MX2A

[Findecanor]

Cherry announced their latest revision of Cherry MX tooling:on their blog today.

The new thing is that the springs are now barrel-shaped and lubed with a lubricant "on par with GPL 205".
They announce 100M actuations and eight guiderails as something new, and not the normal manufacturing standard that it was supposed to have been since Nov 2019 according to the previous entry on their blog.

The MX2A is not just an upgrade; it's a revolution in mechanical switch technology, setting new industry standards and pushing the boundaries of what's possible. It reflects CHERRY's unwavering commitment to perfection, making the MX2A the superior choice for keyboard manufacturers and users who demand the best.

[depletedvespene]

I'll believe their "unwavering commitment to perfection" when their banana space bars disappear for good.

[Muirium]

Guys guys, hey hey! Remember us? We still make MX-compatible-compatible switches too. They're kinda sorta still as good as they ever were, honest.

We're smart! Not like everybody says! We're smart and we want respect!

[mmm]

Bold move to mention a specific trademark when describing the lubrication - I assumed companies did not make direct comparisons with named competing products. I was wrong?

A lot of this reads like standard marketing. But I'm a bit intrigued by the debounce time, described in the blog as:

As with all MX switches, the MX2A continues to employ the unrivaled Gold Crosspoint Contact System. This industry-leading feature promises reliable, precise keystrokes and an unbeatable debounce time of less than 1 Millisecond.

But described like this on the page linked from the blog:

Through a meticulous 2-point welding process, our contacts are attached and integrated with robust resilience. This precision ensures a breathtakingly rapid debounce time of typically less than 1 ms and a lifespan exceeding 100 million clicks for select switch types.

(emphasis mine)

Does this mean 1 ms, or do this mean "probably 1 ms, but if you set set the debounce protection to less than 5 ms in your firmware of choice, you will probably get chattering, just like in any other MX style switch"?

Anyway it's great to see switches keeping up. I'm looking forward to seeing how these are received when they get into the hands on enthusiasts.

I'll believe their "unwavering commitment to perfection" when their banana space bars disappear for good.

Do Cherry produce keycaps still? Are the ones that comes with their standard G80 / membrane series warped?

[Findecanor]

I'll believe their "unwavering commitment to perfection" when their banana space bars disappear for good.

All my Cherry PBT keycaps are pre-2010 but, I never noticed that any space bar was warped.
Edit: OK, of four spare space bars, three of them are a little bit warped, but I can notice that only when placing them on a very flat surface. It is practically nothing.

Has Cherry skimped on quality in more recent times?

[depletedvespene]

…

I'll believe their "unwavering commitment to perfection" when their banana space bars disappear for good.

Do Cherry produce keycaps still? Are the ones that comes with their standard G80 / membrane series warped?

Damn, I was thinking of GMK. Easy to mix up, given how much the latter milks off the (putative) prestige of the former (and how much their marketing relies on their self-evaluated superiority, which is also present in this Cherry announcement).

Now, if it should turn out that GMK's attitude is indeed inherited from Cherry itself, I won't be surprised at all.

[depletedvespene]

…

Through a meticulous 2-point welding process, our contacts are attached and integrated with robust resilience. This precision ensures a breathtakingly rapid debounce time of typically less than 1 ms and a lifespan exceeding 100 million clicks for select switch types.

(emphasis mine)

Does this mean 1 ms, or do this mean "probably 1 ms, but if you set set the debounce protection to less than 5 ms in your firmware of choice, you will probably get chattering, just like in any other MX style switch"?

The “typically” word is the most important one in that entire sentence. I'll opt for your second possibility until hard evidence proves otherwise.

[kbdfr]

Bold move to mention a specific trademark when describing the lubrication - I assumed companies did not make direct comparisons with named competing products. I was wrong? […]

https://en.wikipedia.org/wiki/Comparati ... pean_Union

I'll believe their "unwavering commitment to perfection" when their banana space bars disappear for good.

All my Cherry PBT keycaps are pre-2010 but, I never noticed that any space bar was warped.
Edit: OK, of four spare space bars, three of them are a little bit warped, but I can notice that only when placing them on a very flat surface. It is practically nothing. […]

Even so never found even the slightest warping in any of my (of course original) Cherry space bars.

