I read the note at the bottom about the recordings coming from the community, but I think that variation limits the value significantly.
The Cherry Blue vs Cherry Blue (Full Travel) for example. I would expect the full travel to be louder, the normal sound plus the bottom out, but it seems quieter and more generic. The Cherry Browns were the same way.
Having recordings where there is a lot of control around the recording (same room, mic, distance, levels, etc) and the only variable is the keyboard, would be much more interesting. As it stands, I don’t feel like it’s giving me a true representation that I can use. I’m sure some are, but if I haven’t used a particular keyboard before, I can’t be a good judge of if the sound is accurate or not.
On every keyboard I get, I swap the switches out for silent tactiles[0][1] that I've selected through trial and error. Quiet is really nice (my mouse clicks are louder), but the way they feel is fantastic.
I wish there was a brick and mortar that let you try out a good range of these switches. Places like microcenter have the popular standard choices, but there's so many other switches out there that are just worlds different.
yeah silent tactiles are a gap, the tactile section is all loud stuff (pandas,
banana split), nothing quiet. boba u4 should be on there, adding next round.
brick and mortar thing is real. microcenter's basically it for physical. novelkeys and kbdfans sell $8 switch tester packs with ~10 switches each, not a store but at least you feel them before committing.
Looks like a cool website, but after I test a few different keyboards it prompts me to subscribe. After pressing 'maybe later', it comes back again and again, only letting me test three different keyboards before bugging me again. Completely unusable.
Someone should add a grid of mics and speakers inside a keyboard to record impulse responses for different kinds of keys. In this way, one can first do noise cancellation for the currently used keys, and then apply an IR to make the keyboard sound like another one.
The UX of the website is kind of horrible. After a few clicks it prompted to subscribe. I understand it takes time to build the thing up, but the disruption is huge.
As someone who doesn't pay much attention to the world of mechanical keyboards, very happy that I can use "thock" as a filter.
EDIT: A quick Google shows it's a pretty popular term, so I guess that's how I even know about it, the only other mechanical keyboard term in my vocabulary being "Cherry MX Blue clicky switches" for the ones on my AliExpress mechanical keyboard that prevent me from using the keyboard around other people. Unfortunately it also makes it difficult to hear the keyboard sounds without clicking on the letters instead :(
The HHKB was a Niz Purple Hybrid as a Topre stand-in — flagged as a "proxy" in the methodology but the name was still HHKB, which was misleading. Swapped it for a real HHKB Pro Hybrid recording (CC0 sample from grcekh on Freesound).
Unicomp was outright broken, a single file mapped to every key, which is why it sounded very wrong. It now uses the bucklespring recording from the Model M entry, which is actually authentic because Unicomp builds these on the original IBM tooling. Both fixes are live now.
I made (read: assembled) a lot of keyboards and tuned the sound to my liking, sometimes i would record the sound to share it with friends but the recordings always turned out horrible, not at all what I was hearing.
The frequency response of the mic probably has a lot to do with it; and unless you use stereo mics on a dummy head, the spatial effects won't be quite the same.
I read the note at the bottom about the recordings coming from the community, but I think that variation limits the value significantly.
The Cherry Blue vs Cherry Blue (Full Travel) for example. I would expect the full travel to be louder, the normal sound plus the bottom out, but it seems quieter and more generic. The Cherry Browns were the same way.
Having recordings where there is a lot of control around the recording (same room, mic, distance, levels, etc) and the only variable is the keyboard, would be much more interesting. As it stands, I don’t feel like it’s giving me a true representation that I can use. I’m sure some are, but if I haven’t used a particular keyboard before, I can’t be a good judge of if the sound is accurate or not.
> (same room, mic, distance, levels, etc) and the only variable is the keyboard, would be much more interesting
can't agree more with this.
I have the novelkey creams for example, and they sound nothing like in the sound representation.
People forget how much the plate, material of the keyboard etc vary the sound.
On every keyboard I get, I swap the switches out for silent tactiles[0][1] that I've selected through trial and error. Quiet is really nice (my mouse clicks are louder), but the way they feel is fantastic.
I wish there was a brick and mortar that let you try out a good range of these switches. Places like microcenter have the popular standard choices, but there's so many other switches out there that are just worlds different.
[0] standard preference: https://a.co/d/03j6Boy0
[1] low profile preference: https://a.co/d/06yVB6jg
yeah silent tactiles are a gap, the tactile section is all loud stuff (pandas, banana split), nothing quiet. boba u4 should be on there, adding next round.
brick and mortar thing is real. microcenter's basically it for physical. novelkeys and kbdfans sell $8 switch tester packs with ~10 switches each, not a store but at least you feel them before committing.
Yes, silence should be the default
Looks like a cool website, but after I test a few different keyboards it prompts me to subscribe. After pressing 'maybe later', it comes back again and again, only letting me test three different keyboards before bugging me again. Completely unusable.
"Completely unusable" is an over-exaggeration. By default, assume good will.
Removing it
Someone should add a grid of mics and speakers inside a keyboard to record impulse responses for different kinds of keys. In this way, one can first do noise cancellation for the currently used keys, and then apply an IR to make the keyboard sound like another one.
The UX of the website is kind of horrible. After a few clicks it prompted to subscribe. I understand it takes time to build the thing up, but the disruption is huge.
Why do they want my email though? Well, I can guess why. But why would some random email be useful for such a site?
Building a newsletter, so I can keep sending cool pieces to your inbox
Everybody hates this and you know it, worsening the experience of real people for an absurdly small chance of personal profit
removed it
what profit?
I wish you could play it without typing. I can't hear it over my Cherry MX browns.
Added "Play" button in the sidebar
As someone who doesn't pay much attention to the world of mechanical keyboards, very happy that I can use "thock" as a filter.
EDIT: A quick Google shows it's a pretty popular term, so I guess that's how I even know about it, the only other mechanical keyboard term in my vocabulary being "Cherry MX Blue clicky switches" for the ones on my AliExpress mechanical keyboard that prevent me from using the keyboard around other people. Unfortunately it also makes it difficult to hear the keyboard sounds without clicking on the letters instead :(
Thock is a filter if you type in the bar, and adding "Play sample" button, didn't think of it, but a good idea!
Hmmm. Someone tell the mynoise people. Having a sound ambience of a typing room might be awesome ? Or not ?
The Selectric sounds pretty nice. I should really modify one of mine to be used as a terminal one day.
Tons of ten-hour-long typing sound videos on YouTube.
I always bought the non-clicky keyboards...
Really nice stuff. But the mobile ux is horrendous.
fixed it
HHKB sounds nothing like mine. Weird to type on it and hear a totally different sound.
Model M sounds reasonably close, Unicomp Classic sounds very wrong.
The HHKB was a Niz Purple Hybrid as a Topre stand-in — flagged as a "proxy" in the methodology but the name was still HHKB, which was misleading. Swapped it for a real HHKB Pro Hybrid recording (CC0 sample from grcekh on Freesound).
Unicomp was outright broken, a single file mapped to every key, which is why it sounded very wrong. It now uses the bucklespring recording from the Model M entry, which is actually authentic because Unicomp builds these on the original IBM tooling. Both fixes are live now.
i like the sound when im pressing the buttons but i can't stand listening to them in a recording its horrible
I made (read: assembled) a lot of keyboards and tuned the sound to my liking, sometimes i would record the sound to share it with friends but the recordings always turned out horrible, not at all what I was hearing.
The frequency response of the mic probably has a lot to do with it; and unless you use stereo mics on a dummy head, the spatial effects won't be quite the same.