I’d been wanted to build myself a new desk for a while. The current one I had, while being a good size, was pretty terrible as a desk. The top surface was all flaking off, and it was a simple trestle style. Many years of heavy monitors had given it a distinct bow.
The place I’ve got my desk is also quite large, I can fit in something bigger than a standard 1.5m or 1.8m long desk. Trying to find bigger desks is hard, and quite expensive. The solution came to me when I was looking around Bunnings one day. They sell extruded aluminium tubes, and a bunch of angle connectors to join them together. You can basically make anything that’s of square design. I sketched up a design. The main requirement was strength, I didn’t want the desk to bow, particularly with its long 2.3m length. So I added a third leg, a third beam along the top and lots of cross-bracing.
Here’s the pile of beams from Bunnings:
I then cut them into shorter lengths with a hacksaw, and started joining them together.
Here you can see a couple of the angle-connectors.
Partially-assembled leg parts in a wine box:
The top surface is a giant piece of 22m MDF. This is definitely overkill, I should have gone with a thinner piece of MDF.
The two sides of the desk, with top. You can see the triple-leg design.
Here’s the final desk in action, complete with huge piles of crap.
The two vertical posts that you see on the right-side edge are for future upgrades – I’m thinking of putting some shelves above it. All up the total cost of the desk was about $220. I’m very happy with it. It’s large and strong, and doesn’t bend.
One of the games I really enjoy playing is called Warframe. I wrote this article for Wikia on my builds for the game. I thought I’d cross-post it here for my records.
I’ve been wanting to write an article for a while on my current Trinity builds, but I have been too lazy. I was finally jarred into action by this article on Reddit, which contained so many god-awful builds that I just had to jump in with my own (god-awful :)) builds.
Like many, I was originally disconcerted by the changes made to Trinity last year, but I didn’t join with many of the others who took up pitchforks to storm the DE offices. The usage of self-damaging Bless builds at the time was definitely unhealthy for the game as a whole and I agree that the changes needed to be made. However, since my main build at the time was Bless, I found myself a bit lost for a while, until I settled down on a new main build, which I’ll discuss below.
I love Trinity as a frame, so much so that I have eight of them (so far). At my current stats, I’ve got about 45% of my playtime with Trinity, with the next closest being at 10%. Her survivability is right up there with some of the frames that avoid damage completely, like Loki, and she can transfer most of this survivability to everyone in the squad.
I still remember the first time that Trinity ‘gelled’ for me: I was about MR8 doing a Infested survival mission with some friends. We were doing it tough, with one or the other of us going down fairly regularly. I was down in the pit that you get in the Eris tileset, completely surrounded by infested, and all the pieces of Trinity just came together for me. My train of thought basically ran like so:
I can build my energy and the energy of the team, that’s good, super useful.
I can use Link to reduce the damage I take.
If I wait until my health gets low, then hit Bless, I get great damage resistance.
Oh, wow. Now my whole squad is completely invulnerable for fifteen seconds.
These infested aren’t scratching me, and I’m obliterating them.
I can keep doing this forever!
At that point, I put away Oberon (my previous favourite frame), and mained Trin. I’ve never looked back. No other frame has come close to winning me over as Trinity has.
In my view, there are a few different character types.
Team survival support
Team DPS support
Just about all frames will fit into one or two of these categories. For example, Banshee is a classic DPS and ‘DPS support’ frame. Irradiating Disarm Loki is ‘Individual survival’ and ‘Crowd Control’. I believe that Abating Link Trinity (as described below) covers four of these five character types, making her (in my opinion) the most useful frame in the game. Her pure self-containment makes her robust enough to take into any mission type, and also play a leading role in the squad.
Whereas previously you needed two separate frames to have an ideal EV trin and an ideal Bless trinity, these days, with Bless being less important, you really only need the one frame to handle both builds, as they have the same polarisations. This is good, as really no-one but us die-hard Trin-heads will want to have more than one Trinity in their arsenal.
A note on auras.
You’ll note in both my builds below that the aura polarisation is ‘wrong’. I like to have builds which are independent of the aura polarity so that you can use whichever aura most suits the team you’re with. You can save a couple of forma if you go with a Corrosive Projection – defined slot, but I like being able to swap out with Steel Charge, or Rifle Amp, or whatever suits the current mission you’re on (Dead Eye, anyone? Anyone?)
I thought I’d start off with an EV build, since it’s been unchanged for years. Min duration, Max range, High Power Strength. The classic Energy Vampire build will keep your squad absolutely awash in energy, but will have no survivability whatsoever, other than that provided by overshields and Quick Thinking. A single stray bullet will take you out in high-end content, since it’ll ignore QT. Your Bless will heal your squad, but have an damage-resistance duration of just over one second.
EV Trinity build
Without a doubt, it’s useful, if very narrowly focused. Fortunately, the advent of the Zenurik Energy Overflow passive has made EV Trin largely obsolete, with the exception of extremely cheesy squads.
Abating Link Trinity
Now on to the ‘real build’ of this article. As I mentioned above, after the Trinity nerf in mid-2016, I was a bit lost for a while on how best to play Trinity. My current Bless build still had 30 seconds of duration, but was only doing a flat 50% damage resistance. I could get it up to about 70% without losing much duration, but it just didn’t have the same useful feel as before.
