It's just a shame there's only one P-touch tape printer that's actually supported under Linux using "official" drivers -- the PT-9500. My PT-2450DX is sitting here gathering dust while I learn enough Libgd to write a driver library.
For anyone who's currently engaged in an argument with a PT-2450DX: the protocol is basically the same as the PT-9500PC in "CBP-RASTER (PTCBP)" mode. The documentation for this particular mode is on .
There was an OSS/FS implementation of Battle.net -- Bnetd. Problem is, Blizzard didn't like the idea of that (OH NOES THEY'LL USE IT TO PIRATE OUR GAMEZ! O WOE IZ US!) and sued the Bnetd developers.
There's PvPGN as well, but I haven't had a chance to play with that. Looks like it's still being updated, though.
Probably the Canon 1000D or the Nikon D40 for digital.
That said, there are plenty of film SLRs on the used market. If I was starting over I'd probably get something along the lines of a Canon EOS 33V (Elan 7N) and a "thrifty fifty" (Canon EF 50mm/f1.8 Mk.II) to start with. After that, improve the lens -- EF 28-135mm f3.5-5.6/IS/USM or EF 24-105 f4 L/IS/USM.
If for some reason I decided to go down the Nikon route instead... I honestly don't know what camera body I'd buy, but I'd stick with a 50mm as a starter -- 50mm lenses were for many years the staple of lens manufacturers. They've had years to perfect the optical quality (prime, aka fixed-focal or "fixed zoom", lenses tend to be quite sharp anyway) and get the price down -- an entry-level Canon or Nikon 50mm lens can be bought new for about a hundred pounds Sterling.
Can you tell I've been doing this a while?:)
Re:Any recommendations for a digital point-n-shoot
on
Kodak Kills Kodachrome
·
· Score: 3, Informative
You're probably not going to get RAW mode in any compact in that price range... Not with stock firmware, anyway. The first compact that comes to mind with RAW mode is the Canon G10 and its predecessor, the G9.
Alternatively most of the PowerShot and Ixus range can run CHDK, which adds RAW mode, a live histogram, and a few other really neat toys to the Canon firmware.
(From the article) "Eastman Kodak Co said it will retire Kodachrome color film this year, ending its 74-year run after a dramatic decline in sales."
The problem with Kodachrome (when compared to E6-process slide film) is that the developing process (the "K14" process) is quite elaborate and complex -- it involves seven different chemicals, exposure to light during different stages of processing, and a ton of monitoring. There are only a few companies that still have working K14 processing machines and the chemistry and expertise necessary to run them (Dwayne's Photo Service in Kansas). From what I've heard you can still process it with black-and-white chemistry, though obviously without the colour.
As a point of comparison, E-6 is a far simpler process -- half a dozen steps in the "pure" E6 process, or three (or four, depending on manufacturer) for the "simplified" E6 kits sold by e.g. Fuji-Hunt Chemicals and Tetenal. As long as you can keep the chemical bath temperatures within spec, it's possible to do E6 (and C41, the normal colour negative process) at home. Getting the chemistry isn't easy, and the chemical heaters are getting thin on the ground (but you can always use a sink filled with warm water and a couple of mixing jugs/flasks).
For what it's worth, Fujifilm are still making Velvia. E6 process, and about the same tonal response as Kodachrome. Admittedly it isn't exactly the same, but it's close enough that for most people it really doesn't matter (and there are far more E6 labs and pro-labs than there are K14 labs)...
I suspect the main issue was more that they didn't even bother to download the exploit data while it was still private, even to take a peek for curiosity's sake:
06/04/2009 - nothing heard from vendor, and the private resource containing the vulnerability info still does not appear to have been accessed
It doesn't take two WEEKS to grab a text file off a HTTP server and have a quick skim through it. The problem wasn't that they were trying to fix it, it was that they didn't WANT to fix it, and were probably hoping that they could make the problem go away simply by ignoring the security bod. The "Ostrich" solution, in other words.
(Although ostriches don't actually bury their heads in the sand, but that's OT for this thread...)
Yep, the Thunderbirds were 2nd-gen, the K7 and K75 were the 1st-gens.
