Cover photo
Paul “LeoNerd” Evans
Lives in London
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Other profiles
Contributor to
Basic Information
Other names


BBC Micro, TRS-80, ZX81, VC20, TI-99, and something else we have no idea; at the Science Museum in London.
Alex Taylor's profile photoEd S's profile photo
Ed S
The buying of the Spectrum clone from an outside market in Novosibirsk:
"The clones come in a variety of shapes, colors, and designs, and bear little resemblance to their Western counterparts. Their motherboards were made unofficially in state electronics plants by underemployed workers, who then assembled the computers at home and sold them in ones or twos either privately or at flea markets. We end up purchasing two Sinclair clones; one of them comes with a guarantee -- a handwritten note with the telephone number of the teenager who assembled the device. Cost: the equivalent of $19 U.S."
Add a comment...
I'm considering how to drive a bunch of WS2812B LEDs (otherwise knowns as #NeoPixels) over DMX, with likely some kind of AVR chip inbetween. But see, I have this problem.

WS2812s want a very precise async serial signal to control them. A single bit takes 1.25µs, so a 24bit word for one LED is 30µs.

DMX-512 comes at 250kBaud (1 start, 8 bits, 1 stop), meaning a new byte arrives on the UART every 40µs.

Meaning: there's nowhere near enough time to multiplex them both at once. The moment I start talking to the LEDs I'll have to ignore DMX, and vice versa.

This means talking to 120 of my LEDs takes 3.6ms, reading an entire DMX frame takes 20ms; during either of which I can't be doing the other.

My current options seem to be:

 a) Read DMX continuously, ignoring the LEDs, until DMX says I should do something; do that thing by talking to the LEDs and ignore DMX, then swap back

 b) Split DMX and LEDs across two chips, because in practice I can ignore most of the DMX most of the time, and just pull out the single intensity channel I need to control the LEDs.

Does anyone have any thoughts?
Electra Flarefire's profile photoPaul “LeoNerd” Evans's profile photo
+Electra Flarefire: Well, I have just bought a long strip of 2812 tape.
Annoyingly, I did just discover the lovely APA-102, which would also solve this problem; they're basically SPI.
Add a comment...
Clicking the publish button on ConnectBot v1.8.2 at the pub!
6 comments on original post
Add a comment...
Folks into #electronics and #PCB design: Know any articles about when to use vias, and when to avoid them in favour of longer snake tracks?

I don't feel I have a good grasp of when to prefer one of over the other in any particular situation. Looking for some guidance. Nothing high-frequency or high-voltage - mostly at-most 12V, microcontrollers, sometimes up to 20MHz or so.
Lars “Ebi” Nielsen's profile photo
Add a comment...
After I was done ranting about feature requests, someone asked me if I wanted a pony with it. Here is my response
Emma Skyra's profile photo
Just wondering, but r u a 12 year old from ny with a identical twin named raphiel? Sorry just need to know if u r my friend from school
Add a comment...
I entirely agree.

There's a growing trend lately where the more automation and technology we have, the fewer workers we really need; and yet we're still growing more and more people.

We're already starting to see the strain now, but sometime soon, maybe in a decade or so, something is bound to snap. We'll hit a point where there's more people than we can possibly manage to employ doing useful work.

As a society we need to start asking some serious questions now, so we have the answers ready before that happens.
JP Sugarbroad's profile photoSergey Shepelev's profile photoPrzemysław Szczepaniak's profile photo
We need to redefine useful work -- again. We've done it in the past, implicitly, but now things are moving too fast for that.
Add a comment...
I've been trying to get this SSD1306 OLED display to work. All attempts so far have failed to actually get my data on the display.

I can easily write to the control registers - e.g. to set brightness, scrolling, invert, etc... Just I can't make any data appear. All I seem to get is random noise that was in GDRAM at the time. This both from custom code talking I2C via the Bus Pirate, and in a last-ditch attempt, this photo here showing the display hooked up to an Arduino Mini, running Adafruit's example code (see

I begin to wonder if my display unit is defective somehow and is unable to write to GDRAM.

Anyone have any last ideas before I give up and buy another one?

Edit: Not shown in the photo is the 4.7uF decoupling cap or the two extra 2.2k pullup resistors I just added to the I2C lines. Neither seems to have made any difference.

#electronic #microcontrollers #arduino #Adafruit  
Paul “LeoNerd” Evans's profile photoElectra Flarefire's profile photoMorgan Gangwere's profile photo
Have you double-checked the datasheet? 
Add a comment...
In preparation for the v1.8 release, we are asking for help translating the Play Store descriptions. Every effort is appreciated!

If you don't want to register for Launchpad, you can post the translations as comments on this post and we'll put them in.
2 comments on original post
Add a comment...

Paul “LeoNerd” Evans

Discussion  - 
We've just released ConnectBot version 1.8.2. This beta-testing release contains more bugfixes for the keyboard, along with some new options to control the behaviour of sticky modifiers. This should make keyboard input better across a range of devices including hardware keyboards.

We're hoping to promote this out of beta into being the real production release - so please get testing and let us know of any new bugs since the last release.
Add a comment...
So I think I worked out why I was having trouble with my #soldering last night... Someone tell me if I'm going crazy here, or if this sounds plausible

The solder I have doesn't seem to have flux in it, because no amount of melting it ever produces any white smoke. However, if I cut through it with a sharp knife I do still see what I think is an inner core that looks "different" - still grey, but less shiney than the outside. I also observe that I need to get the iron quite hot on the temp. control in order to melt new solder off the reel, but having melted it once and let it cool into a ball, I can remelt that at a much lower temperature; even at a temperature which will not melt fresh stuff.

I think therefore, that what I have is 63/37 solder comprised of a solid core of 37% lead, surrounded by a jacket of 63% tin - needing, therefore, to be hot enough to melt each on their own the first time, at which point they mix and alloy themselves together into the actual eutectic mixture.

Does that sound plausible?
Paul “LeoNerd” Evans's profile photoIngvar Mattsson's profile photo
So not stale flux. :) Does setting the tip temperature to 250 Celcius produce any evidence of melting anything? That should be enough to melt the tin, but not the lead.
Add a comment...
Dear world:

I have a computer, A, on which is a big disk with lots of music. I have a computer, B, which is attached to my screen, keyboard, and speakers. Ohyes, and B is a laptop which moves from place to place.

Get music from A onto B.

 * Mounting filesystems is far from reliable, often gets upset over laptop suspend/resume

 * mpd is great for transport control but doesn't carry the audio path

 * Pulse's network audio solutions are all crazy, don't work over routers, spin my router CPU

 * trx can't cope with mismatched audio rates

 * Shoutcast copes terribly with packet loss, building up many many seconds of audiolatency.

Someone please explain how - I'm refusing to believe it's as difficult as I'm finding it.
Steven Thurgood's profile photoIngvar Mattsson's profile photoPerry Lorier's profile photoOliver Brammer (toastedtruth)'s profile photo
Plex / XMBC server on machine A, and machine B streams via the browser or client? It's an elegant solution. 
Add a comment...