An off-hand comment on Reddit provided the solution. Someone mentioned Abating Link, but said that it was only okay, as it was too compromised. Too little duration, or not enough armour strip. I set about fiddling with my builds, trying to get a good build going.
The build I eventually ended up with is fantastic (IMHO). It gives excellent duration, with 100% armour strip. A combined Link / Bless will give you 92% damage resistance – right up there with the old Bless builds, but it’s available at will, and with none of the risk that surrounded the self-damage builds. (Hands up those who died because they were stuck in a QT stagger-lock after getting hit by a mob after throwing your glaive. Thank you.)
If you haven’t experienced it before, let me say that 100% armour strip is amazing. Grineer and Corrupted just absolutely melt from underneath you. In long survivals and the like, I almost always end up with the highest damage percentage.
I take this build into all missions, including those with Corpus and Infested enemies. You may ask why, as those factions don’t have armour. A few reasons:
Laziness. The most powerful force in the universe. Yes, even more powerful than a Mirage with a Simulor.
Bursas. While Corpus doesn’t have armour in general, Bursas do. A lot of it. However, a Bursa – immobilised by Energy Vampire – with its armour stripped will just disappear underneath your melee weapon.
Juggernauts. As with Bursas, Juggernauts without armour are very squishy.
So here’s the build:
Abating Link Trinity build
As you can see, the key is getting up to the critical 223% Power Strength that you need to give you 100% Armour Strip. This build gives 224%. On top of that is keeping the range steady and upping the duration as much as possible. Efficiency is the dump stat, since you’re Trinity, and you can recover energy at will, and the high power strength means that you get a lot of energy for each EV cast.
Abating Link: This is the linch-pin of the build. It doubles the practicality of Link, since it turns it from straight damage-resistance to damage resistance and damage-boosting, greatly amplifying the DPS capabilities of Trinity.
Growing Power: This isn’t a mod you see a lot in people’s builds, but it’s extremely useful in this one. It steps up the armour strip from a very-useful ~90% to an enemy-melting 100%. It gives you that extra 25% Power Strength that you need without taking up a regular slot. It’s easy to trigger, as you only need to generate a status proc. Since, in high-end combat, you’re almost always using a weapon with at least a moderate status chance, the counter is running all the time. (If you’re using a weapon that has no real status whatsoever, bring along Taxon, and it’ll keep growing power continually running).
It’s a fairly new mod to the game, but very useful, as it now ups the maximum possible power strength to 374%. (As an aside, once you get it, make a 374% power strength Eternal War Valkyr. It’s amazingly fun.) It’s not the easiest mod to get, but as long as you take your scanner with you on every mission, you’ll easily pick up the plant scans that you’ll need in a reasonable time-frame.
(Primed) Continuity, Constitution: Duration. Duration is good, as you don’t want to have to recast too often. With the Arcane Aura helmet, the duration for Bless is up over 20 seconds.
Quick Thinking, Natural Talent, Primed Flow: These mods are pretty straightforward, and speak for themselves. I put Quick Thinking on any frame that has >600 energy pool, as it greatly increases survivability. Natural Talent is also an essential on all heavy-casting frames.
Exilius Slot: For me, this is the only ‘variable’ slot in the build. As I see it, you have two options here – Cunning Drift for the increased range, or Speed Drift for reduced cast time. I’ve gone with Speed Drift, as casting time is critical, since you’re casting a lot. I don’t see the extra range as being particularly useful. If you’re on a big map, you’re primarily relying on Bless to give you the damage resistance you require while you close in on the enemy.
Weapons: I prefer to go melee with this build. Since you’re focussing on Link to provide you with both damage boost and damage resistance, you’re in fairly close range already. Close that gap a little further and go melee. The primary weapon that I use with Trinity is Mios. Good status chance, enormous range and great attack speed. The range and the rag-doll that Mios does gives the CC boost that Trinity needs. Another top weapon option is Lesion with its pure elemental damage and bonus toxic proc. It’s got some advantages in that it doesn’t need the ‘spool up’ time that a critical-damage weapon like Mios has. Primary and Secondary are whatever you feel like, but focus on Toxin and Slash damage.
I prefer the Mios for missions in which there is a continuous stream of enemies, like Survivals, and also missions which have a lot of Nullifiers. The enormous range helps pop the bubbles safely.
The Lesion works better for missions in which the enemies are start-stop, like Defense, where you can’t keep your combo counter up continuously.
Energy Reduction Sortie
This is a significant weakness with the build above. In an energy-reduction sortie, you’ve only got 150 energy, which is not enough to pop Link, Bless and then have enough energy left over to cast Energy Vampire to top your energy pool back up. In these sorties, I swap out the Blind Rage with an Intensify.
‘Poor Man’s’ Abating Link build
If you would like to try out Abating Link, but don’t want to have to sink five forma into Trinity and get Growing Power, try this build. It’ll give you 97.2% armour strip, which is great, but won’t give you those tasty red bars.
Abating Link build 2
The main loss here is the 15% casting speed, and the 10% power strength.
