Did they fix font rendering? if they did I will go back to windows. If they didn't. No thank you.
Just fix fucking font rendering, I did use Windows machine as my workstation for 4 years, But I had to switch to Ubuntu, because of ridiculous font rendering they introduced in DirectWrite. I don't have money to pay for new monitor every 2 years, and I am quite happy with my dual monitor 27 inch 1080p until windows 10 changed font rendering to DirectWrite. It is fucking disaster in low dpi monitor.
As always Microsoft ignored their current users in pursue of new users. I haven't seen this attitude from any other company. I haven't seen apple make font rendering on old MacBook worse to force people to switch to new MacBooks. Or I haven't seen something similar from Google.
This is fucking total disaster.
As someone who reads text for whole day (programmer) I care about my eyes and it fucking hurts my eye to see font rendering in new UI (UWP) and Edge.
p.s. Old/win32 application do use old rendering engine and are acceptable.
Not the parent poster, but basically, they dropped RGB subpixel rendering, and now use plain anti-aliasing for all font sizes. They may have changed hinting and anti-aliasing parameters, too. No difference on a HiDPI screen, since you don't need subpixels there, but on standard definition screens many find it looks a bit blurrier (me included).
Supposed benefits are that it is easier to render text on colored backgrounds (you can just render with alpha to a buffer and combine it with the background, with RGB subpixel rendering you have to consider the background while drawing), and you can scale text without color artifacts.
To be honest, we have plenty of computing power today, and our GPUs are underutilized most of the time anyway. I wonder why we can't just have RGB subpixel aware rendering of everything (not just text, but also lines and polygons) implemented as a kind of shader pass. Render everything at 3x horizontal resolution, and merge using an appropriate shader. And if you have a HiDPI screen, just drop the shader.
It's actually not possible to accurately composite subpixel AA using the GPU since you can't configure blends per-channel. To do it accurately, you need three different alpha channels.
It is actually possible to do using a feature called dual source blending, which unfortunately is so rarely used that drivers tend to fail to implement it properly. But it does give you the possibility to use 3+ alpha channels with full hardware acceleration, at some memory cost.
Though since DPIs are going up and 3 alpha channels is such an unusual thing, I guess they just figured it wasn't worth the trouble.
Yeah I noticed this the other day. I couldn't believe the fonts looked so terrible. This is coming from a Fedora user where we typically put up with lots of complaints about fonts.
Exactly. BTW It is not the fonts makes me crazy. It is much broader than that. It is culture behind it. Lock people up, then you have freedom to what do you want to do.
If there was a serious competitor in desktop they wouldn't dare to ignore their current user in pursue of new users.
Using Windows at work, and Fedora at home, the font rendering on Windows always drives me crazy. My monitor at work is kind of awful, but the font rendering just doesn't have to be as bad as it is.
The font rendering was my biggest beef with Windows, I like them smooth while most Windows users seem to prefer sharp. I used to use GDI++, then MacType, for over 5 years until two years ago I switched over to OSX. I have been quite happy since then.
I feel like there is this secret turf battle between designers and developers, constantly trying to one-up each other, and ultimately take controls away from each other.
Developers come up with frameworks and systems that take controls away from designers.
Then designers come up with frameworks and systems that take some controls back.
Back and forth. Back and forth.
Taking control away may not be the primary goal, and it may even be a byproduct. Nonetheless, it's pretty amusing to observe the trend play out.
Then you watch how the rest of the world design and code their sites at fraction of the cost with better result. I don't know how we get away with this nonsense in America.
Consistent design guidelines coupled with cross platform libraries is what Google did right about Material design. It seems like both Apple and Microsoft have failed to properly build this before and I have always disliked iOS and Metro designs for this reason.
I also don't like its aesthetic, plus I find it lacking in discoverability and one-handed usability on phones. But consistency of execution is the one thing it gets right, as GP says.
It's hard to see what this is actually going to look like, but if it gives their UI back some more "definition" (or in UX jargon probably discoverability of affordances) I'm looking forward to it.
I know skeuomorphism (as in: imitating physical materials, not in: using metaphors from the physical world) is out, but considering we have these awesome hidpi, high contrast screens (and cameras and light and motion sensors), why can't we have convincing different materials in our UIs?
