Dell’s 32-inch 8K UP3218K Display Now for Sale
anandtech.com
anandtech.com
Going from 4K to 8K for a 32" monitor may seem like a small improvement, but it is a subtle sensory improvement that just makes using a computer more pleasant. Until displays reach 1/60th of an arc minute from standard viewing distances (roughly the discerning power of the human eye under assumptions on contrast), I will always want higher resolution.
Other than resolution improvements, it would be nice if someone would attempt an HDR light field display. This would ultimately lead to a monitor that is indistinguishable from a window.
I would like to see OLED's at this size/resolution, which would accomplish something similar.
Frankly, a 4K 32" monitor is quite crappy..
This matters mostly with tiny type. The difference there is astounding.
I recently argued that 1080p is no good at anything larger than 24" for a monitor, but that it was just right for 24".
Personally I like larger pixels, larger display at a longer distance (maybe it's my vision - gettin old?)
For me the resolution of the hardware doesn't matter a bit until the OS can handle it properly. I don't know what you're using but I would guess that it's not Windows 7.
(Being able to snap windows to half of the screen was one of the biggest UI quality of life improvements I can remember in a long time. MacOS is lagging on this one, as they are optimizing the experience for one screen.)
As far as I know, the Mac version of the feature requires two windows to be selected to share the screen, which is less flexible than the Windows implementation where I can temporarily pop a window into half-screen mode to peek at another window that's beneath it.
I haven't used Windows 10 on a high DPI external monitor, but it certainly works well on the Surface Pro 3/4's high DPI screens.
I've heard the Creators Update has some improvements to display handling, so we'll see how that goes.
This is on Microsoft, no question. Apple handled the transition much better with the hard coded 2x scaling factor. Windows is more flexible with dpi in theory, but a PITA to code against in practice.
You must have really good eyes. When I sit a normal distance away from my 32" I really have to make an effort to see the pixels. Unless you are a graphics professional I would say it's more than enough. I also work on a 40" 4K and there you can really see it's pixelated, but if you use your computer for coding/mail/browsing, it's still fine.
Low dpi screen is "fine", sure - until you get used to better ones.
What would this translate to in PPI or any other unit / measurement we might be familiar with.
I think the 1 arc second threshold is too strict by about an order of magnitude.
https://www.quora.com/How-do-you-convert-arc-seconds-to-mete...
"One degree is an angular measurement. 1/60th of a degree is an arc minute. 1/60th of an arc minutes is an arc second, a rather small angle but an angle none the less.
(...)
1 arc second subtends 1 meter at a distance of 205,787 meters (I did say it was a small angle)."
Which should mean that 1 arc second subtends one millimetre at ~200 meters? Do we really have that high visual acuity?
100 pixels per millimeter at 2m? 200 at 1m? That's ~25 * 200 = 5000 pixels per inch at 1m. I would think 1200 dpi would be more than sufficient at 1m...
[ed: missed the bit about: "from standard viewing distances" - still sounds rather extreme]
1 arcminute is a more realistic size for what people can reliably resolve I think. I guess we need significantly better than that to avoid a subjective feeling it is blurry but, that is about what we can actually reliably tell the difference between I think. We cannot tell the difference below 25 arcseconds or so though, it is not possible without larger pupils.
http://hyperphysics.phy-astr.gsu.edu/hbase/phyopt/Raylei.htm...
The human eye is normally rated at 1 arc minute == 1/60 degree rather than "1 arc second == 1/60 arc minute". This is around 1/30th the size of the full moon.
8K 32in at > ~.32m is "Retina" quality since it's > 60px / deg. 4K achieves this at >.6m. The former is ~275 dpi which is close to the ~300 dpi Apple used to define "Retina" for their iPhone.
Personally, I would prefer 120+ fps rather than higher resolution. I have both a 120Hz & 60Hz LCD on my desk and the difference when scrolling or dragging windows is quite noticeable.
Side note: sizeUp for osx is very good, I wish there were something as good for Ubuntu.
