Heatmiser WiFi thermostat

I mentioned a while back that I was waiting for a Heatmiser Wifi thermostat to test, and thought it about time I gave an update, though you’ll have to wait a little longer before my full review. I’ve had the thermostat installed for a couple of weeks, though it’s a beta version and I’m just waiting on an updated revision to arrive.

The PRT TS WiFi RF uses a remote switch to control the boiler

To recap, the important thing about the unit I’m testing (besides its WiFi geek factor) is that it needs only two wires to the main thermostat, which makes it a fairly straightforward replacement for a lot of domestic UK thermostats. The version I’m testing is the ‘PRT TS WiFi RF’ and should be available in March; it’s a two part kit made up of the thermosat itself and a remote switch, which is controlled by a separate wireless link.

So, there are two things to wire up, but at least you don’t have to run a new cable all the way to the thermostat location, if you only have a two wire+earth connection at the moment.

The remote switch is a standard UK plug/switch size, so can easily be surface mounted or fitted in a wall box near the boiler. It needs a mains supply – which I’ve taken from the same spur that powers the boiler – and then a switched output goes to the appropriate contacts to turn the boiler (in my case, Vaillant combi) on and off.

DIY horrors

The thermostat itself is a little bigger. The rear part is designed to fit in a standard wall box, but it is actually about an inch larger in each direction, so that there’s sufficient space for the touchscreen. Most thermostats that I’ve seen tend to be surface mounted, so you’ll probably need to drill and fit a back box – if you were to put the thermostat in a surface mount box, it would look decidedly odd.

This is the sort of horror that lurks inside my walls

Fitting the back box for that was one of the more interesting parts of the process; I’ve never drilled into a mains cable before… my flat was a pretty simply done 1980s conversion and almost all the electrical points are surface mounted, with the exception of the light switch next to where the thermostat is mounted. I checked that the wires from that switch ran upwards, away from where I was planning to drill for a new wall box, and started work.

After a while, all the lights went out (thanks to the RCD I installed shortly after buying the flat). And this photo shows the reason. The wire that’s been damaged is actually for the next room – instead of going up the wall in that room, it goes all the way through the brickwork and then up the wall in the hall.

And it’s only a few mm below the surface; the original lathe and plaster walls seem to have been simply covered with wire mesh; the new wiring was held in place on the mesh by nails banged in either side of it – I found the same in the kitchen, with the 30amp feed to the cooker – and then about 5mm of concrete render laid on top of the mesh, with a skim of plaster.

So, a bit of a mess to sort out, but in the end I got there, with a back box installed and all the wiring for the thermostat; the existing twin+earth cable was disconnected from the boiler, and connected to the switched spur to provide a 240 volt feed to the thermostat.

Simple setup

The config tool is available for Mac and Windows, but Linux users should have no problem, either

Before connecting the thermostat to the mains, you need to configure it on your computer. There are Mac and Windows setup programs for this, but Linux users should be ok too; when you connect the thermostat via a USB lead, it simply mounts on the desktop as a drive containing a CONFIG.TXT file, and this file is what’s read and written by the setup utilities. It’s plain text with lines like

KEYMODE=WPA2

GATEWAY=10.0.1.1

so a quick read through and you’ll be able to set the options, even without the config program.

Save the config, unplug the USB cable, and then you just need to snap the front of the thermostat onto the back plate – there’s a simple jumper-style connector that mates the front part with the backplate that’s screwed to the wall box – and you’re pretty much done. The instructions say you need to pair with the remote switch, but my units found each other anyway, so I was all set, and everything worked pretty well right out of the box. There’s an iPhone app – which is due to be revised soon, along with an Android version – and of course the built in web server. The protocol used by the iPhone app is documented, and essentially the same as the serial protocol that the company’s equipment uses, and there are third party tools that work with it. For example, if you head over to http://code.google.com/p/heatmiser-wifi/ you’ll find some scripts that will log your temperature in a MySQL database, and plot graphs on a web page, which I’ve set up too, and work fine with the RF stat.

When I’ve finished reviewing, I’ll talk more about usage, configuring your firewall, and so on, but for now, here’s a screenshot of the current wifi interface.