I was really disappointed that "Holo" didn't look like I imagined holograms, and "Material" doesn't look more like paper or any other material. I'm disappointed that the blur effect in macOS is so unphysical and illogical (sometimes blurring the background, sometimes the titlebar, ...).
I wish MS or any future UI would have different types of (subtle!) material effects. Frosted glass, thick refractive glass, reflective surfaces, matte paper... not just "flat" or "blurred". And if pulled of correctly, it would not look gaudy, it would be as convincing an illusion as force touch is on MacBooks. Making me tilt my phone or laptop because I can't believe that there is not actually a piece of paper beneath my screen!
Edit: Yay, it seems they are legitimizing the blur effect that is available as an undocumented API now as "acrylic", and more materials are to come! But oh no, it is the same as macOS vibrancy (also with potentially illogical background and foreground modes) https://docs.microsoft.com/de-de/windows/uwp/style/acrylic
Skeuomorphism is coming back in a measured way — originally it was kind of a "this notepad app looks like a notepad" which ultimately seemed a bit absurd because a yellow-ruled paper texture doesn't really mean anything in a non-physical environment. It was more noise than signal.
...but now as you mentioned — these elements we're interacting with should follow some sort of universal constants and are picking up associations of physical reality that actually mean something. Elements move in from one direction and out from another at set speeds, light sources are coming from consistent places, animations radiate out from the source of interaction...
The video is all eye candy with quick cuts. It doesn't show anybody doing anything. The discussion indicates that Microsoft wants to bring back pens, since finger-pointing is so low-rez. They also have that new movable knob input device. Both of those give tablets more desktop-like attributes.
Part of the idea is that you can write on anything. Didn't Apple try that once about 20 years ago? I think it was called "Ink".
The video is amazingly inhumane. Things that presumably took many person-years to create are shown for such a small fraction of a second, and in such rapid succession, that it hurts my brain to watch.
I know that they are doing some fascinating work --- I own a Hololens, and I think some of their research work is potentially game-changing.
But who is the audience for a spot like this? Do non-devs care about their "design system"? Would devs not rather see something substantial, instinctively dismissing this kind of razzle-dazzle?
That seems to be the trend in videos these days. It drives me crazy when they show a car and it's 0.2s in a tunnel, 0.3s on a mountain and so on. You can watch a whole video and still don't known it looks. But together with music and sound effects it's cool.
- Long single-shot videos of someone doing something complicated. Boring, but useful. Typical use: how-to videos.
- History Channel style. Video of important stuff with occasional interpretive graphics and explanation. Tough to balance but can be done well. Expensive to produce. Tends to oversimplify.
- Classic industrial film. See anything from the Jam Handy Organization. The Internet Archive has hundreds of these films. Good at getting concepts across. Blah by modern standards. Narrated, but not much talking head time. Minimal animation of details. Many stock shots.
That field needs some design work. If the goal is to explain (not sell), the industrial film model, modernized, has potential. It takes more production effort than a blipvert or a PowerPoint video, but less than a History Channel video.
Top of the line in explanatory videos: "Victory through Air Power" (1943) [1] This was made by Disney for one purpose - to convince President Roosevelt to go with heavy land-based bombers as the main weapon in the Pacific war.
Classic Jam Handy: "Around the Corner", which explains how an automobile differential works.[1] (Skip the first three minutes.) If you watch this, you will understand how a differential works, and won't forget it.
An explanation of Microsoft's new UI with similar clarity would be helpful.
Fluent is "Modern Design Language 3.0" (aka "Metro 3.0"). It seems to be an evolution of UWP, not a brand new UI Kit, and the code demos shown show a very easy crossover from existing UWP XAML to add Fluent tools/design.
I am a designer, and I am not sure what most of this on the surface level is. I dug a little deeper and saw some pretty basic component library stuff. I'm not sure if this is just weirdly presented, I am behind the times, or both.
Am I the only one who just wants their desktop to look like Windows 95 for the next 50 years? Except maybe a bit more bit depth and DPI. This design is simply unsurpassed and everything new that comes out is a step back. There is one feature that I would like from modern Windows window manager though: I really like Windows 10's full-text search for settings.
I agree. Maybe I would take XP as the standard. But in the last 15 years I don't see much real improvement but just moving things around and making them look different.