How to Play
It’s fairly straightforward. The most dangerous part of the mission for a Trinity is literally the first 20 seconds, until you can start building your energy pool. With a small starting energy and low efficiency, you don’t have enough to cast Link or Bless, and then cast Energy Vampire. So you have to cast EV first, which means getting close to an enemy without any protection. So find your first enemy, EV, then drop Link. After that, it’s pretty smooth sailing. Just keep Link active all the time, for the protection and armour strip. Pop Bless when you or your squadmates need it, and keep topped up with energy. Have fun!
Here’s a video of the build in action, versus twenty level 145 Corrupted Heavy Gunners.
For the last few years, I’ve been using Quadra Hosting for my web hosting needs. Their servers are great, with good speed and connection, and they’re reasonably priced. They also have the best customer and technical support that I’ve ever seen, bar none. I highly recommend them if you’re looking for a new web host.
Let’s Encrypt is a project that has set out to ensure that all web traffic is encrypted. To that end, they have changed the traditional process of obtaining SSL keys to a much simpler one, and provide them for free.
Also, Chrome will soon be warning users anytime they visit an unencrypted website. With these two factors, I thought I’d set about trying to install a Let’s Encrypt key onto my Quadra hosting account. As you can see from the green lock in the address bar for this site, you can see that I was successful!
Let’s Encrypt’s preferred method of installation is via CertBot client that runs on the server, and installs and renews the keys. However, a quick check with the technical support team at Quadra, and I realised that this wasn’t going to work on shared hosting, where I don’t have low-enough level access to the server.
Fortunately, Let’s Encrypt and others have provided a variety of different ways to install some keys. They have scripts for a variety of different languages. One that I came across was a bash script that can run on shared hosting without root access. It’s simply called acme.sh. It worked really well.
Here’s the process that I went through to set up Let’s Encrypt with the acme.sh script.
Go to your Quadra Hosting control panel, and turn on SSL connections, by going to ‘Domain Settings’, then ‘SSL’ and then clicking the ‘Generate self signed SSL certificate’.
Click ‘submit’ on the next page, then just ignore the page of keys that comes up after that. You won’t be using those keys, but it has activated the SSL functionality for the server for your domain.
2. Next you’ll need to install the acme.sh script. To do this, and most of the rest of the steps, you’ll need shell access. It’s not turned on by default, so you’ll need to fill out the shell access request form and send it through to the Quadra Hosting support team. Then, when you’re logged on, enter
curl https://get.acme.sh | sh
This will download and install the acme script. At this time, it tries to install an alias to acme.sh, but it failed for me. I had to manually edit the .bashrc file and added in the following line:
This can be done simply by entering the following line
This contacts the Let’s Encrypt servers and will generate some SSL keys, and copies them to the acme install directory. It’ll give the location of these new certificates:
[Thu Nov 10 01:46:53 GMT 2016] Your cert is in /hsphere/local/home/xxxx/.acme.sh/systematictechnology.net/systematictechnology.net.cer
[Thu Nov 10 01:46:53 GMT 2016] Your cert key is in /hsphere/local/home/xxxx/.acme.sh/systematictechnology.net/systematictechnology.net.key
[Thu Nov 10 01:46:53 GMT 2016] The intermediate CA cert is in /hsphere/local/home/xxxx/.acme.sh/systematictechnology.net/ca.cer
[Thu Nov 10 01:46:53 GMT 2016] And the full chain certs is there: /hsphere/local/home/xxxx/.acme.sh/systematictechnology.net/fullchain.cer
4. Next up, you need to install the newly created certificates. This is done in two ways. Firstly, through the control panel, then through the command line. This is so that both Quadra and the script know the certificates.
Firstly, log onto the server with an FTP client and download or view the four certificates which were listed in the key creation message (above).
On the control panel, go to ‘web options’ then select ‘edit SSL support’.
This will bring up a page with some areas in which you can paste in the keys. Put in:
Install yoursite.key to ‘Certificate key’
Install yoursite.cer to ‘Certificate file’
Install ca.cer to ‘Certificate Chain file’
The Certificate Key and Certificate file can be uploaded at the same time, but the Certificate Chain File will need to be uploaded separately.
Once that is done, go back to the command line. You’ll need to change the permissions on the ssl key folder:
Note that this will throw a few errors: The script will also try and install some backup keys, which it won’t be able to do. I don’t consider this a problem, as the keys primary location is in the acme.sh directory anyway.
5. Reboot your apache instance to reload the new keys. There is no direct to do this, but the Quadra support team suggested a simple workaround. Go to Web Options for your domain in the control panel, then click on the ‘Mime Type’ button. On the window that pops up, enter some dummy data, hit ‘submit’
Then click on Save / Apply in the Web Options window. Wait a few minutes, then click on the red ‘X’ next to the new Mime type. This will delete the new entry.
Click Save / Apply again. This will reboot the apache instance, and load up the new keys.
6. Test the installed certificate. Wait for five minutes for the server to reboot, then go to https://yoursite.com . If all the steps above have worked then it should be working nicely. Next, check that the cron job to renew the certificates has installed correctly. This should have been done when the acme.sh script was installed. Check it now.
This can be done through either the control panel (go home -> tools -> cron) or through the command line:
7. Celebrate! You’ve now got SSL working on your site, celebrate that you’re doing your small bit to make the whole web a bit more secure.
