ThicThock Konpeitou Switch Review

Here we are, once more, for a third consecutive week of full-length reviews! While I’ve had several people now reach out to me to tell me that they are much more a fan of this pace than my normal, bi-weekly writing schedule, I’ve got to say that this pace is quite much for me to keep up with even though I’m not taking classes or doing research right now. What many people don’t recognize is that depending on the length of the review, depth of the backstory, or complexity of the switch characteristics, a full-length review starts on average around 10 hours’ worth of work between writing, edits, photographs, and associated documentation. By no means am I complaining about the amount of work that goes into these reviews, but it’s a lot of time to spend a fair chunk of every weekend on. I genuinely do enjoy seeing these reviews go out when they do happen, though, and I don’t think I could have written the other 51 of them on this website if I didn’t feel that way.

That being said, though, this shift towards 3 consecutive weeks of full-length reviews rather than having a Scorecard Sunday has actually allowed me to put some real work in on a content project that I’ve had my mind on doing for some time now. Contrary to what some content creators and fans may think, or maybe even in support of what they think depending on you follow, I am by no means a perfect reviewer of switches. Often times people have this view of my content as infallible or even purely objective because of the way that I write or analyze switches past what other content creators choose to do. While I truly strive to be perfect in my execution and accuracy of reviews, not only do I occasionally make mistakes, but I also am capable of growth and developing a deeper understanding of switches as I try more out and time moves on. Thus, as my scope of understanding about switches has evolved over the past year of doing scorecards, I am excited to announce in the next week or two that you will be seeing information about the ‘Scorecard Update Project’.

Figure 1: Here’s exactly how much art budget I have here to hire an artist for a super fancy announcement photo. — **Figure 1:** Here’s exactly how much art budget I have here to hire an artist for a super fancy announcement photo.

Believe it or not, some of my oldest scorecards are over 11 months old now and I’ve become at least a tiny bit more nuanced and well versed in my understanding of the metrics I initially set out to track and assign numbers to. In time, I’ve looked back and not quite felt the most comfortable about some of these scores in particular, so I’ve begun the process of updating scorecards that are “old”, tweaking the scores ever so slightly. To you all, this Scorecard Update Project will entail:

- Updates of old scorecards including minor score tweaks to some. These tweaks are by no means drastic and at the time of writing I have only had one instance where these scores were tweaked by more than 1 or 2 points.

- Complete rewording of all old scorecards regardless of a change in score or not. Additionally, new photos will also be taken for these older scorecards.

- A new, updated scorecard including Manufacturer and Switch Type, for easy viewing by readers.

Figure 2: New scorecard layout featuring Switch Type and Manufacturer Name sections. — **Figure 2:** New scorecard layout featuring Switch Type and Manufacturer Name sections.

As for the old scorecards:

- All old scorecards that have been edited will be saved, as first presented under the ‘Archive’ tab of the website and will always be viewable.

- Under this ‘Archive’ tab as well, a full page showing the changes in scores between V1 and V2 scores for these switches will be posted as well as an explanation of rationale for each score changed.

- Changes to any scores which have full length reviews attached to them will have those sections updated on the switch’s respective review page.

Needless to say, all of these changes will not take place overnight as I have just shy of 100 scorecards currently written up and this was meant to be an ongoing project of content renewal. Next weekend, for Scorecard Sunday, I intend to implement the first of these changes for reviews which are between 10 and 12 months old now and to slowly continue updating these as time moves along. As well, new scorecards produced (including the one in this review) will feature the new updated scorecard style with the manufacturer and switch type listed on there.

While I will continue to be open towards criticism about my effectiveness as a reviewer, I do want you all to know that I am trying my hardest to be as accurate, precise, and consistent sans criticism. Even though many people like to be dismissive about minor differences between plastics, springs, and molds of various switches and to lump them all into the same category, I genuinely feel like this is the only part of this hobby where these minutiae are ignored. And so, I will continue to do my part in illuminating these differences for all of you, and especially so given all of the love and support my audience, my sponsors, and this community at large has shown me.

