http://blogs.pcworld.co.nz/pcworld/tech-guy/ en Copyright 2013 Thu, 31 Jan 2013 13:21:48 +1200 http://www.sixapart.com/movabletype/ http://blogs.law.harvard.edu/tech/rss The idea that desktops might change forever is enough to send geeks into a spin. Something of a shockwave rippled through the hardware geek world as things began winding down for the Christmas and New Year’s break. Rumours flew around claiming that Intel’s upcoming ‘Broadwell’ processor would only come in ball grid array (BGA) packaging. In BGA packaging, tiny balls of solder are used to surface mount processors, rather than the more usual desktop land grid array (LGA) arrangement of having processors with pins that fit into motherboard sockets. Processors with BGAs have plenty of electrical, physical and manufacturing advantages over other forms of chip packaging but once they’re soldered onto a motherboard, that’s it. They don’t come off without a huge amount of effort involving heat guns and hacked toaster ovens. Desktop processors without sockets is heresy of the worst kind to geeks who have become used to the open architecture of personal computing that lets anyone (well, almost) swap many of the crucial electronics out for repair, upgrade, better performance - or just for the fun of it. It’s also not really what businesses built around open PC architecture want to hear either, as the BGA move would close down most smaller operators. As Broadwell is due out in 2014, I asked Intel if it was true that it’d be BGA-only and whether next-year’s ‘Haswell’ range would be the last in land grid array or LGA format. I got what amounted to a non-response, but in the meantime panic had apparently set in with Intel’s large customers: they had to be assuaged with official messages. A spokesperson for ‘Big I’ told American media that the company will continue to offer socketed parts in LGA packages, ”for the foreseeable future„, whatever that means. Leaked Intel processor roadmaps show that in 2014, Broadwell will go into mobile devices and use BGA, but there appears to be a Haswell part in LGA format over the same time period. Fellow chip maker AMD got in on the act to say it will offer socketed chips for 2013 and 2014. Fingers crossed, AMD will still be around to honour its promise. Reassurances aside, the industry would have to be blind not to see this train coming. Controllers, chipsets and processors are already being perma-soldered to motherboards, and the same is starting to happen with memory. What’s more, increased integration and shrinking components mean separate processors will be replaced by ‘systems on a chip’ at the consumer level at least. Already, Intel is seeking to move graphics and I/O controllers onto the main processor die and that movement of key parts onto the die is likely to continue. That evolution is driven by economics and a desire to control the market. Apple and Android vendors have shown the way with devices that are cheap, reasonably powerful and have the hardware closed off to user customisation or upgrades, thanks to the high level of integration. However, I’m not convinced that offering fewer choices will help slumping PC sales and I think vendors need to start thinking what it was about that particular architecture that had people buy computers in droves. Hint: extensibility and flexibility were some of the reasons. And while I’m talking about hardware that shuts you out, I’ve had a few issues with the new Extendible Firmware Interface (EFI), the new standard to start up and control your computer that replaces ye olde BIOS. EFI has some really useful features, improved security and no more text-mode VGA configuration screens, but it can also be incredibly inflexible, especially with Windows. I discovered that with EFI-based computers, you can’t just shift a bootable hard drive from one SATA port on the motherboard to another as you could in the past. The computer sees the hard disk, but it won’t start up. It turns out that EFI ties the Globally Unique Identifier (GUID) in the partition table with a particular SATA port, and Windows Boot Manager isn’t smart enough to work out it should look elsewhere on the system if the drive isn’t where expected. This behaviour causes further headaches when you clone disks - say when you move from a small SSD to a larger one - and dual-booting becomes dependent on the main bootable disk. Luckily, it is possible to sort it out and there’s no need to start digging in with the spudgers, but it’s a manual process that seems like a heap of work just to move a drive from one SATA port to another. Maybe there is a point to the pain, but I fail to see it myself - beyond, of course, making it more difficult for people to control the hardware they’ve paid for. Time to start hoarding old gear… Juha Saarinen]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2013/01/small_balls_of_solder.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2013/01/small_balls_of_solder.html Thu, 31 Jan 2013 13:21:48 +1200 Juha Saarinen, juha@pcworld.co.nz ]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/08/the_c_word.