After I posted this article, I sent a link through to Quadra support with a note along the lines of ‘here’s a guide I made up, please feel free to send it to any other customers who might be trying to do the same thing.’
I shortly got a response from the support team. The team member who’d been helping me with this setup not only responded and corrected some mistakes I’d made, but he also wrote: “You could also mention that we would be happy to do all this for them if they asked us to, e.g. if they are not used to shell commands.”
This surprised me no end. The Quadra Hosting team sell a one-click SSL implementation, which only costs $30, and requires no technical skills at all. Rather than suggest “You could also mention that we have a cheap one-click SSL implementation if they are having difficulty with Let’s Encrypt.”, they instead suggest that they could forgo making that money, and instead spend time help you implement a no-cost alternative.
If you’re not with Quadra yet for your hosting, change. This is just a small example of how good they are.
In the CB400 thread on Netrider, Positron007 speculated that the CB400 (which accelerates really quickly to 70kph) accelerates as fast as a 600cc sportsbike, at least up to the first gearchange.
He speculated that the limiting factor (up to 70km/h) was not the power of the bikes (which are vastly different) but the fact that you’re trying to prevent the bikes from doing a wheelie for this whole time.
So today, I thought I’d use my nerd powers for good instead of evil, and take a crack at explaining a wheelie formulaically. I couldn’t really find anything explaining the physics behind a wheelie on Google, so I thought I’d have a crack at it myself.
Disclaimer: May be wrong.
TL;DR: Positron is right, 600cc bikes don’t really accelerate faster (to 70kph) than the CB400.
Assumption 1 – That both the CB400 and the 600cc sportsbike (which I’ll call the CBR600RR from now on) are both at their ‘liftoff point’ for the whole drag-race (which will be up to 70km/h, after which the CB400 is not capable of wheelie-ing).
Assumption 2 – The CB400 weighs 270kg fully laden with 76kg rider, and it’s COG (with rider) lies 600mm and 500mm from the rear axle.
Assumption 3 – The CBR600RR weighs 273kg fully laden with 76kg rider, and it’s COG (with rider) lies 650mm and 550mm from the rear axle.
Assumption 4 – Gravity is 9.81 m/s2
Why is the liftoff point important? A bike is accelerating at its fastest when the front wheel is at the point of becoming weightless. If the bike is accelerating any faster than this, then the front wheel will continue to lift off the ground, until the rider is on his back. (See endless YouTube videos to see this effect in action).
At liftoff point, both the liftoff and the gravitational turning moments are equal. What do I mean by this?
To describe this as a formula:
Moment (downwards) = Moment (upwards)
I don’t think that this formula applies to drag-cars and other very-high performance vehicles. The principles do apply, but they also have some massive lifting forces created by the torquing action on their drivetrain. Bikes do not have torque-induced lift, especially at the low-power levels we’re looking at here.
The turning moment is the force multiplied by the perpendicular distance to the turning point, and can be described by the formula:
Mt = F * D
To show the turning moments visually:
Horizontal Turning Moment
Vertical Turning Moment
Force is described by the formula
F = m * a
So, to combine all these equations
Solving for acceleration provides:
This is quite an interesting result, as the mass cancels out of both sides. Ergo, the mass of the bike has nothing to do with the bike’s ability to keep the front end off the ground. (Naturally, the bike’s mass will have a lot to do in limiting the amount of acceleration available to the bike.)
So it all comes down to the D(h) against D(v) ratio. In retrospect, this makes sense, as it is well-known that an adequately powerful cruiser can out-accelerate a sportsbike, due to its low COG and long wheelbase (more forward COG).
So comparing the CB400 against the CB600RR:
So the CB400 will actually out-accelerate the CBR, though there is not much in it.These accelerations will mean that the bikes will do 0-70km/h in 1.7 seconds. NOTE: The numbers for the vertical and horizontal distances are pure guesswork. Any assistance to provide real numbers would be greatly appreciated.
The maximum acceleration threshold for a bike is limited by its turning moment.
At any point below this threshold, acceleration will be limited by other factors, such as power, gearing, weight, etc.
The turning moment is a factor of the height and distance of the COG of the Bike (and rider) from the rear-axle.
Any two bikes which have sufficiently similar COG distances will have the same maximum acceleration threshold.
After getting Mac OSX working fine on the Lenovo, I decided to reach higher and go for a real multi-boot setup – five OS’s! I wanted to put on Windows 8, so that I could play Steam games, and I wanted Linux to see how Linux has progressed over the last few years, and BSD just to see if I could.
My first step – upgrading the hard drive. The 320GB Seagate that I was using was fine for two partitions, but wouldn’t really cut it for more. So I bought a Seagate 7200rpm 750 GB Hybrid HD. With that done, I had to partition it up, which was quite a feat:
170GB for the original Windows Partition
120GB for Mac OS X
200GB for Windows 8 (pretty much only for Steam)
60 GB for Linux
50 GB for BSD and
180 GB for general storage (formatted FAT, so that all the OS’s can share it.)
I used Mac OS’s Disk Manager for the partitioning – it does a good job, and it’s a lot harder to make some fatal mistakes.
