Microsoft moving away from desktop Windows

It seems like I'm not the only one that thinks Windows on the desktop is on its way out - even Microsoft agrees... see The Microsoft Albatross

To quad-core or not to quad-core, watt(s) is the questions

I've been thinking of getting a new home computer and even considering getting a retail one from Dell vs. building it myself as I did last time. The reason I did a home build system last time was because I wanted tight control to make it as quiet as possible, and that off the shelf systems just didn't have the specs I wanted at the price I wanted. Last time I spent about $2000 in total, but this time around I'm looking for a sub $1000 system.

In fact Dell has just dropped their price on quad-core processors, reflecting the recent drop in price from Intel so I can actually get such a system for around $900 (assuming I can find a $350 off coupon that works - the current one is "sold out", ie. too many uses). That is pretty amazing since it also includes 2Gb of 800Mhz memory, and a 8xxx series nVidia graphics card. So I'd be getting a new system with a processor that has four cores each as fast as my current system for less than half the price.

But once again I find myself in a system specification quandary. You see I've been on a "low power" jihad recently, measuring the power consumption of all my electronics in a vain attempt to get my average consumption to less than the current whopping 1kw per hour. I'm not sure how that happened, but the reality is really high power bills! I know that an always on computer doesn't help - my current system sucks down 138w when idle, and 174w when pegged - combined with sundry peripherals that's probably 200w worth of power being drawn 24/7/365 which adds up to around $200 a year even at the cheapest rate. The reality is since my baseline usage is > the PG&E's minimum of 342kwh per month then most of my power costs me 20 and 30 cents per kwh, so actually system running costs are more like $400 to $600 a year!

What I do know is my Dell Inspiron 700m laptop sips around 18w when idle (and the screen is off) or a max of 41w when pegged (but not charging the 65Whr battery). That's impressive and I know that the latest laptop processors can probably do much better. So, do I try and build a state of art low power system or just go with the quad-core system and try to keep it off as much as possible? Or is there some happy compromise?

I know for a fact that the Q6600 processor I'm considering has a minimum power consumption of 50W currently, but there is a new "G0 stepping" version just out that reduces that to just 24W (see Extended Halt Power figures) although I know you still have to add all the motherboard, graphics card and other overheads before you get a true idle figure. If I'm really going to true low power than I'm either looking at Core 2 Duo laptop processors (of the T6000 and 7000 series) which idle at around 13W or in "Enhanced Deeper Sleep" mode go all the way down to 1.2W. As it turns out the standard desktop dual core processors now have an extended halt power of just 8W - that applies even to the top of the line E6850 processor.

The reality is I have a feeling that motherboard, graphics card and peripheral power usage will be something I have little control over and will probably far exceed the 20W variation I'm looking at in bottom line processor power consumption. So I think its going to be really hard to design a complete system with any assurance of what its total power consumption will be.

For this reason I think I'm going to focus on reducing my system's up time allowing it to go into suspend or even hibernate mode for more of the day. My current webcam monitoring system requires my system to be up 24/7 and my backup system keeps it occupied at least from the hours of 2am to 6am. The system is also a PVR and spends a few hours a day recording HDTV and recompressing it MP4 format. I'm hoping that eventually it can get down to only 1/3 of a day uptime which would probably dwarf any savings I could get by selecting a dual core over a quad core processor.

Planned obsolescence vs. low impact living

This Treehugger article on planned obsolescence strikes a chord with me but being a gadget and tech freak it does give me problems. In everything else in life I make my stuff last and last and last. Most of my shoes are years old and I prefer leather soled shoes that can be repaired - some I've had since before I left England and that's a long time ago. My car is 13 years old and looks it, but still drives great and I don't have to throw lots of money away on comprehensive insurance, in fact I think I should get a deduction on what insurance I have because no one is going to steal what looks to be a real clunker. And I have lots of clothes and other small things that are definitely this years latest fashion.

Of course when it comes to tech that kind of breaks down... but I'm usually not on the bleeding edget, or not continuously. Once in a while I'll break down and buy something really current, like last time I built a computer and splurged over $300 on the processor. But that was 2002 and state of the art was a 2.4Ghz P4 with 533Mhz FSB. I didn't actually get the fastest - there was one part faster at the time but the extra $200 for another .4 Ghz clock speed just didn't seem justifiable.