The HeatMiser wifi interface, running on my home network

Maxing out the MacBook Pro

My current main computer is, like a lot of people’s, a laptop; since I sometimes work at home, and sometimes elsewhere, it’s a good solution, and at home it’s plugged into a wonderful TactilePro keyboard and an Apple 20 inch cinema display. I used my previous PowerBook for three years, and this one will be four years old in April; it’s an early 2008 MacBook Pro 15.4 inch, 2.5GHz Core 2 Duo, for the more technically inclined.

Some MacBook Pros can actually take 6GB of RAM, rather than the 4GB Apple claims

Memories

When I bought it, 2GB seemed sufficient, and so too did 250GB of disk space. Over time, as OS and applications have become more bloated, I’ve needed more memory, first upgrading to 4GB, which is supposedly the maximum allowable. However, even that is tight these days – especially with bloated applications like Adobe DreamWeaver CS4. And somehow, no matter how many times I try to remember not to, I always end up launching it while running Parallels Desktop at the same time, which is not a pleasant experience.

A bit of hunting around on the internet, however, reveals that this model of MacBook Pro can actually handle 6GB of memory, even though Apple claims to support only 4GB. So, that was last year’s upgrade, and it certainly means things are a little smoother, though two memory-hungry programs at the same time is still a recipe for plenty of swapping and general sluggishness. Definitely an upgrade worth doing if you have one of these machines, but I’m afraid I didn’t think to do before and after timings.

Storage

The hard drive in the MacBook Pro is a 250GB 5400rpm unit; it’s doubtless good for battery life, but if it starts being used for swap space, performance suffers. And, thanks to various bits of video editing, it was getting a bit on the slow side.

My first experiment was adding a 48GB Wintec Filemate card (Amazon affiliate link) to the ExpressCard slot on the side of the machine; it is, apparently, possible to set things up to boot from the card slot, though that’s not a route I went down. I did experiment with using it to store the files for Parallels virtual machines, but found that really didn’t work too well, with the VMs seeming to wait from time to time. I guess virtualising another OS is going to make it hard to optimise performance for an SSD. So, I have been using the card to store other documents, including all my web design files – one of the sites has over 20,000 images – which helps DreamWeaver switch sites more easily.

I’ve also recently moved all my working copy on to the ExpressCard; I’ve done that not so much for performance reasons, but because as well as being used in SATA mode via the slot (which is fast, as unlike some cards, it’s not got a PCI<->USB bridge in the way), the card has a USB socket, so it can be removed and used as an external drive too. After my experience last year when the laptop had to be sent away for a replacement motherboard, I feel much happier knowing that I can simply pop out the data with all my current projects on, and plug it into another system.

To make more space, I’ve also shifted all the video files that were on the internal hard drive on to an external Firewire 800 drive; in truth, I probably won’t need all the raw video again, but I dislike deleting stuff, so that’s a more rational place to store it than on the laptop’s own hard drive.

Having done that, I ended up with about 50GB free, but a machine that’s still feeling a bit sluggish at times. A new MacBook Pro would set me back around £1500; sure, it would be faster, but most of the time I’m just using tools like Word, and the difference wouldn’t be significant. Also, if I buy a new system, I’d end up with OS X Lion, which I just don’t want, and which won’t work with some of the Quark tools I absolutely rely upon. No sale there, Apple (you can’t help wondering if they realise how conservative corporate IT can be).

Enter the SSD

The price of SSDs is much more reasonable now than it was last year – a 512GB unit can be bought for not much over £500. But, tempting though that is, it would be around a third of the cost of a new laptop, to extend the life of a machine that’s almost four years old, and that struck me as possibly not the best way to spend money.

So, since my rearranging of data has left me with 50Gb free on a 250Gb drive, I decided that the best thing to do was to go for a 256GB SSD, with the intention that it will help me get a couple of years more out of the MacBook Pro, by which time there might be a version of the Quark tools I use that actually works with whatever the latest version of OS X turns out to be.

With pretty good reviews, and a reasonable price, I went for a Crucial m4 SSD, supplied with a data transfer kit, comprising a cable and software CD. Ordered on Friday, it arrived Saturday morning, and I spent the afternoon transferring data using SuperDuper, which is a pretty simple process.

The Crucial m4 SSD comes with SuperDuper for Mac users

It’s apparently pretty smart, and doesn’t copy some files that Apple advises against; the net result was that I ended up with a drive that had 60GB free after the duplication. However, it was pretty slow, taking over four hours to copy the files over.

My update process

The drive was supplied with firmware 0009, and a note in the box indicating that there was a firmware update available on the web site; that’s supplied as a disc image that you can burn to a CD and then boot from, whereupon it will search for SATA-connected SSDs and update their firmware.