Although to be fair, the failure mode was a complete loss of the heatsink. In the real world, a fan failure is a distinct possibility (and the motherboard should ideally pick that up and start beeping like mad or shut down if that happens) but the complete loss of a heatsink is incredibly unlikely. I can see it happening on a Socket7 board that's been used as a CPU testbed, but not on a modern LGA775 or AMD board. Well, maybe on one of the AMD boards (if memory serves they still use the clip-down mounts).
Some of the "overclocker" type heatsinks (e.g. Zalman CNPS9500AT) have moved to screw-down mounts. You have a *big* plastic frame that mounts onto the motherboard (to stop the heatsink flexing the board and to serve as a mount) and the heatsink screws into that. So that's six screws that could fail, but only two of which (the two on the heatsink clip) would make the HSF fall off if the screwhead broke off.
AIUI, almost all modern processors have some form of overtemperature protection. On the Intel Core chips it's a clock-throttler which kicks in at about 70C, and IIRC AMD have something that ties into the "Cool 'n' Quiet" clock-throttler. You're probably not going to kill a modern chip just by pulling the heatsink.
I believe it's possible to alter the USB power with a software driver. Just set the power level to over 9000, and your peripherals will fry.
Um, no.
The voltage on a USB connector is fixed at 5V. The controller starts up the slave device (a mouse or whatever) at 100mA, reads off the device descriptor, then kicks it up to 500mA if the device needs it, and if enough power is available.
The rule is, having too much voltage will blow stuff up, but a device will only take as much current as it needs. If you have a chip that needs 500mA at 5V, then plug it into a 1A 5V power supply, then the chip will only 'take' 500mA from the PSU.
Thing is, the USB host controller only has a 5V supply and a 5V output for slave devices. The absolute worst you'd be able to do is turn a couple of devices on and off at random (which could be fun to do as an April Fool's joke)...
That was the 1st-gen Athlons, i.e. the Thunderbird class. The thermal sensor couldn't handle rapid increases in temperature (I think the limit was one degree C per five seconds or something like that) so if the heatsink failed or you forgot to put thermal compound on there...
What you got was a puff of smoke, and a dead CPU and motherboard (more specifically the CPU socket usually melted, and the core voltage regulators cooked). Still a bit of an expensive cockup though; this was in the days where a Tbird would cost around $200, plus another $160 to $300 for the motherboard.
Personally, I'd boot off a Linux live CD (either one of the Ubuntu LiveCDs or the System Rescue CD) and then zero-wipe the first megabyte or so of each partition, and finish off by zero-wiping the boot sector.
That said, if I'm doing a PC repair for someone else, I'll get a USB external drive and pull the contents of My Documents off the HDD with the LiveCD before zapping everything. If it's my machine I'll zap everything and restore from backup...
In the UK, a lot of ISPs block outbound SMTP, or at least heavily restrict it as an anti-spam measure. Typically you can connect to the ISP's mailserver, but not any other SMTP server.
Why not do the same for SMB (ports 138 and 139 if memory serves) and RPC (sorry, can't remember that one)?
A lot of viruses seem to use security holes in the Windows file-sharing and remote procedure call services to run arbitrary code. If you stop machines sending or receiving SMB/RPC packets over the Internet, then you eliminate one of the more prevalent attack vectors used by viruses.
Admittedly that doesn't help when dealing with viruses that spread via web and email links or bugs in web browsers or plugins, but it should virtually eliminate Blaster, Sasser and the like.
This is one case where I'd advocate ISPs monitoring in- and outbound traffic. If a machine is showing signs of virus infection or hijacking, kick it offline and send the owner a letter.
Problem is, Joe Typical User doesn't know anything about computer security, probably doesn't care, and isn't likely to go out and spend £50 to have his local computer shop deal with the virus. If it was free then he might, but not if there's a cost (or even *potential cost* involved). If the ISP keeps booting him off, then he's just going to change to an ISP that doesn't do any form of virus scanning and go back to living in blissful ignorance...
Unless the machine physically won't work, "don't care" is the order of the day.
I've had Logitech mice for ages -- going right the way back to my Acorn RiscPC (technically an "Acorn" mouse, but actually made by Logitech)...
My two latest acquisitions were a V450 Nano (with the nano-IR receiver) that I bought to use with my two laptops (I can't stand the Trackpad on my Eee 1000H or the Trackpoint joystick on my Thinkpad) and an LX8 Laser. Seriously, these two track on damn near anything, including the awful glossy desks at university and my varnished pine computer desk at home.
