Mashed Articles

The end is nigh for the modern graphics chip. It genuinely pains me to say that. After all, I’m an unapologetic chip aficionado, someone who loves the technology of integrated circuits for the sake of it. But it’s becoming increasingly apparent that GPUs are over-engineered, increasingly irrelevant and almost definitely not long for this world.

The background here involves a confluence of technological trends. The most ominous of these in terms of the GPU’s longevity as a discrete component is the architectural convergence of CPUs and GPUs. However, one of the most debilitating symptoms of the graphics chip’s terminal malaise is complexity – sheer, pointless complexity. Take Nvidia’s latest uber pixel pumper, the GeForce GTX 480. It weighs in at three billion transistors. That’s getting on for triple the size of Intel’s beefiest PC processor, the six-core Core i7-980X.

If the GTX 480 was any use, that monster transistor count would actually add to the allure. But the harsh truth is that it isn’t – for almost anything. And that makes it dumb. You see, despite the hype regarding running non-graphics applications on GPUs, there’s still very little outside of games that makes more than passing use of a desktop or laptop GPU. More to the point, the number of games demanding a really high-end GPU that are actually worth playing isn’t merely a small number. It’s zero.

Put it all together and you have a terminal mismatch between the cost and complexity of GPUs and their real-world utility. In truth, I’ve felt this way for some time. But it’s the apparent emergence of a radical alternative to established 3D rendering technologies that really brings home how bloated and ludicrous graphics chips have become.

This alleged revolution in rendering comes from a small Australian software startup known as Unlimited Detail. It’s not actually brand spanking new, having been in development for a year or three. But thanks to the random nature of web-based content aggregators, Unlimited Detail was lifted from obscurity recently in a flurry of YouTube-powered publicity.

Anyway, as far as I could tell the basics of this new rendering technology involve ditching polygons in favour of atomic points in 3D space. The claimed result is quite literally unlimited geometric detail. Oh, and the whole thing runs in software at smooth framerates on a conventional PC processor. The GPU doesn’t get a look in until it’s time to spit out the final 2D images.

You hardly need me to point out it all seems too good to be true. So, there was nothing for it other than to go straight to the source and speak to the guys at Unlimited Detail. The technical brains are provided by Bruce Dell, a former supermarket manager, while the business nous comes courtesy of Greg Douglas, a games insider formerly of developers Auran.

The idea of using atoms or points is not new, of course. The really clever bit in UD is the 3D search algorithm developed by Dell. The precise details are UD’s big secret. But according to Dell, “The algorithm takes point cloud data and files it in a certain way so that it can be quickly sorted and accessed.”

When the algorithm searches for points, it doesn’t do so indiscriminately. Instead, it only pulls up a single point for each on-screen pixel being rendered. “We only grab the atoms we need for each pixel, we don’t touch the others,” explains Dell. In other words, the workload depends on screen resolution, not the underlying geometric detail of the scene being rendered. Thus, an impression of unlimited geometry is created.

The UD guys claim the algorithm is so efficient it runs in real-time in a single thread on just one core of a conventional PC processor. Apparently, it will even scale down to simple CPUs in mobile devices.

So far, the only hard evidence for these incredible claims takes the form of a few pre-recorded videos of dubious quality. However, having spoken to the UD pair, I’m happy to confirm they’re not only incredibly passionate, but strike me as completely genuine. It’s potentially extremely exciting stuff.

Still, even if UD works exactly as advertised, the established players in graphics are hardly going to embrace a technology that instantly renders several decades and billions of dollars of investment obsolete overnight. You have to assume Nvidia, and to a lesser extent AMD, will resist the idea strongly. But if Unlimited Detail’s technology gains any traction at all, GPUs really will look sillier than ever.

Linux doesn’t have a CEO. Consequently, there’s no annual keynote hosted by a charismatic alpha male. But if it did, and if there were a conference covering the first half of this year, the first speech would start with three words: ‘Linux is winning’.

Firstly, a market research firm in the US called The NPD Group revealed that sales of Google’s Android platform overtook those of Apple’s iPhone in the first quarter of 2010, propelling itself into second place behind the waning RIM. Android is becoming increasingly competitive, spanning both the smartphone and the emerging tablet markets, with devices from Dell and HP on the near horizon. This might be why Apple has started a patent infringement lawsuit against HTC, using many of its Android-based phones as physical exhibits in its litigation.