I would really like an OS as simple to use as MacOS 9 but with a modern filesystem, full text search, and keyboard shortcuts for switching apps. That's it. That would be fantastic.
What made Classic MacOS so nice was that it was persistent. Each folder remembered where its window was, what size/shape, icon size settings, icon layout, sort/list view, coloured labels, etc.
Human spatial memory is a fantastic thing. It's so frustrating that we've decided to throw that all away with modern UIs that don't persist anything and windows that get reused (browsers). It's no wonder non-technical users struggle so much that they've more or less given up on desktop PCs.
I'm pretty sure Windows has had the ability to remember folder-specific settings since Windows 95. I don't know when window-sizing and location memory came into play, but it definitely worked in Windows 7+. Until Windows 7, if I remember correctly, folder-specific customization was all or nothing: every folder had individualized settings, or all folders were treated the same. The file-specific libraries (Documents, Videos, Pictures, etc) let you strike a balance between those two extremes (among other things).
It does, but it's not the default. The default in Windows is that you have a browser and opening folders reuses the same window. Classic Mac OS is not like this. Every folder has a 1 to 1 relationship with the window showing its contents. A folder cannot be open twice, and a window cannot change what folder is displayed inside it. Applications worked like this too. An application could only be open once. Trying to open it again would switch to it.
The goal of all this is to help the user build spatial memory for everything on the computer. By not having applications, documents, or folders move around the screen of their own volition the user is better able to keep track of them.
This mirrors a real-life desk where objects like papers, pens, folders, and your calculator can be found right where you left them. If every time you sat down at your desk those objects were rearranged you'd probably complain to the cleaning staff.
Am I the only person who thinks that the UI peaked at Windows2000?
It seems to me that in Win2000 the entire 2D interface/look was perfect, it refined the Win95/98 interface. Sharp lines allowing for easy visual separation between elements. Excellent colour scheme. Buttons were buttons and an obvious visual clue that they were 'clickable'.
"Clicking" on the little "x" to close the application window is near impossible with anything other than the mouse. We can't pretend interfaces like touch are gimmicks anymore. They're pushing a decade now in mass market.
While on one hand I agree that yes, touch is a major component in how people interact with computers, I feel like we are being dragged down to the lowest common denominator of user interface.
I cannot think of legitimate business case to implement the Metro UI/touch interfaces on servers.
Studies have long shown that what is good for the touch gander is good for the mouse goose. Fitt's Law has been known forever and touch makes considering it a requirement rather than just a good idea (bigger targets are easier to touch; bigger targets are easier to mouse to and click). The same applies to proper DPI awareness, in touch it is a requirement, for mouse users it has often been considered a secondary consideration.
Not every click in a business app or on a server needs to be a headshot to get work done. The business case should be clear: Fitt's Law is a productivity gain even for mouse users; DPI awareness is a good idea when even desktop screens are going 4k and beyond.
Yes, because one part of the demographics uses its fingers, we got to piss of the other part of the demographic who doesen't.
Man, i wish i could be that lazy as a coder.
Yes, boss - we kicked the ethernet support. Focus test shows a near majority of our customers comes in via WiFi.
I would be so ashamed on this inability to combine form and function.
When I first heard about this a few months ago, I was excited. I don't know why. Microsoft is cursed with terrible design and always will be. If they ever designed something nice I'm pretty sure the universe would have to unbalance itself to compensate.
I personally prefer the Windows 7 and Windows 10 interface to OSX. I really liked metro; I also prefer the windowing system in general with the snapping and other features. The designs of the surface studio, the surface book, and surface pro are much more compelling than anything I've seen out of Apple in recent memory. My goodness, that surface studio..
IMHO MS is a MUCH more exciting company ATM from a design standpoint; or from just about any other point I would stand.
I like the "Motion" section with a video showing the Xbox One guide menu. Anyone who owns an Xbox One knows the guide menu is laggy garbage with confusing menus. I don't trust a company to suddenly get this stuff right if they can't even provide a smooth and intuitive menu on a game console with 8 CPU cores.
Granted, it got a little better recently, but there's still so many improvements needed.
Microsoft has a history of screwing up every other OS releases. After being slight successful with window 10, It's their chance now to screw up again. https://news.ycombinator.com/item?id=4415403
UI Looks back to Aero/Glass effect of Windows Vista: I wish they had learnt their lesson.