I absolutely love the V450, but I'm very, *very* tempted to get a screwdriver and some superglue and permanently disable the two side buttons on the LX8. Single most annoying mouse feature *ever*. At least you can turn them off in the Setpoint control panel, it's just a shame that said control panel is a 60MB chunk of bloatware...
But eight months on a set of Duracells is still pretty decent. I'd rather that than having to deal with replacing some custom Li-ion pack two or three years down the line (cue: "I'm sorry sir, that product is no longer manufactured and spare parts are no longer available.")
... was from a SuSE 5.2 CD that came with PC Plus magazine *years* ago. I installed that on a Pentium P100 with 64MB of RAM, and just about got it working. Then I gave in and went back to Win98 for a bit.
In early '01 I had a play with Linux From Scratch, installed that on a Toshiba T2130CS laptop (486/75, 8MB RAM, 2GB hard drive and an 800x600 colour passive-matrix LCD) and used that for a couple of years. Did a few upgrades on the home desktop machine (which ended up becoming a 500MHz K6-II box with an 8GB HDD) and pretty much left the laptop alone. Truth be told, it ran Linux far better than it ran Win98... The desktop box ended up running Slackware.
Some time in '05 I upgraded the desktop machine -- that became an Athlon64 3200+, with 512MB RAM, the same 8GB HDD, and some other bits. I ran Slackware on that for a while, then switched to Fedora in mid-2006.
Last year I switched to Ubuntu (after seeing it running on a friend's machine and exclaiming "I don't believe it, they tamed Debian!") and I've been running that on two laptops (an IBM T42 and an Eee 1000H) and the desktop (overclocked Q6600, 4GB, Geforce GTX260, 2x500GB fakeraid-1 + 500GB + 500GB USB).
On top of that, I've got a neat little Jetway mini-ITX board in a 1U rack-case, running Ubuntu Server on a 2x500GB softRAID. That thing is basically running my home network -- Windows domain control, file sharing and network routing. Installation took about half an hour, and most of that was down to my repeatedly screwing up the RAID configuration and partitioning...
I still use Windows, but not as much as I used to (I basically run Linux all the time on the laptops, the desktop runs Windows most of the time because I'm too lazy to reboot into Linux after I've finished playing with XP).
Truth be told, I'd love a reduction to only 99% spam.
Last time I checked, the inbound spam stats on my server came up at 99.8% or so spam. This is a server that's shared with five or six friends, and is basically used for email and web hosting.
Outbound spam is still zero, though. Can't argue with that.
Anybody who thinks that RAID=backup is going to learn an exciting lesson
Except those of us who run RAID arrays with mismatched drives:)
In my case, it's a 500GB RAID-1 array, with a 7200.10 and a 7200.11. The 7200.10 (date code "07394" which I guess is 2007 week 39, made in Thailand) still spins and talks, the 7200.11 (d/c "08364", also made in Thailand) spins but won't talk to the PC...
Interesting, I've had the opposite experience. I had two MSI motherboards (MS-6147 - Slot 1, Celeron/P3 class Mini-ATX) fail, plus I've seen a few of their graphics cards fail (both owned by a friend). Every time it's the same thing, faulty capacitors.
On the other hand, I've got an ASUS A8V Deluxe that's nearly five years old and still works perfectly. The Geforce card in that machine needed a new fan, but the motherboard just keeps on going. It's like the frickin' Duracell Bunny...
I can't comment on current hardware, but the P5N72-T in my current main machine (and the Eee 1000H sitting next to me) work fine. Time will tell, of course, but at this point in time I really don't have anything bad to say about them. In fact, I don't have anything bad to say about Gigabyte either -- I have (somewhere) a GA-586ATM motherboard -- one of the old Pentium boards. To my knowledge, it still works fine.
In any case, I learned long ago not to buy cheap computer components. Buy it cheap, buy it twice.
First thing's first, you'll want some decent reference material.
The Forrest Mims books are a great starter -- grab yourself a copy of "Getting Started in Electronics" to start with. RadioShack should have that one.
I'm also somewhat surprised that nobody's mentioned The Art of Electronics by Paul Horowitz and Winfield Hill. It's pretty much the standard reference work, and a lot of electronics design courses use it as the primary course textbook. It covers damn close to everything you need to know. Bit on the expensive side though, unless you can get a used copy.
