F104+SSK+122+62+77+50+Ergo orders now open! New Kishsaver+Industrial Model F Keyboards

[gliucf]

Ellipse, I think it's time you address that rolling deadline. The practice reads as a bit shady to me.

Prospective customers should be better informed on two points:

1) There is no official deadline

2) Shipping times are long and can not be determined. On this point, and please correct me if I am wrong, but if someone were to order a keyboard now, it would be at least 2022 before it is delivered. That fact is not transparent on the website as it stands.

I just want to echo these important points, which reflect my understanding as well, for any prospective customers who may be browsing this thread (as I did for some time before deciding to purchase a keyboard).

For what it's worth, the "rolling deadline" on the website definitely deterred me from purchasing a keyboard. It came across as deceptive, which made me hesitate to drop $400. I am very glad that I eventually decided to buy one anyway, in spite of the mixed impression. As everyone says, it feels wonderful to type on.

[Bjerrk]

I absolutely agree.
And, Ellipse, your response a few posts back came across as slightly politician-y. Didn't actually address the concerns that were being raised.

[russypt]

Am I the only one who understood the deadlines, after being pushed back to account for delays (which were extensively documented and photographed), were thereafter basically synchronized against rounds of production so Ellipse could tell the factory what to make? Maybe I am just filling in the blanks differently (and apparently more charitably) but I didn't really see any intention to deceptively foster FOMO. I've had two very functional new Fs for a year with continually improving firmware (thanks community!) and I couldn't be happier, at least not until I get my dye-subs. If I have been deceived somehow I struggle to see how.

Not trying to invalidate anyone's feelings, just my perspective. I'm just a happy customer who is more excited about the firmware developments in the community than the precise semantics of 'please have your orders in by then!'

[ytirohtksek]

Am I the only one who understood the deadlines, after being pushed back to account for delays (which were extensively documented and photographed), were thereafter basically synchronized against rounds of production so Ellipse could tell the factory what to make? Maybe I am just filling in the blanks differently (and apparently more charitably) but I didn't really see any intention to deceptively foster FOMO. I've had two very functional new Fs for a year with continually improving firmware (thanks community!) and I couldn't be happier, at least not until I get my dye-subs. If I have been deceived somehow I struggle to see how.

Not trying to invalidate anyone's feelings, just my perspective. I'm just a happy customer who is more excited about the firmware developments in the community than the precise semantics of 'please have your orders in by then!'

The deadline in discussion is the "please have your orders in by then!" date on the project website that automatically updates at the end of each month. The concern is that it's a textbook use of false advertising. Creating a sense of urgency is a well-documented way to coerce customers into ordering a product. Seeing the fake deadline invokes images of scammy diet pill websites and the like, which do not align with the nature of the modelfkeyboards project.

For those reasons, I'd suggest it be removed and replaced with the real deadline, or an explanation of how the deadline works.

[cods]

It’s difficult to tell if you’re actually feeling aggrieved, or you’re just concern trolling. If it’s the latter, which I suspect, then I’m guilty of feeding the trolls. Damn.
If it is the former, here’s an idea: stop searching the world for thing to get upset about.
¯\_(ツ)_/¯
Buy one. Don’t buy one.

[XMIT]

I've had a chance to give Ellipse's new one piece caps a test drive on an F77.

Serial Number 835

IMG_20210321_103059.jpg (1.04 MiB) Viewed 10466 times

Getting reproductions just right is fiendishly difficult. I've also never typed on a brand new 4704 keyboard. I only have my F122, Unsaver, F AT, and F107 for comparison.

On the F77 board the Ellipse key caps do improve key travel feel substantially. They are less scratchy and more smooth than what was on here before.

The key top textures are different. The tooling, I think, leaves the texture just a bit rougher (think lower grit of sandpaper, larger peaks and valleys) compared to the old school F107. The key tops are a hair wider, the cylindrical top scoop is just a hair deeper, the edges are even tighter radius than the originals.

Key tops: New 'Y' (left) and F107 'Y' (right)

IMG_20210321_101649.jpg (247.46 KiB) Viewed 10466 times

Some of the legends are off. The 'Y' is the biggest offender. 'C' was slightly off as well. In general, the legends are heavier weight than an original F107.