However I did make it last - that computer is still running and my main home workhorse. With a recent but by no means top of the line graphics card, plus a SATA2 controller card it now is my high-def TV "Tivo" courtesy of BeyondTV sucking down gigs of TV goodness per hour and playing them back at our leisure. However it is straining now especially if I try doing something else significant at the same time. So I now find myself looking at a new computer while simultaneously regretting having to junk a perfectly good machine that I've nurtured, like some techno-child for years.

Of course I wont actually junk it - it will at the very least get donated sans hard drive. It may also turn into a low power home server so I can have my new system off most of the time. Since the case is rather huge for the later - I plan something with a micro-ATX and a low power laptop Core 2 Duo chip - I think its probably going to get donated. Hopefully just because it is a non-name DIY system it will not get torn apart - with a fresh install of Ubuntu Feisty Fawn its a great machine for web browsing and email with plenty of horses for some gaming and video.

To be honest if it wasn't for the progress in the uses we find for computers there would be no need to upgrade. I'd expect to have to shove in a new disk drive every few years since they fail every few years, but other than that it should keep going, going, going. And that is the general problem with technology perse - it does keep obsoleting itself, if it didn't then we'd all still be using an abacus, driving model T cars (or just a horse and cart) and using 5 pound monster cell phones (or just sending telegrams and writing letters).

Economists will no doubt tell you this is good for the economy since it drives demand. But is it really? It seems to me if that is true then our measures of what is "good for the economy" are seriously out of whack. If all the technology we had today was good enough and was built to last couldn't we just divert all our energies and efforts to something new, or maybe solving long unsolved problems like 100% effective contraception, cures for malaria, safe and low emissions transport systems and well, maybe peace on earth and good will to all mankind?

Now I think I would be foolish to deny that all progress is bad and that fast modern transport is what lets me fly here there and everywhere, cell phones do save many lives and just plain make things easier, and fast computers sure beat a mechanical adding machine when it comes to doing taxes. So there is no point in us all keeping our trusty horse and cart in the garage alongside the Toyota Prius. However I do think we frequently miss out on opportunities to donate that older tech to someone who would now appreciate it far more than I do, in addition tech with actual bona-fide builtin obsolescence really shouldn't be tolerated.

One example would be a UPS I bought three years ago. A hefty (quite literally) investment that served well and then one day gave up the ghost because one of its two lead acid batteries inside, a $15 part, had died (it was actually worse - that battery was fine except for a single dead cell in it). Unfortunately the battery was designed to be non-replaceable and returning it to Belkin for servicing (if that was even an option) would cost more than buying a new one. So I'm sure most of them just ended up in a dumpster somewhere - a big old hunk of steel, lead and PCBs for future archaeologist (or miners) to dig up.

Fortunately I have a degree in physics and spent many years dinking around with electronics. Hence I was able to disassembly the UPS, remove the batteries, retest them, identify the faulty part, and order new ones (yes, I'm smart enough to know I'd should replace both and keep the other as a spare). Total cost was $50 including tax and shipping but saved me buying a new but otherwise pretty similar UPS. It turns out that UPSes haven't progressed much in three years - getting a new one for new features wasn't even worth considering, but at least it looks like manufacturers have got the message and almost all have end user replaceable batteries and recycling programs for the old ones.

Since I'm just the least bit cynical I would wonder if Apple thinks making the iPhone battery non user replaceable will cause more of them to end up in drawers or dumpsters after a year or two, forcing people to upgrade to the latest and greatest. Really I fail to see how they could have not made it a really easy to replace part, if I was not cynical I'd probably say they were doing it for our own good - so there was no danger of cheap fake iPhone batteries flooding the market, perhaps exploding and perhaps leading to expensive time-wasting class action suits.

Perhaps all this talk of recylcing and longevity of systems is why I'm enjoying using Ubuntu Linux - I'm just reusing tried and tested code and designs that have been around since the 70s and I've personally been using since the 80s. Why throw all those millions of many hours of work and experience away just because we can? For that reason I'm not upgrading to Windows Vista. Sure it looks real nice, but really it doesn't do a whole lot more than Windows 2000 ever did. When I see a truly radical worthwhile change in Windows or someone gives it to me for free (that's proper free, not bundled free) then I see no reason to abandon my old XP license. I don't wish Microsoft any ill will (and I have enough friends who profit daily from its existence) but I do wish they would concentrate on incremental improvements and not breaking everything every few years so users end up being forced to upgrade.

