Can a $9 Computer Spark a New Wave of Tinkering and Innovation? | Hacker News Reader

Can a $9 Computer Spark a New Wave of Tinkering and Innovation?

npr.org

52 pointsevo_968 comments

Sigh, I believe the answer is "no". Cost is not the problem with tinkering. I don't think access to computing gizmos is an issue at all in the US. For me I surfed in on a wave of tinkering and innovation by building things that my friends were impressed by and played with (programming calculators, text adventure games, even some electronics). Where is it now? I'm not sure (and I would love to know) how kids these days are building things that are able to compete for valuable attention span. Especially if they are going head to head with psychologists who design smartphone games.

I think children having their own general purpose computing devices is pretty important. Time sharing and the obligation not to break "the family computer" make it pretty useless as a tinkering platform, unless your parents are technical and comfortable with it. The individual devices children own are compilation targets, maybe, but they can't run text editors and compilers.

Cost of the actual computer is not the problem. All of these $50, $20, $5 computers conveniently ignore that fact that you also need $150+ for a monitor, USB hub, keyboard, mouse, WiFi adapter, etc. This quickly raises the price from "something a low-income family can easily spend on a child" to "no fucking way." I found OLPC a lot more compelling.

Still, I'm thankful for cheap hardware on which to run pirate AirPlay targets and homebrew IP cameras :).

> Cost of the actual computer is not the problem. All of these $50, $20, $5 computers conveniently ignore that fact that you also need $150+ for a monitor

Chip has an answer to that.[1] Not sure whether it is really usable enough for kids to work on, but I'd did a lot of work on my TI calculator, which didn't even have a qwerty keyboard.

[1] http://getchip.com/pages/pocketchip

> Time sharing and the obligation not to break "the family computer" make it pretty useless as a tinkering platform, unless your parents are technical and comfortable with it.

I follow the logic here, and intuitively agree, but anecdotally it was almost the exact opposite experience that started my interest in computers.

We got a computer mainly to help with personal accounting type things. Beyond what my mother knew of Excel, my parents were not at all technically savvy. Probably because of this (that is, that they didn't know how to restrict/follow what I was messing with) they gave me relatively free reign to tinker on our family computer from the age of about 8 on. After a few years I just fell into the de facto IT of the house.

I wonder if the problem now is that computers are used for so much more by adults? As long as I could fix/hide whatever I broke before my parents needed to update their budget or check their email, hardly even a daily ritual at the time, I was in the clear.

I'm not sure about CHIP, but a RasberryPi can hook into our TV and I have several old USB keyboards and mice to plug into it. If you don't, you can get one for really cheap. It's the monitor that is most of the extra cost. (I also have a spare monitor, but I'm a computer guy so of course I have extra parts.) Anyway, if the idea is just to allow a kid to have their own computer to break/tinker with, then sharing the family TV to run it is OK. It doesn't solve the time-share issue. (I remember my little sister monopolizing the TV to watch Dumbo The Flying Elephant over and over again.)

About the cost:

A $9 C.H.I.P. (virtually un-google-able name, btw, absolutely terrible for noobs) comes with a standard RCA video cable. It has built-in WiFi and 4GB of NAND storage.

You need peripherals. Using USB stuff means you'll need a powered USB adapter (<$10 on Amazon). You can also use bluetooth. I like the bluetooth keyboards with a built-in trackpad. That's $20 and great for ease of use and flexibility. You can pop the chip on an empty video input (because most other devices are using HDMI) and just keep the keyboard on the coffee table. No cables, no mess.

You'll need a USB power supply for the device itself. I find it hard to believe that most people can't scrounge one up from an old smartphone, or is able to find one at Goodwill, but let's call that $10.

For monitor, used is definitely the way to go. CRT displays are practically free, as well as older liquid crystals. If you really need a new device, you can definitely spend less than $50 as the resolution of the device is still going to be the limiting factor.

So, with shipping for the CHIP, that's $35 for the average person that does have a USB power supply and junky TV.

No need to spend $150 plus on a monitor/keyboard/mouse.

The tech stall in my local market was selling a 17" TFT that had DVI for £15 the other day (about $22.50), mice, keyboards etc are a few pounds, You could get everything you needed for about the price of the RPI if you shopped around for it.

I was tempted by the TFT, removed from it's case and built into something it would have been a useful thing.