Once that was done, it was a pretty straightforward, though time-consuming task to put all the OS’s on it. The Mac OS and Win 7 partitions I could copy straight across from the 320 GB drive. Windows 8 installed without too much hassle, but only once Windows 7 was on. It wouldn’t install into the non-first partition without another boot partition being in place.
For my Linux partition, I chose Ubuntu, which is definitely ‘flavour of the moment’. I’ve installed Linux in the past – Red Hat 3 on my old Pentium 2 300Mhz desktop computer, and Gentoo on a Sony Vaio Picturebook. Compared to those old installs, Ubuntu installed like a dream, I’d say fractionally easier than the Windows 8 install, and getting close to that of OSX.
I wasn’t completely thrilled with Ubuntu. It’s probably the Gnome-based Unity desktop, but I found a few of the OS choices to be quite annoying, particularly the dock on the left. There was also a surprising lack of configuration options in order to change the user experience. Sleep / wake worked just fine without any setting or configuration changes, which impressed me a lot.
One thing that impressed me though was the wireless connectivity. With my last Linux install – Gentoo on a Picturebook – getting wireless networking running required a lot of hard work, and a lot of editing of configuration files. However, Ubuntu made the experience as easy and fast as Windows or OSX does.
I was trying to go for five OS’s – with BSD being the fifth OS. Unfortunately, I wasn’t able to get it to work. It was very frustrating. The first time I tried to install it, I then discovered that it would only install into a primary partition, not into an extended partition. After much grinding of teeth, I re-partitioned the hard-drive and started from scratch.
I then moved Windows 8 to an extended partition, so that I could install BSD on a primary. Then, for some reason, I just could not create a bootable USB installer with BSD on it, it just wouldn’t work. The X220 doesn’t have a DVD-drive, so I had to go with a plan B – move the hard-drive across to another laptop which has a DVD-drive, then install, and move the HD back across for first boot.
Using this method, I was able to get PC-BSD installed and working, at least if they had full control of the bootloader. Once I tried to install another bootloader, to boot the other OS’s, then BSD would no longer work. I tried both the FreeBSD and the PC-BSD variants, but neither would boot. I eventually gave up. The BSD partition was only meant for ‘fun’ – a bit of icing on the cake, but it was proving to be more hassle than the rest of the OS’s put together.
Only after I had gone through all of that drama with installing on another computer that I realised that I had another install option. I could have used Virtualbox, and booted up the installer in Virtualbox, then installed on one of the other drives.
Getting the boot loader to work is generally a pain with these multi-boot installs, but I was able to get that sorted pretty easily. I used my Windows-to-go bootable USB to boot into Windows, and then I used EasyBCD to configure the boot loader on the computer. So my boot loader now looks like this:
After I had got it all working, I went back and had another look at Ubuntu. I wasn’t completely happy with the Gnome environment. Fortunately, there’s an officially supported branch of Ubuntu called Kubuntu, which uses the KDE environment. I found this to be much more to my liking, and a lot more configurable. The only downside is that the wireless networking isn’t quite as easy or smooth as Ubuntu’s, and takes a bit longer to re-initialise itself after sleep.
All up, I’m now very happy with my work-issued Lenovo X220. It does absolutely everything I want, and can run any program I want. It’s got a good size and is extremely robust. I think that when it’s time to hand this computer back, I’ll go and pick up a 220 for myself.
One of the things that really annoys me with modern computers is that they don’t use their RAM as much as they should – particularly with video and music playing.
Case in point. I was watching a movie on my new Hackintosh the other day, and the hard disk light was flashing continuously. The computer wasn’t doing anything else except watch the video, and I had more than 6GB of free RAM on the computer. However, the video player (VLC in this instance, but they all do it) wasn’t using that RAM, it was continuously reading from the HD. That’s okay for a desktop computer or a laptop with an SSD, but for a laptop with a regular HD, that’s a real waste of battery power. The 800MB video file could easily be read into RAM, then accessed by the video program.
One of the best ways to get around this problem is to create a RAM disk, something that used to be a regular feature on Apple computers. The obvious ways of doing it were removed with the introduction of Mac OS X, but the underlying capability is still there.
I was able to find out how to do this pretty quickly, through a Terminal command:
This creates a RAM disk that’s about 1.6 GB in size. Modify the last number to make the disk larger or smaller.
One thing that the articles didn’t mention is how you can easily turn a Terminal command into an ‘executable’ that you can click on to run. So I thought that I’d post a quick ‘how to’ to show how to do this.
Open up a text editor (I use TextWrangler) and paste the command into it.
Save it as a text file (I call it ‘ramdisk.sh’) to the drive.
Once this is done, do a ‘get info’ on the file. Go to the selection ‘open this file with’, and select Terminal.
Once this is done, when the script is double-clicked, it will open up in Terminal and run:
This will create the ram disk, and put it on the desktop, like any other type of network or local drive.
You can then copy your video or other files to it, and get extremely fast access to those files. It’s great as well if you want to do some video editing / photo manipulation, but you have to remember it’s volatile memory. If you lose power for any reason, you’ve totally lost the contents of the drive.
As I mentioned in my last post, one of the remaining issues is that the Lenovo X220’s wifi card isn’t recognised by Mac OS X. One of the ways around this is to replace the wifi card with one that’s recognised by OS X, and this is the path most take, as the new card only costs about $15, and works seamlessly with the computer.