66 comments
[ 3.1 ms ] story [ 107 ms ] thread> Most of the changes include subtle additions like blur effects, which is a type of design we saw back with Windows Vista
And so the wheel turns...
Just fix fucking font rendering, I did use Windows machine as my workstation for 4 years, But I had to switch to Ubuntu, because of ridiculous font rendering they introduced in DirectWrite. I don't have money to pay for new monitor every 2 years, and I am quite happy with my dual monitor 27 inch 1080p until windows 10 changed font rendering to DirectWrite. It is fucking disaster in low dpi monitor. As always Microsoft ignored their current users in pursue of new users. I haven't seen this attitude from any other company. I haven't seen apple make font rendering on old MacBook worse to force people to switch to new MacBooks. Or I haven't seen something similar from Google. This is fucking total disaster. As someone who reads text for whole day (programmer) I care about my eyes and it fucking hurts my eye to see font rendering in new UI (UWP) and Edge.
p.s. Old/win32 application do use old rendering engine and are acceptable.
Supposed benefits are that it is easier to render text on colored backgrounds (you can just render with alpha to a buffer and combine it with the background, with RGB subpixel rendering you have to consider the background while drawing), and you can scale text without color artifacts.
To be honest, we have plenty of computing power today, and our GPUs are underutilized most of the time anyway. I wonder why we can't just have RGB subpixel aware rendering of everything (not just text, but also lines and polygons) implemented as a kind of shader pass. Render everything at 3x horizontal resolution, and merge using an appropriate shader. And if you have a HiDPI screen, just drop the shader.
Though since DPIs are going up and 3 alpha channels is such an unusual thing, I guess they just figured it wasn't worth the trouble.
If there was a serious competitor in desktop they wouldn't dare to ignore their current user in pursue of new users.
There's nothing dealbreaking about the platform for me, well there's hardly even anything nagging about it.
I feel like there is this secret turf battle between designers and developers, constantly trying to one-up each other, and ultimately take controls away from each other.
Developers come up with frameworks and systems that take controls away from designers.
Then designers come up with frameworks and systems that take some controls back.
Back and forth. Back and forth.
Taking control away may not be the primary goal, and it may even be a byproduct. Nonetheless, it's pretty amusing to observe the trend play out.
Then you watch how the rest of the world design and code their sites at fraction of the cost with better result. I don't know how we get away with this nonsense in America.
Still you are right though about the cross platform thing
I know skeuomorphism (as in: imitating physical materials, not in: using metaphors from the physical world) is out, but considering we have these awesome hidpi, high contrast screens (and cameras and light and motion sensors), why can't we have convincing different materials in our UIs?
I was really disappointed that "Holo" didn't look like I imagined holograms, and "Material" doesn't look more like paper or any other material. I'm disappointed that the blur effect in macOS is so unphysical and illogical (sometimes blurring the background, sometimes the titlebar, ...).
I wish MS or any future UI would have different types of (subtle!) material effects. Frosted glass, thick refractive glass, reflective surfaces, matte paper... not just "flat" or "blurred". And if pulled of correctly, it would not look gaudy, it would be as convincing an illusion as force touch is on MacBooks. Making me tilt my phone or laptop because I can't believe that there is not actually a piece of paper beneath my screen!
Edit: Yay, it seems they are legitimizing the blur effect that is available as an undocumented API now as "acrylic", and more materials are to come! But oh no, it is the same as macOS vibrancy (also with potentially illogical background and foreground modes) https://docs.microsoft.com/de-de/windows/uwp/style/acrylic
...but now as you mentioned — these elements we're interacting with should follow some sort of universal constants and are picking up associations of physical reality that actually mean something. Elements move in from one direction and out from another at set speeds, light sources are coming from consistent places, animations radiate out from the source of interaction...
The video is all eye candy with quick cuts. It doesn't show anybody doing anything. The discussion indicates that Microsoft wants to bring back pens, since finger-pointing is so low-rez. They also have that new movable knob input device. Both of those give tablets more desktop-like attributes.
Part of the idea is that you can write on anything. Didn't Apple try that once about 20 years ago? I think it was called "Ink".
I know that they are doing some fascinating work --- I own a Hololens, and I think some of their research work is potentially game-changing.