Look for the Babani Publishing books too, they usually sell for a few quid on Amazon. Worth every penny - they're basically little pocket books that cover one subject (e.g. oscillators, filter circuits, etc) and cover it in some depth but without the mathematical baggage of Horowitz and Hill.
Start by learning Ohm's Law! V=IR, P=IV and the reciprocals. They're tremendously useful and state the relationship between voltage, current, resistance and power (Watts).
OK, now we move onto tools... Get yourself a decent multimeter - don't go overboard and spend $300 on a top-end Fluke, you don't need it. A $20 corner-shop digital multimeter will do fine. Just keep the battery charged up, they tend to drift when the battery starts to drain. Needless to say, the better ones don't do that:)
You'll also want a solderless breadboard and a half decent 5V power supply. The little switch-mode "universal" supplies are usually pretty good, Maplin have one that does 5V (among other voltages) on sale for £9 at the moment.
Components-wise, get yourself a selection of resistors, capacitors and transistors. FastComponents are worth a look if you're in the UK, they've got capacitor kits, crystals, some ICs, stripboard... Resistor packs can be had for not much money, you want a basic E12 kit, preferably consisting of 0.25W metal-film resistors, and 100 of each value. Expect to pay about £40 (there's a company called Dannell Electronics -- again UK based -- that sell these on eBay for a bit less). Bear in mind that it's sometimes cheaper -- especially with Farnell -- to buy the individual resistors than the overpriced "E12 Resistor Kit".
Suppliers... Well, there's FastComponents, Dannell, Bowood Electronics and the like in the 'small guys' category. These places tend not to have a very wide range of parts, but are often ridiculously cheap. They're usually run by people who actually know a bit about electronics, so they might be able to offer suggestions as to what you need.
Next you have the big guys -- CPC, Farnell, Maplin and RS. CPC and Farnell are basically the same company -- CPC have historically been a spares supplier and don't really do raw parts, whereas Farnell stray more to the raw components side of the line. Prices are reasonable, but not great. Rapid Electronics are worth a look - cheaper than Farnell, but I've never ordered from them. YMMV.
If you're in the USA? DigiKey rock (I've ordered from them a few times), and there's also Mouser and one other company whose name escapes me at the moment... Also search for BG Micro, they're a surplus supplier and often have some pretty good deals on odd parts.
Surplus suppliers are the 'odds and end shops' of the electronics hobby community. There used to be Greenweld, Mainline and a few others on this side of the pond; Greenweld has gone "Innovations Catalogue" and no longer sell components, and Mainline's warehouse burned down. The latter have risen from the ashes and now sell primarily on eBay UK.
Surplus kit is what it is -- sometimes you see fantastic bargains on things like graphics LCDs that normally sell new for £50-£100, but usually hit the surplus market for far less. YMMV though, some stuff is new-old-stock and generally quite nice, other things are pulled from broken equipment. Yet more stuff is cheap because it's weird
At the risk of getting flamed into oblivion by the "DRM BAAAAD" crowd... surely if you PPV a movie, you're paying for the right to watch it once. Like going to the cinema - if you want to see it again, you buy another ticket. This is pretty much how Sky advertised their 'Box Office' PPV service - "it's like a cinema in your living room" (IIRC).
Except in this case you can live-pause the movie while you refill your tub of popcorn, grab another beer, or whatever. Then if you missed a few seconds you can kick it back and watch it again so you don't miss any of the plot.
For what it's worth, I can't honestly think of any movie I've watched in the last ~4 years that I'd actually watch again...
(I've never had a TiVo because I refuse to pay $15/mo for the rest of my life for a few kilobytes of guide data. When I moved to a no-dish apartment, I built a PC-based DVR that kicks ass). A MythTV box, or something else?
Actually, DVI can carry HDCP-encrypted video, just as HDMI can carry un-encrypted video. I don't consider HDMI itself to be 'evil' or 'DRMy', just the DRM scheme itself (HDCP)...
DVD (which has laughable DRM)
Not really an alternative to BluRay, the video is SD-only (768x576 for PAL 16:9). And the DRM's only "laughable" because someone broke it:P
Alternative to iTunes - DRM-free MP3 download(amazon, etc), CDs that are not protected(harder to tell)
Get a LiteOn CD/DVD drive. Watch almost all forms of CD "copy protection" disappear into the background. I've yet to find a CP'd audio "CD" that my LH-20A1H (or indeed the LDW-851S before it) couldn't read.
