If anyone reading this is interested in throwing caution to the wind and doing a hardware startup: I'd love a 24" eInk display or two, connected via USB. Ideally usable in either portrait or landscape orientation. Sure, the refresh rate would be super slow, but I'd use it for documentation and other reference material while actually coding on my macbook or one of my external LCD displays.
I really wish there were more screens like the Pixel Qi. If it was an option for my laptop I'd buy it right away.
Pixel Qi's big problem as I see it is that they can't seem to get their production up to speed.
The few (two?) options they have are so outdated they really only make sense on the OLPCs. Except of course that they are rather pricy at 275 USD.
And if you're going to use the screen entirely in monochrome anyway, what you really need is just the LCD stripped of the color filter and the backlight replaced with a simple mirror. Should be much cheaper than what PixelQi is doing.
I'm even tempted to try this as a DIY project at some point. That would be a nice monitor for coding on.
I was watching a bunch of YouTube reviews of some of the current generation of ereaders. And what struck me the most is how good the refresh rate was on some of these devices.
In the some of videos they show how even pinch zooming and panning around is possible, which I think is really impressive.
Surely it's not the rate of an LCD but I bet it would be near usable for coding.
I guess what I would miss the most is color syntax highlighting.
I'm curious, since you mention USB, if you've done any calculations on USB vs Eink bandwidth?
Would USB2.0 have enough to keep up with 24" worth of "pixels" even at Eink's relative low refresh rate? Or would USB3.0?
Let's say it's 24" at twice the resolution of 1920x1200 or four times the pixels, with 8 bit grayscale (because this is a fancy e-ink display, after all). USB 3.0 can do 5 Gbps which means 67 Hz would be an upper bound.
>USB 3.0 can do 5 Gbps which means 67 Hz would be an upper bound.
That's assuming that it's uncompressed, and if you're talking about using it largely for text, not compressing would just be crazy. For that kind of input, you can easily get 10x compression ratios losslessly. And at that resolution you can get away with extremely aggressive lossy compression, and ratios on the order of 50x are totally plausible. There's no reason for bandwidth to be a problem here.
No need to use interframe compression. Just JPEG will get you around 50x ratios with virtually no discernible artifacts at pixel densities that high (for mostly-text input). And there's no slow-motion about it; the whole point of compressing is to allow for very high framerates. If you can't get the E Ink to push 30hz, then there's probably not much reason to use lossy compression anyway. If you can get over 30hz, then lossy compression will keep you from totally saturating the USB 3 bus.
There are eInk specific things that a graphics controller needs to know about, like flashing the screen (or a region) after an update to eliminate ghosting.
So it would probably make more sense for the communication between the host and the panel to be at a higher level than a bitmap of the entire screen - something more like the Windows GDI, Display Postscript or Xlib. Sure, you'd occasionally need to blit in the whole screen, but you could limit the number of pixels to update, and limit the number of updates per second.
I'd much rather have a display that doesn't suck up all my USB bandwidth. If you are going to do that, you might as well use the Thunderbolt port. But then you just have a big, expensive grayscale screen with a low refresh rate.
That's nice in theory, and would have worked in 1995, perhaps even as late as 2003. But nowadays, almost every toolkit on almost every platform does rendering to an off-screen bitmap, and then copies (or "composes") that bitmap onto the screen - so even if only one pixel is changed, that's often lost in the pipeline.
However, something like VNC's and RDP's "tile" coding (that is, remembering for every non-overlapping n by n rectangle whether its content has changed since last update) would still work.
Been there, the project stalled when we couldn't get any panels of that size so it had to be custom made, and considering any panels bigger than the standard 6 inches were already expensive the price of these other panels was going to be plain crazy, and we needed no less than 1,000 people to signup for it.
One point against matte screens that you need to consider...
Matte screen coatings eliminate reflections by diffusing them over the entire field of view. As the total brightness of your room increases, matte screens diffuse all of it evenly, eliminating reflections but also losing contrast ratio.
As the total ambient light increases, your screen's contrast ratio can drop from 800:1 to 50:1 (or worse). Any contrast ratio lower than about 300:1 can cause eye strain.
End result: with a typical screen of about 200cd/m^2 a matte coating is totally unusable if your room is brighter than about 1000 lux but falls below 300:1 contrast around 450 lux (typical office lux is between 350 and 500).
In summary... you need to know your environment. Matte screens work great in dark rooms with lots of spot lights to cause reflections. Glossy screens work great in bright ambient conditions with even lighting. Semi matte works well if you need a balance.
No, I don't have data. I'm running off memory from when I worked at a printing and display company (8 years ago so memory could be faulty).
The point is actually related to the way your eye adjusts to ambient brightness levels. It's not about the contrast on its own but about the contrast, the number of brightness levels of the display and the brightness level with respect to the ambient conditions.
The rule of thumb was that less than 300:1 starts to cause eye strain on an 8 bits per channel (256 levels) gamma power curve (non-linear) device when the device had a limited maximum brightness output (for LCDs at the time that was 200cd/m^2) and the room was bright (for an office).
Kindles are 4 bits per channel (16 levels) linear and their contrast ratio isn't a complete lie (unlike most LCD makers contrast ratios who report only the absolute best case). But that's not the biggest reason why they're different. Kindles don't emit light, they reflect it (not including the Kindle White and Fire). This means that they maintain their brightness level with respect to the environment much better.
Of course, Kindles start to suffer at low light (for similar reasons -- they lose their apparent contrast level).
It’s just opinions. There isn’t any answer there. (I’m not even sure there is one. The difference may be too small to be able to say that in general, one or the other is better. Maybe it’s really down to opinion.)
