Typing speeds on keyboards

I’m curious, how fast are the forum members at typing - ideally with how much accuracy?

I’d like to know both those with and without TextBlades and, for those with TBs, how it has changed, if you can remember.

For me, I was in the mid 50s/minute at 97% before the TB and typically mid-60s averaging about 99% with the TB.

There are many typing test sites out there which count differently as well as give different kinds of tests. So, before I recommend one, just for consistency sake, let me explain some of these.

Some sites count every letter you get wrong. Some only count how many words are wrong (thus two errors in one word only count as one error). Some count corrected mistakes while others don’t (though you obviously take time to correct an error). Some do normal sentences, but with no odd symbols. Some include numbers. For those last two items, a somewhat professional typist probably isn’t affected much, but for those of us who type more casually, we simply never practiced those so we suddenly slow way down.

So, I think the best way to make decent comparisons, that we should stick to the basics even though my opinion is that the TB has considerable advantages with numbers and symbols (don’t have to reach way up for them anymore, thus I touch type them instead of stopping to find them).

Thus this is my recommended site for such basic testing:

https://typing-speed-test.aoeu.eu/?lang=en

It just uses common words (no punctuation or symbols - not even sentences).

It will give you the calculated characters/min (the tests are 1 minute so it’s all of them minus the adjustment for errors) and calculated WPM (based on 5 characters/word). It also tells you the total number of characters. It does NOT give you the accuracy rate but tells you how many words you missed (no matter how many errors are in the words). So you can calculate that yourself. Of course, there are different was of calculating as I mentioned above. You can see each word you missed listed together with how you spelled it so you can count them yourself. But it is easiest just to assume any missed word is one error.

My last test had 339 characters after error adjustment (68 wpm). I actually typed 359 characters, but there were four errors. So my error rate would be based on 4/359 (1.1%) so my accuracy would be 98.9%. Thus my report would simply be:

68 wpm at 99% accuracy (if I didn’t want to bother with decimal points).

In my case, I’d do more than one test - only a minute apiece so doesn’t take much time. Which is why I would then say 65 wpm at 99% accuracy.

I don’t have much in the way of old speed records.

Today: 415 CPM (that is 83 WPM), 419 CPM, but made 2 mistakes (out of 77 words).

I didn’t really take the test seriously, or sit properly, and it’s my 1st test on that site. Plus AutoCorrect appeared twice and ruined my flow.

In the past I did a few tests in a sitting and kept the best score. 94 WPM on an Apple Aluminium Keyboard with Numeric Pad, and 94 WPM on TextBlade a few months later.

Objective measurements are important, but I wanted to mention, I no longer have forearm pain on a workday, something that bothered me every workday prior to TextBlade. Also, in terms of productivity, it’s hard to measure differences, but I certainly do a lot more arrow keys, selection, cut-and-paste, screenshots, and use strange shortcuts to start/terminate a debug session and step through code. All of those tasks are now done without my fingers leaving home row. I also no longer stretch my hands to reach backspace, forward delete, or even braces and brackets.

Also, I don’t “practice” typing since I do it every workday (and as here, non-workdays too). My speed at certain patterns would improve with practice, but there isn’t really a set of practice that is suited for coders.

For example, with Visual Studio Intellisense, if I wanted to type:

using Sytem.Reflection;

By the time I got to

using System.R

it would suggest, and by

using System.Ref

the suggestion would certainly be correct.

Prior to TextBlade I would type a lot more rather than use the arrow keys to choose between suggestions, because moving my hands would incur a greater penalty than just typing more characters. Now, I’m free to use arrow keys without leaving the home row, so I use Intellisense suggestions far more.

And though I use keyboard shortcuts almost exclusively, to this day, I will struggle to remember whether it is Opt-Shift-Arrow, or Ctrl-Shift-Arrow, to select the previous word, etc. But with TextBlade, I just use the selection layer, and I don’t even think about it.

It’s almost “magical” how I just select exactly what I want, without interrupting the thought process.

These productivity gains are hard to design a quantitative test for, and, for those who care that much about productivity, hard to engage volunteers to run through batteries of tests.