But yeah, I use iTunes (+myFairtunes) for the weekly free music, but anything I buy is always on CD. I have too many portable players (Asus Eee, and Sony MZRH1 Minidisc to name two) to make DRM'd music worth bothering with...
To be honest, I haven't seen any protected CDs in a while. Maybe it's because I tend to ignore anything that's in the top 40...
Slashdot Mirror
Are you related to my boss?
It's just a shame there's only one P-touch tape printer that's actually supported under Linux using "official" drivers -- the PT-9500. My PT-2450DX is sitting here gathering dust while I learn enough Libgd to write a driver library.
For anyone who's currently engaged in an argument with a PT-2450DX: the protocol is basically the same as the PT-9500PC in "CBP-RASTER (PTCBP)" mode. The documentation for this particular mode is on .
Obligatory Armageddon quote:
American components, Russian components, all made in Taiwan! :)
There was an OSS/FS implementation of Battle.net -- Bnetd. Problem is, Blizzard didn't like the idea of that (OH NOES THEY'LL USE IT TO PIRATE OUR GAMEZ! O WOE IZ US!) and sued the Bnetd developers.
There's PvPGN as well, but I haven't had a chance to play with that. Looks like it's still being updated, though.
Probably the Canon 1000D or the Nikon D40 for digital.
That said, there are plenty of film SLRs on the used market. If I was starting over I'd probably get something along the lines of a Canon EOS 33V (Elan 7N) and a "thrifty fifty" (Canon EF 50mm/f1.8 Mk.II) to start with. After that, improve the lens -- EF 28-135mm f3.5-5.6/IS/USM or EF 24-105 f4 L/IS/USM.
If for some reason I decided to go down the Nikon route instead... I honestly don't know what camera body I'd buy, but I'd stick with a 50mm as a starter -- 50mm lenses were for many years the staple of lens manufacturers. They've had years to perfect the optical quality (prime, aka fixed-focal or "fixed zoom", lenses tend to be quite sharp anyway) and get the price down -- an entry-level Canon or Nikon 50mm lens can be bought new for about a hundred pounds Sterling.
Can you tell I've been doing this a while? :)
You're probably not going to get RAW mode in any compact in that price range... Not with stock firmware, anyway. The first compact that comes to mind with RAW mode is the Canon G10 and its predecessor, the G9.
Alternatively most of the PowerShot and Ixus range can run CHDK, which adds RAW mode, a live histogram, and a few other really neat toys to the Canon firmware.
URL for the latter is: http://chdk.wikia.com/wiki/CHDK
(From the article)
"Eastman Kodak Co said it will retire Kodachrome color film this year, ending its 74-year run after a dramatic decline in sales."
The problem with Kodachrome (when compared to E6-process slide film) is that the developing process (the "K14" process) is quite elaborate and complex -- it involves seven different chemicals, exposure to light during different stages of processing, and a ton of monitoring. There are only a few companies that still have working K14 processing machines and the chemistry and expertise necessary to run them (Dwayne's Photo Service in Kansas). From what I've heard you can still process it with black-and-white chemistry, though obviously without the colour.
As a point of comparison, E-6 is a far simpler process -- half a dozen steps in the "pure" E6 process, or three (or four, depending on manufacturer) for the "simplified" E6 kits sold by e.g. Fuji-Hunt Chemicals and Tetenal. As long as you can keep the chemical bath temperatures within spec, it's possible to do E6 (and C41, the normal colour negative process) at home. Getting the chemistry isn't easy, and the chemical heaters are getting thin on the ground (but you can always use a sink filled with warm water and a couple of mixing jugs/flasks).
For what it's worth, Fujifilm are still making Velvia. E6 process, and about the same tonal response as Kodachrome. Admittedly it isn't exactly the same, but it's close enough that for most people it really doesn't matter (and there are far more E6 labs and pro-labs than there are K14 labs)...
I suspect the main issue was more that they didn't even bother to download the exploit data while it was still private, even to take a peek for curiosity's sake:
06/04/2009 - nothing heard from vendor, and the private resource containing the vulnerability info still does not appear to have been accessed
It doesn't take two WEEKS to grab a text file off a HTTP server and have a quick skim through it. The problem wasn't that they were trying to fix it, it was that they didn't WANT to fix it, and were probably hoping that they could make the problem go away simply by ignoring the security bod. The "Ostrich" solution, in other words.