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/08/the_c_word.html Fri, 31 Aug 2012 11:13:07 +1200 Juha Saarinen, juha@pcworld.co.nz ]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/08/what_tech_guy_did_on_his_holid.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/08/what_tech_guy_did_on_his_holid.html Mon, 27 Aug 2012 12:30:41 +1200 This column was originally published in the April issue of PC World. I've been to more demos of Long Term Evolution - the nonsensical name for the next generation of cellular broadband - than I can recall. The one I recall best was with Huawei in Shanghai: I got to go on the Maglev train to the airport and punt packets at 30Mbit/s while we zipped along at 430kph. That was in 2010 and as far as I can tell, Chinese telcos have yet to launch LTE. The technology is cropping up elsewhere though, like the US and Canada. LTE is also tantalisingly close to New Zealand since last September’s launch of Telstra’s Australian 4G service. I got a chance to test drive it in February this year. Telstra’s 4G isn’t a trial network, in other words. I wondered whether 4G would be able to deliver the high speeds I’d seen in my trials, once the real, live network was being used by an estimated 100,000 customers. The first few tests in busy areas in Brisbane CBD showed 20Mbit/s downloads and just over 10Mbit/s uploads; nice, but I expected more. I found a good spot at Brisbane Airport and ran some tests using speedtest.net. While the downstream speed (39-44Mbit/s) is impressive, it’s nice to see that Telstra has paid attention to the upstream which at 22Mbit/s is faster than my artificially-limited VDSL2 connection. I was watching Task Manager’s graph of the network throughput, and it was smooth without the peaks and troughs that indicate packet loss and ensuing TCP slowing. What’s more, the latency of the LTE connection is nice and low too. In fact, the speeds I saw exceeded Telstra’s claimed ”typical speeds„ of 2-40Mbit/s down and 1-10Mbit/s up. They’re a long way off the theoretical maximum for this variety of LTE (100/50 Mbit/s) but even the ‘real life’ figures exceed what you get from most fixed broadband connections nevertheless. Currently, there’s only a single device for the 4G service, the Sierra Wireless U320 USB modem. It costs A$299 on its own, or is bundled for free with Telstra’s two-year contract plans. The U320 is a chunky number, 50mm wide, 64mm long and 11mm thick. It’s equipped with two external aerial connectors, as well as a GPS and a MicroSD slot nobody will use. The U320 is as big as it is because it supports not just 1.8GHz LTE, which Telstra uses for its service, but also 2.6GHz LTE and 850MHz/2.1GHz dual-channel and single-channel HSPA+ and HSPA. There’s a fair bit of electronics inside the modem, as well as aerials for the different frequency bands. I was very happy with how tenaciously the U320 locked onto data signals. Roughly five kilometres away from CBDs and airports, the LTE signal disappears and you’re on Telstra’s NextG 3G service; where I was, the U320 ran in dual-channel mode, and gave me 15-20Mbit/s downloads, and 1.5-2.5Mbit/s uploads with 45-60ms latency. Again, very good considering this is a cellular service. The LTE network itself was put in place by Ericsson for Telstra and the service uses 10 or 15MHz blocks configured as frequency-division duplexing (FDD), with separate bands for upstream and downstream data. LTE can also be configured as time-division duplexing. This sends and receives data in the same frequency band, which uses less spectrum. However, if you want the best performance and lowest latency, FDD is the way to go. Telstra’s pricing is pretty good if you can tolerate the two-year contracts, starting at A$20 a month for 1GB of data going to $80 for 15GB. The data caps are about the only fault I can find with the service: based on my experience on speedy 4G, Telstra should multiply them all by ten. One gigabyte of data is ridiculously low and won’t last a day. It would be great to get LTE, Telstra-style, in New Zealand too. However, our competition-loathing telcos aren’t likely to do anything until 700MHz spectrum is allocated to them with maybe some further subsidies on top. LTE should also have been mandated for the Rural Broadband Initiative to ensure a service with a modicum of future-proofing. Sadly, despite clear evidence to the contrary, the government pretends LTE isn’t ready yet. For that reason, Vodafone built the RBI offering as a bung 5Mbit/s ”peak speed„ service with low 5GB data caps, using 3G gear. To think what could’ve been, if the RBI had been designed with vision and foresight… Juha Saarinen, juha@pcworld.co.nz ]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/08/generation_next.