'Phonebloks' pitch video describes LEGO-like modular gadget idea

Instead of buying the 5$ electronics component for 10$ at a phone retail shop, an increasing number of people orders it from China for 5$ (including shipping fee).

Customers could have their own phoneblock parts custom-made directly in the factories. You could design an add-on, have 100 of those made, just order one for yourself, then let the manufacturer sell the other 99 world-wide and you will receive a small designer’s fee per sold item.

I feel that the weakness of this proposal is that the modularity as shown would address only computational capability and no affordances of adapting physical dimensons, ergonomics and material properties, or mounting it to other devices.

Thunderbolt’s connector is small; but the silicon real-estate required to implement is is massive by cellphone standards.

Check out the specs on the controller chips (and remember that, even if you go with a dubiously-compliant flavor that doesn’t have Displayport, that controller needs to be hanging from a few PCIe lanes, of which weedy cellphone SoCs generally don’t have an abundance). The smallest chip in the lot, weakest specs, no host capability, is a 5x6mm die that draws .7watts. By comparison, a fairly beefy ARM SoC (Exynos 15 in the Nexus 10 in this case) will hit maybe 8watts if you drive both the CPU and GPU as hard as you can, roughly half that in less contrived ‘fairly heavy load’ scenarios, and substantially less when idling.

Thunderbolt is impressive in that Intel managed to find room in the (already rather tightly packed) mini-displayport connector for a few PCIe channels; but low power it is not.

I should also remind folks, I suppose, that external battery packs are already available. Modified “bump guard” cases that carried a second battery were one of the earlier accessories sold for the iPhone.

Yes, batteries wear out. Yes, that means devices with batteries sealed inside them will need Really Annoying maintenance eventually. That’s the engineering tradeoff – if you can’t live with that, buy a device which makes swapping the batteries less painful.

According to the video, a couple of screws lock everything in place. I imagine the idea is that breadboard contains a perforated plate that’s springloaded to push it into perfect alignment with the breadboard perforations; screwing in the screws pushes the plate upward by half a millimeter or so, locking it into the channels in the pins that you can see in the video.

I was going to bring up a similar point: if electronics manufacturers actually wanted to support a “modular upgrade model” then a PC would be the logical place to start. Swapping out many components can be a daunting or impractical task even for the more-technically-proficient-than-average among us.

Note the comparitive size of a tower PC versus a laptop versus a palmtop. Flexibility is great, but it imposes costs in space and power use and so on.

Pick your tradeoffs. Don’t expect to be able to optimize everything at once.

That was kind of my point. If this hasn’t happened with PC towers (arguably already the most modular consumer electronics devices out there) then I’m not going to hold my breath for portable gadgets just yet.

It is mostly exciting to those of us that were addicted to the Maemo/Meego full linux phone experience…

If you are willing to accept a fixed level of performance, we mostly can. HDDs, being a moving part, have a slightly higher failure rate than one would like; but the rest is mostly a software problem, one that is even frequently fixable by non-techies with the restore partition (though, if you go that route, your computer won’t hang on to your files any more than your TV will).

The humble TV’s great advantage is that it has had long periods where buying a new one got you virtually nothing in terms of improvements, and that it has never been expected to save state (aside from, perhaps, some channel-scan data that can be recreated but are faster to store) so software trouble is kept to a minimum.

1 Like

The one argument in favor of camera swap, specifically, is how much of a (good) camera is in the optics and the sensor size (not necessarily resolution; but more physical pixel-sponging goodness). Both of those things are expensive, of interest only to a subset of users, and occupy additional space; but electronically, crunching the output of an X-megapixel sliver of bottom of the barrel silicon behind a shitty plastic lens is identical to crunching the output of an X-megapixel DSLR-quality silicon sensor mounted behind the finest in multi-element fancy optics.

This concept piece carries modularity to an arguably pointless extreme, given that modularity has costs; but for specific components(like the camera) where logically-equivalent parts can differ wildly in performance depending on how much you care, modularity might actually work:

Consider the following rough-cut proposal: mechanical mount (parallel rail or circular twist, to suit) with MIPI CSI-3 interface broken out into contacts, along with power and ground).

It wouldn’t make sense for the real cheap seats, the physical connectors alone could cost more than the ‘.3 megapixel’ wonders in your basic feature phone; but for more photography-oriented cameras(like the Lumias) it’d allow you to trade off anywhere between ‘minimal bump, easily pocketable’ to ‘performance equal to the finest sub-DSLR’ (since there’s no actual room for a mirror assembly) as occasion suits.

This topic was automatically closed after 5 days. New replies are no longer allowed.