I'd like to take the opportunity to show you the trailer video of the Ultimate Hacking Keyboard, a high-end mechanical keyboard of which I'm the lead developer. Our keyboard is going to be kickstarted soon so you're more than welcome to share it, follow us and subscribe to our list to get notified when our campaign starts. See you on UltimateHackingKeyboard.com! Cheers!
geekhack has the best article every written on N-key rollover. The issue is more complicated than one might think.
I've started Coder Keymaps a long time ago to create a special keyboard mapping that's the best for me. That idea is to map Hungarian characters in a special way using the Windows key. Take the standard US layout, keep a Windows key pressed and press an alphanumeric key which produces an accented character on Hungarian keyboards and voila: the key will produce the relevant Hungarian character.
That was the basic idea but I went further about two years ago when I realized that hand travel distance is much longer than it's supposed to be in many cases. When writing code one's right hand must move often between the alphanumeric keypad and the navigational keypad. To alleviate this problem I decided to map the whole navigational block to the alphanumeric block through the Windows key.
I've used the above configuration with great pleasure and it improved my efficiency for almost two years. Unfortunately the X keyboard drivers must have been changed in the last two Ubuntu releases because my xmodmap keymaps stopped working. I knew it in the beginning that xmodmap is outdated and XKB is the future but it wasn't really urgent to port Coder Keymaps to XKB so I didn't do that.
Due to the pressing need to use my beloved mapping I've made some efforts and had some online chat with Sergey Udaltsov who is very knowledgable about XKB.
Long story short, it seems that it's impossible to create such an exotic keymap on Linux. Not that it's not possible to create it with XKB, but various GUI toolkits, such as GTK+ interpret the mappings in strange ways and the mapping wouldn't be consistent accross toolkits. I'm sure that this can be solved by modifying the X keyboard driver or the toolkits but as you may suppose it's a heroic work. Not only that, but this is an OS-specific problem and there are no standard solutions that truly work.
I finally decided to attack the problem differently by creating a keyboard hardware that has limitless power regarding remapping. It's actually not a new idea of mine, it's about two years old. The prototype is in development and it's very innovative in many ways. I've gathered a small, but knowledgable team and we're progressing rapidly. I wanted to have a working prototype by the end of this year but I'm not sure we get there in time because rapid prototyping is expensive and the delivery of rare electronic components take time to arrive to Hungary. But no matter how long it will take, we'll never give up.
As a result of the above I don't wanna devote any more time to Coder Keymaps. The project has been closed.
There's a pretty interesting comany I've heard about lately named Matias. They create various products, but some of their keyboards are especially interesting because of the special layouts they're using.
Their Optimizer layout that is used in the Matias Optimizer Keyboard really hits home for me. The idea is very good, but they're not the first to invent such a special layout and probably neither me. I've seen a similar layout a long time ago on a site that I don't remember, but a guy basically made a customized X Server layout with the core idea of using the JKLI keys and the Windows key as a shifting key to produce handy cursor navigation. I've seen his work after I came up with the Ultimate layout which is similar to these layouts. The core idea of the Optimizer layout is very good, but their implementation is suboptimal for a number of reasons which I won't talk about now because I don't wanna share the details yet.
On the other hand their Half-QWERTY layout that is used in Matias Half-QWERTY One-Handed Keyboard and Matias Half Keyboard is new to me and I haven't heard about it yet. It may truly be their own invention although it doesn't take a genious to came up with such an idea I believe. I think the Half-QWERTY layout works wonderfully in practice because it's easy for the human mind to mirror the keys of one half of the keyboard. One thing that is hard to understand for me is why these products are so bloody expensive. They sell them about $600. One reason I can imagine that the target audience is very limited and they cannot sell many of these and want to make additional profit with these products but such high prices are very discouraging for customers I believe.