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/08/generation_next.html Mon, 27 Aug 2012 12:09:37 +1200 Why thirteen is lucky for broadband speed tweaking Net nostalgia: One of my early columns for PC World explained how you could speed up your internet connection by tweaking some low-level network settings within the operating system. This was thirteen years ago (gulp), just as DSL arrived in New Zealand. Few people could afford "D$L" back then, thanks to Telecom’s outrageously high pricing and internet connectivity was primarily via 56k modem. There was quite a bit of mileage to be had from tweaking, providing you were willing to dig through obscure Control Panel and Registry settings. Kiwis were keen tweakers mainly because much of the internet is such a long way from New Zealand. Internet data packets take a while to reach their destination here and this inevitably affects performance. DSL made things easier, and we no longer worry about modem UART buffers and similar tech-arcana. In fact, today’s broadband is dead simple to operate in comparison. The big difference between now and then is that line speeds have increased for most people. I was rapt at seeing speeds of 1.8Mbit/sec in November 1999 from local servers even though the first-generation Nokia M10 ADSL modem actually connected at 5.3Mbit/sec. Fast forward to 2012, and ADSL2+ means 15 to 20Mbit/sec downloads for many people (but only 1Mbit/sec up). My VDSL2 connection is set to 70Mbit/sec down and 10Mbit/sec up. Thank goodness, New Zealand ISPs have built their networks and improved quality and performance hugely to match the higher speeds - I don’t think anyone uses high-latency satellite links for DSL traffic backhaul as they did back then. We still have that distance problem to deal with in New Zealand however. Ping tells me that the US West Coast is 140ms away and data packet round trip times to Europe are over 300ms. The internet protocol used for the vast majority of data, Transmission Control Protocol (TCP), requires confirmation messages before itsends packets to be received. While it waits for those messages, TCP simply stops. This isn’t much of a problem on a low-latency local network, but once round trip times hit 50-150ms, TCP just can’t keep the data pipe full without compensating for the big latency somehow. For optimum performance to deal with packet latency of 200ms, I would have to set a receive window buffer for my connection well over a megabyte - much more than TCP’s original 65,536-byte maximum. Making sure TCP worked well on high-latency networks used to be a manual process that involved poking around in the Registry on Windows or setting sysctls in *BSD. Such ju-ju is, I’m pleased to report, no longer necessary. Windows 7 has an auto-tuning TCP/IP stack that identifies high-latency connections and enables the right settings, ditto Apple OS X. Testing is easier too, thanks to local and overseas Measurementlab.net nodes. These are a joint effort by Google, Skype, Amazon, Bittorrent and others and provide some very handy open source tools with which to diagnose your internet connection. The good news: with a new TCP stack, you’ll get good performance despite high latency. I tested with DiffProbe and other tools against servers in Sweden (305ms) and elsewhere overseas and hit 56 to 60Mbit/sec down, and 9.5Mbit/sec up. Pipe full, in other words. Modern TCP seems to work and won’t - as I’ve heard a surprising number of people state - limit your throughput from the US to around 9Mbit/s. The fact that you don’t get full speed from your broadband connection is more likely due to other choke points than TCP. ISPs rate-limit certain traffic, webservers often can’t dish stuff up fast enough and older routers can’t handle new TCP features and ignore them, putting everything into go-slow mode. This means that we should fix broken stuff on our internet so we’re not disappointed when the UltraFast Broadband connections become available. This could mean anything from a router upgrade to a new version of your operating system. Don’t expect superfast broadband through Windows XP and early-generation Wi-Fi for instance. To some extent, you can work around the above issues with caches and content delivery networks (CDNs) that shorten paths and reduce latency. That way your box won’t be talking to the actual server overseas, but instead fetches stuff locally. However, these break the end-to-end principle of the internet and create their own set of headaches. For example, Vodafone and TelstraClear’s transparent caches prevent their customers from using anti-geoblocking services such as unblock-us.com. Telco and ISP gear vendors will take out a contract on me for saying this, but to blithely spend hundreds of millions of dollars on caches and content delivery networks around the country to deal with what’s essentially a maintenance issue doesn’t make as much sense as getting to the root of the performance problem. UFB providers may wish to think about this. Juha Saarinen]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/03/pumping_ultrafast_packets.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/03/pumping_ultrafast_packets.html Mon, 05 Mar 2012 13:04:48 +1200 Juha Saarinen]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/02/what_does_your_mobile_say_abou.