> Where is it now? I'm not sure (and I would love to know) how kids these days are building things that are able to compete for valuable attention span.

I'd say it's videogames. I got into programming through glitching games, then scripting games, then programming games. I know a lot of other people my age (mid-20s) who had similar experiences.

Look at kids making various contraptions in Minecraft -- and look at how many of them go into the modding scene.

I don't think that tinkering experience has been lost; people just forget what a small fraction of the population was interested in this stuff in the first place.

Continuing on your thought; one thing I've noticed is that, even being only a couple years removed from the "kids these days", it's hard to appreciate how ~big~ things like minecraft are. Talking with my older coworkers, the same spark the parent comment spoke of is now being channeled into these shared digital creative experiences.

I know a 12 yo who made a simple game in Unity. I think he plays with making games in Python and Scratch too, though I haven't seen them.

Tools change, but interested kids still find a way to tinker.

>Sigh, I believe the answer is "no". Cost is not the problem with tinkering.

More to this point, my first tinkering was on an outdated, free, hand-me-down 286 from a neighbor. It was enough, though, to probably set in motion what has become the majority of my income later in life by working in computers throughout the years

Olimex, who are in a strong position to know, say that the cpu alone (Allwinner R8M) costs $16, direct from Allwinner[1], in volume. That's just the CPU. None of the other components, the PCB, the labor, etc.

The $9 price just isn't sustainable. It appears the the CHIP team is doing a combination of padding shipping prices, and probably taking a loss for the sake of initial marketing.

[1]https://olimex.wordpress.com/2015/06/05/how-to-get-in-the-ne...

What about the Raspberry Zero? The specs looked similar, it costs $5, and I haven't seen any complaints about inflated shipping.

The cpu for that is a broadcom BCM2835. The best guesses for that part is somewhere in the $3 range. Unlike the CHIP, it has no onboard flash, or wifi. It does have HDMI.

Personally, I don't see how the Zero could cost less than $5 to make. Even if the BCM2835 is $2, there's the 512MB ram, various connectors, PCB, etc.

From what I understand, assembly is your real cost when you start getting into BoM costs that are that low.

NTC worked directly with Allwinner and I believe they have a special agreement that made the CHIP possible. There business model is kind of like the Gillette Model. Sell the CHIP for cheap and get people to buy the accessories which have much higher margins.

>>Sell the CHIP for cheap and get people to buy the accessories

This, combined with inflated shipping, is probably the key. However, even assuming some aggressive discounts, the BOM must still be higher than $9.

Even with crazy discounts, the CPU + Realtek Wifi + PMU alone would have to be close to $9.

There's a lot of other parts as well: https://github.com/NextThingCo/CHIP-Hardware/raw/master/CHIP...

I priced the BOM and it came to 50 bucks USD. So yeah there is definitely aggressive subsidizing going on, but the company is small and growing so they must have something in place that makes their plan feasible.

That sound a lot - how did you price it?

Went through the schematic and digikey/taobao(chinese eBay)/googled the most significant ones. Came out to roughly 45 bucks and added 5 bucks for the passives i didn't look up.

I mean, wholesale lot prices, or individual component prices?

Individual component prices to give a sense of how much it would cost you alone to make one versus NTC.

badly

>cpu alone (Allwinner R8M) costs $16

highly unlikely, price point for entry level tablet SoC with PMIC is ~$5. $16 might be retail one off Gweilo prices.

https://usd99allwinner.wordpress.com/2014/07/30/launching-th...

http://www.tomshardware.com/news/allwinner-5-dollar-64bit-ch...

The R8M has integrated flash and DRAM...$16 is the high quantity price.

That was the original CPU CHIP was going to use. They switched to the R8 (no flash/dram) for some reason. The R8 is $5 in high quantities, but of course, they would now need to add the flash and dram, neither of which is cheap.

This article gets brought up in every post about CHIP. Maybe there's something to it but coming from a competitor this sounds a bit fishy. It was also before the Raspberry Pi Zero (made in the UK!) launched, which lacks quite a few of the CHIP's features but at $5 sets a high bar in any case.

The bom cost for the RPI zero is clearly lower. No flash, no wifi, much cheaper cpu, etc.

It is almost a certainty that the BOM cost alone for the CHIP exceeds $9. That's the unassembled pile of parts.