(Although ostriches don't actually bury their heads in the sand, but that's OT for this thread...)
Ah, good old sticktion!
the one that says BORK
Or perhaps BORT?
+1 to you sir, right on all counts.
Yep, the Thunderbirds were 2nd-gen, the K7 and K75 were the 1st-gens.
Although to be fair, the failure mode was a complete loss of the heatsink. In the real world, a fan failure is a distinct possibility (and the motherboard should ideally pick that up and start beeping like mad or shut down if that happens) but the complete loss of a heatsink is incredibly unlikely. I can see it happening on a Socket7 board that's been used as a CPU testbed, but not on a modern LGA775 or AMD board. Well, maybe on one of the AMD boards (if memory serves they still use the clip-down mounts).
Some of the "overclocker" type heatsinks (e.g. Zalman CNPS9500AT) have moved to screw-down mounts. You have a *big* plastic frame that mounts onto the motherboard (to stop the heatsink flexing the board and to serve as a mount) and the heatsink screws into that. So that's six screws that could fail, but only two of which (the two on the heatsink clip) would make the HSF fall off if the screwhead broke off.
AIUI, almost all modern processors have some form of overtemperature protection. On the Intel Core chips it's a clock-throttler which kicks in at about 70C, and IIRC AMD have something that ties into the "Cool 'n' Quiet" clock-throttler. You're probably not going to kill a modern chip just by pulling the heatsink.
I believe it's possible to alter the USB power with a software driver. Just set the power level to over 9000, and your peripherals will fry.
Um, no.
The voltage on a USB connector is fixed at 5V. The controller starts up the slave device (a mouse or whatever) at 100mA, reads off the device descriptor, then kicks it up to 500mA if the device needs it, and if enough power is available.
The rule is, having too much voltage will blow stuff up, but a device will only take as much current as it needs. If you have a chip that needs 500mA at 5V, then plug it into a 1A 5V power supply, then the chip will only 'take' 500mA from the PSU.
Thing is, the USB host controller only has a 5V supply and a 5V output for slave devices. The absolute worst you'd be able to do is turn a couple of devices on and off at random (which could be fun to do as an April Fool's joke)...
That was the 1st-gen Athlons, i.e. the Thunderbird class. The thermal sensor couldn't handle rapid increases in temperature (I think the limit was one degree C per five seconds or something like that) so if the heatsink failed or you forgot to put thermal compound on there...
What you got was a puff of smoke, and a dead CPU and motherboard (more specifically the CPU socket usually melted, and the core voltage regulators cooked). Still a bit of an expensive cockup though; this was in the days where a Tbird would cost around $200, plus another $160 to $300 for the motherboard.
Personally, I'd boot off a Linux live CD (either one of the Ubuntu LiveCDs or the System Rescue CD) and then zero-wipe the first megabyte or so of each partition, and finish off by zero-wiping the boot sector.
dd if=/dev/zero of=/dev/sda[1..n] bs=1M count=1
dd if=/dev/zero of=/dev/sda bs=1M count=1
That said, if I'm doing a PC repair for someone else, I'll get a USB external drive and pull the contents of My Documents off the HDD with the LiveCD before zapping everything. If it's my machine I'll zap everything and restore from backup...
In the UK, a lot of ISPs block outbound SMTP, or at least heavily restrict it as an anti-spam measure. Typically you can connect to the ISP's mailserver, but not any other SMTP server.
Why not do the same for SMB (ports 138 and 139 if memory serves) and RPC (sorry, can't remember that one)?
A lot of viruses seem to use security holes in the Windows file-sharing and remote procedure call services to run arbitrary code. If you stop machines sending or receiving SMB/RPC packets over the Internet, then you eliminate one of the more prevalent attack vectors used by viruses.
Admittedly that doesn't help when dealing with viruses that spread via web and email links or bugs in web browsers or plugins, but it should virtually eliminate Blaster, Sasser and the like.
This is one case where I'd advocate ISPs monitoring in- and outbound traffic. If a machine is showing signs of virus infection or hijacking, kick it offline and send the owner a letter.