Legends: New 'Y' (left) and F107 'Y' (right)

IMG_20210321_101701.jpg (323.21 KiB) Viewed 10466 times

Legends: New 'C' (left) and F107 'C' (right)

IMG_20210321_102209.jpg (344.08 KiB) Viewed 10466 times

The stabilized keys are quite good. However, using Ellipse's stabilizers is mandatory. I found a fitment issue whereby the new horizontally oriented stabilized keys are NOT compatible with original Model M stabilizers - the posts are too wide and they bind. The vertically oriented stabilized keys (Numpad Plus and Numpad Enter) do not appear to have this limitation.

New "Numpad 0" key with original Model M stabilizer. This stabilizer binds. Note complete lack of clearance in the photo.

IMG_20210321_102815.jpg (230.06 KiB) Viewed 10466 times

The tooling is definitely different. The spring holder divot on original keys is solid, whereas on the newer tooling it's hollow. Also, the original tooling has stabilizing wings extending radially from the inner cylinder to the outside of the key.

Tooling comparison: Ellipse key (left) and original F107 (right).

IMG_20210321_102328.jpg (114.98 KiB) Viewed 10466 times

I'd really like to see these keys on a Model M for comparison. I'll order an additional set just for this.

[Ellipse]

Thanks for your detailed review XMIT. Here are my notes:

Key texture - the key top is more textured on purpose! Keys from newer molds that haven't seen much usage (like the new molds for this project) as well as keycaps that haven't seen much usage have a fresher texture. My NIB Model F's have far more key texture than my other (used) Model F keyboards from around the same time. Once millions of shots have been made from the molds the texture wears down, but the factory can re-texture the tops every now and then as part of maintaining the molds.

The keys are a replica of the one piece F122 and XT originals in my collection. IBM's keycaps varied over the years and I have interestingly found the later production keycaps (including those on many F107s) to be not as good as the 1983-1985 period IBM originals.

IBM's later 4704 and other Model F caps had noticeably different fonts and had looser quality control - thinner in stroke, more condensed, and less nice looking in my opinion. The project's goal for the aspect of dye sublimation was XT quality fonts, not reproductions of the 4704 fonts (even though the rest of the project is 4704 repro!).

The filled in hole inside the key is possibly a later design by IBM compared to the hole. (The hole is for the ejector pin of the mold!) The earliest keys in my collection have a hole for the most part, as well as the 4 circular designs on the bottom edges as shown, so I put it there. IBM's key designs varied across production lines, across time, and possibly across their different manufacturing facilities.

They probably stopped that design because it puts a lot of stress on the one ejector pin during injection molding. During injection molding of the keys, one of the ejector pins of the most-used type of key (1U) actually broke and needed to be repaired.

In summary, all of the legends are accurate to the 1984 F122 and XT caps in my collection and the keycap part tolerances are within my required +/- 0.1 to 0.2 mm. Forum member Zed did an excellent job recreating the fonts exactly from high resolution scans I sent over.

Regarding the stabilizer inserts, agreed it is best to use the project inserts with project keys and non-project inserts with the non-project keys. The inserts are also more difficult to remove. As many of you probably know, given PBT being more difficult to work with due to shrinkage, getting the PBT parts just right so there is not too little, and not too much space between the key post and inside of the stabilizer insert proved very difficult a few years back, but getting the insert working with the new project keys was the priority over getting a looser fit like the originals.

2018-02-27 Key Molds and key mold repair.jpg (174.07 KiB) Viewed 10449 times

[XMIT]

Thanks, Ellipse.

I'll dig up an XT or AT for another comparison at some point. The 'Y' key seems super wide and the 'W' key seems quite narrow.

[Ellipse]

Yep the key legends are definitely different from even some contemporary IBM keys on other keyboards.

Here is a 2016 scan of my originals done on an Epson V700. Keys sourced from a 6110344 F122 from mid-1984 and an XT from around the same time. Also some Model M keys are in the scan; they were thickened up and redesigned a bit to match more consistently with F keys.