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/02/what_does_your_mobile_say_abou.html Thu, 09 Feb 2012 11:03:31 +1200 Juha Saarinen]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/02/technology_isnt_for_techies.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2012/02/technology_isnt_for_techies.html Wed, 08 Feb 2012 14:30:33 +1200 Juha Saarinen]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/11/the_mixed_legacy_of_steve_jobs.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/11/the_mixed_legacy_of_steve_jobs.html Mon, 28 Nov 2011 11:52:48 +1200 - Juha Saarinen]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/11/the_painful_road_to_ipv6.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/11/the_painful_road_to_ipv6.html Thu, 03 Nov 2011 10:58:17 +1200 Much as I find Apple people annoying, I like Cupertino hardware a great deal. With the move to Intel hardware the UNIX-like Mac OS X finally has room to stretch its legs too. If Macs are essentially PCs, which in turn is a standardised platform that runs Windows, Linux and BSDs happily, could you not run Mac OS X on anything? If you really want to know, head down the ”Hackintosh„ route - running Mac OS X on hardware that isn't Apple-made. Make sure before you begin that you're armed with a good amount of technical nous and even more patience. A word of warning right from the start though: getting a Hackintosh up and running can be seriously Deep Geek. I’ve tried Hackintoshes before and while I admire the community effort going into working around Apple’s quirks and foibles, I came away realising they’re not for average users. For starters, OS X uses the Extensible Firmware Interface (EFI) rather than the BIOS used by most PCs. This means you have to use a boot loader such as Chameleon to emulate EFI, and also make sure that your BIOS settings are correct with Advanced Host Controller Interface (AHCI) for storage and the 64-bit High Precision Event Timer (HPET) enabled. If they’re not, you get a kernel panic when you boot up OS X. What’s more, under certain conditions OS X resets your BIOS by clearing the CMOS. That’s not the end of the FLAs though. OS X uses ACPI - the Advanced Control and Power Interface - like PCs to set up devices. The important part of ACPI is the Differentiated System Description Table, or DSDT, that tells OS X what the base system is. The DSDT is stored in the BIOS of the computer, and you can grab it, disassemble and recompile it with freely available tools to make sure your hardware is recognised by OS X. Some hardware can be animated by using drivers, which are called Kernel Extensions or Kexts. Many Hackintosh installations load the operating system, and then you head over to aites such as kexts.com or tonymacx86.com, or even insanelymac.com and snag some drivers. I did say you need to be intimately acquainted with your hardware, didn’t I? Not just that, but you also need a large amount of reading and Googling detailed information and potential problems before you install anything. Install the wrong Kext or DSDT, and boom, kernel panic strikes. In previous Hackintosh setups, this used to happen each time you fired up Software Update, unless you took precautions and manually upgraded stuff. However, this time around there is an easy way. The excellent Kakewalk [kakewalk.se] set of scripts will do all the heavy lifting for you, provided you make sure to use a compatible motherboard. Gigabyte boards seem to be popular with Hackintoshers, and I happened to have a spare EX58-UD4 model handy, with a 3.2GHz Core i7. Cutting a huge saga short, Kakewalk 4.0.1 extracted the necessary disk image from the copy of OS X 10.7 Lion that I plonked down $39 for, added a boot loader, hardware drivers and installed everything to a bootable, 8GB USB stick. Installing Lion on the Hackintosh machine took maybe an hour all in all and afterwards everything worked aside from the audio and the ability to put the computer to sleep. The first problem was easily solved with a VoodooHDA driver, but the second was trickier as Apple’s CPU power management driver doesn’t like PC ACPI implementations and has to be disabled. Getting this to work requires DSDT work. It’s not impossible however, and after a bit of trial and error plus the obligatory kernel panics, I had a Mac Pro-like Hackintosh. Fast and stable, I can almost pretend I’ve got a Mac Pro if I don’t look too closely at the fairly ugly tower case and the PC keyboard that’s awkward to use with OS X. Because Apple’s OS X code is left alone, you can now run Software Update without too many worries. Provided you’re willing to put in the huge effort, are Hackintoshes legal? I can’t tell for sure: the underlying operating system is Darwin which is free and open software, and the same goes for the XNU kernel, which incorporates Mach, FreeBSD and NetBSD code. But, despite being FOSS-based, Apple’s end-user licence says you can’t install, use or run OS X on non-Apple-labelled hardware. Any commercial use is an absolute no-no: Apple has cracked down on commercial Hackintosh operations like PsyStar in the past, and will do so again. That said, word amongst the OS X x86 crowd is that Apple doesn’t care that much about amateur Hackintosh-ery. I imagine this is because it’s hard to imagine anyone bar myself and a small band of techies with a masochistic bent actually putting together a Hackintosh and getting it up and running. Everybody else will happily continue to sell both their kidneys for that lovely, tested and stable hardware that just works, and I totally understand why. http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/09/kext_massaging.