Hey /diy/ I have been wanting a Raspberry Pi-like for a while,

>expensive in my country

Auto fix'd.

Going to cost you more to make than import.

But if you still want to, start with the requirements.
Obviously USB.
HDMI?
Ethernet?
WIFI?
Linux?

And no idea how you're going to solder any kind of decent chip manually without experience.

Unless by "Raspberry Pi" you mean a generic dev board in which case it's a whole lot easier.

>>1043115
This will cost more than its worth by several orders of magnitude.

If you cant afford a raspberry pi then get an orange pi. And if you cant afford that then you got some serious problems mate.

>>1043136
Yeah, sure, I would like to start with a small project and make it scalable perhaps. Or just a small project to learn and make it better next time.
USB, Linux and Ethernet is more than fine and I'm sure it will be a big challenge.
I have decent solding skills, also. No problem to destroy some components as well.

The main difference between a generic dev board would be running Linux I guess? Not so sure what you mean with generic dev board to be honest.

Thanks for the response!

>>1043140
The money isn't really an issue, I can afford it, I just want to make this a small challenge since I don't actually need a Raspberry Pi, it would just be a cool think to have and to tinker with.

>>1043146
If you're in it for the learning experience, don't. There's barely any thinking involved.

You buy a chip and necessary I/O and solder it to some board. Although I don't believe RPI chips come in DIP so you'll need to pull off some serious witchcraft to solder it onto a board that you'll need to get fabbed elsewhere ($$$).

Better off making a diy-duino.

Crap these days is so integrated that "building your own x" is not that useful beyond learning how not to screw up a PCB design. And SoCs generally have NDA'd datasheets and require PCB tech that's very expensive in small quantity.

Personally, as a EE, I think I would've been better served by learning a little bit more about software engineering. Stuff like networking and OS concepts.

>>1043209
>Although I don't believe RPI chips come in DIP
hahahaha, you are very funny little man. You can't buy the chips for raspberry Pi's. DIP? Stop you are killing me. RPIs use custom made SoCs.

Get an Arduino instead of a Pi if you're into electronics.

>>1043115
1) Find a mcu that is well-documented. Most manufacturers provide no documentation at all unless you pay lots and sign a "loyalty" paper.
2) Check u-boot and Linux kernel support.
3) Read design rules for high-frequency signals, especially bus related (e.g. RAM bus has strict requirements)

File: Capture.jpg (116KB, 1274x869px) Image search: [Google]

116KB, 1274x869px

>>1043115

I'm making my own Nixie clock right now, and let me just reiterate a point already made:

You're going to spend more money on making the board than buying one online.

The components aren't super expensive, and since you're a student your time is worth basically 0, but the tools and having a board printed (even if you do it yourself) is going to cost about what you would pay for one bought online.

Sorry, I know it's probably not what you want to hear, especially on /diy/, but unless you're doing it for a hobby/shits and giggles/project, you're better off buying one.

>>1043427
>components aren't super expensive
Don't listen to this guy.
RPi-level CPU are expensive, and so are RAM memories of a reasonable size.

>>1043115
Sounds like a pain in the ass, probably will cost you the same. How much is it to import?

>>1043427
Is there a chunk missing from that diagram? There are two 5v leads from the power circuit.

Also been wondering, is not isolating the uC in a nixie clock actually safe? Can't help but have this feeling that something might go wrong and fry it.