I didn’t want to do this though, since the laptop belongs to my employer. I didn’t want to open up the case and break the warranty. Instead, I went to ebay, and bought the smallest USB wifi dongle I could. It was a Edup RTL8192C 802.11n dongle.
It’s remarkable the miniaturisation that they’ve been able to do for this card. The fact that they could pack the electronics and an aerial into this tiny package is amazing. It’s only 150N networking, and the range isn’t great, but that’s to be expected.
Naturally, the drivers that came with the dongle don’t include Mac drivers. It took a bit of hunting to find some, but I eventually found them. I’ve included them below in case anyone finds this page when looking for drivers.
When installing on Mountain Lion, the installer says that it fails, but it actually works. To initiate the wireless connection, you have to use the ‘Wireless Network Utility’ app which is installed. This app replaces all the built-in wireless connectivity
An interesting side-effect is that the wireless dongle also works in Windows, so you can connect to two wireless networks simultaneously.
For the last few weekends, I’ve been doing a little project, trying to get Mac OS X running on my work-issued Lenovo x220 laptop. It’s quite a good laptop with an i7 processor, 8 gigs of RAM.
However, I can’t really use it. It’s got a pretty locked-down copy of Windows 7 running on it, and I don’t have admin rights. That’s fine, my workplace pays for this computer, they can set it up how they want. So I decided to experiment, to see how I could use the computer in other ways without affecting the ‘work’ nature of the computer at all.
The first thing I tried was to swap out the hard drive. The Lenovo makes this pretty easy, with just one screw on the outside case. With one of my spare laptop drives in, I then installed Windows 8. Naturally, that worked just fine, but switching out the HD every time I wanted to do something different is a bit cumbersome.
So then I tried a new feature of Windows 8 called Windows To Go. It’s basically a full install of Windows 8 that can run from a USB stick or external hard drive. I then created a WTG external drive. This worked fine, and allowed me to play Steam games with no hassle, and without affecting the work hard drive at all.
After a few weeks, I thought that I’d see what other possibilities might be achievable. I’ve really missed my MacBook Pro since the motherboard died about a year ago now, but I was never quite able to justify to myself to buy a new one. I’ve always been an Apple and Mac fan, so one day, I did a search, to see if it was possible to create a ‘Hackinosh’ with the Lenovo.
I found a few pages that had some methods on how to do it. The main one that was very useful was from the ThinkPad Forums.
The main issues I had to work through were:
Creating a boot USB stick.
Installing Mac OS X
Copying across the Windows 7 partition and fixing the bootloader.
Creating the USB boot disk gave me a bit of grief at first. I tried creating one using various instructions that I found on various pages, that mostly involved copying files across from the Mountain Lion installer DMG. This method copied the files across, but didn’t leave me with a drive that the Lenovo would see as bootable. The next method I tried was from using the Lion Disk Maker tool. However, while this tool works really well with genuine macs, it doesn’t work so well for Hackintoshes.
The method which ended up working for me was by using TonyMac’s UniBeast tool. This worked well, and got me to the installer window. Then I got stuck at the next step. The installer only works for GPT-formatted hard drive’s and the Lenovo (since it uses Windows 7) only allows MBR. (Windows 8 now works with GPT). Fortunately, the main Thinkpad Forum page came to the rescue again, pointing to this page on the OSx86.net site.
On that page are some hacked installers, which allow you to install onto MBR partitions. Once these were copied across onto the USB installer, it then allowed me to install, and get to first boot.
At first boot, things got a little tricky. At this point, you need to install various extensions to get the computer to recognise all the hardware, change a few configuration files, set up the boot manager and patch the DSDT, so that control of the processor is correct. At this point, I got a lot of kernel panics. Much fun.
The forum posts lists two methods for first-boot setup, one by ‘Fraidos125’, the other by ‘Superkhung’. Some people record success using one method, some people record success using the other. Unfortunately, neither worked for me. I kept reading through the thread, trying to figure out why I wasn’t having any success. Then, toward the end of the thread, I saw that someone suggested using both methods in conjunction, one after the other. I gave this a try and had success!
Here’s a photo of Mac OS running on the Lenovo.
There are only three residual issues with the ‘Hack OS X’ install – the wifi doesn’t work, the bluetooth keyboard installer pops up every time you boot and 3rd USB port doesn’t work.
Wifi not working is a recognised issue with the 220’s. Mac OS just doesn’t have a driver for this wifi card. Most people just buy a new wifi card and install it into their laptop, since a new card only costs about $15. I don’t want to do this with a work-issued laptop though, so I’ll find a USB alternative.
The second issue – the bluetooth installer is slightly annoying. The keyboard and trackpad is PS2, and work fine with the right extension. However, on bootup, the computer doesn’t quite seem to recognise that a keyboard is connected, and starts up the bluetooth keyboard installer every time. It’s easy to quit out of it, but it’d be better if it didn’t show up at all.
The third problem is strange. The x220 has three USB ports, two USB2, and one USB3 port. The USB3 port doesn’t work at all. I suspect that there’s something wrong with my configuration files. Most x220’s only have 2 USB ports, with the i7 computers having three. I think that it’s one of the files that I downloaded from the Thinkpad forums that’s causing the issue. It hasn’t bugged me enough yet that I want to fix it, but I’ll probably add that to my list for later.
