicker-free/sample-and-hold display, and will have lots of motion blur.

There is never going to be a point where reducing the strobe duration is going to reduce flicker.
Reducing the strobe duration only widens the gap between strobes.
The only thing that continually reducing the strobe duration will do is that there is probably a point where the strobe duration is so short that we can no longer perceive an image at all, and the screen will just look black.
thatoneguy wrote:The speed at which the frame is displayed. Or maybe I should say the time a frame is displayed on the screen for.
Say a frame is displayed for 1 nanosecond at 60fps@60hz, at 60@120 it would be slightly more than a nanosecond because it would have to display the same frame twice but at this point the frame is displayed for such little time either way that the human brain cannot detect the double-image because it is below their detection threshold.
Whether the strobe duration is 1ns or 5ms, at 120Hz the strobes are still 8.33ms apart.
Reducing the strobe duration only increases the amount of off-time between strobes.
  • With a 1ns strobe duration, the off-time will be 8333332ns
    With a 5ms strobe duration, the off-time will be 3333333ns
Both are still going to display double-images if it's a 60 FPS source being displayed at 120Hz.
Assuming that brightness is kept constant, the double-images are going to be worse with the 1ns strobe since you have increased the gap between the images.

Falkentyne
Posts: 2793
Joined: 26 Mar 2014, 07:23

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by Falkentyne » 23 Apr 2017, 02:56

I'm completely rambling here because I'm sleepy, but,

I think flicker is determined purely by how fast the strobe frequency is happening, since the fewer # of strobes that happen in a second, the more "time" that you will see the monitor turned off, and thus the more flicker you will have as the amount of off time will be rather high at 60 time a second. For example at 60hz strobing, the strobe frequency (timing) happens every 16.7 milliseconds. With the strobe persistence at 1.0ms, the backlight is on for 1ms and off for 15.7ms. And this happens 60 times a second. However this on/off time has less effect on flicker than having it occur only 60 times a second (60hz) rather than 120 times a second (120hz). The 60 times a second will have a lot more flicker because the off and on times don't occur as often.

Now yes, raising persistence will reduce the flickering, but you will also be decreasing the amount of blur reduction by too large of an amount. You will also wind up increasing the brightness to uncomfortable levels as well. Also on Benq monitors, you can't make the persistence at 60hz high enough to remove flicker anyway, as the maximum allowable persistence is 5ms. I forgot the reason for this, but it has to do with the backlight LED's. Benq blur reduction boosts voltage to the LED's to compensate for the dimming that occurs during strobing, which is why BBR is brighter than ULMB (ULMB does not increase voltage to compensate for strobing). A direct 1.8x voltage increase to the backlight without strobing would burn out the LEDs at 100 brightness, so there is a limit to how high the persistence can be to keep the total drive of the LED's in spec.

Likewise, at 120hz, lowering the persistence and shortening it to very short levels like 0.5ms does not increase the flickering to any noticeable level because the 0.5ms "on" time for the backlight still occurs every 8.3ms (it's off for 7.8ms).

Now a double strobe occurs when the strobe timing happens twice per frame, instead of once per frame. So at 60hz, you wind up having two 8.3ms strobes instead of one 16.7ms strobe. Thus you see a double image because the backlight is flashed on "twice" in the same frame. So doing 120hz strobes at 60hz causes very bad image quality but the same amount of almost imperceptible flicker as 120hz strobing at 120hz.

At 120hz itself, double strobing is nowhere near as bad anymore. You will still see a double image, but it will look far less distracting, as you will have two 240hz strobes (4.25ms strobes) in the 120hz (8.3ms) frame.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by Chief Blur Buster » 23 Apr 2017, 10:50

Very good, Falkentyne... Just a terminology fix.
Falkentyne wrote:I think flicker is determined purely by how fast the strobe frequency is happening, since the fewer # of strobes that happen in a second, the more "time" that you will see the monitor turned off, and thus the more flicker you will have as the amount of off time will be rather high at 60 time a second. For example at 60hz strobing, the strobe frequency (timing) happens every 16.7 milliseconds.
Terminology nit: The strobe frequency (in Hz) and the strobe timing / period (in milliseconds) are two different things, although frequency is convertible to period, and vice versa...
Falkentyne wrote:At 120hz itself, double strobing is nowhere near as bad anymore. You will still see a double image, but it will look far less distracting, as you will have two 240hz strobes (4.25ms strobes) in the 120hz (8.3ms) frame.
"4.25ms strobes" should be "srobes every 4.25ms" to prevent confusing "strobe ON length of 4.25". Since there'd be an ON period and OFF period in the 4.25ms timespan.

Other than the minor terminology mix-up, you're correct indeed...