Secondly, Google announced its intention to open source the VP8 video codec. This was acquired when it bought On2 earlier in the year and it will be used alongside Vorbis and the MKV container to create Google’s WebM video format. This is vitally important for Linux. The nascent H.264 format, as used by Apple and many HTML5 video streams, is encumbered by patents, and current open-source implementations live under the shadow of legislation. VP8 and WebM have the potential to match it for quality, and while WebM will undoubtedly attract similar litigious trouble, having an umbrella the size of Google should satisfy many Linux distributions, especially when Mozilla, Opera and Adobe have already pledged their support.

Finally, the UK’s new coalition government has published its Programme for Government. There are two points in the section on Transparency that are great news for free software. One states, “We will create a level playing field for open-source software,” while the other adds, “We will ensure that all data published by public bodies is published in an open and standardised format, so that it can be used easily and with minimal cost by third parties.” If these promises come true, it will transform attitudes to open-source software and Linux, and hopefully open the door for its use within government and schools, two areas where it’s ideal.

Many of us used to think that for Linux to be judged a success, it had to be installed and running on more desktop computers than Microsoft Windows. And there are great swathes of Linux users who still feel the same way. But the world of computing has changed. There’s more than one way of judging the success of something that started as just a good idea.

Windows, Linux and OS X are survivors. They’ve lasted this long because they exist within their own ecosystems. Linux, for example, is fed by a curious mixture of enterprise investment, embedded hardware vendors and a community brimming full of zealous commitment. There’s a low-cost threshold to entry and a subsystem that maintains itself with very little investment. It’s these factors that have shaped how it looks, how it feels and how it’s operated.

The ecosystems inhabited by both Microsoft and Apple are equally well-adapted to their environments. The former is the domain of the utilitarians, offering straight functionality for an up-front price. The latter is an increasingly important fusion of fashion and function. But things have changed. The borders between the ecosystems have become indistinct. Apple has surpassed Microsoft in market value, winning thousands of new fans through it’s no-fuss interfaces and lower prices. There’s a shift in the balance of power.

And thanks to Google, Linux is becoming less free and less open, proving that in the new markets where it’s having the most commercial success, it’s becoming more like Apple. ROMs are encrypted and need to be rooted for user-hacking, third-party applications have to be sold through a single vendor and personal information is held in the cloud by a sole provider. If Linux wants a taste of similar success, it might find it if it makes similar concessions to a user’s freedom.

But then we’d have failed. The Linux ecosystem would have become too polluted, bogged down by sponsored kernel additions, paid-for support and short life cycles. It may be a commercial success, but no longer an active one. Our hypothetical CEO might make further compromises, and make judgements against the interest of Linux users. Which is exactly why we don’t have a CEO, and exactly why the success of open-source software is so difficult to judge using the same language as its competitors.

Any operating system that contains the letters B, S and D usually conjures images of geeky elitism, arcane interfaces and the undead world of UNIX. Despite its similarity, this is an image Linux has largely been able to shake off, thanks to its friendly graphical installers and configuration tools. But BSDs can offer a unique insight into what has made Linux popular, as well as an opportunity to hone your command-line and trouble shooting skills in a world that might be getting too easy. And while you do need a little technical confidence to get any BSD system up and running, it’s not half as difficult as it first appears.

FreeBSD is not as demonic as its logo might suggest. Honest.

FreeBSD is a the most popular implementation of version 4.4 of the Berkeley Software Distribution. This was the original BSD, a version of UNIX that was developed between the late-70s and the mid-90s and used a famously liberal licence. This licence has meant that anyone can use, copy and redistribute and re-implement its code and APIs. Which is exactly what FreeBSD attempts to do, alongside other projects like OpenBSD and NetBSD. In turn, there are many projects like Apple’s OS X that build upon the foundations in FreeBSD, all thanks to the liberal licences of the original.

Step 1: Prologue

There are several important differences between FreeBSD and Linux, but the most fundamental is the kernel. The term ‘Linux’ is most often used to refer to the entire operating system, from the boot code and drivers to the desktop and the applications. We’d call Ubuntu, Fedora and OpenSUSE different versions of Linux, for instance. But this definition isn’t accurate. ‘Linux’ should only really refer to the kernel – the chunk of code at the heart of the system that deals with hardware, networking, drivers, storage, CPU and process management, and the BSD kernel is entirely different.