Once I had Mac OS installed, I decided to try and make the computer dual-boot, so I wouldn’t have to swap out the hard-disk everytime I went to work. Doing this was almost as tricky as the MacOS installation. The main problem is that I don’t have administrator rights on the work partition (Windows 7), so I had to work completely ‘hands off’ from it.
One of the things which helped this was the WTG boot drive, and this came in very handy for the dual-boot set up. It enabled me to boot up the 220 without having to touch the internal drive, allowing me to manipulate it as I needed.
The first step was to copy an image of the Win 7 partition from one HD to the other. I tried a couple of methods, with no success, but then I found a method that worked. For this I used a tool called DriveImage XML. It works like the image copy-and-move capabilities of Mac OS’s Disk Manager. It allowed me to copy across the Windows 7 partition from one drive to another, with no changes.
Once the partition was copied across, I then re-installed MacOS into the other partition. Then I just had to get a bootloader working so that I could select which partition to start. The Hackintosh installer uses a boot-loader called Chameleon, but I couldn’t get this to work with the Win 7 partition. I tried creating a Win 7 rescue USB and fixing the Win 7 partition, but that didn’t work either.
The easy way to get both to work is to use a program called EasyBCD on Windows. It works very well, but I don’t have admin rights on the Win 7 partition. Fortunately, the WTG drive came to the rescue again. Running EasyBCD on it, I was able to fix the bootloader on the laptop’s drive. It boots into the Windows Boot Loader (the finest of text-based user interfaces), which then enables me to select which partition I want to boot from. If I select MacOS, it then goes to the Chameleon boot loader, which allows me to select boot-options for the MacOS partition.
Now both partitions are working great. I’ve got my pristine work partition, and a Mac OS partition for fun. It makes me want to install a few other OS’s, just to see if I can. Maybe a Quin-boot Work / MacOS / Win 8 / BSD / Linux setup.
Motolegion are a brand of jeans designed by Australian brand RHOK. I recently purchased a pair, and this is my review of them.
For the purposes of this review, I’ll be comparing the jeans against some work-issued jeans that I have – some King Gee jeans.
King Gee’s (left), RHOK (right)
Ordering and Delivery
Ordering was very simple from the RHOK website, though the website seemed confused as to how much the jeans were – the page said $145, but the Paypal checkout said $146.
Once I placed my order, I waited for 11 days, with no jeans in sight. I sent off an email, and was quickly replied to by Marinko. They were out of stock, but had forgotten to let me know. They said that they would ship the jeans in a couple of weeks. A couple of days later, however, I received another email, saying the jeans had been mailed, but they had probably sent the wrong size.
When the jeans arrived, they were indeed the wrong size. Marinko was very apologetic, and immediately sent up the correct size jeans, and a pre-paid express post satchel to send the wrong jeans back.
My first impression of the jeans was how thin the material was! I was expecting a very heavy, thick denim, but the denim felt a lot thinner and flexible to my fingers than the King Gee’s. A measurement with my micrometer confirmed by suspicions – the RHOK’s measured at 0.77mm vs the King Gee’s at 0.93mm. This obviously doesn’t matter from a protection standpoint, as the Kevlar does the protecting, not the denim, but the jeans may not last as long as I’d hoped.
What the jeans lack in thickness, they definitely make up for in weight. The King Gee’s weigh in at 804 grams on the scale, the RHOK’s are 1.10kg! I don’t know where the weight comes from – probably the kevlar panels, and the huge amount of thread which has gone into stitching these jeans together.
The stitching is very heavy duty. I’ll comment more on it later.
The jeans are not subtle at all. Red stitching all around, and a giant logo on your ass. If there was one aspect of the style of the jeans that I could live without, it’s the logo. Given that I usually wear work-issued jeans most of the time, and these come with the corporate logo on the back, I’m not totally adverse to this idea. However, the sheer size of the logo is off-putting.
Giant Ass Logo
Some of the panels, such as above the knee, and the seat also come pre-distressed. The ‘blue’ colour is also very dark. While not black it’s definitely on the ‘navy’ side of blue.
The jeans are quite long, about 1cm longer than the 87R King Gee’s, which are a very traditional 501’s – style cut.
They’re also quite wide at the boot. You’ll have no trouble at all fitting these over your Sidi race boots. Each leg averages about 1cm wider when laid flat.
Width and colour difference between the two jeans.
The belt loops are a little bit narrow in length. My belt just squeezes through.
The stitching is extremely heavy-duty. All of the important seams are triple-stitched. One area of concern is that all the stitching is external, and doesn’t have the ‘two under, one over’ style that is found on other jeans – like the King Gee’s. This raises a slight concern that in a bad slide the stitching may possibly wear through, though this seems highly unlikely, given the strength of the stitching.
Internal Stitching, RHOK’s
Internal Stitching, King Gee’s
Each stitch has a massive amount of thread going through it. I’d like to know just what percentage of the weight of the jeans is made up of thread!
On the bike – When on the bike, the jeans are perfectly comfortable, and don’t get in the way at all. However, I found that one of the seams appears to settle right on the inside of my knee, which is uncomfortable when hugging the tank. When I sit on the bike, I need to spend a second to make sure that the leg is rotated so that the seam is out of the way.