Raspberry Pi Foundation is a charity sponsored by Broadcom, maker of the chip.

$5 is an arbitrary price, and in no way shape or form comparable with a business making a single board computer and having to buy components at market rate.

For the lazy:

   - 1 GHz processor
   - 512 MB RAM
   - 4 GB Storage (No SD card support)
   - WiFi b/g/n and Bluetooth 4.0
   - Composite output by default but HDMI and VGA are supported with an adapter (Sold separately)
   - 1 USB Port and 1 USB Micro with OTG
   - 8 GPIO

If you're not using the display pins, you can use at least 23 more pins as GPIO.

We've had cheap computers in every generation, remember the Sinclair ZX81 ? ($100 when a PC was $1500+)

What sparks tinkering and innovation is open minds and creative thought processes.

(as well as laws that don't get in the way of tinkering)

Those kinds of minds find the tools, not the other way around.

> We've had cheap computers in every generation, remember the Sinclair ZX81 ?

I would dispute that. That was the 8-bit era. An entry level programmable computer in the sub $100 price range has been missing for a long time.

A $10 smartphone today is as powerful if not more powerful than most of these attempts at low end computing.

Just needs a keyboard and HDMI output.

Do $10 smartphones exist today?

In the USA there is a $10 smartphone sale practically every week of one brand or another.

This week: http://www.bestbuy.com/site/tracfone-alcatel-onetouch-pixi-g...

It's very low end for a smartphone but can run almost any android apk and use wifi even without airtime. Some $10 phones can be rooted and then you almost have a full linux-like system.

For $20 you can get quad-core with 1GB of ram: http://www.walmart.com/ip/Straight-Talk-LG-Prepaid-Destiny-L...

I'm not sure a $10 smartphone is going to be bought for a elementary school child to learn to program on. Nontechnical parents are not going to buy something like that, root it, and teach their child to program. The glory of the 8-bit era was the ready to go nature of the programming environment and the cheap price.

On the second hand market, as a locked down android phone, absolutely.

I really don't have any faith in the company, I have an otto camera. They abandoned all support and development for it as soon as it was shipped to backers. (one update) Felt like, "we've got your money, here is the minimal amount of stuff to keep us legal....now go buy or other stuff"

The problem with these kind of articles and releases is that for most people, $9 vs $100 isn't that big of a difference... At least in the U.S. and other first-world countries where an initiative kid can earn either in a relatively short period of time, and most parents could gift it to them if not.

As others have and will point out, it isn't just $9, there's the need for another system to be able to read documentation, and work on while doing projects on said $9 computer... there's also the peripheral costs, kb, mouse, monitor which would likely run it up to close to $100 anyway.

I think the issue is sparking ideas and interest in younger people... most are more interested in writing phone/tablet apps than they are in touch with tinkering. There's a reason that Radio Shack had to shift gears, and in turn died off...

Don't get me wrong, the availability of cheap computers with decent i/o pins is really nice, and for anyone interested in integrating electronics and software, there's never been a better time... But having yet another <$100 option really isn't going to spur extra tinkering at this point... And getting exposure into schools is particularly difficult given that most people with the skills to actually work and teach for these things are likely employed at a much higher salary than they'd get in a K-8 role (at least 2-3x).

> The problem with these kind of articles and releases is that for most people, $9 vs $100 isn't that big of a difference.

What world do you live in? Median household income in the U.S is $51k.

I think you are missing the entire picture. I have to agree that it is not a big enough cost difference because if you go spend the $9 and then try and use it without $100 more in equipment or already existing computer you will be disappointed. So it simply is not going to spur that much more interest.

Actually, that isn't true.

People lacking basic means can often work, trade, scrounge, repair, etc... to get a working, usable system. And they will do that too.

Even in the 1st world, most kids would have to wait longer to ask for something that costs $100 vs $9.

Not counting IOT which is already huge on its own and just focusing on children, maybe the point of things like this isn't the 1st world?

but what if I wanted to make a 100 computer/robot network now it's a big difference. or my drone army to ward off neighbors...

Is it really that huge compared to options before CHIP?

I'm going to go out on a limb and say it's not the price of the electronics (which are very cheap), or the availability of them (they are plentiful), or even the knowledge to learn (this is the age of the Internet on mobile devices), but is actually the price of another thing that that is going up: time (and by extension, focus).