Below that is a scan of the new reproduction keys on the same scanner. Pardon the colors being a bit off in my old scan.

2016 09 scanned keys.jpg (1.84 MiB) Viewed 10434 times

2021-02-11-new-Model-F-scanned-keys-2-1536x1332.jpg (252.46 KiB) Viewed 10434 times

[XMIT]

Thanks.

I've seen your scans before. I don't know why there appears to be parallax. Is it a lens based scanner? This would introduce a tiny amount of distortion along the X axis for things not precisely on the bed of the scanner.

The 'C', 'J', and 'Y' do agree with these scans, at least.

[blur410]

I have a question, is there a way to toggle N-key rollover with the F62, for some reason my keeb is stuck in 6key rollover mode.

[darkcruix]

I have a question, is there a way to toggle N-key rollover with the F62, for some reason my keeb is stuck in 6key rollover mode.

I assume you have QMK as firmware. If that is the case, did you compile it yourself or did you use the QMK Configurator for it (or left the original firmware on it).
You can hard code NKRO as default with a compiled firmware. Another option is the shortcut key that enables / disables NKRO. The key should be NK_TOGG to toggle it. With QMK you need to create this key e.g. on layer 1 with Fn+N to enable / disable it.

[blur410]

I'm using the stock layout from the factory with QMK on it. I assume the base loadout doesn't have this key combo? And also does QMK flashing follow the same general flow as a standard QMK board? I don't know which of the board choices on QMK to use and I don't want to risk flashing my board with an unusable layout or something. It is the newest factory F62.

[Lumiloop]

Got my solenoid+driver yesterday and installed it last night. Holy moly, this thing is amazing! It took some time getting it fine tuned to my liking, but now it's working flawlessly and I can confidently say this is the best tactile typing experience I have ever had. If anyone else is on the fence about installing a solenoid in their keyboard, I'd highly recommend it. It was that last little piece that the keyboard needed to become truly amazing.

For the people who have installed the solenoid in their keyboard, I'm curious how you mounted the driver? I would like to screw it in somewhere, but could not find a place where clearance would allow it. Right now I just have it stuck to the case using a combo of electrical and mounting tape.

[Ellipse]

Lumiloop please do share what you did to fine tune your solenoid and driver, and maybe some installation photos if you don't mind! So far not too many people have shared their settings even though a bunch have already shipped out.

[blur410]

I am curious as it says in the QMK config 4 different options for the New Model F62, which is the compatable variant that works with what my keyboard shipped with? My controller is a green pcb with an x on the back, and I don't know if that means its universal, or wcass, or if it uses the base version. How exactly do I know what will not brick my board?

[pandrew]

The brand new Model F keyboards by default come with the "wcass" controller. (The controller was designed on this thread: viewtopic.php?t=13479 )
The 'xwhatsit/brand_new_model_f/f62' configuration is just a pointer to wcass configuration, so it's the exact same. It's there precisely to show what the keyboards come with by default. It's imaginable that someone might destroy their original wcass controller and replace it with another type, that's why I provide the other options too.
Note:
* if you flash a plain '/xwhatsit' controller firmware onto the wcass controller, your layout will be messed up, because the wcass board shuffles the rows for easier layout routing (for this reason layout files are actually not compatible between wcass and xwhatsit controllers, when using the original non-qmk xwhatsit firmware). You will still be able to use util.exe to enter bootloader mode, but it may be harder to figure out how to press leftshift-rightshift-b to enter bootloader mode if you don't want to use util.
* if you flash a universal firmware onto the wcass controller, then you will have to open up the keyboard, and use the pad-shorting method to enter bootloader mode. This is because universal controllers use a different chip, and usb stack code is not cross-compatible.

I don't believe that there's a way to permanently brick the keyboard, except for mechanical damage, like tearing off the USB port, and such. Flashing wrong firmware cannot permanently brick it, but it is annoying, cause doing the pad-shorting method is not as comfortable as the other methods.