Off the bike, the jeans are surprising uncomfortable. The main culprit is the velcro panels inside the knee pockets. Because they face towards the outside of the jeans, the velcro strip basically sits against your leg, tempered only by the thin mesh material. The sharp edges (especially the corners) of the strip poke through the mesh, and onto your leg.
This isn’t unbearably uncomfortable, but it’s pretty much continuously noticeable. I’m hoping that they soften up with wear and washing.
The Kevlar panels themselves are completely comfortable. You won’t even notice that they’re there.
The Kevlar protection is excellent, huge panels of kevlar weave covering all parts which are likely to slide along the ground in an accident. All the panels are the same size as the denim panels, and stitch into the same seams.
The hard protectors are highly weird, when compared to my other pair of riding pants (Motoline textiles). They are a standard ‘hard plastic over breathable foam’ style of protector. The knee protectors are a standard shape.
The shape of the hip protector, however, seems all wrong. They look like they have come out of a jacket, not a pair of pants. For comparison:
This is a side-by-side with the knee / elbow-forearm protectors from my Spidi jacket next to the shoulder / hip protectors from the jeans. The shoulder / hip protectors are nearly identical in shape.
Perfectly fitting ‘hip’ protector
When the hip protectors are in place, it makes you look like you’re wearing jodhpurs – it gives you fantastically wide hips. The protectors just stand out from your body, and don’t conform at all. There is a good centimetre of air from your body to the protector in the middle of the hump.
By comparison, here’s the hip protector from my Motoline textile pants.
One of these is designed for human hips.
The knee protectors can be inserted and removed through cleverly designed hidden zips below the knee.
The knee protectors go into mesh pockets that have ‘hook’ velcro on the bottom of the pocket. In addition to causing the comfort issues mentioned above, they also destroy the pocket itself. Every time you open the pocket, the hooks on the velcro tear at the mesh, which in my case is already starting to fuzz up from the abuse.
Knee pocket with Vecro ‘hook’ side up
The knee protector pockets are massive in the ‘up’ direction – they go halfway up my thigh. However, they’re not long enough in the ‘down’ direction. When I fit the pads so that they’re comfortable with a bent knee, the pads stick out the bottom of the pocket, so that the pocket can’t be done up! This is an epic flaw in the design.
They fit just fine if I rotate them 180 degrees, but I think they were meant to protect the shin, not the thigh. At this point in time, they’re just not usable. I’m going to either have to cut the knee protector with a Dremel tool (not a generally good idea, since it’ll leave some sharp edges), or buy a more standard foam protector and cut it to shape.
I bought these jeans hoping to ride them to work every day, to pop out the knee pads, and then work all day. As it stands, I can’t see myself doing that.
The knee pads are too difficult and time-consuming to put in. The hooks on the velcro also tear up the mesh extremely rapidly. Taking the protectors out once or twice a day is going to destroy the mesh in very short order.
I have two options – ride to work with the protectors in, then change to a different pair of pants. If that’s the case, I may as well wear a pair of waterproof textile pants. My second option is to ride without the protectors in place. That’s the option I’m taking at the moment, though riding without knee protectors is a concern.
The style and cut of the jeans seems aimed at a younger market segment. This is not a bad thing, but something to be aware of if you’re interested in purchasing these jeans.
Overall, I’m disappointed in the jeans. They’re not as good as I was hoping for, in comfort and practicability. I honestly can’t see myself buying another pair of these jeans, unless some of the recommended changes below are made.
Please note, that these recommendations are only based on my personal experience and tastes – YMMV. These are in order of my preference.
1. Fix the knee pad pockets or protectors, so that they actually match each other in size.
2. Remove the ‘hook’ velcro on the inside of the mesh pocket. Replace it with ‘loop’ velcro on the top side of the mesh pocket, then allow the owner to decide if they want to attach the ‘hook’ side to the protector.
3. Have a hip protector that is actually designed for human hips.
4. Have a second ‘subtle’ jeans style, for those who don’t want a giant logo on their ass.
5. Have a jeans style that has a more traditional tapered leg cut.
I really enjoy a good home-brew ginger beer, it’s a perfect summer drink. I’ve made a few different varieties from the different pre-mixed cans that you can buy, but I can never remember which ones are the best.
A month ago, I made up a fresh batch of ginger beer. This time, I took a photo of the pre-mix can, so I’d know which one was good.
I had a bit of a disaster as I was making it. I accidentally poured the wort over the hand-mittens that I was wearing to protect my hands from the heat. I was certain that I’d infected the batch, but I went ahead anyway.
After two weeks fermenting, I opened up the fermenter. There was a nasty black mould around the thread of the fermenter, but the beer looked and tasted great.
I bottled up the beer. After a week of gassing, I gave it a first taste. It was really excellent. The temperature-controlled fridge had done another great job. There were hardly any off flavours to the beer at all. This kit has only a light ginger flavour, and comes out tasting quite similar to the famous Bundaberg ginger beer.
Next time, I think I’ll try making a ginger beer from scratch. To do that, though, I’ll have to buy a wort chiller, to cool it down enough for fermentation.