I feel that students have to start planning for college earlier and earlier these days, even planning how courses line up as early as middle school. When all academic scheduling decisions (sports, classes, extra curriculars, clubs, etc.) are always viewed through the lens of college admission, there is very little time to turn brains off and just play with the knowledge they've gained thus far. If an activity isn't directly linked to job training of one or two strong focuses that could potentially lead to a major in college and a job afterward, they aren't seriously considered as an activity worth spending time on, and that is a pity.

I would add lastly that as a counterbalance to this extreme need for academic focus, instead of seeking invigorating hobbies to which students at the end of the day will lack the energy for, they instead seek more mindless and passive activities which are readily available on their mobile devices. The harder they work, the dumber they want their hobbies to be.

It certainly lowers the bar to entry.

Somewhere out there, there's a kid with an idea that will one day become the next Zuckerberg or Woz. Whereas it was next to impossible for my peers to convince our parents that we needed a $3,000 computer setup just for learning, these young people can earn the money for their own supplies with a summer job.

I think the biggest improvement of the CHIP over the Raspberry Pi, besides the price, is it's got on-board WiFi and flash storage, and is pre-loaded with an OS. Those are nearly always the first 3 things I need to take care of when starting a new Raspberry Pi project.

I think these computers (Rasp Pi, CHIP, Beaglebone, BBC Micro) are doing what the one laptop per child couldn't do. That is to be the affordable computer to the 3rd world. Almost everyone in the west can afford a $100-$500 computer, but that is still beyond the reach of many people in the 3rd world. That means basic computer and programming skills will be within reach for a large number of the population in the world.

Also, currently, most of the development work available are for PCs, Mobile, and Tablet devices. These chips are the building blocks of the Internet of Things, meaning development work needed in the future will go beyond just programming for PCs, Mobile, and Tablet devices.

I think the biggest positive aspect of widespread distribution of raspberry pi and devices like it is to ward off the fear from adults in the school system.

If the school board comes out and says, we're purchasing these things for an entire district - this is an official movement from the school board top down - the governor signed off on this and we're getting these out to every single kid in our schools, teachers can no longer point to microelectronics and say that it's a bomb.

Hopefully fewer students get sent to the police station for tinkering with electronics.

People do know it's possible to get $1 ones on alibaba?

an Arduino clone is something like $1.70 on aliexpress and has a fraction of the processing power so it's not exactly a fair comparison

An arduino clone is almost infinitely more useful as a teaching tool than these arm based computers are, though.

Source: I teach people to program microcontrollers every day.

completely agree, the complexity of Linux actively hinders learning. Much easier and useful to learn on a bare-bones microcontroller.... I certainly learned on a 4MHz z80 and have a much more in depth understanding of software and hardware as a result.

I believe I would have been better off with better math courses in junior high and high school than I would have been learning computing. Perhaps that's too path dependent, and based too much on the sheer level of pain experienced as an undergrad having to learn it properly.

But I was hacking (analog) electronics ( at a very shallow level ) so maybe the little RasPi class machines make up for that.

A $9 computer that cannot be shipped for less than $10 is a $19 computer.

From getchip.com, shipping to Canada comes out to $6.22 for a single CHIP, and $11 for 5 CHIPs

The VGA/HDMI converters ($10/$15) cost more than the computer. (!)

> And unlike Microsoft products, CHIP has both open-source software (Linux) and hardware.

That seems like an unnecessary dig.

Not really? It's a valid point, it's generally easier to tinker with open source software.

I agree. Maybe it should read "unlike Apple, Microsoft, Samsung, X, Y, Z, CHIP has both open-source software..."

actually that seems like a lie, it shipped with usual arrangements of allwinner crap binary blobs

NPR covering the news in true NPR style... Five years too late.

doesn't this still require a monitor and keyboard? they should provide "starter-kits" for the uninitiated... but yeah $9 once you know how to use it is great.

Self-contained handheld version. $49

http://getchip.com/pages/pocketchip

plus you're going to need a second computer to read the online documentation and forums and so on.

We should make them run code on a cardboard computer [1] first :)

[1] https://en.wikipedia.org/wiki/CARDboard_Illustrative_Aid_to_...

Reply on news.ycombinator.com