Andrei

[darkcruix]

I am curious as it says in the QMK config 4 different options for the New Model F62, which is the compatable variant that works with what my keyboard shipped with? My controller is a green pcb with an x on the back, and I don't know if that means its universal, or wcass, or if it uses the base version. How exactly do I know what will not brick my board?

I actually have this planned in my technical manual as an extra section with the photos of all controllers and how to identify them (there is already a section at 4.10.1 "Controller Options"). You can download the draft manual from https://www.bucklingspring.com.
Like Pandrew said, the default controller in the Brand New Model F is the wcass and it is also the default one that links to it.

[blur410]

Cool, thanks for the help! I'll possibly experiment with VIA at some point but it seems QMK is most reliable for now.

[Ellipse]

To follow up on solenoid adjustments, I have posted a 30 second video on adjusting the solenoid settings:

Some solenoid settings may result in a lower sound, or no sound coming from your solenoid. This video compares the sound with two different solenoid settings. With maximum throw, extend 20ms, retract 30ms, you get a full volume solenoid click sound. With extend reduced to 15ms, retract 30ms, you get a reduced volume solenoid sound. You can also adjust the physical solenoid settings with the two screws found on the solenoid - another solenoid video in this channel discusses how to physically install and adjust the solenoid in your keyboard.

[Freedomplot8]

Is there a baby idiot's guide on how to program the keys for this specific keyboard (F77)? I just can't get this to work. I think this has QMK. I get the .json and .hex file and try to use QMK toolbox, but it doesn't recognize my keyboard. I followed those QMK videos on using bootloader mode but nothing works. Along with all that, the keyboard only has 6KRO and I want permanent NKRO, not a toggle. Any reason that's the case?

[darkcruix]

Is there a baby idiot's guide on how to program the keys for this specific keyboard (F77)? I just can't get this to work. I think this has QMK. I get the .json and .hex file and try to use QMK toolbox, but it doesn't recognize my keyboard. I followed those QMK videos on using bootloader mode but nothing works. Along with all that, the keyboard only has 6KRO and I want permanent NKRO, not a toggle. Any reason that's the case?

When you have QMK Toolbox open (and loaded the hex file), press Left Shift + Right Shift + B (all at the same time) on your keyboard. This will bring your controller into DFU mode. You can see a yellow text in the toolbox regarding this. Once this is the case, just hit the button called Flash. And it will push the firmware (hex file) to your keyboard.
Why there is not NKRO by default: When the USB Human Interface Device protocol was designed, they had a limit of 6 keys in there. This is the minimum all systems agree on (including BIOS etc.). NKRO is not standardized and there is a good chance, the keyboard doesn't work in certain circumstances. In fact, on a Mac you can't get into the recovery mode or NVRAM. On a PC, it depends on the UEFI designer, but I have seen problems with HP laptops.

[Viridae]

Hi everyone,

I recently received my F62 and have been unsuccessfully attempting to map the keys. I am complete noob at using both QMK and VIA. I downloaded both but neither would recognize the keyboard. I read on reddit and here that I would need to flash my keyboard in order for it to be recognized, but I could not seem to do so without the keyboard being recognized (I believe this is the step I am messing up). Regardless, my end goal is to configure my F62 in HHKB split backspace configuration.

I believe I have the hex file and the JSON files that appropriate for the the HHKB split backspace configuration. So I 'THINK' I am good to go, once I can solve the flash issue. Please let me know if you can help. The VIA program looks much more user friendly than QMK, so I was hoping to use that instead, but perhaps this is a mistake?

My H key also does not appear to be working, but I think that needs an entirely different fix.

Thanks for all and any help!

[darkcruix]

Hi everyone,

I recently received my F62 and have been unsuccessfully attempting to map the keys. I am complete noob at using both QMK and VIA. I downloaded both but neither would recognize the keyboard. I read on reddit and here that I would need to flash my keyboard in order for it to be recognized, but I could not seem to do so without the keyboard being recognized (I believe this is the step I am messing up). Regardless, my end goal is to configure my F62 in HHKB split backspace configuration.

I believe I have the hex file and the JSON files that appropriate for the the HHKB split backspace configuration. So I 'THINK' I am good to go, once I can solve the flash issue. Please let me know if you can help. The VIA program looks much more user friendly than QMK, so I was hoping to use that instead, but perhaps this is a mistake?