Matte is hands down superior in almost all situations except games and movies.
People who own computers that don't give them the option often formulate excuses or justifications, but I dont think anyone with extensive experience with both would ever choose glossy. Matte wins hands down with everyone i know.
Agree, looking at a matte screen is much easier on my eyes than any glossy screen has been. Everything looks smoother, the colors softer, and zero glare.
Agreed on all counts, in addition, I think matte allows you to lower the screen brightness way lower than glossy.
You have to maintain a certain brightness to overcome the reflections on a glossy screen, and I feel brightness is one of the biggest factors in causing eye strain.
I'm not in any way an expert, but ergonomics of a workplace is not just opinions. [It is actually important. Poor workplace ergonomics can cause adverse health effects. And I'd guess, that a glare-type screens is a joke in the eyes of an ergonomics export.]
Whatever idiosyncratic stuff you do that helps you get into the zone for a long coding session is perfectly fine.
Shortly after landing my first job out of college, I picked up Orbital's first two albums and listened to them nonstop with a bit of Autechre and Aphex Twin while coding. After I left that job, I misplaced the Orbital albums, but kept listening to Autechre and Aphex Twin while doing other stuff.
Last week I picked up Orbital 1 and 2 again, and put them on while coding. BAM right into the zone. Probably won't work for you, but it works for me.
> Is a glossy or matte LCD screen better for long coding sessions?
Better for you than screen choice: standing desk - you will code better and faster
Better for you than screen choice: inject 2 minutes of getting up and walking around every 15-20 minutes of your "long coding" session - you will code better and faster
Better for you than screen choice: start your long coding session after 30 minutes of moderate exercise (walk or bike ride) - you will code better and faster
I've recently had the pleasure of purchasing the 30" matte Dell U3011 IPS and the 27" semi-glossy Samsung S27B970 PLS.
There are pros and cons to both. The anti-glare coating on the Dell makes black text on white background look absolutely rubbish, fuzzy and hard to read. You can sort of get used to it, but every time you switch back from using the semi-glossy Macbook it looks terrible. Black themes don't suffer from these problems nearly as much.
With the Samsung, colours look much more vibrant and text is sharper. Reflections are very much an issue, especially with black themes. In very bright conditions with visible light sources the screen becomes almost unusable.
Overall, if I could control my lighting, I would go with semi-glossy. Otherwise I would suffer the Dell's anti-glare.
I find that both fuzzy text and reflections lead to eye strain and headaches. If they could tone down the anti-glare strength (but still leave some AG coating) I would probably go for that in all circumstances.
On an unrelated note, I do prefer the 30" pixel pitch (.25 mm) to the 27" (.23 mm). At the given viewing distance the .25mm pixel pitch more closely matches my Macbook's, meaning you can use the same font sizes for both.
The mention of "wide screen" in the second most voted comment reminded me of my biggest pet peeves for monitors in 2012 - the disappearance of 16:10 monitors.
Get a big 16:9 one, and use a wide, vertical taskbar. It'll cut you down to 16:10 and you get the joys of using a wide, vertical taskbar: http://snag.gy/0pZP1.jpg
I have both and well, glossy blows: not only it reflects nearly everything around it more than a CRT would but its also a dirt magnet.
Too bad these days a lot of laptops, specially mid to highend models come with glossy screens. Unless you're working in a cave you'll get reflections all the time.
I've noticed that at least some of the MacBook Air screens -- and the Retina MBP screen? -- are glossy, but with an anti-reflection coating. I've been wondering for some time why Apple hasn't been using using AR coatings on their glossy screens. These coatings used to be standard on high-end CRT monitors.
Anyway, I'm glad to see Apple coming around on this. Interesting, too, that they don't seem to be talking about it at all. Maybe they're worried that advertising some of their screens as AR-coated will draw attention to the fact that others, such as their 27" Thunderbolt monitor, are not. (I tried to use one of those for a few days, and gave up; the reflections were driving me nuts.)
I came here to say this. The glossy MBP 13" screen has driven me nuts, I did the whole online petition dance, and it turns out the MBA 13" isn't half as bad. Dropping the glossy black bezel helps, too. The only thing better would be a matte, black bezel (to hide the camera hole).
I think the general quality of the screen is far more important than glossy or matte. It's like asking if a 2 liter engine is better than a 4 liter one. In what chassis? At what weight?
The latest Apple displays I'm using that are glossy easily trump my 30" Cinema Display in terms of being pleasant to read. They might be better in matte but the underlying display quality and technology has proven to be far more important than the coating.
Did everybody get so busy that they forgot how to find things out for themselves?
Try both. Figure out what variables you have to control for and control for them. If you can't tell, it probably doesn't matter.
Obviously some people prefer one or the other. If glossy gives you headaches and matte doesn't, does it matter how many people prefer glossy? So why ask them when you can get the data yourself?
If you think the difference affects your health and performance, isn't it worth the experiment? If it doesn't, why bother asking?
Asking is cheaper than performing the experiment. The asker probably can’t afford buying one matte and one glossy monitor just to try them out. They also might not know anybody else with the other kind of monitor to compare with. That is especially likely if they’re buying a laptop with a built-in display – it’s unlikely that they’ll know two people with that model of laptop, with the same size of display, but one display matte and the other glossy.
If glossy gives you headaches and matte doesn't, does it matter how many people prefer glossy?
-- Some experiments are non-reversible
Damage to your eyes, for example. This is one area that asking for peer comps should not be crtiticised as just being lazy. Pay it forward, and all that. IMHO.