[1] http://accessories.us.dell.com/sna/productdetail.aspx?c=us&c...
At that distance you cannot see the whole screen anyway, as the 1 arc second resolution is limited to a very small spot at the center of your vision.
Personally, I'd prefer a taller aspect ratio than 16:9, which seems designed for movies. 4:3 or even 1:1 (square) would suite desktop and medical imaging work better.
I might just not be very sensitive but I don't notice the positive difference when switching to 4K, but the negative ones such as various apps not respecting dpi scaling settings are immediately visible. On 15" any app that not scaled is unusable.
Light field displays just mean even more resolution increases because they're usually created by throwing more pixels at the problem and sticking microlenses over it. Although it doesn't have to be planar, afaik you can also achieve light fields by modifying the phase by altering the properties of the medium in its depth dimension. But you still need to encode a lot more informaation, so whether you have 2D or 3D pixels... you still need more resolution.
Or an arc second, a second being 1/60th of a minute
(as an aside, I was prevented from posting this for some time due to 'submitting too fast,' but I've only posted five times today, twice in the past hour & thrice in the hours before that — how strange!)
I've been coding on a 40" 4K monitor for a while now and it's awesome to be able to see most of my code without scrolling. I would love to get better resolution but I don't want to go back to 32" (4K is good enough for coding but it's definitely pixelated at 40").
You can buy TVs in humongous sizes so I'm sure there is no technical limitation, but apparently they seem to think there is no market for large monitors. The first one on the market with a 40" monitor with a resolution above 4K has my money.
Wiki[1] notes that DSC is visually lossless per ISO/IEC 29170-2 test method, but I'm neither familiar with DSC nor the ISO/IEC standard. What are the practical implications of calibrating against a $200+ colorimeter accessory that Dell is so intent on adding to your cart?
[1] https://en.wikipedia.org/wiki/DisplayPort#Display_Stream_Com...
Also, any idea what DP cable lengths will be pragmatically limited to at such a resolution?
P.S. The naivete of this type[2] of marketing tactic never fails to blow my mind away...as perceived from an inferior Dell U2415.
[1] http://www.dell.com/en-us/shop/dell-ultrasharp-32-8k-monitor...
[2] http://i.dell.com/das/xa.ashx/global-site-design%20WEB/2a9f3...
I'm really struggling to wrap my mind around how sensibly pushing 100W of power out of a wee little USB-C port on a laptop might look though.
[1] https://www.displayport.org/what-is-displayport-over-usb-c/
It could also be possible that faster displayport alternate modes could be possible, but are just not available on current-gen transceivers.
I've never looked at LCD panel power usage before ... are those numbers high or low or ... ?
If I go to best buy and buy a 50" samsung 4k TV, what would that power usage be, roughly ?
(just trying to get a sense of comparison)
For the tv comparison, choosing the first 55" 4K tv on Samsung's site: "Typical Power Consumption: 65, Maximum Power Consumption: 170".[2]
[1]: https://www.asus.com/us/Commercial-Monitors/PA328Q/specifica...
[2]: http://www.samsung.com/us/televisions-home-theater/tvs/4k-uh...
This is tangential to this thread, but can anyone in here explain the state of the art in eye tracking? Actually rendering 16k quality for my peripheral vision seems insane to me, so I'm really interested in the barriers between today's tech and a good foveated headset.
Can anyone in the field give a précis on what is the state of the art?
So: 60 arc-minutes to a degree, 180 wide: 10800 pixels.
60 * 135 wide: 8100 pixels.
An 8192x4096 is not there, but a 16384x8192 is.
Now double that so you can get stereo vision (though really there's a lot of overlap that's going to be hard to arrange), update it 100 times a second in 48 bit color... 420 billion bits per second. Double that if you believe that we need 200Hz updates.
A Titan XP/1080 Ti is finally just about enough to push 4K/60Hz with a single card. 16K would be 16 times as many pixels per frame, and 144 Hz is more than twice the refresh rate.