Before delving into Das Keyboard I'd like to take the opportunity to show you the trailer video of the Ultimate Hacking Keyboard, a high-end mechanical keyboard of which I'm the lead developer. Our keyboard is going to be kickstarted soon so you're more than welcome to share it, follow us and subscribe to our list to get notified when our campaign starts. See you on UltimateHackingKeyboard.com! Cheers!
And now on to Das Keyboard...
I felt the need to disassemble Das Keyboard and see what makes it tick. Just to make things clear, I think that Das Keyboard is the best traditional PC keyboard ever created, but not the best keyboard that is possible to create because the standard PC layout requires too much travel distance for various keys and it's suboptimal from the typewriter's perspective. No, I don't mean DVORAK, but something that I could call QWERTY Compact that is very easy to learn. But that's the topic of another post.
First of all, I want to apologize for some of the crappy pictures. My friend, Dömi and I did everything we could do to make reasonably good pictures but the lighting was less than optimal and we're pretty unexperienced at making such photos. Anyways, let's see what we have!
This is the back of the keyboard clearly showing where the screw holes are. You have to harm the central warranty label so say bye-bye to the guarantee if you proceed further. You also have to remove the upper rubber feet to access the screws, but you'll be able to easily glue them back later.
Now comes the tricker part. There are some hooks that hold together the front part and the back part of the keyboard. It looks like this:
I think it's probably impossible to disassemble the keyboard without breaking some of these hooks, but you don't really have to worry because the screws alone will hold the front part and the back part pretty tightly so you'll be perfectly able to use your beloved Das Keyboard after assembling it. You have to pressure the hooks from the sides with some handy tool to disassemble the front and the back parts.
Now let's see what's inside!
After removing the keycaps (which is a pretty easy job) you can see the soul of Das, the Cherry MX tactile keyswitches. They are both extremely durable and extremely enjoyable to type on. I'd like to note that it's a misbelief of some people that different pressure sensitivity is required for various switches, namely that less force is needed to press a switch as the distance from the center gets longer. Every key needs constant pressure, about 50g. I've measured the constant pressing force by placing some coins on different keycaps.
We have lots of those switches, this obviously does not make Das a cheap keyboard. The switches are panel mounted, not PCB mounted. Panel mounting is a more robust solution than PCB mounting. Panel mounting also make Das more expensive, so think about what does this keyboard provide for its price before saying that it costs too much.
The keycaps are not ordinary either. When seeing Das from its side it's easy to notice that the top of the keys form a slight U shape. This shape is more optimal for typewriters than the much more common linear shape used by the rest of the keyboards (not counting the Model M and Unicomp keyboards). Because of this shape, each row has to close a different angle with the base panel than the other rows.
The keys are attached to the PCB (but they sit on the aluminium panel shown above). Most dome switch keyboards use foil, but we're not dealing with mediocre quality here.
Das uses diodes to implement N-key rollover. Let me note that it doesn't implement true N-key rollover, only 10-key rollover, but it's hardly a limitation as most of us have 10 fingers.
The controller PCB is definitely not an ordinary one. The left IC is a Genesys GL850G USB hub controller and the right baby is a Holtek HT82K95E USB Multimedia Keyboard Encoder 8-Bit MCU.
After using Das for some weeks I could hear some squeaky noise when pressing the space bar. I was more than surprised because everything else was so terrific about the keyboard. After I applied some lubricating grease to it the irritating voice gone away. Some grease should be applied to these wires during the manufacturing process. Apart from that I cannot say any bad things about Das.
You probably pretty much know my opinion about Das after reading this post, but let me say a word of wisdom at the end. You can buy many crappy keyboards throughout your life or one real keyboard. The choice is yours. You know which I opt for.
You can check out higher resolution photos in a separate Google+ album.
Thanks for Werner Heuser for linking this post on Repair4Keyboard.com.
I'm using a Das Keyboard for two months and it's an incredibly elevating experience. It's so much better than any traditional keyboards that it's hard to describe it with words. I'll write about Das Keyboard in a long post, but I've already made some research to find out whether all mechanical keboards are superior to their dome switch / scissor switch counterparts.