Problem is, Joe Typical User doesn't know anything about computer security, probably doesn't care, and isn't likely to go out and spend £50 to have his local computer shop deal with the virus. If it was free then he might, but not if there's a cost (or even *potential cost* involved). If the ISP keeps booting him off, then he's just going to change to an ISP that doesn't do any form of virus scanning and go back to living in blissful ignorance...
Unless the machine physically won't work, "don't care" is the order of the day.
I've had Logitech mice for ages -- going right the way back to my Acorn RiscPC (technically an "Acorn" mouse, but actually made by Logitech)...
My two latest acquisitions were a V450 Nano (with the nano-IR receiver) that I bought to use with my two laptops (I can't stand the Trackpad on my Eee 1000H or the Trackpoint joystick on my Thinkpad) and an LX8 Laser. Seriously, these two track on damn near anything, including the awful glossy desks at university and my varnished pine computer desk at home.
I absolutely love the V450, but I'm very, *very* tempted to get a screwdriver and some superglue and permanently disable the two side buttons on the LX8. Single most annoying mouse feature *ever*. At least you can turn them off in the Setpoint control panel, it's just a shame that said control panel is a 60MB chunk of bloatware...
But eight months on a set of Duracells is still pretty decent. I'd rather that than having to deal with replacing some custom Li-ion pack two or three years down the line (cue: "I'm sorry sir, that product is no longer manufactured and spare parts are no longer available.")
... was from a SuSE 5.2 CD that came with PC Plus magazine *years* ago. I installed that on a Pentium P100 with 64MB of RAM, and just about got it working. Then I gave in and went back to Win98 for a bit.
In early '01 I had a play with Linux From Scratch, installed that on a Toshiba T2130CS laptop (486/75, 8MB RAM, 2GB hard drive and an 800x600 colour passive-matrix LCD) and used that for a couple of years. Did a few upgrades on the home desktop machine (which ended up becoming a 500MHz K6-II box with an 8GB HDD) and pretty much left the laptop alone. Truth be told, it ran Linux far better than it ran Win98... The desktop box ended up running Slackware.
Some time in '05 I upgraded the desktop machine -- that became an Athlon64 3200+, with 512MB RAM, the same 8GB HDD, and some other bits. I ran Slackware on that for a while, then switched to Fedora in mid-2006.
Last year I switched to Ubuntu (after seeing it running on a friend's machine and exclaiming "I don't believe it, they tamed Debian!") and I've been running that on two laptops (an IBM T42 and an Eee 1000H) and the desktop (overclocked Q6600, 4GB, Geforce GTX260, 2x500GB fakeraid-1 + 500GB + 500GB USB).
On top of that, I've got a neat little Jetway mini-ITX board in a 1U rack-case, running Ubuntu Server on a 2x500GB softRAID. That thing is basically running my home network -- Windows domain control, file sharing and network routing. Installation took about half an hour, and most of that was down to my repeatedly screwing up the RAID configuration and partitioning...
I still use Windows, but not as much as I used to (I basically run Linux all the time on the laptops, the desktop runs Windows most of the time because I'm too lazy to reboot into Linux after I've finished playing with XP).
Truth be told, I'd love a reduction to only 99% spam.
Last time I checked, the inbound spam stats on my server came up at 99.8% or so spam. This is a server that's shared with five or six friends, and is basically used for email and web hosting.
Outbound spam is still zero, though. Can't argue with that.
Anybody who thinks that RAID=backup is going to learn an exciting lesson
Except those of us who run RAID arrays with mismatched drives :)
In my case, it's a 500GB RAID-1 array, with a 7200.10 and a 7200.11. The 7200.10 (date code "07394" which I guess is 2007 week 39, made in Thailand) still spins and talks, the 7200.11 (d/c "08364", also made in Thailand) spins but won't talk to the PC...
On the other hand, I've got an ASUS A8V Deluxe that's nearly five years old and still works perfectly. The Geforce card in that machine needed a new fan, but the motherboard just keeps on going. It's like the frickin' Duracell Bunny...
I can't comment on current hardware, but the P5N72-T in my current main machine (and the Eee 1000H sitting next to me) work fine. Time will tell, of course, but at this point in time I really don't have anything bad to say about them. In fact, I don't have anything bad to say about Gigabyte either -- I have (somewhere) a GA-586ATM motherboard -- one of the old Pentium boards. To my knowledge, it still works fine.