To put that another way, we haven't even started to think about the connector standard that you could gang together to push that many pixels, let alone having a GPU that could actually push them. At this point you are pretty much talking about some kind of lossy compression being involved ("visually lossless" sure is a great euphemism). With lossy compression you might be able to get away with ganging together a couple of whatever DisplayPort 1.5 or 1.6 end up being.
In fact with the end of Dennard scaling there's some fundamental problems with how you would even use a GPU like that as a consumer. Your average (US) household circuit is 120V/15A peak, and can be run at 80% load continuous (12A), which works out to 1440 watts at the wall. With an 80% efficient PSU that works out to 1152 watts inside the case.
A Titan XP/1080 Ti pulls 270 watts at stock clocks [0]. So hypothetically even if you stacked four of them on an interposer you're now pulling 1080 watts inside the case, which is almost your entire circuit capacity. So by a naive calculation (16 times the pixels twice as fast = 32x divided by four GPU dies) we need at least an 8x improvement in overall efficiency before this is viable.
(Four big GP102-sized chips on an interposer is technically possible right now - they only need to be on the interposer where you need interconnect bumps - they can hang off onto a support substrate, and you can have shared memory controllers/HBM2 stacks/etc on the interposer that make this appear to be one big GPU chip instead of SLI/Crossfire. Since the interposer is actually a chip all on its own, some of these auxiliary functions can actually be built into the interposer itself (the "active interposer" concept), with the biggest obstacle being getting the heat out since the interposer has other chips stacked on top of it...)
220V users obviously have things a little easier here (~twice the capacity per circuit). This cuts the overall efficiency improvement necessary down to 4x.
There will also be some efficiency improvements on the software side. For starters we can cut quality a bit (that's 4K/60fps ultra), foveate rendering, and other software magic. Speeds don't scale perfectly negatively with increased resolution so you will get some savings there. Assume some speedup from DX12/Vulkan too.
Oh and since this is 16K/144 Hz per eye I guess there's an implicit assumption here that we can make NVIDIA-style Multi Viewport Rendering work at near-100% efficiency, otherwise that's another factor of (up to) 2x that needs to be accounted for.
So let's say that we need a 4x improvement in overall hardware/software efficiency. Let's say, something like a doubling in GPU throughput-per-watt and to double software efficiency before an absolutely state-of-the-art rig could even feasibly do this task on its own dedicated 220V/15A circuit. And there's a few fairly optimistic assumptions built into that 4x number.
Take your best shot at how long it will take to quadruple efficiency in the post-Dennard era. Let's say 20-30 years, unless there's a massive breakthrough in materials science or optical computing or something.
[0] https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1080_...
Actually, in Europe, you get 220V/25A from normal wall sockets, or, for stuff like washing machines or stoves, you get up to 400V/63A.
With that – roughly 3.5kW from a normal socket, or 25kW if you use a socket for a large appliance, compared to the 1.4kW available in the US – you can easily power a fuckton more.
As far as I can see it's either colo or run a 220V circuit, which is a bit of a non-starter for a hobby project.
Heck, you can even get uncomfortably close to maxing out a circuit with a single box. SLI 1080 Tis = 540W, plus 250W for an OC'd Intel HEDT processor is 800 watts inside the box, or 1000W at the wall (8.25A of 12A continuous). Hope your wife doesn't start the hairdryer while you're gaming...
This all works because you have between one and three 400V/63A triphase circuits per housing unit, but if your US circuits are this tiny...
As a line might be hardwired to multiple sockets, or to an appliance, or might be connected via a CEE plug instead of Schuko, it might support more than 16A.
If you build a large-scale GPU Cluster, you won’t connect them all to a single socket, but each to a separate socket, although they might be on the same line. That’s where the 25A can be useful.
690 V is industrial (this the logical progression dictated by star-delta connection of motors), and not used in residential contexts. 125 A CEE is the largest standard size. Beyond that other connectors are used. Industrial use also sees other connectors / kinds of connections for higher voltages (up to medium voltage [couple kV]).