Switch Background

I’m just going to go ahead and pre-emptively apologize for this section before starting it. Even after having spent way too long researching the background of this switch, I’m still confused about some parts of its history that I’ve tried to shine some light on.

The very first announcement of the Konpeitou switches came on June 18th of 2020 in the form of a Discord announcement on the ThicThock server. ThicThock, the vendor who is responsible for designing the Shogoki and Marhsmallow switches, announced these as a “progressive smooth linear” option coming in October of 2020 that year. While no more information was teased as to the release of these switches other than a rendered photograph of “ThicThock switches coming eventually”, we can glean that the name was based on a Japanese sugar candy which is roughly pronounced in the same fashion as the spelling of ‘Konpeitou’. This association would doubly make sense as it continues alongside ThicThock’s candy theme first started with the Marshmallow switch name.

Figure 3: Initial teaser render of the four switches to be released by ThicThock eventually. — **Figure 3:** Initial teaser render of the four switches to be released by ThicThock eventually.

Figure 4: Konpeitō, a sugar based Japanese hard candy. — **Figure 4:** Konpeitō, a sugar based Japanese hard candy.

Two months later on August 1st of 2020, ThicThock made another, more substantial announcement about the approaching availability of the Konpeitou switches. In this announcement, the first groupbuy for the Konpeitous were announced as a China exclusive, starting that day, and featuring Konpeitous with 64g. progressive springs and factory lube. At some unstated point shortly thereafter, GBs outside of China were to be hosted by vendors other than ThicThock with Konpeitous featuring 68g. progressive springs and no factory lube. Upon attempting to check out these other sources, though, I was met with a flurry of groupbuy dates for March and May as well as a 3-4 month lead time for production. Given that these switches were both slated to ship out around Q3-Q4 of 2021 and actually did so, this leaves me ultimately confused as to the timelines presented by these vendors as they don’t quite add up. My best understanding of the groupbuys and types of Konpeitous released are as follows:

R1 Konpeitous

China Exclusive
Groupbuy Date: August 2020
64g. Progressive Spring
Factory Lubed

R2 Konpeitous

Worldwide
Groupbuy Date: Between Aug. 2020 and Jan. 2021
68g. Progressive Spring
No Factory Lube

R3 Konpeitous

Worldwide
Groupbuy Date: After January 2021
68g. Progressive Springs
No Factory Lube

Unfortunately, to only make matters more confusing with respect to R2 and R3+ Konpeitou groupbuys, a difference in the in-switch stem material changed as well. R2 Konpeitou switches (as well as R1 for that matter) featured stems made of a mystery, properietary material mix known as ‘JX’. However, for R3 and beyond, the Konpeitous featured POM stems as the stock option due to reported material difficulties and costs of JX, and only offered JX stems as an additional, add on package during the groupbuy. As best as I can tell, for all of these groupbuys the price of Konpeitous were $0.70 per switch and the JX stems for R3+ groupbuys were sold at $33 per pack of 110, or $0.30 each.

Figure 5: Screenshot of Reddit GB announcement for Konpeitous made by Davis of 3DKeebs. — **Figure 5:** Screenshot of Reddit GB announcement for Konpeitous made by Davis of 3DKeebs.

While I’ve never encountered such complexity regarding a switch and its subtle variations over a few groupbuys, my best guess for this decision is due to regional preference differences between China and the West. In general from my experience, eastern audiences tend to prefer lighter spring weights and light levels of factory lubing. In context, this makes sense when you consider that brands like TTC and LCET are very popular in the East as they both hit either one or both of these metrics extremely well. Western audiences, on the other hand, tend to take a more hobbyist and heavy handed approach, enjoying ultra-strong tactiles, unlubed switches for lubing by hand afterwards, and so on and forth. This doesn’t necessarily excuse the confusion regarding the release of these switches by ThicThock, though, and they could have done much better with differentiating between these different releases. In order to help deal with this, somewhat, if you are reading this from anywhere other than Asia, there is a 90% chance you have 68g, unlubed Konpeitous on your desk currently. If you bought them as soon as they came out in 2020, there is a good chance the stems are made of this ‘JX’ material, and if you bought yours anytime after that, then they are likely POM based.