The Linux kernel that was originally developed by Linus Torvalds, and it’s still the only part of the whole operating system he has control over. The remainder of what makes a complete operating system, the windowing environment, the desktops and the applications, are pulled from open source project that are mostly using one of the the GNU Public Licences. Hence, the official name for the entire Linux operating system is really GNU/Linux to show that there are two parts of the whole project. Which is why replacing the kernel isn’t a trivial operation.

Don’t be scared of text mode. It’s part of Linux’s legacy, and serves a very useful purpose.

Any new kernel needs to be broadly compatible with Linux so that the remainder of the software stack can be ported without too much difficulty. Fortunately, both FreeBSD and Linux are UNIX-alike, which means there are many similarities, and the result is that with a standard installation, you’ll find many of the tools you’re already familiar with, albeit in a different configuration.

Many system administrators feel FreeBSD is has been a more stable choice for servers over the years, and that it can out-perform its cousin on certain tasks. It’s also a great choice if you want to run a server on limited hardware, as the requirements for a BSD-based system are often less than for the Linux equivalent. FreeBSD, for example, lists its minimum requirements as a 486 CPU with 24MB RAM, which is quite staggering in today’s world of terabytes and quad core processors. FreeBSD is also a i386-based platform. There are ports to other processors, but the project’s focus has always been compatibility with Intel’s standard architecture, and as a result, could be better suited to the majority of machines that parts of the Linux kernel.

Step 2: Installation

While there are Live CD versions of FreeBSD, the traditional install disc is still the most common medium for getting hold of the latest version. But you’ll need to steel yourself against its antiquated installation mechanism.

When you boot your machine with the disc in the drive, the first thing you’ll see is the black and white ASCII art of the install menu. You won’t see any other graphical embellishment until you’re able to boot successfully into a working desktop. For most installs, you’ll need to choose option 1 from the menu, but if you’re using an older machine, you may want to try 2 (with ACPI disabled) to avoid any potential problems.

Unlike trial Linux installation packages like Wubi, installing BSD has a tendency to be destructive. Make sure you’re not going to obliterate anything important before continuing.

You’ll then have to wait a few moments while various kernel messages scroll by before you’ll see the text-based installation and configuration screen. If you’ve used Debian, this kind of text installer will feel familiar. There’s no linearity to the install process. You can move backwards and forwards through the various options, and continue to make adjustments to the installation until you quit the installer and restart the system.

For a basic, working environment, you will need to do at least the following. Select the second option in the menu to initiate a ‘Standard Install’ and read the information on the screen that follows. The next page displays ‘fdisk’, the disk partitioning and formatting tool. Press F1 for documentation, but if you’re using your entire hard drive for this installation, press ‘a’ to select everything, followed by ‘q’ to apply the changes. This will delete everything currently on the drive, so be careful.

The next page will ask if you want to install a boot loader, which is the menu that lets you choose between whatever operating systems you have installed. Choose the second option (BootMGR), and on the following page, you need to create the various partitions used by FreeBSD using the same fdisk-like interface we’ve just seen.

Once again, if you’re using an entire drive for the installation, you can just press ‘a’ to let the installer create the most appropriate array of partitions followed by ‘q’ to make the changes permanent.

The next page will ask you to choose a distribution. Unlike a Linux distribution, FreeBSD uses the term to refer to the default selection of packages that are to be installed. Select ‘Custom’ and add ‘base’ and ‘kernels > GENERIC’ to your installation.

This will give you enough package to get a working system, and we’ll need to add the desktop environment at a later stage. Return to the ‘Exit’ option at the top of the list and press space to jump into the package installation routine.

Step 3: Post-Install

After all the preliminary configuration has completed, you’ll be asked whether you want to configure any Ethernet or SLIP/PPP devices. Select ‘Yes’ if you are connecting to the internet through your machine’s ethernet port, and you should see your adaptor listed in the top of thew connections list. Choose the adaptor, say no to IPV6, say yes to DHCP and skip through the configuration page to the OK button. Say ‘no’ to your machine being a network gateway,’no’ to enabling any inetd services or running SSH, FTP and NFS server and clients, and don’t edit the console settings. You can safely setup a timezone for your machine and enable the PS2 mouse emulation if you’re using one.

Don’t worry too much about your initial selection of packages. You can easily add more later.