This is really the fundamental problem with saying "X ppi is enough for the eye" or "you can't see faster than X Hz". This is a digital approach to an analog system. Your brain is tuned to identify resolution in the center of your vision and movement in the periphery, and has all kinds of "special case" circuitry to react quickly when necessary.
Classic example, fighter pilots are capable of recognizing silhouettes flashed on a screen at some ungodly rates. But if you were, say, playing a game, 100 Hz is very smooth already (even CounterStrike), and you might not be able to visually distinguish that from 144 Hz.
It's like a psychovisual model with a temporal aspect incorporated.
The eye clearly has frequency limitations, so we just need to find the sampling strategy which accurately codes all perceptible frequencies.
In an ideal world, we would also be using hexagonal grids rather than square, which gives a resolution increase for the same area and a given pixel size, and also reduces off-axis aliasing.
As for hexagonal grids, there are a number of complications introduced by that, especially when it comes to using general algorithms based on regular 2D euclidean space.
Agreed about the complications of the hexagonal grid. We make implicit assumptions everywhere. Though I'm not sure they are insurmountable for some purposes; with graphics APIs like OpenGL, texture coordinates and samplers could be using a hexagonal grid relatively transparently. That would allow use of hexagonal displays even if the texture data is square.
The eye clearly has frequency limitations
While you certainly could use 16K displays for VR, I think it would be entirely unnecessary.
Err what? Of course they do!
My current 16:9 24" 4k display in portrait is too tall---have to move my head up and down a lot, and too skinny---hard to put two windows side by side comfortably. Have to fiddle with overlapping windows a lot. In landscape it would be too short, and I rarely watch movies on my desktop. One movie trailer a quarter approximately, so optimizing for that use case is absurd.
I would prefer a fatter 16:10 instead, and was happy with the 22" 1920x1200 that was replaced, though I love the increased resolution of the new monitor.
Edit: The rendered viewport per monitor is 5120x2880 (=2560x1440 Retina), so everything is sharp.
In fact my old monitor was smaller (22") and I liked it better except for the lower resolution. Why? Because I could tile two windows side by side in portait. Docs and code, no fiddling, very little scrolling, and very productive.
I guess I could ignore the top few inches of my monitor but that is somewhat of a pain because of what happens on window maximization.
Windows has lousy built-in tools for laying out windows. Try AquaSnap, or script something with AutoHotKey.
This seems like it will be possible one day.
My predictions aren't holding up [1]. I feel too many customers are satisfied with small form-factor and/or are tolerant of using multiple displays and suffering the inconvenience of bezels.
I'm not saying those are the only two options, I'm just curious what the downside is, specifically with productivity use. I feel like if I had a 50" concave display, I'd want my window manager to have some kind of logical organization similar to what I'd have with multiple monitors anyway.
But 8K/UHD-2 might take a while, since the marginal gain from such a resolution is probably not going to be huge for your typical TV viewing experience, and it would cause problems at many levels.
I have never understood what's the point of investing in such an expensive and new tech for videos for YouTube, most of the tech channel videos on YouTube will be non relevant really fast so it's not like you are future proofing...in two years from now still very few people will have 8k screens, and cameras will cost at least 50% less. Storing 8k videos will increase storage costs and processing bandwidth too.
- Cropping can be more flexible.
- Higher resolution = better IQ at lower resolution than sensor at that resolution, because sensors have a bayer pattern.
these people are professional content makers. they make their living from buying/receiving gear and reviewing it on youtube. this isn't a hobby. youtube is big media now. very, very big. "just because you don't use it doesn't mean nobody else does."
if a film editor, or professional photographer, or rich guy, or web developer, or options trader, or magazine designer, or whoever is in the market for an 8k display and/or camera rig, they go to youtube and watch the reviews.
but in general... you know we used to have 640x480 screens and 20MB hard drives, right? that was 'normal', and 1024x768 seemed excessive. welcome to the forever now.