Don’t waste your time doing this before you copy your system files to the SSD; yes, the release notes tell you that it’s quicker to update the firmware on an empty SSD, but the updater won’t see the drive anyway, unless it’s connected to the motherboard. The USB/Sata cable can’t be used for the update.

So, step 1 was to connect the drive to the MacBook Pro using the USB Sata cable; a pop-up asked if I wanted to initialise it. I called mine ‘m4′ to differentiate it from the Macintosh HD, and accepted the default options – SuperDuper erases the disk again anyway.

Install that from the CD supplied with the drive, and then download the zip file of the firmware updater from the Crucial website. Inside the zip is .iso file, which you can burn to CD using Disk Utility.

Step 2 is to tell SuperDuper to copy all the files; this will be slow. Use the “Backup all files” option and select your main hard disk as the source, and the SSD as the destination. SuperDuper also takes care of making the SSD bootable. Go and have a lie down, or something; this bit takes time.

When the copy is finished, you can try booting from the external drive, if you want, by restarting, and holding down the Alt key to see a list of available drives. I just took it on trust, frankly – I could see the files were there by browsing the SSD.

Step 3 is the physical installation; for this, I recommend downloading and printing out the PDF from iFixit. It’s a pretty straightforward task. You can start up from the SSD right away if you want, but I didn’t. Instead, I popped in the CD I made earlier, and held down the Alt key.

When you do that, you’ll get a boot menu, and the CD is labelled ‘Windows’; click to select it and FreeDOS starts, leading eventually to a prompt asking if you want to update the drive firmware. Type yes and press Enter, then wait a few minutes. You’ll see a message when it’s done, and you can then hold down the power key to turn the MacBook Pro off.

Step 4 is starting up from the SSD. It was here I had a minor wobble, after removing the CD and then seeing a message telling me there was no bootable volume found. Don’t panic if you see this; just hold down the Alt key when you start the laptop, and select the drive, by whatever name you called it.

When it starts up, open the Startup disk control panel and make sure you select the SSD. I chose to rename mine back from ‘M4′ to ‘Macintosh HD’ just in case there are any scripts that have that name hardwired into them. And then, all was well, with the system starting up and shutting down perfectly.

Job done

So, was it worth it? So far – and I’ve only been using it for a little while – I’d say a resounding ‘Yes.’ I expected a speed increase, but not as dramatic as I received.

For these timings, I started with the system turned off, and timed how long it took until the desktop wallpaper appeared, and then until all the startup applications had finished loading, indicated by the appearance of the TigerLaunch icon in the menu bar.

With the hard drive, the wallpaper appeared after 57 seconds, and the TigerLaunch icon after a total of 91 seconds. With the SSD, those times have shrunk to 25 and 26 seconds, respectively. So, a saving of 1 minute 5 seconds in system startup time.

Next, I launched Firefox, which I have starting with a blank page. This would take fifteen seconds using the hard drive, and is now at most two seconds using the SSD; I say at most, because I have to press the stopwatch button, and it’s near enough instant.

Finally, while leaving Firefox open, I launched Dreamweaver CS4, and timed how long it takes until the splash screen listing recent files appears in the workspace. With the hard drive, that used to take 51 seconds, and now it takes just 11.

Definitely, then, a dramatic improvement, and though I’ve not timed before and after, running Parallels VMs is much snappier now too. If you have an old MacBook Pro like mine, and don’t fancy being forced into a Lion update or spending the cash on a new machine, I can definitely recommend boosting the memory to 6GB and replacing the hard disk with an SSD.

Netflix or LoveFilm?

If you’ve been asking yourself that question, my article comparing the two is now live on RegHardware.

Broadly speaking, I think that I’m more likely to keep paying for Netflix after the end of the free period, but that’s because I tend to enjoy working my way through some of the ‘box set’ type material that they have, and I don’t personally think that LoveFilm has a great TV selection right now. If I were more into watching films, then I’d likely make a different choice – but during the week, I don’t tend to watch TV that much in the evenings, and for me delving into another episode of something on Netflix when there’s nothing more to watch will be a lot more likely than sitting down to watch a whole film.