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/09/kext_massaging.html Mon, 12 Sep 2011 13:11:37 +1200 Recently, I’ve had to deal with files of sizes in the region of 50-100BGB, whereas normally a big file would be 2-5GB or so. The problems that you’d expect with this kind of file size is in opening them, compressing and uncompressing them and moving them. The Xeon W3550 CPU and 12GB of DDR-3 RAM in my HP Z400 workstation crunch through those gigabytes of data without breaking a sweat. Ditto the new SATA-3 OCZ Vertex 3 solid state drive with its 550Mbyte/s reads and writes. Compressing and decompressing even massive files is fast thanks to the quick I/O and multi-core processor ripping through the files in no time at all. Moving or backing up the data to other systems or external drives over the network or an external bus is another story. 'Painfully slow' describes the situation the best. What’s more, there doesn’t seem to be anything affordable on the horizon that’ll make things go faster, not for a while at least. For example, the 802.11n Wi-Fi connection in the computer reckons it’s talking to the router at 270Mbit/s which is amazingly good. Unfortunately, that translates into real-life performance of 95-110Mbit/s which is horribly inefficient. Move away from the Wi-Fi access point or have other devices join in and share the bandwidth and the speed halves. The spotty performance of Wi-Fi today doesn’t fill me with confidence that new wireless technologies such as 60GHz WiGig will deliver anything near the promised 6-7Gbit/s. Besides, with a reach of a mere ten-fifteen metres and the high signal frequency providing poor penetration of walls and objects, it’ll be curious to see how useful the likes of WiGig are in reality. Going cabled via a USB 2.0 external drive isn’t much better. The over-optimistic 480Mbps including overhead that the technology promises shrinks down to 25-35Mbyte/s and that’s a best-case scenario as well. The new backwards-compatible USB 3.0 standard promises a dizzying 5Gbps peak speed and you can expect something like 200Mbyte/s throughput with fast devices and ideal conditions. This is good stuff, as it’s almost twice as fast as 1Gbit/s Ethernet. However, it’s not as quick as even a 3Gbit/s SATA port, let alone the new 6Gbit/s ones, and USB 3.0 drives are still fairly rare and small (I don’t have one). I suspect some of the above were the reasons behind Apple’s decision to use Intel’s new Thunderbolt interface, which provides 10Gbit/s speeds. Not just in one direction, but both up and down, for a total of 20Gbit/s. Plus, you can run displays off a Thunderbolt cable. That requires active cabling with transceivers at each however, and so far the only major manufacturer using Thunderbolt is… Apple. Nobody else seems interested, which is a shame as Thunderbolt speeds are what we should be expecting right now. While I am totally impressed that Intel’s getting 20Gbps out of copper over three metres, was more interested in Thunderbolt’s first incarnation, Light Peak. At the Intel Developers Forum demo some two years ago, Light Peak was shown off pushing data and high definition video streams over thirty metres of optical connection. In optical form, Light Peak is supposed to scale up to 100Gbps. Couple that with the possibility to run long cable lengths and you have, in my mind at least, a winner. Maybe an expensive one but dammit, would you say no to 100Gbit/s transfer rates? Ideally though, I’d like to keep everything networked and not futz around with external devices. My LAN’s been gigabit-ised since 2004 and everyone’s network should be at least that fast in 2011. When I first installed gigabit Ethernet NICs, motherboard chip sets simply weren’t fast enough to drive them, ditto hard drives. It’s not until lately with gigabit Ethernet interfaces being connected to PCI-Express buses and fast drives that I’m able to hit 110-120Mbyte/s. Now, if I could upgrade to 10Gbit Ethernet, and scale performance accordingly, I’d be happy and have some margin for the future. In fact, I expected to be at 10GE by now, but that technology still isn’t at affordable levels even though the specification was fixed in 2006. Installing 10GE would give the same reach as 1GE, or around 100 metres over copper unshielded twisted pair albeit with new, Category 6A cabling that looks like a pain to install. Cat 6A cable also happens to be really expensive, costing up to three times as much as Cat 5E which I normally use. Switches and NICs are also very pricey, and Intel is doggedly sticking with 1GE on their roadmaps, possibly because it’d be difficult to furnish the full 1.25Gbyte/s that 10GE can handle at consumer pricing levels. What this means is that 10GE isn’t likely to come out of the data centre any time soon, leaving our LANs as the bottleneck. Given that we store more and more data on our terabyte and bigger drives and want to share that, this is something vendors should remedy sooner rather than later. http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/09/bottlenecks_interface.