Mechanical Keyboards: An Honest Guide for Typists

A mechanical keyboard uses an individual spring-loaded switch under every key. A membrane keyboard — the kind bundled with most desktops and built into most laptops — presses a rubber dome against a shared circuit sheet. That single difference in construction explains almost everything people notice when they move between the two, and why the effect on typing is smaller than enthusiasts tend to claim.

What the switch actually does

Every mechanical switch registers a keypress at a specific point in its downward travel. That point is the actuation distance, and it usually sits around 2mm into a total travel of about 4mm. Rubber domes generally register only when fully collapsed, at the bottom of their travel.

This changes where you can stop. On a mechanical switch the key has already registered halfway down, so a trained typist can release and move on without driving the key into the plate. On a membrane board you must bottom out on every stroke, because bottoming out is the registration. Over thousands of keystrokes, that difference in impact is felt in the fingers.

Switches fall into three families.

Linear switches move smoothly from top to bottom with no bump and no click. They are quiet and consistent, and give no feedback about where actuation occurred — you either learn the depth by feel or you bottom out.

Tactile switches include a small bump at or just before the actuation point. You feel the key register. For typists this is usually the most useful family, because the bump tells your finger the job is done and it can start lifting. Most people who type prose or code all day settle here.

Clicky switches add a sharp audible click alongside the bump. The click is a separate mechanism, not a by-product of the bump, and it does nothing that the tactile bump does not already do. It is an auditory preference with real social costs in shared offices and on calls.

Actuation force and the fatigue trade-off

Switch weight is quoted in grams-force at actuation, and the common range runs from about 35g to 80g.

Heavier switches genuinely do reduce accidental keypresses. If your fingers rest with weight on the home row, a 45g switch may register unbidden, whereas a 67g switch will not. Typists with heavy hands, or those who rest rather than hover, often find their error rate drops on a stiffer board.

The cost is cumulative. The difference between 45g and 67g is trivial for one keystroke and substantial across 40,000 of them in a working day. Heavier switches produce more forearm fatigue over long sessions, and that fatigue tends to show up as declining accuracy in the last hour rather than as pain. Most people land between 50g and 65g for all-day use. If you are unsure, err lighter and improve your hand position instead.

Hot-swap sockets

A hot-swap board has sockets soldered to the PCB so switches can be pulled and replaced by hand. Without them, changing switches means desoldering several dozen joints. The practical value is that switch preference is hard to predict from description — weight and tactility feel different in a sustained session than on a switch tester, where you press a key ten times out of context. Hot-swap turns an expensive guess into a reversible one.

Keycaps and stabilisers

Keycap profile is the shape and height of the caps across the rows. Sculpted profiles vary the angle row by row so the keys curve towards your fingertips; uniform profiles use the same shape everywhere. Sculpted is generally easier to touch-type on, because the row differences give your fingers positional information without looking.

Material matters more than most people expect. ABS is softer and develops a shine where fingers contact it, usually within months on heavily used keys. PBT is harder, more textured, and resists that polishing for years. The difference is not decorative: a grippier surface makes off-centre strikes less likely to slide.

Wide keys — space, shift, enter, backspace — need a stabiliser to stop them tilting when struck off-centre. These are the most neglected part of a keyboard and the most common source of disappointment in an otherwise good one. A poorly tuned stabiliser rattles, loudest on the space bar. Lubricating them is a fifteen-minute job that improves the feel of a board more than most switch swaps do.

The honest verdict on speed

Mechanical keyboards improve the typing experience considerably. They are more pleasant, more consistent, better built, and they last a decade rather than three years. Those are real benefits, and they are worth money.

The speed gains are small. Reported differences tend to be a few words per minute at most, and much of that comes from the board being consistent rather than mechanical — every key feeling identical, with no dead spots and no keys needing a firmer press than their neighbours. Cheap membrane boards vary from key to key; that variation costs accuracy, and removing it helps. There is also a confidence effect that is hard to separate out: a board that registers reliably encourages you to type at your actual ceiling rather than hedging. That is worth something, but it is not the switch doing the work.

If you type at 45 WPM, a new keyboard will not take you to 80. Technique will. Fix your hand position and accuracy first, then buy hardware that makes the hours more comfortable.

Where to start

Measure where you are before changing anything, so you can tell hardware effects from training effects — the typing test gives you a baseline in a minute.

If technique is the real gap, the typing lessons work through finger assignments and home-row discipline in order. Developers weighing a board for code should also consider symbol placement, which the programming typing practice drills expose quickly.