In any case, I learned long ago not to buy cheap computer components. Buy it cheap, buy it twice.
They're like kids that enjoy destroying other kids' sand castles.
I'll go back into my cave now.
The Forrest Mims books are a great starter -- grab yourself a copy of "Getting Started in Electronics" to start with. RadioShack should have that one.
I'm also somewhat surprised that nobody's mentioned The Art of Electronics by Paul Horowitz and Winfield Hill. It's pretty much the standard reference work, and a lot of electronics design courses use it as the primary course textbook. It covers damn close to everything you need to know. Bit on the expensive side though, unless you can get a used copy.
Look for the Babani Publishing books too, they usually sell for a few quid on Amazon. Worth every penny - they're basically little pocket books that cover one subject (e.g. oscillators, filter circuits, etc) and cover it in some depth but without the mathematical baggage of Horowitz and Hill.
Start by learning Ohm's Law! V=IR, P=IV and the reciprocals. They're tremendously useful and state the relationship between voltage, current, resistance and power (Watts).
OK, now we move onto tools... Get yourself a decent multimeter - don't go overboard and spend $300 on a top-end Fluke, you don't need it. A $20 corner-shop digital multimeter will do fine. Just keep the battery charged up, they tend to drift when the battery starts to drain. Needless to say, the better ones don't do that :)
You'll also want a solderless breadboard and a half decent 5V power supply. The little switch-mode "universal" supplies are usually pretty good, Maplin have one that does 5V (among other voltages) on sale for £9 at the moment.
Components-wise, get yourself a selection of resistors, capacitors and transistors. FastComponents are worth a look if you're in the UK, they've got capacitor kits, crystals, some ICs, stripboard... Resistor packs can be had for not much money, you want a basic E12 kit, preferably consisting of 0.25W metal-film resistors, and 100 of each value. Expect to pay about £40 (there's a company called Dannell Electronics -- again UK based -- that sell these on eBay for a bit less). Bear in mind that it's sometimes cheaper -- especially with Farnell -- to buy the individual resistors than the overpriced "E12 Resistor Kit".
Suppliers... Well, there's FastComponents, Dannell, Bowood Electronics and the like in the 'small guys' category. These places tend not to have a very wide range of parts, but are often ridiculously cheap. They're usually run by people who actually know a bit about electronics, so they might be able to offer suggestions as to what you need.
Next you have the big guys -- CPC, Farnell, Maplin and RS. CPC and Farnell are basically the same company -- CPC have historically been a spares supplier and don't really do raw parts, whereas Farnell stray more to the raw components side of the line. Prices are reasonable, but not great. Rapid Electronics are worth a look - cheaper than Farnell, but I've never ordered from them. YMMV.
If you're in the USA? DigiKey rock (I've ordered from them a few times), and there's also Mouser and one other company whose name escapes me at the moment... Also search for BG Micro, they're a surplus supplier and often have some pretty good deals on odd parts.
Surplus suppliers are the 'odds and end shops' of the electronics hobby community. There used to be Greenweld, Mainline and a few others on this side of the pond; Greenweld has gone "Innovations Catalogue" and no longer sell components, and Mainline's warehouse burned down. The latter have risen from the ashes and now sell primarily on eBay UK.
Surplus kit is what it is -- sometimes you see fantastic bargains on things like graphics LCDs that normally sell new for £50-£100, but usually hit the surplus market for far less. YMMV though, some stuff is new-old-stock and generally quite nice, other things are pulled from broken equipment. Yet more stuff is cheap because it's weird
Except in this case you can live-pause the movie while you refill your tub of popcorn, grab another beer, or whatever. Then if you missed a few seconds you can kick it back and watch it again so you don't miss any of the plot.
For what it's worth, I can't honestly think of any movie I've watched in the last ~4 years that I'd actually watch again...
Actually, DVI can carry HDCP-encrypted video, just as HDMI can carry un-encrypted video. I don't consider HDMI itself to be 'evil' or 'DRMy', just the DRM scheme itself (HDCP)...
Not really an alternative to BluRay, the video is SD-only (768x576 for PAL 16:9). And the DRM's only "laughable" because someone broke itTo be honest, I haven't seen any protected CDs in a while. Maybe it's because I tend to ignore anything that's in the top 40...