Say ‘Yes’ to the next question, and you’ll now be looking at the package manager. This is where you choose what applications you want to be installed on top of the default option we chose earlier, and there are thousands of packages to choose between. For a simple setup, jump into the ‘x11’ menu and select the ‘kde4-4.3.1′ package. It’s exact name will depend on the version of FreeBSD you’re playing with. If you’re not a fan of Gnome, you could also choose ‘gnome2-2’ from the same list of packages. Selecting either will also mark their dependencies for installation. You also need to select xorg-7, and any other packages you know you’re going to need.

When you’re ready to go, jump back to the top package list, select ‘Install’ and press space. You’ll need to wait a while for all the packages to install. The next step is to create a user account. You can do this by saying ‘Yes’ to the option, then selecting ‘Add User’, and entering a user name in the page that follows. Select OK to make the change permanent and exit from the users and groups menu. You’ll then be asked for the system manager’s password, and you’ll need to type this twice.

After that, you can say ‘No’ to the post-install configuration request and wait for your machine to reboot. You’re now at the point where you should have a basic, working installation, and you can quit from the installation menu and restart your system.

Step 4: Configuration

When your machine re-appears, you’ll be greeted by the sombre monochrome of the command line. Login as ‘root’ with your system manager’s password. For both Gnome and KDE, you need to add the following two lines to the ‘/etc/rc.conf’ configuration file:

dbus_enable="YES"
hald_enable="YES"

Unfortunately, you’re going to need to use the ‘vi’ text editor. Type ‘vi /etc/rc.conf’ to load the file. Press ‘i’ to enter insert mode, move to a new line and type the following. Press escape to exit insert mode, followed by ‘:wq’ (without quotes) to save the changes and quit the editor. Next type ‘reboot’ to restart your system.

FreeBSD doesn’t come with a desktop activated by default, but it’s simple enough to change.

When you get back to the login screen, enter your user account details this time, and when you get dropped back to the command line, type ‘vi .xinitrc’ and add the following line to the file:

exec /usr/local/kde4/bin/startkde4

This is telling your system that when the X.org graphical system starts, you want KDE to be used as your desktop environment. Save and exit vi.

Step 5: Launch Desktop

Usually, at this point, you need to create an ‘xorg.conf’ file to define the display properties for your machine. But recent releases of the X server are able to create a working configuration without any further editing. Which means typing ‘startx’ is all you need to do to launch a graphical environment running KDE. If this doesn’t work, then you will need to create create a working /etc/X11/xorg.conf file.

But with FreeBSD 8, it’s more likely that you are now looking at KDE running through its Akanadi porting routines as it builds up a configuration for your desktop. After a couple of minutes, this will leave you with a KDE desktop running on-top of FreeBSD, and you’ve just earned another trophy for your awards cabinet.

At long last: a GUI! And one of the more stable interfaces you’ll find. Here’s hoping BSD serves you well.

This is exactly the same KDE you’ll find on Linux, and you’ll be hard pressed to find any difference between the way it works on FreeBSD and the way it works with Kubuntu. It’s only when it comes to system configuration that you’ll notice because FreeBSD doesn’t have any graphical configuration tools, which means if you need to change anything, you’ve got to be prepared to go back to the command line. But that’s another story.

See also: PC-BSD 8.0

If you’ve followed the main text to install a shiny new version of FreeBSD, you might have noticed that the install mechanism really wasn’t all that shiny or new. It reality, it feels ancient. But this doesn’t mean that the operating system has been languishing unloved and undeveloped, it just means that making the installer easier to use is low on the priority list.

Fortunately, this being open source, demand for a better way of doing things has led to several alternatives, the best of which is PC-BSD, which you’ll find at www.pcbsd.org. It does several impressive things. Firstly, it replaces the tepid monochrome installer of FreeBSD with a graphical application much-more in-line with its Linux counterparts. It will also automatically install and configure a recent version of the KDE desktop, which should mean you can get productive with a FreeBSD system as quickly as possible, without touching the command line, and there’s a wonderful wiki full of helpful documentation.

This means you can install PC-BSD by placing the disc in the drive, rebooting your machine, answer the questions that appear and wait for the operating system to install. You won’t even need to worry about manually partitioning your drive unless you want to create a custom configuration.

Another important difference is that it PC-BSD doesn’t use the same package management as FreeBSD, although you can still get to it if you need to. Instead, package are available as single files with the ‘.pbi’ file extension, which can then me installed with a simple click. It’s more like how packages are handled on OS X, and is far better than the weird world of dependencies you find on Linux.