When it lists the number of words, that is apparently based on the actual number of words rather than the standard 5 characters = a word including spaces. The WPM is based on the standard.

I should have also added that this test doesn’t count corrections as mistakes so, ideally, folks should try not to correct, just for consistency sake. Sometimes I do it automatically, especially when I can feel myself moving to the wrong key.

I’m a bit puzzled by your stat: “415 CPM (that is 83 WPM), 419 CPM, but made 2 mistakes (out of 77 words)”

Because it has always seemed that any error subtracts 5 from the CPM. But it looks like it only subtracted 2 for each of your errors.

I also haven’t noticed auto-correct kicking in on the site. Maybe I have a setting in my browser to turn it off (I looked quickly, but couldn’t find it).

Of course, you are correct that speed is not the only issue. If someone loses speed but eliminates pain, that is a good trade-off. But it’s still good for people to get an lot of feedback on the speed and accuracy. Likewise the ease of cursor movement, reaching numbers and symbols, etc. Just as it will be important for a subset of people to know if some coding situations don’t work as well with the TB - or if it just requires some customization. I certainly can’t remember all that has been said about this, but I’m pretty sure at least some have said they had problems, but found customization workarounds.

Absolutely. So I figure we can at least cover the normal typing that most do. Worst case, if the situation for coders was awful with a TB but for other typing it was great, we’d still have a great market for the TB. It would be the opposite of a lot of unique keyboards which cater to special situations, but aren’t so good for most in the market.

1 Like

I usually use the following site for testing typing speed: http://typingmaster.research.netlab.hut.fi

Before receiving the TB I did get some ridiculous scores there where at least several sentences turned out to be over 100 WPM. I think that I might have let those values rather than the average be put on my mind and made some claim about it here on the forum.

After receiving the TextBlade I have struggled a bit more to reach a speed that I feel confident with. Using that site now gives me around 48 WPM for the average and peak sentences around 90 WPM.

The site specified in this thread gives me an average around 55 WPM, then I focus on fixing errors, I had a streak that required no fixing which gave me a 57 WPM. I think that it is easier to get a consistent flow of typing for a bit longer with this threads site, but I do really think that the other site measures something that is more similar to how I usually type.

I feel like I am still a bit slower now on the TB than I used to be on my laptop keyboard, but it feels like a fairly small difference now. I think that the bigger problem is that I make some more errors on the TB which still makes it slightly less appealing to use when taking public notes. But I think that the error level is low enough now. I just need to train myself to learn using the “delete word” key.

1 Like

Would also be cool if one of these sites could give data on wrist and hand fatigue, but they have no inputs to measure and compare that.

Deckaddict - yes, the edit layer to select last word is very popular. For words you want to change, it’s way faster than hitting backspace sequentially. You just select last word with one command, and type new one.

It automates all the cursor positioning, and eliminates all the backspace keystrokes.

6 Likes

I didn’t like the test because it has you reading a single sentence and then typing it, then resetting for the next sentence. Just too far off from the most common typing to me. Which also explains why some sentences can be MUCH faster than average.

I believe longer tests create their own stress which can lead to errors. But I hate taking the time to do long tests. But 1 minute tests are a decent compromise.

So, on the site I recommended, you average 55 wpm with unknown error rate, right?

This site http://patorjk.com/keyboard-layout-analyzer/#/main has some amount of key distance measurement built in. It’s based on a standard keyboard but supports multiple layouts (including customizing a layout). I can’t see any way to program anything like a TextBlade into it.

Edit: actually it looks like you could program a textblade into it. Someone took the time to build a Kinesis model for example.

What I have so far:

TextBlade / Other keyboard
DBK: 65 wpm - 99% accuracy / 55 wpm - 97% accuracy
colinng: 83 wpm - 99.5% accuracy
deckaddict: 55 wpm - 99% accuracy

Typically I messed up 1-2 words when I had 55 WPM. Many times I had it clean too. I got several 58 WPMs when everything was all clean.

