deskthority

It's time to see how stiff the keys are on your keyboard!

A USA nickel weighs 5.00g according to the US mint (if you don't believe the US Government see this pic with $10 of nickels). A EU Dime weighs 4.1g. Just start collecting them up and balance them on a key. It's easier to glue them together into a 50g stack but not necessary if you don't have glue around.

IBM Model M - 14 USA Nickels or 70g.



I'll take the results and include them in the database of keyboard switch stiffness, I call this the RipOmeter. This guide also includes other hints if you want to be more accurate but it's more fun if you aren't tainted by previous results.

Editors Note: Be sure to center the stack or Webwit might not accept the result.

And be sure to test at least 5 or 6 keys or the result might not accepted by The Solutor

And be sure to test with coins from different Eurozone countries, who knows what the @^~\#}¤& use as an alloy...

EU dimes should be consistently 4.10.


Plus they are the right size.

It needs more 1g coins, otherwise you still have accuracy errors from 5 to 10g, differences between keys not withstanding...

I should check how much small euro coins weigh.

Why try to fix a broken method by using different coins, glued coins as suggested earlier, whatever, if you can do it right at once? Forget the ripoffmeter, it's not accurate. Just get a scale and it doesn't matter what your weights consist of as long as it's small, and it's accurate. Here I use a scale with 0.05g accuracy, a plastic cap, and 3 euro worth of fishing weights. Fill the cap until the key activates, then weigh the cap. Endlessly better. Most switches out there you already know the activation force, because it is published, so stacking some coins where one guy measures the Cherry Blues as 9, the other 10 and the other 11 doesn't add a single thing. Only by more accurate measurements you learn things, such as the consistency of the Cherries vs the buckling springs, or which switches are consistent no matter where you hit them, in a center or in a corner. And the actual activation force, so ripster can verify his method (otherwise he would never know if his method is or is not close and what results to pick).

Lead is bad for you. In fact it wouldn't surprise me if the WitOmeter is banned in California.

You COULD use Yen. I did that for a while but didn't really learn much.

Yen=1.00g


Plus I have found people on the Internet to generally be pretty stupid. In fact at least half of the people on this planet are at or below average intelligence. So simpler is better when trying to gather lots of data points.


BTW I see what you did at Reddit. I still like the comment, "WHO cares how stiff a keyboard is!".

You don't need lots of bad data points to get the activation force of, say, a Cherry Blue switch. Keyboards with these are known to have been sold before, and the required force at the activation point is published by Cherry and measured by better 3rd party methods. It's an exercise in futility. Also "dithering" does not apply to measurements of this nature, it's nonsense. Many bad data points don't make a good one. It's a basic error to think of the deviations as randomly distributed. You can't even measure if it is. High school material. Just as the fact at least half the data points of something don't need to be below average. Here's an example: 999 999 999 999 999 1. So there you have it, the ripoffmeter is for people who are "pretty stupid", by someone who is pretty stupid. In the end you got lots of people fiddling with coins. I guess it keeps them off the streets.

Lets have a contest to see who has the STIFFEST by balancing nickels.

And THEN we can try keyboards later.

We don't have to, the dithered results are already in.

There are no Euro dimes, only Euro cents

webwit wrote:We don't have to, the dithered results are already in.

Being the US results averaged with black people, immigrants from Latin America and so on, would be nice to know what is the WASP's average result

Curiously enough, the central point of the map seems to be at the border between Mali and Niger, for both of which apparently no values are given. Draw your own conclusions. My guess: 42

By the way, when I was a boy, farthest peeing contests were already quite common.
I just cannot remember what they were good for.

Specific switches without published force charts or actuation numbers I'd like RipOmeter results from the Deskthority.net community are:

Acer


Fujitsu Peerless


Futaba


ICLs



Marquardt


NMB Space Invaders - have always been curious that the colors mean.


Omron

pics courtesy of hisao.jp

SMKs - Whites

pics courtesy of Sandy55

SMK - Browns and other colors


Ya Sheng Buckling Spring


*** Pics from geekhack community posts unless noted

I can measure pretty much all of them, if I have time. Some are impossible to measure due to not being PC compatible. For example that ICL 10key pad.

Bottom out forces are interesting for the linear switches as well.

webwit wrote:Here I use a scale with 0.05g accuracy, a plastic cap, and 3 euro worth of fishing weights. Fill the cap until the key activates, then weigh the cap.

Better only because your weights are in smaller increments.

Anyway. What is important for how a key really feels is not only the force it takes to activate a key, but the entire integral under the travel/force graph. That is something that you can not measure this way.

