Latest blog entries

It has been a few years since I last put together unofficial LEGO sets using MOC instructions from Rebrickable and ordering parts from BrickLink. Today is when I clicked publish on one of my own designs.

Once of the core podcasts that I listen to is Do Go On, a fact based comedy podcast hosted by these three Melbourne comedians:

I don’t recall exactly when I first saw it, but there was LEGO set for creating a custom BrickHeadz figure of yourself. The marketing images for that set include a figure that is a spot on Matt Stewart, and another that is close to Jess Perkins:

There was no immediate example that was close to Dave Warneke, but to be honest he is a white guy that records a podcast, so how hard could he be to represent?

I sat on this idea for a while, until a few months ago I decided to finish off some incomplete MOCs. If I was going to be ordering pieces, I might as well order more.

I didn’t go straight to building, first I downloaded the instructions for the Go Brick Me set, this is broken up into the variety of hats, hair, facial features, clothing and other choices. After a few hours of tinkering in Studio I had a design I was happy with:

I added these parts to my orders, waited for them to arrive and then put them together:

More photos can be found at Do Go On as LEGO BrickHeadz on Flickr.

But what now? How should I get these to the podcast? I have seen them at a number of live podcasts in Melbourne (and in Koh Samui back in 2019) but I knew that their next Melbourne live shows wouldn’t be until October, which I would not be able to make it to. I could post it to them, but then I wouldn’t be able to see their reaction.

For Do Go On I am on the Patreon tier where you can submit a “Fact, Quote or Question” that they will read out at the end of the podcast. I have submitted a number of these before and I could do so again, making sure that the package had instructions to wait for the FQQ and the FQQ opening with a note to ensure they had the package.

This worked as today (episode 568, skip to around 2:26:00) the package was opened. In the video version of the episode the reaction from Jess is exactly what I was after.

I have now shared this information and photos with the other Patreons, and also hit publish on the design file I had uploaded to BrickLink.

I have written before how my computer mouse of choice has been the Microsoft IntelliMouse Opitcal 1.1A, so much so that I have a total of seven of these mice, all of which have been repaired over the years. However late last year I discovered a new problem in that the rubbery coating on the mouse wheels is starting to deteriorate. I tried to order some replacements, but they were not good enough.

My search for a replacement has meant that for the last couple of months I have had three mice connected:

I wanted to keep the cost low which meant excluding wireless options, but when you get into it there are not that many wired options. I also wanted a simple black mouse without flashy RGB effects.

My first candidate was the base model Razer Cobra which was working well, however was noticeably smaller than the IntelliMouse which I kept switching back to.

My second candiate was the Cooler Master MM310. This is closer in size to the Intellimouse and I found that I wasn’t switching back.

Today, after three months of only using the MM310, I gave my desk a tidy and kept only the MM310 connected. I also placed an order for another one, so I would have a spare.

(this is mostly a post for my own reference)

At the camera club we use IrfanView to display our monthly competitions entries. The images to be presented end up in a number of folders which are copied to the club’s laptop, then batch scripts run IrfanView with the /slideshow option.

Recently I reset that laptop back to a fresh install of Windows 11 (the only way to uninstall all Adobe software) so I also needed to reinstall programs such as IrfanView, which I did.

But then I tested running a slideshow, oh I need to turn off the text which shows the image path. No problem, there is a page in the settings for slideshow:

I turned off “Show text” and turned on “Hide mouse cursor”

Tested the slideshow again… that didn’t work.

I loaded a single image and went full screen, that doesn’t show the text. What is happening?

I then spent an hour or two trying all sorts of options like reverting to an older version and targeting an ini file in the local directory, none of them worked. Though I could change the position of the text, so some settings were working.

I had been so fixated on the text showing that for a while I didn’t notice the other issue, that the images were advancing automatically, not waiting for mouse/keyboard input.

Where is that setting? Oh, those are when you go to start a slideshow:

There it is at the top left, “Sequential after mouse/keyboard input”

Hold on… why is there also a “Show text” setting here?

Ah, if I change the setting and then Cancel, it still changes settings for slideshows, so now when I run from the batch file it is doing what I want.

Not the best design, mixing settings for running a new slideshow with settings for future slideshow.

Is is also only now that I have taken these screenshots that I see that “Full screen / Slideshow” settings page is split, most of the settings apply to both full screen and slideshow, but the bottom settings (where “Show text” is) is below a “Full screen ONLY” divider…

When I originally setup Zigbee2MQTT on a Pi I followed the recommended path for Linux, which was to install Node.JS, clone the Zigbee2MQTT repo and then start it up. That first startup would install dependencies and build the app.

Over time I was able to at least once use the provided upgrade script to upgrade to the latest version in-place, but it then got to a point where the upgrade would crash the Pi and fail. Due to the Pi I am using only has 512MB (some reserved for video which I don’t use) one recommendation was to set memory limits for the build, but that just caused the build to fail with an out of memory error.

So I didn’t upgrade.

I knew I needed to get the upgrades working so when I started considering upgrading the Pi I use for Kodi an additional benefit was that I could replace the Pi 3 A+ with a Pi 3 B+ that has double the memory. But then what would I use the current A+ for? It felt a bit of a waste to upgrade the hardware for just the build process, because otherwise the A+ is running Zigbee2MQTT fine.

I preparation for a potential new install of Zigbee2MQTT I went back and reviewed the install instructions, which is when I found that their recommendation is now to just use Docker. I am fine with using multiple docker containers on my linux box to run various services, but it still felt a bit wrong to have a computer for a single task running a full operating system which then runs docker for a single container…

But it works and I have now upgraded Zigbee2MQTT to the latest version without upgrading hardware.

Later this month I am heading up to Sydney and have built in three full days to wander around with my camera. As well as researching new places to go I was also looking back at my past photos, only to find that on a previous trip in 2008 (the early days of me having a DSLR) I had only done something with two images, and then left the rest.

While there were no spectacular shots, there was enough there to end up with a decent selection so I put them up as an album: Sydney 2008

I then looked further into my older photos and found a few more trips or events where I had more unprocessed photos, so I worked my way through them and put up four more albums:

• Puffing Billy 2009
• Yarra Valley Hot Air Ballooning 2009
• Cape Otway Day Trip 2010
• Phillip Island 2011

My photos after these have been suitably processed, so now I need to get back to what I have taken over the last year.