My H key also does not appear to be working, but I think that needs an entirely different fix.

Thanks for all and any help!

I would suggest you download the following parts into one folder:
* QMK Toolbox (https://github.com/qmk/qmk_toolbox/releases)
* VIA (https://github.com/the-via/releases/releases/tag/v1.3.1)
* Firmware for VIA and F62 (https://www.bucklingspring.com/download ... model-f62/)

Next steps:
* Unzip the Firmware and Via files
* Startup QMK Toolbox
* Load the Via Firmware (hex file) into the Toobox
* Press and hold left and right shift and the key B on your keyboard - a yellow text appears in the QMK Toolbox
* click the button "Clear EPROM"
* click the button Flash, once the text appears that you can now flash the keyboard
* Once complete, close the QMK Toolbox

Next steps (also the steps you need to do in future, when you want to change the layout):
* Open the VIA application
* Click on Settings and enable "Show Design tab"
* Click on Design Tab
* drag and drop the JSON file that was part of the VIA zip file into the square
* Go to Configure, Layout, and select your keyboard layout

Above is also part of the PDF included in the ZIP.

Hope this helps

[ashtronaut]

Hi, I just got my F62 Model F keyboard in the mail the other day, installed the keys and everything was working really well . One small issue that came up today is that I stopped getting feedback on the left control and it tends to bind as well. If I take off the keycap and gently tug on the spring, then it will stop binding, but if I let go of the spring, it continues to bind. Sometimes just turning the board vertically like if I am inserting a keycap stops the binding as well, but if I let it sit flat, it binds again. I've looked at the troubleshooting steps in the manual, but I'm not sure. Is there a recommended way to fix this issue?

[Ellipse]

ashtronaut please check out the manual on the project web site for a fast way to fix any key binding - works 100% of the time for me!

[ashtronaut]

ashtronaut please check out the manual on the project web site for a fast way to fix any key binding - works 100% of the time for me!

hmm ok, I'm just not sure because it isn't a key with a stabilizer insert, so I've tried all of the relevant fixes in the manual except lube. I also find that it frequently doesn't register, and when it does is when it binds. Could there be an issue with the spring? It sounds very different from the other keys, with no click.

[Ellipse]

Yes the same fixes apply to the smaller keys though I've never heard of an instance of a non-stabilized key getting stuck unless there's some other issue. Do the flipper and spring move freely when no key is installed? Maybe the flipper got stuck - this does happen occasionally if the keyboard was jostled around in shipping (!). Not sure what you're saying with something binding when the key is not installed. Maybe the spring was damaged/bent in which case you'll need to re-seat the spring as described in the manual, or replace it as a last step.

[ashtronaut]

Yes the same fixes apply to the smaller keys though I've never heard of an instance of a non-stabilized key getting stuck unless there's some other issue. Do the flipper and spring move freely when no key is installed? Maybe the flipper got stuck - this does happen occasionally if the keyboard was jostled around in shipping (!). Not sure what you're saying with something binding when the key is not installed. Maybe the spring was damaged/bent in which case you'll need to re-seat the spring as described in the manual, or replace it as a last step.

Yes the flipper and spring move freely when the key isn't installed. I tried re-seating the spring with no luck. One thing that I've noticed is that it does seem to get a bit stuck when I press the key, which can be fixed by pulling on the key lightly up from the board. Then I hear a really faint click and I can press it again.

[ashtronaut]

Yes the same fixes apply to the smaller keys though I've never heard of an instance of a non-stabilized key getting stuck unless there's some other issue. Do the flipper and spring move freely when no key is installed? Maybe the flipper got stuck - this does happen occasionally if the keyboard was jostled around in shipping (!). Not sure what you're saying with something binding when the key is not installed. Maybe the spring was damaged/bent in which case you'll need to re-seat the spring as described in the manual, or replace it as a last step.

Yes the flipper and spring move freely when the key isn't installed. I tried re-seating the spring with no luck, but replacing it seems to have done the trick