As of the writing of this review, Konpeitous are still sparsely available in stock in their R3+ form for roughly the same prices mentioned previously. Regardless of the changes that were made to stem material, spring weight, and factory lubing options throughout this tumultuous history, the Konpeitous have always had fully nylon housings and came as linear switches with an extended stem pole. (The dimensions of the pole did not change as a result of the material according to sales pages.)

Konpeitou Switch Performance

Note: For this review, my exact Konpeitous used are the R2 groupbuy ones which feature a JX stem, no factory lubing, and 68g. progressive springs. Any comparisons made using the POM stems, which I also picked up, will be explicitly stated as such.

Appearance

Regardless of the color saturated teaser renders from the initial announcement of the Konpeitous, these switches arrived from JWK featuring a light pink housing and lilac-purple colored stem. The purple of the stem is mildly close to something like JWK’s Lilac linears and the pink housings are not particularly close to anything currently offered on market. The pink housings were made entirely of nylon for all rounds of groupbuys of these switches, whereas the lilac-purple stems were made of either the proprietary polymer mix ‘JX’ or POM depending on which round they were bought in. The springs featured in these switches, regardless of their weighting, were golden in color and featured a relatively unique pattern of coils discussed below.

Figure 6: Konpeitous compared to other modern pink offerings (left) and Lilac linear switches (right). — **Figure 6:** Konpeitous compared to other modern pink offerings (left) and Lilac linear switches (right).

Given that these switches were manufactured by JWK, I will be referring to the more intricate details of the molds of these switches regarding the mold chart I first introduced here in the Naevy V1.5 Switch Review. Looking first at the top housings of these switches, they are the second of three total switches to feature the ‘ThicThock’ nameplate, with the other two being the Marshmallows and recently released Shogoki switches. In terms of mold features, these are most similar to Type A molds which feature a two-letter interior code to denote the molds being used. Unlike Type A molding, though, the inside of the LED strip features only a narrow window rather than a full open slot, as can be seen below. Aside this difference from Type A molds by way of the underside of the LED slot, this would make these most similar to top housings like that of the Alpaca V1.

Figure 7: Internal of Konpeitou bottom housing showing two letter mold markings (left) and shortened LED access slot (right). — **Figure 7:** Internal of Konpeitou bottom housing showing two letter mold markings (left) and shortened LED access slot (right).

Figure 8: Stupidly difficult to photograph ThicThock nameplate present on Konpeitou, Marshmallow, and Shogoki switches. — **Figure 8:** Stupidly difficult to photograph ThicThock nameplate present on Konpeitou, Marshmallow, and Shogoki switches.

Looking next to the stem in these switches, these have features most similar to a Type B stem in that that they have a tapered stem rail and large backplate mold circles. Aside the length which hasn’t been documented here for a Type B stem prior, even though it likely has occurred somewhere, these are pretty cut and dry and don’t have much that is all that interesting about these. As well, it should be noted that the mold patterns on both the JX and POM stems would indicate that they used either the same molds or identically designed molds.

Figure 9: Konpeitou stem showing angled slider rails and large backplate mold circles. — **Figure 9:** Konpeitou stem showing angled slider rails and large backplate mold circles.

Moving on to the bottom housings of the Konpeitous, these actually feature a mix of many features having been previously noted in other Durock/JWK switches. First, looking at the interior of the bottom housings, these feature the most recently common upper rim design with ten mold mole circles around the outer edge as noted in Type A molds. As well, internally near the base of the bottom housing, these feature four small mold circles like in Type C molds. On the underside of the bottom housings, these have a single letter mold marking between the third and fourth LED/Diode pins, which is most similar to a Type B mold. This mix of designs and mold type characteristics surprisingly make the Konpeitou bottom housings most similar to something like Opblacks, which were the first switch I documented on this website to bare this mix of features.