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/09/bottlenecks_interface.html Mon, 12 Sep 2011 13:09:10 +1200 As far as some people are concerned, they’ve got something that has worked for the last ten years, so why fix what isn’t broken - at least in their minds. Some years ago, I agreed to modernise the IT solution at accountancy practice. The need to modernise hadn’t come about due to curiosity about new tech and a desire to evaluate if it’d make life easier and more productive. The brief that omitted many of the points a system integrator needs to know for a successful migration. Before long, the modernisation turned into a nightmare mission - the gap between the old system and the new was huge. I won’t go into the ‘fun’ I had trying to migrate data from a locked file format with its root in the late eighties, but I still remember some things that I wasn’t able to convince the client to upgrade. The practice used the internet to send and receive data to and from government agencies like the Companies Office. I thought that was a great start to boosting productivity, but then I learnt the process took place overnight… over a modem connection. Night time transfers were necessary because the practice had to use the modem connection for other things during the day. If the transfers failed - and they often did - it was a matter of restarting it the next night. Clients had to wait for things to be filed and processed, just as they had to hold on for email responses from the practice - the messages were downloaded and replied to once a week. At the time, DSL had come down in price and an entry-level plan was entirely affordable for the practice. I tried to explain the benefits of an always-on connection that was much faster than dial-up: increased productivity and happier clients. Another benefit, for them and me was that I could even do maintenance remotely if they used DSL. All my arguments were in vain. The practice felt dial-up was enough and soon afterwards signed up for a two-year fixed contract that cost almost as much as DSL. A telco rep had convinced the practice that DSL was too fast for its needs and threw in some reduced-cost toll calling to sweeten the deal - even though most practice’s clients were in the same city. Even when you point out obvious benefits, there’s still an odd fear of fast broadband around. One person I spoke to said moving to a fast connection would be like getting a Ferrari and who needs that? That person had a new-ish Intel laptop with a powerful Core processor and a fast, cavernous SATA hard disk. That’s a ‘Ferrari’, too. So why would this person not opt for a broadband ‘Ferrari’ instead of the Yugo copper broadband most of us have and will continue to have until the UFB comes into play six years from now. Much of the fear of the fast comes from successful marketing propaganda campaigns that tells us broadband is a limited resource where each bit has to be doled at steep charges. Marketing people don’t tell customers that technological advances have made it much cheaper to deliver much faster broadband to everyone, and in greater volumes. This so much the case that artificial constraints for speeds and data caps have been introduced to create segmentation. The newspeak term for this is ”product and service differentiation„ which means that instead of getting broadband connections that go as fast as the technology allows, the brakes are put on it to different degrees. The more you pay, the less the brake is applied. Remember, we’re talking about the same product here, in different guises. Same bits and bytes, same network pipe and gear, but on one plan, they cost more than on another. In other words, you have a perverse situation in which money and effort is spent on slowing down broadband to create ”entry level„ products. I would imagine that the cost of slowing down broadband service is recouped elsewhere, such as from already expensive premium plans. The notion that artificial constraints are necessary is my biggest disappointment with the government’s Ultra-Fast Broadband project. It’s been done to encourage private companies to invest by creating a market similar to today’s, through the artificial segmentation of broadband. Distorting the market by creating slow 30/10Mbit/s plans to be used in five years’ time when the world’s infrastructure is capable of 1Gbit/s today is denying the obvious, which is that we’re in the Age of Plenty - when it comes to broadband at least. This market-constructed vision is not just blinkered but also one that shows a lack of understanding of technology. Sadly, we’re paying for it, and we didn’t have any real say in the process either. Welcome to your UFB. http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/09/fear_of_the_plenty.