The other factor is that the TV set I’m presently using – a Samsung D8000 – will have to go back to the PR folk soon, and my own set, a Panasonic V10, doesn’t support any of these services. But I can still access Netflix using my iPod touch and play it through the dock on the AV system, and I’m looking forward to trying out one of the Roku media players too, which will be slightly better solution.

BBC details of the Streetdance 3D broadcast

Regular readers will recall that at the start of the month, I wrote about the broadcast of Streetdance 3D on the BBC HD channel, and speculated about the method that was being used to provide support for 2D viewers.

In a blog post today, Ant Miller at BBC R&D reveals how it was done.

What can you watch with IPTV?

The first of my articles for RegHardware’s IPTV week has just been published. It gives an overview of what sort of content is available via the main IPTV services in the UK.

Inview ties up with Acetrax

While many of the video streaming services such as LoveFilm and Netflix are only to be found on the higher end TV sets from companies like LG, Panasonic and Samsung, Swiss firm Acetrax is looking to extend its reach still further.

The company – which offers content on a pay as you go basis, with material to rent or buy – has teamed up with Inview to roll out the service across set top boxes that support their new connected TV platform.

Acetrax, running on Inview's new connected TV platform

Inview is perhaps best known to some people in the UK as the providers of the EPG data that drives the Radio Times Extra (formerly Teletext Extra) and TopUpTV services, but they also have a connected TV platform rolling out, which is available royalty free to ‘tier 2’ equipment manufacturers who don’t want to have to create their own. Inview then makes its revenue as a share of that generated by content providers such as Acetrax.

Acetrax is going to be on that new platform, along with Grooveshark and some other services, around the middle of this year. A spokesperson for Inview wouldn’t reveal the names of the brands involved, but I understand that it’s likely to appear in FreeviewHD set top boxes and connected TVs from the house brands of major UK retailers, and similar brands around Europe. There will probably also be some SD kit that includes it too, though given the closing gap between the prices of SD and HD kit, I suspect all but those on the tightest budget will be looking at HD, especially with the Olympics coming up this summer.

The new platform apparently integrates apps and a recommendation engine with Inview’s 14 day programme guide, rather than splitting content services off into a separate part of the interface, as is common with most of the ‘smart TVs’ I’ve looked at recently. It’s only slated to appear on new equipment; most of the kit that has the existing Inview software isn’t equipped for IPTV anyway.

Inview's TV guide

Aiming at second tier manufacturers, Inview's platform will try to integrate IP services with the EPG

Using Netflix with a Yamaha iPod dock

I’ve been playing with Netflix, like a lot of people, seeing what the new service is like. In short – you’ll be able to read a longer article next week on Register Hardware – I like it. For the sort of things I want to watch, the £5.99 is a pretty good deal.

There’s one niggle; my own TV set is one of the first generation Panasonic VieraCast sets, which doesn’t even support AceTrax, and isn’t going to get an update for things like that, so there’ll be no Netflix. I’ve been fortunate enough to have a Samsung 8000 series set here for testing, which is a great bit of kit, but sadly it’ll be going back soon.

Is that the end of my Netflix usage? Perhaps not; there’s an iOS app for Netflix, and I do have an iPod Touch. I also have a dock for my AV amp, which I’ve used in the past to play video from the iPod through the TV, taking advantage of the free weekly downloads from Orange.

That works pretty well, though I do have to alter the settings in the AV amp to stop video processing, if I want the picture to fill the screen. The amp is a Yamaha RX-V667, and I’m using the YDS-12 dock.

The problem

Unfortunately, it doesn’t work so well with Netflix; when you plug the iPod (or iPhone) in to the dock, you get an on-screen menu that can be used to select movies and music on the iPod. If you’re playing content, like movie rentals from iTunes, that appears in the menus, that’s fine – when you select a movie, the menus go away.

Start up Netflix and start watching, and you get this, instead:

Yamaha's dock menu overlays the video from the iPod - not much use for Netflix

I tried in vain using the obvious options to get rid of this, checking all the settings in the menus, and using that ‘x’ at the bottom right of the screen. No joy. The sound and video play fine, with the dock menu sitting on top.

The fix

Googling turned up the answer, and it’s thankfully a simply one: Just press the ‘rec’ button on the AV receiver’s remote control, and the menu disappears, until the next time you plug your iOS device in. I don’t know why they chose that, instead of the ‘close’ icon on the screen, but at least now you’ll be able to use an iPod touch or iPhone with your Yahama dock to enjoy Netflix full screen, if it’s not built in to your TV set.