Figure 10: Konpeitou bottom housing showing Type A and Type C JWK mold circle markings on the upper rim and internal base, respectively. — **Figure 10:** Konpeitou bottom housing showing Type A and Type C JWK mold circle markings on the upper rim and internal base, respectively.

Figure 11: Single letter mold marking on the underside of Konpeitou bottom housings. — **Figure 11:** Single letter mold marking on the underside of Konpeitou bottom housings.

Taking one last stop in my inspection of these switches for the ‘Appearance’ section, I’d like to note the springs as they are especially different from many progressive style springs people may have seen. While these are the “Magically Progressive” or MP line of springs that ThicThock has sold in standalone fashion for some time now, what makes these particularly interesting is that they have only one end of the spring which is significantly more tightly coiled than the rest. While people have debated on the merits of whether or not progressive springs are truly “polar” and their orientation matters, these are one of the only instances I can recall in which a stock spring for a switch was visually polar. Note, upon opening these switches the more densely coiled side is most commonly found on the bottom rather than on the top.

Figure 12: Konpeitou “Magically Progressive” (MP) 68g springs. — **Figure 12:** Konpeitou “Magically Progressive” (MP) 68g springs.

Push Feel

As I was attempting to compile the history of these switches, amongst all the sources I looked at it was clear that the mystery material in the stems known as ‘JX’ was the major marketing tactic for the Konpeitous. While we’ve recently seen a few other proprietary mixes of material such as Tecsee’s PME and Gazzew’s mystery material in the Boba housings, this is effectively the first special, entirely unknown polymer blend that has been used by Durock/JWK in a switch. This should not be construed however as JWK having developed an entirely new plastic, but instead this should simply read as them having some mixture of already commonly used plastics in production that they are not willing to share the formula for. The chances that any switch manufacturer has developed or is using a never-before-seen polymer in an application as low profile and as cheap as this genuinely doesn’t make much sense. So, with that understanding in mind that ‘JX’ is likely only a mixture of polycarbonate, nylon, and perhaps POM, I was rather suspicious if these switches would actually live up to the marketing hype around them.

Figure 13: “Mystery Material” family photo including Konpeitous, Kingfishers, and Gazzew Boba U4s. — **Figure 13:** “Mystery Material” family photo including Konpeitous, Kingfishers, and Gazzew Boba U4s.

Finally having these in hand, though, I can honestly say that I’m wholly unimpressed. Since I’ve received these significantly later than most of the community, the amount of people who sung me their praises of ‘JX’ really goes to show how easily consumers are manipulated by even a half-baked marketing scheme. The ‘JX’ stems are smooth, don’t get me wrong, but they just don’t provide anything that I think can’t be obtained elsewhere with other already available options. While slightly inconsistent, with some stems having a mild, consistent level of scratch, the majority of my JX-stemmed batch was smooth throughout the stroke but provided an unsatisfying, hollow bottoming out and even thinner, plasticky topping out. While ‘chattering’ is typically a word used to describe tactile switches with leaf-based issues in performance, this is the first time I’ve been compelled to ever describe the topping out of a switch, much less a linear, as ‘chattery’, even though that only does so much to belay the odd texture of the topping out. Hell, I even tried the POM stems in these exact same, unlubed Konpeitou switches and didn’t really notice much if any of a significant difference between them. Perhaps the POM stems, on average, were just a hair more scratchy and leant a more tacky, sharp topping out sound to the switches, but overall, there is a surprisingly large overlap between the performance of these two stem materials. And amidst all of these oddities with respect to the push feel, the progressive springs were hardly noticeable to such an extent that I actually didn’t mention them at all in my first draft of this article.