[Hak Foo]

Bold move to mention a specific trademark when describing the lubrication - I assumed companies did not make direct comparisons with named competing products. I was wrong?

Does this mean 1 ms, or do this mean "probably 1 ms, but if you set set the debounce protection to less than 5 ms in your firmware of choice, you will probably get chattering, just like in any other MX style switch"?

Do Cherry produce keycaps still? Are the ones that comes with their standard G80 / membrane series warped?

Practically speaking, does that level of debounce precision matter? Are operating systems typically set up to handle a keyboard that polls super-fast?

When I was experimenting with firmware, I found that I got good results adding a few millisecond pause after sending a USB packet. I guess my PC didn't like "key down" followed immediately by "key up". So if other firmware behaves similarly, it's likely that you'd still only see the benefits half-way-- the switch could "close" in the eyes of the system in 1ms but it might still wait much longer to reopen.

[Muirium]

Practically speaking: categorically not.

Bro speaking: hell yeah.

[felix12]

I'm quite interested in how they behave in terms of stem wobble and scratchiness

[Findecanor]

I'm quite interested in how they behave in terms of stem wobble and scratchiness

ThereminGoat posted a wordy review of the MX2A RGB Black on his site.
TL;DR: They need to be broken in to distribute the factory lube. There is less wobble. However, premium linear switches from the competition, and even the MX Black Clear Top ("new Nixies") are smoother and have less wobble.

BTW. The "Cherry" logo on the top of the RGB switch housing is now rotated 180° compared to switches with black housing. Makes sense.

[mmm]

Bold move to mention a specific trademark when describing the lubrication - I assumed companies did not make direct comparisons with named competing products. I was wrong?

Does this mean 1 ms, or do this mean "probably 1 ms, but if you set set the debounce protection to less than 5 ms in your firmware of choice, you will probably get chattering, just like in any other MX style switch"?

Do Cherry produce keycaps still? Are the ones that comes with their standard G80 / membrane series warped?

Practically speaking, does that level of debounce precision matter? Are operating systems typically set up to handle a keyboard that polls super-fast?

When I was experimenting with firmware, I found that I got good results adding a few millisecond pause after sending a USB packet. I guess my PC didn't like "key down" followed immediately by "key up". So if other firmware behaves similarly, it's likely that you'd still only see the benefits half-way-- the switch could "close" in the eyes of the system in 1ms but it might still wait much longer to reopen.

It somewhat matters. Allegedly "osu!" players swear to using Wooting keyboard due to their fast response time, and world records have been consistently beaten by people using Wooting keyboards. I admittedly haven't looked into this claim, and it can be a mix of corporate sponsorships and placebo-effect, I don't know. It might also have something to do that rapid triggering of contact based switches cause resonance which cannot be the case for hall-effect switches, but that is pure speculation from my side. Realistically I would not be able to discern the difference between 1ms and 5ms in an A/B test.

But as this is a hobby of somewhat optimizing typing, it at a point where it is technically better, and therefore I would prefer it, even if it made no difference in practicality. There is no reason or logic to it. (Or maybe the reason is capitalism having ingrained into our brains that we must always strive to get better stuff than we already have?)

I'm quite interested in how they behave in terms of stem wobble and scratchiness

ThereminGoat posted a wordy review of the MX2A RGB Black on his site.
TL;DR: They need to be broken in to distribute the factory lube. There is less wobble. However, premium linear switches from the competition, and even the MX Black Clear Top ("new Nixies") are smoother and have less wobble.

BTW. The "Cherry" logo on the top of the RGB switch housing is now rotated 180° compared to switches with black housing. Makes sense.

Thanks for sharing it. Maybe it's just a bunch of marketing claims after all.

[felix12]

ThereminGoat posted a wordy review of the MX2A RGB Black on his site.
TL;DR: They need to be broken in to distribute the factory lube. There is less wobble. However, premium linear switches from the competition, and even the MX Black Clear Top ("new Nixies") are smoother and have less wobble.

Thanks, nice he did a review on these. But I wonder how they can be more scratchy then new Nixies. I have an unlubed sample and it's scratchy as hell.