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/09/fear_of_the_plenty.html Mon, 12 Sep 2011 13:03:55 +1200 Juha Saarinen]]> http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/05/the_copper_comeback.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/05/the_copper_comeback.html Mon, 30 May 2011 15:10:22 +1200 As of writing this, the devastating Christchurch quake that wreaked so much havoc and caused so many deaths and injuries is still on everyone’s mind. All my friends and acquaintances there are safe - some with houses in pieces, and not quite sure what to do next - and not hurt, thank goodness. In finding out what had happened to one good friend whose house was approximately 1.5 kilometres from the epicentre, I had what can only be described as a totally contemporary moment. As you can imagine, my friend’s communications were all severed after the quake. No phones, no email. I asked on Twitter if any of my followers were in the vicinity, and if so, could they perhaps go and check on my friend. Two people responded, but I didn’t hear back from either. Apparently, their mobile phones ran out of battery power. Google set up the People Finder site that does exactly what the name says, and I posted the details of my friend there. That unexpectedly brought me into contact with other people around the world also worried about my friend. In the end, I emailed a mutual acquaintance in the UK at the publication my friend used to work for, and heard that he was fine. The acquaintance had been in touch with a colleague of my friend’s in Australia, who in turn had informed her that my friend was fine and their house stood up to the big shake. Next, I passed on that information to people in various other parts of the world and couldn’t help thinking that again, the Internet has shown what a marvellously powerful tool it is and how much good it can do. And yes, I’ve finally heard from my friend, who now has email access again. Unlike just a few years ago, I’m not in a minority in thinking that the Internet is a crucial communications channel, which is a relief. Our telcos and ISPs active in the region pulled out all stops to restore service in Christchurch, and they’ve done marvellous work. Thanks to them, people can now avail themselves of official government sites and volunteer ones such as eq.org.nz and business.eq.org.nz to find information as well as seek and offer help. Here, I’d like to extend special thanks to Brother’s NZ operation for offering printers to eq.org.nz. The site alone is a great resource, but as Chad Catacchio, co-lead of CrisisCommons Community Working Group (he was also involved in the web-based Ushahidi crisis crowdsourcing of assistance and resource during the Haiti quake) says: ”a printed map not only doesn’t require data, it doesn’t require a battery and also, paper maps are universally understood.„ Fast communications and increased information sharing should help Christchurch recover quicker and more efficiently; to what extent I can’t tell yet, but there’s no doubt in my mind that internetworking is beneficial in disasters. However, I’m not convinced that we’re doing enough to build resilient networks in disaster-prone areas. This is a question of how much you want to spend, of course, but it does seem silly that people with wireless phones couldn’t use them when mains power disappeared. Not because the wireless handsets didn’t work, but because there were no rechargeable batteries in the base stations. Since the 48-50V DC and low current phone lines are supposed to disappear as we move to fibre-optic networking over the next decade, this is now becoming an urgent problem to solve. Do we insist on each property having power storage capacity, and for how long? Six hours? Twelve? A whole day? Apart from power issues, the asymmetric nature of NZ’s broadband is a real hindrance towards recovery. Your company may have had a nice fat pipe to the outside world in the now destroyed office building, but what if you had to temporarily relocate to somewhere served by DSL only to keep the business alive? You’d be limited to 5-800kbps upstream speeds. Try backing up your data to the cloud with those speeds and serving customers, sending email, and more. Sadly, the government’s Ultra Fast Broadband network designers seem to be thinking in DSL terms with asymmetric bandwidth that’s shared with many users so don’t expect this to improve any time soon. We really need to recognise that disasters will strike and that we need to be prepared in every way possible to deal with them. This includes our communications networks: they have to be resilient without single points of power failure, and enjoy route diversity. juha@pcworld.co.nz http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/03/on_shaky_ground.