Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Pockit Modular computing demo [video]

reddit.com

Post reply on HN
Print view View original

11–20 of 34 posts

Re: Pockit Modular computing demo [video]

#11

Post by j-pb »

Solder_Man wrote: post #10
j-pb wrote: post #4

Pretty crazy for what seems to be a one person project. How do you even get your hands on/produce such a weirdly sized trackpad?!

Used to be just me. But two amazing people assisting me now (one of them full-time); immediately took this decision upon recognizing the significant community interest from Reddit, et al. The trackpad is one of my favorite recent Blocks. It's a great instance of circuit scalability. I use standard projected-capacitive-touch with a sensing IC that reads a rectangular grid of electrodes -- but they are all printed on t…

Absolutely amazing!

I hope you keep at it and we'll be able to get our hands on one some day. It's so ridiculously cool!

Currently I'm waiting on a ClockworkPi devterm as my new laptop replacement.

I'd love to replace my phone with a pockit, with a small physical keyboard module, and replace my laptop with a megapockit with a huge backplane and a big modular keyboard.

The possbilities! Thanks again for making something so cool and inspirational!

Re: Pockit Modular computing demo [video]

#13

Post by Solder_Man »

varjag wrote: post #5

Slick. I have a lingering suspicion though that the cross connector pads, despite looking identical, serve their own specialized functions and cannot be freely mixed. I.e. it would be highly challenging to have all the pads accept USB or HDMI.

Only high-bandwidth signal groups (such as HDMI) have the limitation of being usable at only one (or a few) positions. Technically these too can be given more/complete placement freedom, but at the cost of a higher PCB layer count, or exceptional noise-immune routing skill, which I presently don't have. I plan to hire a layout expert to assist with the DFM eventually.

The current PCB routing is done such that about 90% of blocks can be interfaced at almost/all positions. (I'll make a more formal statement later after verifying this number.)

This was in fact a foundational aspect of the design -- "How can the classic breadboard be made (a lot) more powerful at the cost of as little flexibility as possible?"

Re: Pockit Modular computing demo [video]

#14

Post by Solder_Man »

j-pb wrote: post #11
Solder_Man wrote: post #10

Earlier quoted context omitted.

Used to be just me. But two amazing people assisting me now (one of them full-time); immediately took this decision upon recognizing the significant community interest from Reddit, et al. The trackpad is one of my favorite recent Blocks. It's a great instance of circuit scalability. I use standard projected-capacitive-touch with a sensing IC that reads a rectangular grid of electrodes -- but they are all printed on t…

Absolutely amazing! I hope you keep at it and we'll be able to get our hands on one some day. It's so ridiculously cool! Currently I'm waiting on a ClockworkPi devterm as my new laptop replacement. I'd love to replace my phone with a pockit, with a small physical keyboard module, and replace my laptop with a megapockit with a huge backplane and a big modular keyboard. The possbilities! Thanks again for making somethi…

> megapockit with a huge backplane and a big modular keyboard

Look at that... the human mind is phenomenal.

We're not all ultra-skilled, detail-oriented electrical engineers. So why does every person have to go through (a tiring rite of passage of) wiring, soldering, debugging, coding from scratch -- instead of using the same creative mind and effort to come up with ideas, and actually play with their implementation?

I'm not primarily a developer, but it's easy to see how much the software world mushroomed through DRY, modularity, etc. In some ways, Pockit is an effort to do the same for electronics prototyping.

Re: Pockit Modular computing demo [video]

#15

Post by goda90 »

It looks like it has a SO-DIMM-like slot that can fit the Raspberry Pi Compute Module 3. Unfortunately the CM4 went with a different style connector, so in a sense, it's already obsolete. It'd be great if someday there were a standard for these kinds of connectors that all the SBC manufacturers were using.

Re: Pockit Modular computing demo [video]

#17

Post by marcodiego »

goda90 wrote: post #15

It looks like it has a SO-DIMM-like slot that can fit the Raspberry Pi Compute Module 3. Unfortunately the CM4 went with a different style connector, so in a sense, it's already obsolete. It'd be great if someday there were a standard for these kinds of connectors that all the SBC manufacturers were using.

I think 96boards is exactly that, no?

Re: Pockit Modular computing demo [video]

#18

Post by Solder_Man »

goda90 wrote: post #15

It looks like it has a SO-DIMM-like slot that can fit the Raspberry Pi Compute Module 3. Unfortunately the CM4 went with a different style connector, so in a sense, it's already obsolete. It'd be great if someday there were a standard for these kinds of connectors that all the SBC manufacturers were using.

As I mentioned in a deeply embedded comment somewhere on Reddit, wait for the upgraded-PCB version demo coming up in the next week, you're going to love it if you're a fan of the CM4's more powerful Broadcom CPU.

Note: Obviously, this too won't solve the issue of hardware universal-standards; nothing ever will. People just have the tendency to never get along. We are all great at being OK-enough with each other though.

Re: Pockit Modular computing demo [video]

#20

Post by varjag »

Solder_Man wrote: post #13
varjag wrote: post #5

Slick. I have a lingering suspicion though that the cross connector pads, despite looking identical, serve their own specialized functions and cannot be freely mixed. I.e. it would be highly challenging to have all the pads accept USB or HDMI.

Only high-bandwidth signal groups (such as HDMI) have the limitation of being usable at only one (or a few) positions. Technically these too can be given more/complete placement freedom, but at the cost of a higher PCB layer count, or exceptional noise-immune routing skill, which I presently don't have. I plan to hire a layout expert to assist with the DFM eventually. The current PCB routing is done such that about 9…

You can't easily splice balanced signal lines, so it's going to end up more complicated than just a routing issue.

But I like the concept. I understand that the pads are indexed with magnet locks? What happens if a user crams a component in incorrect orientation anyway?

Post reply on HN