I shouldn’t say that my general lackluster experience with smoothness and housing collision feeling is entirely indicative of the performance aspects of JX material, regardless of the lack of explanation around them. Perhaps these hold onto lubrication in the same fashion as something like the Lumia stems, or have some sort of break in period that I’m not going to be able to truly test well in time for this review. But if either of those things were the case, why are they not being marketed as such and why are these only marginally, even questionably better than normal POM stems? I can’t help but feel like I was pitched a brand new, unique feeling linear experience and got struck with a sub-par feeling, overhyped marketing gimmick.

Sound

Much like with my aforementioned disappointment with the push feel of the Konpeitous, the sound honestly doesn’t do much more to boost my feelings about the switch. Falling lockstep in line with the feeling description, the JX-based switches are free from scratch noise, but have airy bottoming and topping out sounds overlapped with a very thin, almost chattery-like topping out experience. In fact, all of these noises are even further accentuated with a spring ping at higher activation speeds, but its not present in maybe 40% of the switches tested from my batch.

Without the POM stems making these switches sound truly any better, and only perhaps increasing the cacophony that is the topping out noise, I am a bit surprised that these switches are said to have nylon housings. In most of the switches that I can recall testing that have nylon housings, these are far and away an anomaly in my brain that don’t particularly line up with my expectation for sound for a nylon housings, as nylon housings tend to produce a deeper, more solid, and less frilly collision sound.

Wobble

Thankfully, amongst all of the oddities of this switch, both related to and not related to its physical properties, the wobble is at least normal. Having no top housing wobble, even after having opened and closed a few switches a handful of times, the only real thing to note here is the potentially troublesome amount of stem wobble. Coming in equally prevalent from the N/S and E/W direction, there’s enough wobble that it may be bothersome to some users and could be thought of as on par with other, lower effort modern switch releases.

Measurements

If you’re into this level of detail about your switches, you should know that I have a switch measurement sheet that logs all of this data, as well as many other cool features, that can be found under the ‘Archive’ tab at the top of this page or by clicking on the card above. This sheet typically gets updated weekly on mail days, and has been a subtle project I’ve been developing for some time now!

In addition to the measurements noted above, which are for a JX-stemmed Konpeitou switch, I also wanted to do measurements comparing the POM and JX stems in order to verify to myself that any noted performance differences were not the result of manufacturing. To do this, I picked five JX and five POM stems at random and carried out my normal stem measurements (A-G) for them and then both averaged the results as well as ran a one-sided t-test on them to determine if there was a statistically significant difference between the measurements for each dimension between each stem.

The one-sided t-test had a relatively boring set of conditions: {alpha value of 0.05, Null Hypothesis of no significant difference between the measurements of JX and POM stems, Alternative Hypothesis of a significant difference between the measurements of JX and POM stems}. From these results, which are noted above, it was found that there did appear to be a statistically significant difference between all measurements except for B (Stem Width), and thus there could potentially be a difference in performance as a function of the stem material. (Statistically significant results are noted when a t-test returns a p-value lesser than the pre-designated alpha of 0.05 or 5.00E-02.)

However, in my testing I didn’t really note many differences between the POM and JX stemmed Konpeitou performance. While I was sure to note the differences that I did see in the section above, those changes I did mention felt quite minimal, at best. As well, It’s not as if I would be omitting any noted changes for sake of brevity either, as that seems completely antithetical to these reviews at this point. Ultimately, these results could be skewed by the relatively small sample group size of 15 measurements per dimension per stem material (3 replicants per dimension by five randomized stems), but I did my best to try and minimize error or skewing of results through replication, randomization, etc. If you would like to reach out to me about this section in particular to hash out the statistics behind this, or push for a more thorough investigation, I’d be glad to discuss such.