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/03/on_shaky_ground.html Thu, 31 Mar 2011 15:59:52 +1200 I quite admire Intel as a company. It’s a mega-corporation with operations everywhere - and not so long ago, a solid presence in New Zealand that would be welcome if it returned to our shores. What’s more, Intel takes risks and screws up every now and then. Who can forget the huge and slow Netburst fiasco with its massive instruction pipeline that saw rival AMD trounce Intel in the performance stakes? It caused the chip giant to revert back to microprocessor architecture - with its roots in the mid-90s - which now powers the Core CPUs. Then there was Rambus RDRAM. It looked as though it would evolve into a double-data rate SDRAM beater, but didn’t. Normally, companies don’t recover from serious missteps like that. Intel not only recovered, but fought its way back as top dog. It’s remained in pole position now for the last four years at least with AMD looking increasingly pale as a processor technology innovator. Intel’s Core processors and adjunct logic up the ante every year and the technological progress shows no sign of letting up. This year’s new big thing, Sandy Bridge, is a 32-nanometre processor micro-architecture and process shrink, with heaps more features integrated inside the processor package. It’s an example of Intel delivering on everything promised by the PhDs I’ve spoken to at various developer forums. Why am I a little underwhelmed by Sandy Bridge then? Don’t get me wrong: I was first in line to take a look at the new tech. I had the Intel-supplied Core i7 2600K and Core i5 2500K processors with P67 and H67 boards for a few weeks over the Christmas holiday (right, now you know what I did during, etc.) Even without souping up the automatic Turbo Mode overclocking, the Core i7 2600K snapped at the previous generation Nehalem top-of-the-range Core i7 980X processor in many benchmarks. The Core i5 2500K wasn’t very far behind either, bar some multi-threaded benchmarks, and for the price of around $350 street, this is a fantastic processor. What is Intel thinking behind releasing K-class CPUs aimed at overclockers with graphics integrated in the processor, though? Especially since you can only overclock either the CPU or the GPU (a little) - and depending on the chipset used, you may not have access to the integrated graphics. The IGP in Sandy Bridge knocks the socks off low-end graphics cards; as a proof of concept, yeah, it’s impressive. However, no tweaker worth his or her salt will use the IGP because it’s just not good enough, and its lack of DirectX11 support compounds this. Then there’s the annoying new LGA-1155 socket - one pin less than last year’s 1156 CPU nest - that necessitates a new motherboard. Sure, you can reuse your old DDR-3 RAM and heatsinks, but LGA-1156 isn’t exactly ancient. The news that there is a bug in the P67/H67 chipsets that could stop SATA ports from working doesn’t help. In the current rotten economy, if Intel is hoping that people on a budget will ditch LGA-1156 and LGA-1366 platforms to jump onto Sandy Bridge, well… shouldn’t there be a little more to lure people? Turns out there is: Intel is apparently prepping the Z68 Express chipset that allows overclocking of not only the CPU but also the GPU (probably a bit pointless but…). The Z68 Express also includes attractive features such as Intel’s Rapid Storage Technology SSD caching - use a cheap SSD as a cache for electro-mechanical hard drives that are slower, but much bigger and affordable. Those features sound seriously cool, and so do the new, yet to be launched, Intel G3 SSDs that would appear to be the natural storage companions for Sandy Bridge. Traditionally, Intel puts the performance boot into the competition with an exorbitantly expensive Extreme Edition processor: not so with Sandy Bridge. Details are hazy yet, but eight- and ten-core Sandy Bridge Extreme Edition processors are apparently due out later this year, with quad-channel memory forthcoming. The hard core crowd might just sit out the current wave of chips, especially since a new 2011-pin socket will be introduced, requiring a motherboard swap again. Intel flew me over to Sydney in January for the official Asia-Pacific launch of Sandy Bridge to listen to the formidable Shmuel ”Mooly„ Eden, vice president and general manager of all things PC. Sandy Bridge is very much Mooly’s baby: he worked on the Timna CPU range, which too had integrated graphics and memory controller. Timna ended up as a cancelled clunker, but I can’t gainsay Mooly as to the advantages that Sandy Bridge brings because they’re real. At the same time, I can’t escape the feeling that Intel seems to have turned down the pace of innovation a notch or two. Perhaps Sandy Bridge was harder to cross than Intel cares to admit? http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/03/a_sandy_bridge_too_far.html http://blogs.pcworld.co.nz/pcworld/tech-guy/2011/03/a_sandy_bridge_too_far.html Wed, 02 Mar 2011 15:57:49 +1200