As an engineer, I’m continuously impressed with how far rtings is pushing objective testing for consumer products. Also shoutout to Anthony for loving the box jades. Easily my favorite clicky/tactile switch.
Thanks for watching and the kind words! At the end of the day, it's always about helping you as a consumer get the right and best info you need about a product!
I would strongly suggest also testing off center! Some switches are incredibly smooth straight up and down, but won‘t be as smooth when pressed off center. I know this is more of a stability problem rather than factory smooth test, but this definitely means something.
yea off center test is important reminds me about a video regarding the worst switches ever. some of them had a tendency to jam wen pressed of-center. in-case anyone is wondering pretty much all the first places were extremely old mitsumi miniature mechanical place 2
what a great premise and i am very glad to see another big channel show there is a galaxy of switch types out there. I'm not too picky, I just like a distinct difference going from keyboard to keyboard; My work machine uses a Topre RGB with BKE domes, where as my game machine uses nice lite kailh box whites. The box jades annoying his coworker got a laugh from me; they are quite a bit louder.
Thanks for watching! 😄 There really are a ton of different switches out there, so we aimed to try and make it easier for people to almost know what to expect of them out of the box!
You should do more tests but with off center keypresses. Most linears feel exactly the same when testing straight on, give or take the spring and leaf properties. Off center presses are where you'll actually see switch materials and designs actually being stressed. Also if you want to make an ultimate test, you should also test the raw components and build a frankenswitch with objectively the best parts, and then use that as a reference.
As a fellow Box Jade user, I appreciate Anthony. If he wants to be even more annoying, there exist the most amazing (and cursed) clicky switches in existence. The "Zeal Clickiez". They're at least twice as loud as the Jades, with a deeper more satisfying sound. If you want the best possible sound, remove any keyboard foam, and buy a larger, preferably metal, case to allow the sound to reverberate. The only disadvantage is that the "light" version requires as much force as the average "heavy" switch category. I'm glad someone is working on improving testing poorly understood components. You're doing great work!
Everyday i thank god for RTINGS and their technical reviews. Otherwise id have to rely on gamer69 who reviews based on vibes (yuck). edit: on a serious note, thank you rtings, you helped me pick the right monitor for me and your keyboard reviews are goated with the technical testing.
What an amazing video yet again from Rtings! Thanks! I'm actually thinking of buying a new keyboard, so this info comes handy, even though I'm not some kind of a keyboard wizard. One suggestion: Does the feel (or scratchiness in this case) of these switches change after longer period of use? If the plastics are rubbing on each other, I would assume, that would change the feel of the switch after loooong period of time, so something to think about setting up for fun to see in the future!
Glad to hear you enjoyed the video! 😄The feel will change the more you use them and break them in, though it can take a long time to get to that point as you mentioned!
Data from LABs will always be tainted by Linus saying that couple of hundred dollars is not worth it for doing the testing properly, cant trust them anymore for numbers.
@dtemp132 I hope LTT won't compete with Rtings, I like them both. Maybe a cooperation would make happier everyone. LTT mentioned Rtings positiveley in multiple videos. 😊
Definitely would want to see housing tightness/ stem wobble in the future. Also potentially more of a sound analysis with various plate materials. Maybe frequency response?
I've had an acryl base like that and the switches weren't very secure in it. some were moving inside of the holes, some poped out by a millimetre. Though it will take more time, i recommend using solder to fix the switches in please and eliminate variables.
An interesting note for testing: I would record how a human would press different switches and record that curve. And I think this pressing curve is different with different keys because of the different resistance, but maybe can be averaged for a more human aproach. I think in real life the keys are not pressed linear like the test machine. It would be an interesting experiment. 🙂 I really appreciate the value you make with these videos. I would really enjoy working on projects you do 😊
Good job, factory lube makes such a a difference, great to remove subjective testing since so many people are wrong, especially without having a wealth of experience to compare with. Even with objective testing though, it's ultimately still the users subjective preference and experience that matters most when selecting a switch. more data is more better though! I wonder how my smoothness ratings compare with your objective measurements
When I was working on developing cycling shifter buttons for feel you needed a high actuation force.... but people connected loudness to click feel. Noise blocking headphones or ear muffs threw peoples rankings way off with the switches.
Can you guys look into laptop keyboards? Some laptops have issues where if you don't press down on the keys correctly, such as on a corner or edge, you can feel the "snap" of the key actuating but it won't actually yield a keystroke. Notable examples would be the ASUS gaming notebooks. I have some video documenting this if you need an example of what I mean, but it should be easy to check for - just slowly apply pressure to the key at various spots and see if you can get the key past that actuation point without a keystroke happening. In my experience, the keyboards where this is very easy to do I experience higher error rate in my typing, more people report keycaps falling off (Cause of bad stabilization maybe?), and many of them tend to be gaming machines. Some of them even aren't (such as some of the Dell Latitudes I tested) but I don't have a good way to quantify this per machine. Seems like you guys have the equipment to better do this than I ever could/was able to. Thanks!
Why don't you test these mini switches against a full size cherry DX-44 mounted in a IL\Happ button for a unbiased neutral control. It's not a Keyboard button but if any mini button can even get close to a full size microswitch mounted to a 2in plunger with a large spring inside it, it means its a decent enough switch.
While we catchup on the backlog of user questions from the holidays (our team was off), we're currently answering all questions via our website to ensure we don't miss any! www.rtings.com/forums
Reason many went for the hearing test is because a switch may feel smooth but due to spring/contact rub sound scratchy. Early kalih switches had a lot of those issues, needless to say switches have come a long way since then.