Other

For all of the points of contention, confusion, and some other ‘c’ word to complete this alliteration, one thing that was absolutely nailed with the Konpeitous was the switch sticker sent alongside my batch. Done by the first notable switch artwork artist, Wintheart, the sticker included in build-sized batches of Konpeitous from ThicThock features Soot Sprites with not only Konpeito candy but also hilariously trying to carry away the switch with them. Say whatever you will about my general anti-weebness, or for other switch stickers out there, but this is by far one of the cutest stickers I’ve seen yet in this hobby.

Figure 16: Konpeitou switch sticker included in ThicThock sold batches. — **Figure 16:** Konpeitou switch sticker included in ThicThock sold batches.

Comparison Notes to Other Notable Linear Switches

Note – These are not aimed at being comprehensive comparisons between all factors of these switches as this would simply be too long for this writeup. These are little notes of interest I generated when comparing these pieces to the Konpeitous side by side.

If you would like to learn more about these switches, I’ve already completed reviews or scorecards for some of them. If the names below are highlighted in a grey color, click them to be taken to my other content or reviews on them!

Raed V1

- Aside being more deep sounding at both the topping and bottoming out compared to the Konpeitous, the Raed V1s also have a much more consistent tone throughout various typing speeds whereas the Konpeitous increase in pitch and volume at higher typing speeds.

- The stem wobble in both N/S and E/W directions on the Raed V1 switches is significantly lesser than the Konpeitous.

- While both sit in that sort of ‘smooth yet subtly consistent scratch’ territory, the scratch in the Raed V1s feels a bit more deep and strong compared to the Konpeitous.

Alpaca V2

- Comparing these two switches side by side is quite frankly unfair, as the Alpaca V2s are standard deviations more smooth than the Konpeitous and only serve to highlight the weak points in them.

- As well, the stem wobble on the Alpaca V2s is arguably the best of any of the switches on this list, and further beat out the Konpeitous.

- Both of these switches have the same ‘pitch up’ in topping out sounds as you increase the activation speeds, though it’s significantly greater in Konpeitous.

KTT Strawberry

- While fairly similar in terms of general magnitude of sound, the KTT Strawberry switches have a much flatter, and more bass-centered sound to their housing collisions than the Konpeitous.

- As well, the stem wobble on the N/S and E/W directions of the Strawberries is significantly better than the Konpeitou switches.

- Overall, given the sheer price difference between these switches, and given the fact the Strawberries are not exactly KTT’s “strong suit” in being four pin tops rather than clamshell style, its incredible to me how much the Strawberries pound-for-pound seem to be better linears than the Konpeitous.

Gateron Ink V2

- Even though both of these switches increase the volume and pitch of their topping out as activation speeds increase, the Gateron Ink V2s do it marginally more so than the Konpeitous.

- Of the switches on this list, the stem wobble in both directions between these switches is the most similar, though it is still ever so slightly greater in the Konpeitous than the Ink V2s.

- Overall, the smoothness between these two switches in their stock forms is surprisingly close to each other.

Cherry MX Hyperglide Black

- By leaps and bounds, the nylon housings in the Cherry MX Hyperglide Blacks produces a much deeper sounding switch at the housing collisions than the Konpeitous.

- The Konpeitous do edge out the Cherry MX Hyperglide Blacks in terms of stem wobble though, as this has been one of the few areas that Cherry has never quite seemed to want to directly improve upon.

- The Hyperglide Blacks though, from a good batch, are definitely smoother than the Konpeitou switches.

Novelkeys x Kailh Cream

- For the first time on his list, I can definitively say that the Konpeitous are smoother than a switch, as they are leaps and bounds more smooth than a stock Novelkeys Cream switch.

- The wobble on the Novelkeys Creams is noticeable and in the same realm as the Konpeitous, though it is still a bit better than the Konpeitous in both directions.

- In stock form, the Creams also are significantly louder in terms of overall sound than the Konpeitou switches, and especially so at higher speeds of activation.

Scores and Statistics

Note – These scores are not necessarily completely indicative of the nuanced review above. If you’ve skipped straight to this section, I can only recommend that you at least glance at the other sections above in order to get a stronger idea of my opinion about these switches.