(And I'm sure that for a professional video person, it's not just "use it like your old camera, but it produces better output." There will no doubt be subtle changes needed for the best output, whether lighting, scenery, filters, workflow, or what have you.)
We programmers just happen to be really fortunate that the hot new tools we want to learn are usually free.
Which is unfortunate.
The article points out the challenge though, all that memory that needs to from where it is into the pixels on the screen at a rate fast enough to not annoy you with repaint lag. Which reminds me that the real 'winner' in this space will be if someone can put 300 ppi where ever you are looking in a larger field of view and leave the rest at 100ppi and 50ppi.
I'd bet against it being achieved before GPU's have caught up with the requirements though.
Frankly, 8k at 31.5" is kinda pointless unless you have eyes of an eagle or are working glued to your monitor. As a tech demo from Dell, it's cool!
I imagine 8k is not the easiest to use on Windows and there do not seem to be any Apple products with the right amount of power to use a display like this correctly (i.e. With the same power you'd put on a custom machine running Windows.).
Linux is the real pain - most things only support integer scaling for UI elements and 200%/2x makes it feel a bit big to me.
That’s actually only the fault of Gnome.
All KDE and Qt programs support fractional DPI scaling – with a different ratio per screen.
Don’t blame all Linux programs if it’s only Gnome that’s broken.
The environment variable used is
QT_SCREEN_SCALE_FACTORS=DisplayPort-2=1.75;HDMI-A-0=1.08;However, I'm not sure what macOS/OSX do, but it feels like they simply render whatever the app wants to render using many pixels instead of one. On Windows it feels like each app has the full control so each one needs to be updated to handle the DPI settings.
Where does your 800ppi claim come from?
If you aren't holding your face up to your monitor, you don't need 800dpi. The original iPhone "retina" display was 326ppi and that was assuming a 12 inch viewing distance. For monitor viewing distance, you'll be 20 or more inches away, making 280ppi plenty for the vast majority of people.
At a normal viewing distance a 100dpi monitor is already decent. A UHD monitor is just great. You have to get very close to see individual pixels. I'd say doing AA is no longer required even then.
8K seems overkill for most purposes. Sure, there is a niche that can take advantage of it, but I don't see advantages for the mass market.
Same as with SACD, CD tech is simply good enough for pretty much everyone so SACD never took off.
I write this despite being a high dpi junkie, I bought a ViewSonic VP2290b (IBM T221 clone, 3840x2400 22") back in 2006 and dealt with a huge hassle of 4 DPI inputs for years.
No it isn't.
And as another poster pointed out, the blurriness of the screen may compound with the natural blurriness of uncorrectable vision loss.
AA not only is still required on my 27" 4k display, but in terms of clarity AA fonts are still MARKEDLY inferior to fonts which are optimized for non-AA rasterization (like Terminus).
I'm guessing that for monitors <= 30" 8K will be the endgame, at least for me.
In any case, I'm super happy to see the resolution race back in full force, it took the industry quite a while to recover from the CRT->LCD switch. It was about time we moved past 1080p as the standard.
It's also amusing that so-called "high definition" is not so high anymore, and SD not so standard. What will 8K be? Ultra HD Alpha Plus?
I just wish Apple had made a retina Cinema display and left it at that (they have the tech, it's in the 5K iMac).
I appreciate the author saying 'maybe' because I can't, off the top of my head, understand why 16K would add any benefit over 8K... Aren't we at 'retina' with 4/8K anyway?
EDIT: at a 32" resolution...
Same applies to desktop monitors and phones. The reason phone screens pushed high DPI first was because you were much closer to the screen to make it fill your view.
Back then, 4K display was still way too expensive. Although I could buy one, I didn't think spending half the price of good laptop computer for a 4K display doesn't worth the money. I was wrong.
4K display greatly improved my productivity. I should have purchased a few of them already.
It's a pity that most operating systems aren't designed for huge screens - Windows' start menu is always in the bottom-left corner, for example.
This year 2 27" 4k IPSs for ~£350 each (both LGs, the first one a better model but cheaper because pre-Brexit-vote).