Oh wow, did not expect to see switches review, always seemed to subjective of a thing. That's awesome that you can find a way to help people choose the best switch for them. Going on feel ranking from an individual is a decent strategy, but its really nice to see the actual numbers too.
Personally I prefer the quietest switches with the least amount of travel. I stopped using my new Logitech gaming keyboard as I hated the long switch travel time as it was delaying my relativeness. Btw I love the idea of Quickie-switches such as on Wooting's keyboards, I hope all keyboards take on this tech for all its keys but some gaming platform's, I think it was Uplay have started banning people if they detect its use as they see it as cheating. But we have always had technological variances where one user can gain advantage over the other based on the products they are able to purchase. I think its wrong to ban people especially as its likely to be a technology all the major brands will include in the near future.
Been lurking around for a long time now. But here I can't contain myself and leave a comment. I'm constantly amazed at how accurately you perform testing - considering factors that really make a difference. Huge thank you for that! ❤ Regarding switches, I'd advise everyone to try pre-lubed BSUN and HMX switches - they are far superior to the 'popular' brands like Cherry, Akko, etc. My favorites are BSUN Roselle, Bunny, Akashi, HMX Cloud, and if you're on a budget: the MZ Z1 switches. All these have been easily accessible recently - lucky us! :)
Thank you for labs testing switches. I would suggest that you also document the known material and manufacturer of the switches alongside the force graphs and scratchiness measurements.
I think adding the audio component was done a little early/hastily for this. A wider selection of switches should've been used to detect at what point the scratchiness factor enters into the noise level/floor of what is distinguishable for human perception, while having the test run in a quiet box that participants would reach into, so that they couldn't hear the switches but had to focus solely on the feel of them. This would additionally require a quiet room for the participants to control for exterior noise and vibrations impacting the testing. Then for the sound, participants shouldn't touch the switches at all but rather have a machine actuate them for consistent reproducible responses, and they should be listening at a set distance. I'm sure you have a weighting between the scratchy factor and audio one for the factory metric, and I think until you individually measure the perception graphs (think SPL A curve) of both, your compound score won't be accurate. It's not enough just to tweak it until it matches what the participants scored in this one test, particularly when noise floor thresholds are hit/involved. (when I say noise, I don't mean sound. I mean the point at which you can't discern from no stimulation/difference at all: the placebo point) I wanted to join rtings a few years back, I'm in the area, but covid really messed things up for me...
Were these factory lubed switches? How sure are you that all these switches were "unlubed"? Factory lubed switches are way more inconsistent than if someone lubed them, and I'm surprised this wasn't highlighted in the video at all. Additionally, I understand you're trying to remove any bit of variance with the decision to not hand-lube switches, but that makes this experiment a little less meaningful since a majority of users do not use these switches straight out of the box. If you ever dedicate the crazy amount of time to hand lube these switches (good luck, lol), using the popular lubes associated with each switch would make this way more impactful.
When I try a new keyboard, whether it's laptop or mechanical, I always press the keys on their side/corner to see if they are scratchy. Because when I type, I don't always press at the exact center.
Loving the work and overview! Looks like you guys have a real knack for engineering projects and objective QA. In the consumer review space, this is truly industry-leading.
RTINGS and Hardware Unboxed are the only sources I fully trust on this stuff. Trying to measure things with as much objectivity and hard data as possible really helps and the effort that goes into this is seriously appreciated. Hope I can pitch in more support in the future.
Ooh keyboard switches are coming?! Love it! I really need more data so I can determine which ones I need to try next :D I'm currently using pre-lubed Gateron Silent Browns, as I fell in love with these while trying out more than a hundred keyboards in South Korea's main keyboard store. Data will definitely help me put my experience with switches into something more tangible. Cheers!
I played around a bit trying different keyboards when I "discovered" mechanical switches. Came to the conclusion I have no preference. Guess my fingers are not refined enough to care. My turtle beach membrane keyboard I got from a bargain bin feels good enough to me even compared to the much more expensive cherry switch keyboard. But I found this video of people putting so much effort into putting a objective numerical data on the scratchiness of keyboard switches amusing.
I wonder how far the cheap outemu switches are from the cherry switches, currently typing on some outemu red switches and they arent too bad imo, but the WORST switches ive ever felt have to be the KRGD blue switches, truly awful feel and sound
Vouching for the off-center test here. Especially hitting LCTRL with a force vector that's not perpendicular to the key. I switched keyboards because of a difference in that, even though I couldn't tell the difference in normal up-down presses.
When people type, they don’t only apply axial loads to the switch (as your tester does). Humans apply lateral loads as well, and this can induce additional binding / scratchiness that manifests independently of the pure axial measurement. There are a number of different ways this could be incorporated into your testing, but going beyond pure “straight down” smoothness would be highly illuminating.
I think you’re doing that force test too slowly, the result won’t reflect what a user would feel as they’ll be pressing the key much faster with their fingers.
i started typing this before I have finished watching the video. I would want to know if the travel time between "home to actuation to bottom to reset to home" and their variant lengths of time between each switch could result in different noise levels or smoothness. You have a slide that shows actuation to reset, but obviously this video and its contents are about smoothness. And the video doesn't cover what i am describing.
"objective". imagine trusting someone who is putting switches (oil kings) that not only have polished molds/polished rails and also come factory lubed at around the same level as switches (mx blacks) which have years old, worn out molds, are so scratchy stock that they are basically 95% of the switches that people send in to get diamond polished and broken in by specific services just to be at an acceptable level of scratch. not even the best batch of MX black is anywhere near as smooth as the average oil king.