But who is the audience for a spot like this? Do non-devs care about their "design system"? Would devs not rather see something substantial, instinctively dismissing this kind of razzle-dazzle?
- Blipverts, like this one. Pretty, but useless. Typical use: Ads, Kickstarter videos.
- Talking heads with PowerPoint. Boring. Typical use: online courses, TED talks.
- Long single-shot videos of someone doing something complicated. Boring, but useful. Typical use: how-to videos.
- History Channel style. Video of important stuff with occasional interpretive graphics and explanation. Tough to balance but can be done well. Expensive to produce. Tends to oversimplify.
- Classic industrial film. See anything from the Jam Handy Organization. The Internet Archive has hundreds of these films. Good at getting concepts across. Blah by modern standards. Narrated, but not much talking head time. Minimal animation of details. Many stock shots.
That field needs some design work. If the goal is to explain (not sell), the industrial film model, modernized, has potential. It takes more production effort than a blipvert or a PowerPoint video, but less than a History Channel video.
Top of the line in explanatory videos: "Victory through Air Power" (1943) [1] This was made by Disney for one purpose - to convince President Roosevelt to go with heavy land-based bombers as the main weapon in the Pacific war.
[1] https://archive.org/details/VictoryThroughAirPower
[1] https://archive.org/details/Aroundth1937
- "What Stops Them" - hydraulic brakes
- "Tough Friends" - automotive steel alloys
- "No Ghosts" - auto frame design for stiffness
- "Horsepower" - shows exactly what one horsepower is
- "Balance" - system integration of the automobile
- "Precisely so" - measurement and manufacturing precision
- "Streamlines" - air resistance
- "Down the gasoline trail" - from fuel tank to exhaust
All these are in the Internet Archive in the Prelinger collection of industrial films.
Chevrolet really wanted to explain how an automobile works. These films are much better at explaining than most of what we see today.
[video displays as just a black screen. no errors.]
Full disclosure: I am not a Designer.
What made Classic MacOS so nice was that it was persistent. Each folder remembered where its window was, what size/shape, icon size settings, icon layout, sort/list view, coloured labels, etc.
Human spatial memory is a fantastic thing. It's so frustrating that we've decided to throw that all away with modern UIs that don't persist anything and windows that get reused (browsers). It's no wonder non-technical users struggle so much that they've more or less given up on desktop PCs.
The goal of all this is to help the user build spatial memory for everything on the computer. By not having applications, documents, or folders move around the screen of their own volition the user is better able to keep track of them.
This mirrors a real-life desk where objects like papers, pens, folders, and your calculator can be found right where you left them. If every time you sat down at your desk those objects were rearranged you'd probably complain to the cleaning staff.
It seems to me that in Win2000 the entire 2D interface/look was perfect, it refined the Win95/98 interface. Sharp lines allowing for easy visual separation between elements. Excellent colour scheme. Buttons were buttons and an obvious visual clue that they were 'clickable'.
(Disclaimer: I use i3 exclusively.)
I cannot think of legitimate business case to implement the Metro UI/touch interfaces on servers.
Not every click in a business app or on a server needs to be a headshot to get work done. The business case should be clear: Fitt's Law is a productivity gain even for mouse users; DPI awareness is a good idea when even desktop screens are going 4k and beyond.
I would be so ashamed on this inability to combine form and function.
Judging by how many people actually use touch UIs for Windows, yes we can.
That feeling hasn't been since in any versions of windows since then.
I personally prefer the Windows 7 and Windows 10 interface to OSX. I really liked metro; I also prefer the windowing system in general with the snapping and other features. The designs of the surface studio, the surface book, and surface pro are much more compelling than anything I've seen out of Apple in recent memory. My goodness, that surface studio..
IMHO MS is a MUCH more exciting company ATM from a design standpoint; or from just about any other point I would stand.
I like the "Motion" section with a video showing the Xbox One guide menu. Anyone who owns an Xbox One knows the guide menu is laggy garbage with confusing menus. I don't trust a company to suddenly get this stuff right if they can't even provide a smooth and intuitive menu on a game console with 8 CPU cores.
Granted, it got a little better recently, but there's still so many improvements needed.
UI Looks back to Aero/Glass effect of Windows Vista: I wish they had learnt their lesson.