Prices have gone up from my POV.
In answer to your question: https://www.overclockers.co.uk/monitors/by-type/4k-ultra-hd?... -- under £300 for 24" 4k IPS. OCUK is a UK subsidiary of Caseking (de), so I'd be surprised if similar prices weren't available across Europe.
In all the above experience, I've found Acer to be very good (3 monitors, 2 excellent and 1 with pixel defects), Benq to be poor (multiple returns before I gave up and bought something cheaper) and LG to be absolutely flawless.
I wish Dell would produce 8K monitors in a much larger format, like 48".
I can't say anything about windows 10's support (The highest I have is 1080p)
Multiple monitor support on Windows 10 I've also found to be very dependent on your graphics drivers and hardware configuration, as it always has been.
Linux support is spottier, especially when using multiple displays of varying DPIs, but even Gnome works pretty well on a HiDPI laptop these days.
Actually, Gnome is the only one that doesn’t do proper HiDPI support. It only supports 96 and 192dpi - not any other ratio.
In comparison, Qt supports different ratios for every screen, all ratios specified as float – you can even scale up, down, whatever.
The environment variable used is
QT_SCREEN_SCALE_FACTORS=DisplayPort-2=1.75;HDMI-A-0=1.08;Qt’s had QT_SCREEN_SCALE_FACTORS=DisplayPort-2=1.75;HDMI-A-0=1.08; for years, but GTK still only supports one global scale factor, and that’s limited to 96dpi or 192dpi.
So JavaFX only scales to these, while Java Swing scales perfectly (but still only handles one global scale factor)
You can even drag a window from a HiDPI monitor to an old fashioned one and it will do what's right.
If it does, you're doing something wrong.
(It does improve legibility of small text, I'll grant you that)
Anyway, on iMac Apple doubled the resolution (quadrupled the number of pixels) when they introduced the "retina" versions so there was no font size change.
I picked 44" as the optimal size for me by standing 3 feet in front of all the 4K TVs at my local warehouse store and seeing what sized TV let me see the entire screen without swiveling my head - 48" was just a little too big.
My 4K screen works really well in Windows 10 other than a couple of old apps that don't to font scaling for their title bar correctly. I have the TV scaling at 125% and the laptop scaling at 250%. I'm not sure what an 8K screen would do for me as a programmer.
I have two Samsung 4k monitors and they are unbearable to use with HDMI at 4k, I had to get a new video card to drive them both over display port.
One thing I also did before I bought the exact model was do a little research -- the TV I bought has 1 HDMI port (out of the 5 available on it) that runs at 60Hz instead of 30Hz like the other 4 ports. I use that port for my laptop. I also read online that that port also skips the "Ultra HD engine" in the TV but I can't find that in the tech specs.
Btw, while I was looking for the tech specs I realized I have a 43" TV not a 44" (https://www.vizio.com/m43c1.html)
Maybe you wouldn't be stuck with monochrome bitmap fonts for good clarity. AA on 4k is still quite uncomfortable in comparison.
I never understand why people say things like this. Why not just throw more pixels at text rendered in a readable font? Does DPI scaling not work on their operating systems?
As my monitor resolutions have gotten larger, my text on screen hasn't become smaller at all. I simply go with a higher fidelity. Isn't that what most people do?
Also, do you know what a larger than 32" display is going to do to your poor neck?
HN's skwirl, right as I posted this: (27", 163ppi) https://news.ycombinator.com/item?id=13949827
I don't think anyone is beating https://amzn.com/dp/B00PC9HFO8 at $540.
If you found this info useful, this is a referral link you could choose of your own free will (no pressure from me!): http://amzn.to/2nvx9XP
Dual 16k @ 144Hz
Granted, phones are ahead of desktop displays in pixel density, and that seems more applicable to VR displays.
Which, incidentally, you could do now that you are in a VR, whenever you feel like.
In most cases, I'd guess that most people mean 3840×2160.
8K is 4320p. UHD 8K is 7680×4320.