That scenario happened to me last year when Intuit took it apon themselves to stop man of its features from working unless I updated to the latest and greatest. Truth is Quicken 2004 did everything I wanted already - as did the previous version I had before that. Instead I had to shell out more money to get the latest and greatest and relearn a bunch of stuff because the entire user interface had changed. I don't find it any easier to use myself - its just different. The funny things is right after I coughed up the upgrade fee I discovered there is an open source project that reads Quicken files and does pretty much all I ever needed (rather like Word 3.0 !!!. So Intuit, I'm afraid next time you inform me that your built-in obsolescence is forcing me to upgrade I may pull out the Joker and go elsewhere.

Still all these endless and forced upgrades must be good for the economy - I mean I make a tidy sum helping small businesses deal with just these upgrade issues. Yet another take on it is that businesses just need to keep innovating to keep up with all the new companies - and supporting all their legacy customers is too expensive since tech support and designing rock solid upgrade paths is time consuming and error prone. If they don't force upgrades or ditch product Luddites then they will become uncompetitive or have to charge their customers exorbitant fees for product support.

In the final analysis I just don't know - I'm not an economist. But my simple view says that built-in obsolescence just doesn't make sence, that products should be designed for maximum life times from the start, should be designed to be recyclable when their life time is over and that it is not a given that all progress is necessary and sufficient for a happy life.

VoIP - close but no cigar (yet) - PART 1

I've been dabbling on an off with computer based telephony since I got involved with my very own dot.com experience ViaFone. Yeah we were going to change the world with a telephone based app but unfortunately we were just way ahead of the times. Only one company from those days really made it in any way shape or form and that was TellMe that recently got scooped up by Microsoft for $800 million. I think they did pretty well, and of the technology at the time they stayed on target with pure voice technology, executed and weathered the storm.

Anyway my dot com ditched its voice centered plans when it realized that real telephone apps require expensive telecoms infrastructure, expensive hardware, expensive software, and expensive telecoms and telephony engineers. You have to understand that telephony is kind of like power technology - its a world unto itself full of regulations, restrictions and just pure arcana. Enter VoIP...

It stands to reason that if you have a 1,3,6, or more megabit data connection to your home that can stream DVD quality audio visual - or even HD now - then you could surely have a decent voice conversation over that data link. I mean just look at what the computer game biz can do with massively multiplayer realtime games - all that data going back and forth brining you split second messaging and data. A bit of voice should be easy.... Well you'd think that wouldn't you.

And yet it seems like it has taken forever for VoIP reality to arrive and really work - just like video phones that still don't exist after decades of promises except in a select few corporate conferencing facilities. The reality is that VoIP the most prevelant voice over internet systems in use by consumers today are actually completely separate from the telecoms based offerings. Just like with webcames for video its been a long, disparate (or desperate) and mostly proprietary road for voice chat.

There was an initial flurry of activity in the sector when DSL became easily accessible to consumers (and AOL's countdown clock started) and that featured a number of small companies with proprietary technology for voice chat all claiming they were superior in some way or the other. Unfortunately most systems just didn't have that many users, and certainly didn't talk to other voice chat systems. There was also the problem that few people actually had a microphone on their PC - it was always an add on for a desktop, and almost no laptops had it built in before Apple made that a standard item (it was an obvious and cheap addition but I'll thank them for showing the way).

Then the big text chat people like Yahoo figured they could add voice to their existing chat systems which had the advantage of large established user bases. The problem was that there were now even more people you might be able to talk to, but couldn't because you were on MSN and they were on Yahoo, or AOL or some other system.

Then a little while after the original Napster went tits up (pardon my English) the people from another P2P app KaZaA, went off and formed Skype which turned out to be yet another voice chat system that also did a bit of text chat. For whatever reason it gained a popular following and I heard about it early on, and even tried it pretty early on. As usual even with its own dedicated band of Skype-ers there was still the problem that I really didn't know more than one or two people who used it and by then it was almost always easier to just pick up the phone. Even when I did use it I seemed to have no end of problems with microphone noise and quality. I almost always had to end up picking up - the phone.