I agree that the test I linked is a bit too slow to perform. But I like the concept of knowing what to write and then measure that.

Not knowing how many characters (or the overall error percentage), it’s hard to know how this all breaks down. 55 wpm with, say, 2 wrong words and assuming there was only one character wrong in those two words, I still wouldn’t know how many total words or characters to calculate the accuracy. But two seems pretty small for 15 sentences so I’ll assume it was about 99% accurate.

The slowness of the test (needing to do 15 separate tests, though each is only one sentence) is a bit annoying to me, by real issue is that it just doesn’t seem to match what most people would be doing. It even says to read the sentence first, then type. Of course, you can ignore that, but who reads a sentence then goes back to type it, etc? But even without reading ahead, people usually aren’t going to be typing a sentence, stopping, type another, etc. May be some who do, but I’m trying to focus on what is most common.

OTOH, it would probably be good if anyone who does special things - like various types of coding - have found the TB isn’t working for them and why (after looking for customization options, of course). But those would be addressing special cases. Not the vast majority.

I type with high accuracy at around 100wpm (that’s typing words from another source, eyes on the source, not the key board.). I type at around 115wmp with high accuracy when I’m just free ballin’ it (and know pretty well what I’m going to say.) I hit 120 or so with less accuracy and few special characters on the reg.

Of course, depending on what I’m typing, a single sentence can take seconds or an hour. lol You know the old adage “If I had more time, I would have written a shorter letter.”

1 Like

Wow, I think what impresses me most is your speed when “free ballin” it. I’m too busy thinking of what I’ll say, often rephrasing from what I start with as I type. So I only get in the mid 40s when typing that way - like this post.

Of course, I don’t know what you mean my “high accuracy”. I assume it is at least 97% but to me “high accuracy” is when I’m at least at 99%. When I get several hundred characters without an error, I’m thrilled.

If you could take a couple of those one minute tests and give the results (and calculate the accuracy), that would be great.

But the point (at least for me) about measuring typing speed is not finding your rate of production from an holistic standpoint. The point is to figure out how much time do you spend on the actual typing. Sure when you write text you think about what to write next, but while typing a sentence you typically have that one almost completely figured out first.

And a very common situation at least in my job is that you sit eight persons in a meeting and make notes maybe in an spreadsheet. For the discussion to be able to continue you need to finish the notes, and then you want that done very quickly and it may very well be someone else’s sentence.

Or say you had a 20 minute discussion in the corridor and you need to sum it up and mail somewhere quickly, at that point you have most of the structure figured out already but just need to really bash out the words and might need to type for five minutes consecutive.

2 Likes

Interesting. Assuming we aren’t misinterpreting what we each mean, I don’t do anything like that. Specifically, I don’t think even what my first sentence will be. I know the point I want to make and the first words are probably known before the actual typing starts, but all the rest I’m thinking as I type. Pretty much the same way I would talk - I don’t think about the full sentence then either. I’m thinking about future words while I say the present ones.

The tb has led me to realize the huge number of differences in how people type that I really don’t think most people realize (not layouts or how many fingers, but little details about how the fingers move). But this is the first I’ve seen about different ways people may think as they type.

I’ve had the TextBlade for roughly two years. I haven’t measured my typing speed either before or after. Subjectively, i feel that my flat-out speed of accurate typing is lower with the TextBlade. I also feel that when I’ve used software typing tools to practice, as I did when I first got the TextBlade, I improved, but it’s boring and I had other things to do, so I stopped. If pure uninterrupted typing speed really bothered me, I’d go back to those tools and work harder at improving speed and accuracy. So why not and further, why do I stick with the TextBlade when I could dust off a legacy keyboard and revert to it at any time? The novelty of this keyboard has long worn enough and only its utility matters any more.

Here’s why. The entire process of committing ideas to written text is easier, faster, and more comfortable.

I’ve absorbed the edit keys into my fingers making it simple and easy to back up a letter or word to make corrections. Or to erase and retype the last word. Bear in mind that most of my errors are of intent (leaving out a word, typing the wrong word, changing my mind entirely on what I want to write), and with the edit keys at my fingertips, I correct as I think. This ease of editing is much of what I love about the TextBlade.