... and you shouldn't measure the keys on the bottom rows.
The keyboard should be angled so that the PCB is perfectly level. Then measure on one of the key caps that are the most level, which are usually on the top row.
Only then will the force vector -- which is the gravity vector -- be in the right direction for comparative measurements such as this.

Findecanor wrote:Better only because your weights are in smaller increments.

False. Up to 50 times smaller. That's the difference between a tape measure of units of 50cm or 1mm. But also more accurate because of the cap with means the force vector will always be centered, and the actual weighing afterwards instead of hoping of accurate coins which are not worn.

That is something that you can not measure this way.

True, you don't get the entire force graph and only the activation point, how observant of you.

... and you shouldn't measure the keys on the bottom rows.

False, level the keyboard.

Findecanor wrote:

webwit wrote:Here I use a scale with 0.05g accuracy, a plastic cap, and 3 euro worth of fishing weights. Fill the cap until the key activates, then weigh the cap.

Better only because your weights are in smaller increments.

Anyway. What is important for how a key really feels is not only the force it takes to activate a key, but the entire integral under the travel/force graph. That is something that you can not measure this way.

... and you shouldn't measure the keys on the bottom rows.
The keyboard should be angled so that the PCB is perfectly level. Then measure on one of the key caps that are the most level, which are usually on the top row.
Only then will the force vector -- which is the gravity vector -- be in the right direction for comparative measurements such as this.

All good points. All documented in the original RipOmeter post. Please continue to give input and I will continue to address it here: RipOmeter.

Findecanor wrote:

webwit wrote:Here I use a scale with 0.05g accuracy, a plastic cap, and 3 euro worth of fishing weights. Fill the cap until the key activates, then weigh the cap.

Better only because your weights are in smaller increments.

Anyway. What is important for how a key really feels is not only the force it takes to activate a key, but the entire integral under the travel/force graph. That is something that you can not measure this way.

... and you shouldn't measure the keys on the bottom rows.
The keyboard should be angled so that the PCB is perfectly level. Then measure on one of the key caps that are the most level, which are usually on the top row.
Only then will the force vector -- which is the gravity vector -- be in the right direction for comparative measurements such as this.

All good points. All documented in the original RipOmeter post in this section:

The RipOmeter Sucks or I don't give a plug nickel for your method

.

Please continue to give input and I will continue to address it here: RipOmeter.

The Professor will answer them there.

Do you know? Do you know?

Actually, that was rude. Let me copy and paste the response here (from OCN posting, now at 7.5K views).

Kinesis Freestyle.


Kinesis Freestyle Force graph.


Measured using this rig. Itronic-Fuchs 8274 Force Displacement Tester


Using the nickel test this guy gets 45g as the peak force.

Same result.

Cheaper too.

According to that force graph, the Kinesis Freestyle would seem very Topre-like. I doubt it would feel the same in real life, but at least it's low-force.

I betcha in real life it feels mushier than Pamela Anderson's boobs.

Her boobs can't be that mushy. They do have pretty substantial rubber domes in them.

I like the SMK Browns. Light, clicky, tactile.

SMK Whites aren't very light.

Laser keyboard - SMK whites - 70g.


For EU members the weights are here. Seem accurate. This dude got 70g on his postal scale and wikipedia gives 12x4.1 + 3x5.74 + 3.92 = 70.3g However I consider 70.3 false precision.

Wonder if Sixty's mailroom ever calibrated their scales after he weighed that Leopold?

It doesn't matter what you think or consider here because you have no basic knowledge of math nor will you ever provide the logic to back up your claims. Also, if you have a scale AND are using coins, you are doing it wrong. I thought the whole point was to get a faulty way to measure without the need for such extra items as the working method with smaller weights and a scale requires. It's like weighing yourself in sacks of potatoes, and then put the sacks on a scale, i.e. an exercise to get an unnecessary wildly inaccurate measurement.

webwit wrote:It doesn't matter what you think or consider here because you have no basic knowledge of math nor will you ever provide the logic to back up your claims. Also, if you have a scale AND are using coins, you are doing it wrong. I thought the whole point was to get a faulty way to measure without the need for such extra items as the working method with smaller weights and a scale requires. It's like weighing yourself in sacks of potatoes, and then put the sacks on a scale, i.e. an exercise to get an unnecessary wildly inaccurate measurement.

He used his postal scale to weigh the coin stack. Didn't trust Wikipedia I guess.

U Mad Bro?

Nope, I'm pretty happy since you proved my points by giving an unintelligent answer. Thanks bro.

I was pretty impressed with his postal scale.

Plus that technique removes observer bias.