that's why I chose not to use a darker purple (like the one of hyperfuse) but something in between purple and magenta

https://www.pantone.com/color-finder/PURPLE-C
Pantone purple-c (pure violet) might be a good option to consider, since the 2070C you chose is still on a darker purple spectrum IMO

I would accept upper left quadrant.

yep, I agree with fohat, top two from left column.
Rest is bull****, like non-binary sexes.

This is nice! I'd like to see it with darker alpha base color though (like, say GMK Slasher) to make the purple really pop more. I always thought that CR and N9 are actually slightly too far apart in terms of tint to work that well together.

Project update: The first Round 2 beam spring keyboards have started shipping, with priority for the low/custom serial skip the line project supporters.  I expect to finish up over the coming weeks.

For those folks who decided to stay with Industrial Gray or Black case colors even though their boards will arrive later this year, feel free to email me if you change your mind and want to switch case colors to avoid additional wait.


A few folks have indicated worry about the longevity of beam spring switches but I would not worry:  the Round 1 beam spring boards have been in use for years with zero reports of it only lasting months.  I look forward to the reviews but a review of a new board is not going to capture longevity data.  These are not new prototype products.  If these products were rated for low keystrokes then they would have started failing within months, but hundreds have been in use since early 2023.  No one's saying that their spacebar key module (or other commonly used key) broke after months of usage, for example (outside of modules that were improperly factory assembled, that the user would have replaced during initial setup or that I would have replaced during QC testing.  I usually replace every R1 spacebar module's part B - the part containing the white plastic plus its attached thin metal, along with others that do not properly function during testing). 

As discussed years ago, metal alloy quality has improved significantly since the 1970s, when only an exotic material may have been sufficient to use in beam spring keyboards.  My understanding from https://www.reddit.com/r/MaterialsScience/comments/1bytxlx/is_spring_steel_essentially_steel_with_a_large/ is that as long as the part operates entirely within the elastic range of the material, well below its yield point, and the stress of operation (especially the most stressful moments of snapping upon key press and release) is less than the fatigue or endurance limit, there will be no permanent deformation or breakage of the thin metal during normal operation and it will have a long lifetime.  In other words, my understanding is that the amount of the curve and its stress as it moves between pressed to released is not so much as to approach its yield or fatigue points.  The 1970s IBM beam spring design is already good enough (and working well within the alloy's limits) that these keyboards work without fatiguing the metal after 55+ years with alloys designed in the mid to late 1960s, so it is likely the case that the thin metal will not be the first thing to go in a new or original beam spring keyboard when any modern metal is used in the original IBM part dimensions and stress levels (for example, the Model F's curved capacitive PCB failed in one of my F122's due to a crack in the PCB and had to be replaced with one of my original IBM spare F122 PCBs).

It is my understanding from reading the fascinating Silo Beam Switch project updates that the goal was to miniaturize the beam module to make self-contained mechanical switches, which put the more compressed components under far more operational stress from bending, flexing, and snapping than the standard IBM beam switch dimensions, and that resulted in unacceptably low longevity during testing.  And it was not certain that the original IBM alloy would have even held up under the higher compression and stress of the compact switch design; maybe IBM tried to make the modules smaller during their early R&D but ran into the same issues.  The reproduction project's switches have the same footprint, travel, and dimensions of the metal parts as the original IBM and so are not under stress that exceeded the specifications of today's standard SS alloys, so longevity was not affected by design changes in the reproduction boards.  The QC issues with Round 1 boards were discussed in depth and related to errors during assembly of some of the modules, with flyplates being bent by hand, causing the thin metal to be bent too much.  Replacing these parts fixed the issue for the round 1 boards.  The Round 2 boards are assembled without that hand bending which caused some of the thin metal parts to exceed their yield points.