A 600 dpi A4 sized tablet would be lovely. It would be a lot harder to get to that effective dot pitch with colour displays.
300 dpi should be the entry level.
A 600 dpi A4 sized tablet would be lovely. It would be a lot harder to get to that effective dot pitch with colour displays.
300 dpi should be the entry level.
(The original K&R style had the opening brace on a separate line for functions, but I believe this is due to the original syntax for declaring argument types - if you declare them on separate lines, like variables, a distinct { clearly ends the argument block and starts the function body.)
http://www.softwarepreservation.org/projects/ALGOL/report/Al...
http://www.algol60.org/reports/algol60_rr.pdf
(Although technically this was actually syntax, not just syntax highlighting, since that's how the reference language distinguished between keywords and identically spelled identifiers.)
Response time is a pretty critical metric for me when it comes to coding. I don’t want to have to wait a couple hundred milliseconds to scroll or switch files, so I’m not convinced coding is a great fit for the existing tech. Code isn’t read top to bottom like a novel.
That said, hell yes I want less eye strain when coding. I’m sure we can all relate to that.
I should probably try something radically different again in another different direction soon. There are generally somewhat better options available for semantic highlighting now than there were five or ten years ago.
That's already as good as a mid-grade laserprinter output, and for text is pretty much entirely sufficient, especially in monochrome.
Other issues still remain: some ghosting (at higher refresh / lower quality), limited greyscale (4--16 shades in most cases), and generally far lower refresh rates than emissive displays (1--8 Hz seems fairly typical). Yes, you can watch monochrome video, if you want. It's in the "sufficient for a general overview", but not high quality.
The place this shows up most especially isn't in video, per se, but on web pages with animations, whether of icons, graphics/charts, or embedded videos. The result is exceedingly distracting at high display quality settings, with a slow, flashing refresh behaviour.
The current typical resolution for 10–13″ tablets (A4 is about 14.3″) is 1404×1872, which is 2,628,288 pixels. You’re asking for 13× as many pixels, and 3× as an entry level. However you arrange it, it’s going to use a lot more power and probably need more memory and a more powerful CPU/GPU.
Call it 400 dpi? 450, maybe? It doesn’t seem too farfetched to me.
Also, there's a huge difference in bandwidth requirements if the display is monochrome. Consider the framebuffer needed for a conventional 8-bit colour depth display at 5000 by 7000 versus a monochrome, two colour, or a grayscale 16 shade e-ink panel.
24 bit RGB: 105 MB
5 bit grayscale: 21.875 MB
2 bit: 8.75 MB
mono: 4.375 MB
Pair that with a low refresh rate and partial screen updates and it's suddenly not so crazy.Partial screen updates can go much faster than that; reMarkable 2’s total latency from marker to screen update is around 23ms if I recall correctly, and I suspect (without confirmation) that it’s updating the screen at 200Hz, the sample rate of the marker.
In theory, yes, you can go down to 1- or 4-bit colour (and at 600dpi I think going hard monochrome and depending on dithering would be quite acceptable), but in practice software stacks are rather attached to 24-bit and don’t tend to like going below 8-bit. Especially general-purpose stacks—if you build something yourself from the ground up (whatever that means), then sure, you can generally choose your own path. There are certainly paths that make better use of meagre resources, but experience says they’re seldom taken.
I think you’re probably underestimating how much power pushing that many pixels and that much screen bandwidth takes. Such a device is feasible (provided they can manufacture a large enough panel with a tiny enough dot pitch), but not straightforward, and will have some significant tradeoffs. I’m not surprised no one’s trying anything even close to it.
You’re talking about a general purpose monitor attached to a PC. How is bandwidth an issue there? The very same computer can push 120 Hz (or more) to higher res screens than the Dasung. It seems very clear to me that the problem is that the e-ink cells physically can’t react in time to show the next frame. The ghosting means they’re pushing it to the limit. If you’re talking about the controller within the display, well that argument doesn’t track. Controllers capable of pushing pixels that quickly work perfectly fine in conventional LCDs, even power constrained ones like an iPad Pro’s. Why not use those?
E-readers have been monochrome for over 20 years, not to mention every other limited-colour piece of hardware. Has that not been enough time for software to adapt and libraries to mature? To be clear, I’m not dreaming of a device to run X11, Firefox, and Doom (like the PineNote people seem to want). I want self-rewriting paper. Of course I want software tailored to that experience, not something off the shelf.
If I am making any ridiculous requests, it’s 600 dpi screens at that size. I’m sure there are a thousand reasons why that’s a hard engineering problem.
Even high-quality refresh is fast, though there is a prominent transition.
With e-ink, there's a distinct difference between paginated navigation and scroll-based, with pagination being strongly preferred.
For web browsing, I recommend EInkBro, which provides paginated navigation as a default. For any longer-form Web reading, that's my preference. It's superior even to Pocket's exceedingly poorly-considered pagination mode.
The other alternative is to print-to-PDF and display that within an e-book reader. Unfortunately, many websites format poorly in that mode, with text being cut at pagination breaks, or omitted entirely.