...About ULMB voltage-boost existing on some monitors and not on others
Also, I think that some ULMB backlights does actually use backlight-voltage-boosting (like LightBoost does), we've never heard a blanket confirmation for/against -- but we know LightBoost does use the backlight-voltage-boosting, and I've seen the corresponding ULMB mode (on some of the monitors) equally as bright. So that one little detail, I'm not 100% sure of. Different monitors use bigger/smaller voltage boosts. The more voltage boost, the more risk to the LEDs, but generally, approximately 3x wattage is safe for most modern LEDs during 'pulsed' operation (voltage times amperage -- Ohm's Law -- the voltage only needs to go up less than 3x in order to output 3x wattage). You still have average wattage the same as before or less -- but you squeeze bigger wattage into shorter flashes. To a certain (limited) extent, this is generally safe for LEDs -- and that's what many, but not all, strobe backlights (including LightBoost) do to attempt to compensate partially for their dimness. There isn't even consistency of always (or not) using voltage boost for a specific strobe brand name -- so a specific monitor (of the same model) may use boosts and another may not.

...About multi-image artifacts (double-strobe, PWM-artifacts, etc)
Personally, I don't prefer double strobing at any refresh rate (240Hz strobe at 120Hz refresh) -- but it could be useful if you are super-sensitive to 120Hz flicker ;) There is still a double-image effect if you strobe 240 times a second at 120Hz. The double-image effect remains even at higher frequencies -- you can still see PWM dimming artifacts at >500Hz, for example. Anytime you flash a image more than once, there's still the potential for double-image or multi-image artifacts, even at much higher Hz. Motion at say, 2000 pixels/second can still have PWM-style artifacts at 500Hz with an image-repeat every 4 pixel apart (2000/500 = 4) -- in the form of repeating edges:

Image
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

StrobeMaster
Posts: 48
Joined: 25 Apr 2014, 01:31

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by StrobeMaster » 24 Apr 2017, 11:32

Chief Blur Buster wrote: ...About ULMB voltage-boost existing on some monitors and not on others
Also, I think that some ULMB backlights does actually use backlight-voltage-boosting (like LightBoost does), we've never heard a blanket confirmation for/against -- but we know LightBoost does use the backlight-voltage-boosting, and I've seen the corresponding ULMB mode (on some of the monitors) equally as bright. So that one little detail, I'm not 100% sure of. Different monitors use bigger/smaller voltage boosts. The more voltage boost, the more risk to the LEDs, but generally, approximately 3x wattage is safe for most modern LEDs during 'pulsed' operation (voltage times amperage -- Ohm's Law -- the voltage only needs to go up less than 3x in order to output 3x wattage). You still have average wattage the same as before or less -- but you squeeze bigger wattage into shorter flashes. To a certain (limited) extent, this is generally safe for LEDs -- and that's what many, but not all, strobe backlights (including LightBoost) do to attempt to compensate partially for their dimness. There isn't even consistency of always (or not) using voltage boost for a specific strobe brand name -- so a specific monitor (of the same model) may use boosts and another may not.
Oops....

  1. It is mainly the LED current that is boosted, not the voltage, although boosting a diode current comes with some increase in voltage too.
  2. Ohm's law is U=R*I and has only little to do the formula for power consumption P=U*I, especially when talking about LEDs.
  3. It is the first time I hear that there are ULMB implementations out there which do not boost LED current. These must be either really dark, or use considerable long strobe durations.
  4. The highest boost I measured so far was for the BenQ XL2540: 360% of the normal level in terms of (momentary) luminance. Since luminance does not increase linearly with current (LEDs work less efficient at higher currents), 360% in terms of luminance corresponds to an even higher boost factor in terms of current, perhaps around 400%.
    When it comes to MBR, the BenyQ 2540 is special in two ways: (1) strobed backlight can only be activated via the service menu, and (2) the LED current is directly modulated through the OSD's Brightness setting, no matter whether MBR is active or not or what the strobe duration is. I guess the reason why BenQ didn't make MBR accessible through the OSD is because they were not sure whether the longest strobe duration at the highest Brightness setting (and possibly in combination with 240Hz) was still in a safe range regrading LED lifetime - I certainly used such extreme settings only for getting the measurements.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by Chief Blur Buster » 24 Apr 2017, 12:58

Thanks for the clarifications, StrobeMaster!
I knew that voltage only went up slightly -- to get a lot more wattage through a LED, but I incorrectly used some terminology.

As for XL2540 -- you might be interested to know the newer XL2546 enables DyAc in the main menus, giving user easier access to blur reduction.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

StrobeMaster
Posts: 48
Joined: 25 Apr 2014, 01:31

Re: OLED 4K 30" 60 Hz - Dell UP3017Q

Post by StrobeMaster » 24 Apr 2017, 14:58

Chief Blur Buster wrote:As for XL2540 -- you might be interested to know the newer XL2546 enables DyAc in the main menus, giving user easier access to blur reduction.
I wonder if this is the only difference between the two models. Would be a shame if it was!

Post Reply