Little of my typing is long form. This is about as long as it gets for a single uninterrupted flow. Usually there’s some kind of physical interruption - other software packages, web searches, shuffling through papers or emails or entirely different activities - that disturb the writing. So the small work to fix typing errors goes unnoticed in that uneven flow.

Plus, I can move between applications and use command keys to mostly avoid the mouse even with interrupted typing.

I’ve found that hand position is critical. On a legacy keyboard my hands can be at any angle. With the TextBlade my hands must be in the same position all the time. It’s comfortable, but I can’t be lazy about it and not get it right. Sometimes I have to slow my fingers on certain letters, or go back and correct, where I feel the strike zone is too small (more software boundary adjustments would help). Other times when I feel the TextBlade is simply wrong about where my fingers have landed, I watch my fingers and see that I’m wrong and have acquired bad habits. Those are, of course, difficult to correct and would be less likely on a legacy keyboard where each key is exactly one character.

Lastly, my hands must be perfectly dry. In an office, that’s easy. But I work at home and can’t run from the kitchen having washed a cup in the sink and rush back to the keyboard without patiently drying my hands. If I don’t do that, accuracy will be awful and uncorrectable.

But in writing this note I haven’t reached for the mouse once, or moved my fingers more than a fraction of an inch, or even glanced at them. That level of comfort and focus is priceless.

9 Likes

I think you’re the first person to say they MIGHT be slower, but still preferred it for other reasons. And they are darn good reasons. Even when I had some issues with the TB after I got it, I didn’t want to switch back.

I never have made any effort to switch apps with it though. Just like I almost never use the edit keys to move by word or paragraph, etc. I almost always just move by character and up and down lines. Just sticking with my habits. Maybe someday I’ll deliberately work on those things.

Yep. When I find myself getting more errors, I almost always realized I’ve gone back to old habits, which means my elbows against my side. That turns the hands, of course, which isn’t ideal for the angled key blades. I also sometimes move the base of my hand too close to the TB which causes other problems with a couple characters.

It’s why, when I first got it, I suggested that it would be nice if a future version let you change the angle. BTW, you can change it a little (too much and it disconnects) by wedging something at the top of where the two blades connect magnetically. I did that for awhile when my bad habits were stronger than they are today.

I had that happen a lot for awhile. I’d even move my fingers (without looking) closer to the key I wanted a couple times and still get the wrong character - convinced the TB was messed up - only to look down and see I was wrong!

1 Like

I’ve been switching up between a number of different keyboards of late (currently typing on a notebook keyboard, but if I only had some sort of mat that would cover these keys, I could use something else :wink:) and I’ve noted some major speed differences. Back when I was in school and we were competing for who-could-type-the-fastest, I maxed out somewhere over 150wpm, but those were tests designed for speed and don’t reflect real world usage. My current speed on a good keyboard is talking speed, i.e. I can talk and type what I am saying at the same time. (This is quite useful for taking notes in meetings, and also for drafting technical docs while explaining a concept to someone.)

I do have a relatively new keyboard that I’ve been using, and my typing speed isn’t great on it yet, not because the typing speed is slow, but because my error rate is still a bit high.

MacBook Pro 13, late 2013 - 53 WPM @ 96%
iPhone 6S built in default soft keyboard - 50 WPM @ 100%
TextBlade on MacBook - 53 WPM @ 100% + watching YouTube

It should be noted that I have to concentrate a lot more when typing on the built in keyboards.
Typing directly on the MacBook I fall in to old habits, consistently looking at my fingers, losing my place, shifting my hands to letters where they don’t belong, all the usual stuff.
The cognitive load typing on TextBlade is next to nil for me. I join conversations and continue to type. My typed note recall is better than hand written notes because I’m more engaged than having to figure out how to get the letters in the right place.

1 Like

You just gotta make one of these. A lot of wear and tear on this thing but works great for my laptop.

Here is the build for it…

2 Likes

awesome. I’ve been thinking about a little cover for my MBP keyboard.