Then still later I started hearing about VoIP and a company called Vonage and another called Packet8, these guys could actually give you a real phone number and let you make a phone call over the Internet with it. The funny thing was you didn't even need a computer - just a box you plugged a regular phone into. In fact you couldn't even use a computer to make a call since there was no software solution. So where ever you went you had to drag you Vonage box, a power supply and plug a phone into it. But the funny thing was, it really seemed to work and pretty quickly I heard about several friends who'd bought Vonage.

Now the real problem was that if you wanted to use it at home you still needed a good internet connection which usually meant DSL back then and because naked DSL didn't exist then (it used to, then they got rid of it, then it came back recently...) you couldn't ditch your phone company connection even if you wanted. If you wanted to use Vonage on the road you'd still have to drag that box around and it just didn't work at internet cafes since you'd never have a ethernet jack to plug into (unless you also had a compatible bridge in your pocket). The finally issue for me was Vonage chose to behave just like a phone company, you could only buy their system if you chose a monthly plan with bundled minutes.

Now for a long time I've had a non-standard long distance carrier at home - they are pretty unique in that they have no monthly charges unless you use the service. That's what I want. The only charge I get from them is all those usury telecoms taxes that people seem to want to slap onto anything that might offer a semblance of free speech - you know "free, as in free-fiddy a minute" not "free, as in beer". I guess someone has to pay for all that phone tap technology. If we all decided to just ditch telephones and start passing post-it notes I'm sure they'd find a way to tax that too...

Anyway, I digress (again what else do you expect from the Long Dark Tech-Time?)

While I was first looking at Vonage and Packet8 I'd learned that there was away to use what was called a "soft phone" that used VoIP and understood this new SIP system that Vonage and others were using (but not Skype). A friend of mine said he'd been able to hack into his box and figure out the settings and subsequently make calls directly from his computer. Now that is what I'm talking about. Unfortunately it seemed like Vonage wasn't really happy with people doing that and it was at best flakey.

At the same time Vonage and others crawled out onto the consumer beach struggling to gain a foothold it seemed that pretty soon (and I wish I had time to research the time-line for all of this - I remember it well but not with exact dates attached) every IT magazine and his dog was talking about big corporations rolling out VoIP systems in their offices and saving bundles.

And yet still, here I was paying $12 a month of a phone line I didn't use other than to let me pay $70 a month of a 6mbps DSL line, and also chuck $100 a month at T-Mobile for two lines of cellphone goodness and a 1000 minute bucket of cellular chat time, plus unlimited EDGE cellular data access and WiFi roaming.

I briefly experimented again with Skype - I even purchased a years worth of subscription for a call in number (30 euros) so people could call me and I could call them without getting a weird 1234567 number showing up on their caller-id. I tried it for some transatlantic calls and it generally stank. Whatever headset I used I'd almost always have problems and my brothers would complain of "that helicopter noise". More to the point it seemed whenever anyone would call me on the computer with Skype (or MSN or Yahoo voice or video chat) I'd have no headset to hand.

The other problem I had was even if I had my headset almost everyone that called me seemed to have just a microphone and speakers so I'd be forced to listen-listen to-to my-my own-own echo-ehco... ouch. I'd be quickly reduced to a stammering fool and they didn't see to realize why because they could hear me just fine (except for the helicopter) with no echo.

One final thing about Skype was that their voice mail system sucked because there is no way to check it without a computer. You actually have to turn on your computer, fire up their software and then listen on that machine. What I really, really wanted was to have my voice mail just go to an email and be able to download and play it from my mail reader, or maybe just visit a website and hear it there. But Skype was and is having none of that for reasons that I just can't fathom.

TO BE CONTINUED...

Fair and balanced - tech sucks?

Looking back at the last few entries of this blog I realized that I've been on a bit of a ranting negative tear against tech of late. Given the title of this blog that should come as no surprise to my readers (is there anybody out there?) but although I'm a pessimist I do like to through in a little positive reportage once in a while. As an ex of mine once taught me, you have to have a little balance once in a while so you can appreciate the other point of view. So, this next one is a bit more middle of the road, mostly positive, with only short and slightly dark tech-times...