It turns out that the answer is a definite no.
I've heard about / tried the following ones: KPT-102, DTK and Everex. It's not a miracle that you probably haven't heard about anyone of them because they are all crap. Mechanical doesn't mean quality. Apart from the Cherry keyswitches and the original buckling spring mechanism all I see is inferiority.
I've been to Budapest in the recent days and had the opportunity to test some keyboards to see how usable they are regarding touch typing.
I was especially excited to try Sony VAIO laptops. Sony's laptops are very elegant and stylish, typically designed for CEOs and other high profile people. Unfortunately this doesn't mean that it's a pleasure to type on them, quite the contrary according to my experience. Flat keys look great, but they make typewriting hard because one's fingers cannot feel their position. Apple neither has a place in the market of typewriter's keyboards with their recent flat-key models. Lesson learned.
I've tried several dozen laptop and desktop keyboards, but my findings are pretty disappointing. Tactile feedback is usually poor and laptop keyboards have much more issues in general. The most frequent problem is that compact layouts usually result in inferior usability and the inventors of these shameful creations are incredibly resourceful regarding totally messing up layouts in the most non-intuitive manners one could ever imagine. I could talk for hours about this one topic alone, but I won't dwelve any deeper into this issue now.
Another interesting aspect of the Ultimate Keyboard which I've became more aware of is the keyboard technology used. Wikipedia has a fascinating page on Keyboard technology which explains most of the things that touch typists should know about the topic.
The article Mechanical Keyswitches, Membrane Keyswitches, Scissor-Switch Membrane Keyswitches is full of insights regarding the workings and efficiency of varous keyswitch technologies. I think that mechanical keyboards are probably superior to scissor-switch membrane keyboards which are superior to ordinary dome-switch keyboards. I'm not sure but I'm gonna test it myself.
There's also an interesting article about the 10 worst keyboards of all time. It's in Hungarian, but the pictures alone will scare the shit out of you.
As a last reference there's a microswitch keyboard made by a die-hard hacker named Tim Tyler who was apparently dissatisfied with other keyboards. I can easily imagine that his keyboard is superior to 99.9% of the keyboards available in the market.
I've always loved laptop keyboards because of their good mechanics and compact form factor. After making some search I've realized that one can order them individually, but I'm surprised to see their prices. I don't understand why they are so expensive compared to PC keyboards.
The Sony Vaio TZ keyboard is certainly a beautiful piece of keyboard. I always thought that such flat keyboards are not fast to type with, but this article might prove me wrong. I'd like to give such a keyboard a try in the near future. Apple also has some keyboards that are similar to this one.
It may be a pleasure to type on an IBM Model M, but it certainly looks very cheap. Its mechanics is interesting, but I don't know any museums around here to have one to try. Maybe I'll look up some auction sites.
I've always loved the mechanics of laptop keyboards, and there are some fake-laptop keyboards for the PC, but none of them mimics the mechanics of the real laptop keyboards. Fortunately, it seems that it's possible to buy laptop keyboards and mod them to connect to other machines.
According to the little research that I've done it seems that the pitch distance (distance between keys) is 18 mm in case of PC keyboards and it's around 17 mm in case of laptop keyboards. The stroke distance (distance a key moves when pressed) is about 4 mm in case of PC keyboards and it's about 2 mm in case of laptop keyboards.
I think tactile feedback is a strong attribute of a typewriter's keyboard which is mostly a combination of stroke distance and the used mechanics. Short stroke distance may result in insufficient tactile feedback and long stroke distance may be uncomforable and unproductive because there is too much path to take for your fingers to press a key.
Unfortunately nobody has published any extensive tests about people's typing speed when using specific keyboards, so I'll try to do such a research by myself. I'm afraid it's gonna cost me much time and money, but I'm sure it's gonna worth my efforts. I'm thinking of creating a frontend for people to populate such a database. Researching this would be a cool project.