Haven't heard the Japanese on it, but I think you can but Princess Mononoke in there too. Didn't the English dub even enjoy a modest Japanese theatrical run, because enough Japanese liked the English voice cast? Not a huge run, but still..
Meh the english dub of Princess Mononoke was pretty bad. They completely change a lot of the subtext in what is going on, who the true villains and who the true heroes are. I suspect it was done to make it more "western" with black/white good/evil rather then shades of grey as in the original version.
The japanese language version (the spoken, not the english subtitles) had a lot more subtlety.
Keeping track of and navigating a few million planes could be done on one server if it was well coded. Which would really cost like 500,000. I'm sure there are a bunch of other things that need doing but i'm so far off of 10billion that i've no idea how they got it that high.
You're living up to your name?
Let's talk about some of the issues:
- Radar is an inexact medium of information. Transponders help a lot, but they only have 4 digits and can be disabled or break. GPS transponders (where the aircraft reports its position) would help, but not every plan has that.
- Communication between the air/ground is not assured.
- Weather, visibility, winds aloft, precipitation, fog are all unpredictable elements that have to be accounted for. You need sensors for all that, or figure out how to do voice interpretation of a pilot's report of bad weather.
- You also need to know the status of the runways. Are they clear of ice, snow, debris or other incursions?
ATC only seems simple on the surface, during good weather, clear skis and everything running on schedules.
While I'm sure the recession has something to do with it, the fact of the matter is, unless you're a hard core gamer, or trying to run Vista, any computer bought in the last few years is "good enough."
Pretty much close on the money.
Back in the 1985-2000 period, computer power basically doubled every 12-15 months. So a three year old machine could be 4x-8x slower then a new PC. But by 2000, this rate was definitely down to 15-18 months to double performance. Now it's slowed down to about 24-36 months to double performance, at least if you exclude doubling the cores.
I think a lot of computer makers (and Microsoft) are still stuck in the 90s, thinking that people will upgrade their computers every 3 years. A more realistic view is that any PC bought since about 2002 is probably still a viable machine for light desktop use.
The dual-core PCs that we put in over the last 3 years? Those have a planned lifespan of 6-12 years. Power users will probably get upgraded at the 5-7 year mark (not counting upgrades like boosting the memory), with the older dual-core machines used by less demanding users.
It's definitely a different world since multi-core hit the scene. I don't think manufacturers or Microsoft have quite grasped that yet.
The big one is Microsoft Access. Even the ancient and venerable MSAccess97 doesn't run in WINE. Setting up a WinXP VM inside of a Linux box just for MSAccess doesn't really solve the issue.
For the type of work that we do (projects that only run for a week or two, every project is different, very little reference of old data), a large central database server doesn't make sense for us. We do have a large central database server for the collection of data, but for archival purposes, it gets put into MDBs which are easy to work with.
(On the upside, SubVersion is extremely good at dealing with binary files like MDBs. So that has not been a bottleneck for us.)
I'll probably take another look at alternatives this year.
e-books are, by far, the dumbest invention ever made.
For story-type reading (fiction, historical, basically anything without large images or detailed drawings), they're fantastic. You'll have to pry my Sony PRS 505 out of my cold dead hands.
- 400+ books on my reader (with room for a lot more)
- No DRM on my book files (Baen publishing, Project Gutenberg)
- 2-4 weeks of battery life
- Weight about equal to a large format paperback
- Reasonably sized screen with adjustable text size
- Cost under $300 (still a tad high)
- Readable in bright sunlight
I've used it for a year now. Usually for an hour each night before bed and sometimes longer on the weekends. The next time I move, that's about 100 kg less in book weight that I'll have to deal with.
If I traveled regularly, I would value it even more.
Frankly, I'm hoping that Win7 is approximately Vista SP2. Because, as much as I would love to switch off of Windows desktops and move to Linux w/ WINE. There are a lot of applications that we have which won't run in WINE, and quite a few of those are show-stoppers if we can't run them.
So I'm hoping that MS manages to create a decent OS that we can deploy when we start retiring WinXP desktops in 2011-2014. (All of our desktops are dual-core CPUs with 2GB of RAM, easily expanded to 3.5GB. We're expecting 5-8 years of useful life from them. We may even drop in quad-core CPUs if it seems like we're CPU-limited by 2012.)
Vista was a good generation to skip - there was too much new technology in there (UAC, new driver model, restricted permissions).
I'm still hoping that we can start shifting to Linux desktops, but it will require that WINE is able to run MS Office Professional. Or that ODF actually lives up to its promises (unlikely).
Face it, what kind of attention do you get when your servers never fail? When you never lose a database?
One good way to stay proactive is to write up a quarterly report on what happened during the previous quarter and what is planned out for the next year. Works well for keeping your value on the radar.
Corporate work sucks. Rules, HR policies, having to go through dozens of managers to get something changed, and politics out the wazoo. Degrees and certifications and other paper mean more then actual work experience or know-how. Been there, done that, wasted 8 years.
I'm in the same boat as Gorobei. Jack of all trades means that my job never gets boring. There's always something new to tackle. If I need time off, I get it. (In fact, the less of a fuss you make about time off or "getting what's owed me", the easier it is to get extra time off.)
On the upside, I'm senior enough at the company that I don't get abused. I know what needs to be done, what is urgent, what is nice-to-have, and what is bullpuckey. The CEO listens to me, and we periodically re-arrange priorities.
I'm hoping more for a quad-core CPU that is a 45W TDP and is under $100. All of the current quad-core CPUs out there are 95-125W (including Intel's) and are $130-$200.
I don't get such comparisons, for really low-end stuff like that wouldn't a P3 1 GHz be enough to?
For the past 2 years, I would consider the purchase of any single-core CPUs to be "penny-wise pound-foolish". Typically, you can get a dual-core CPU for only a few dollars more, and you end up with a much more responsive machine.
Our low-end work desktops have all been dual-core X2s, usually a 45W design where possible. The power savings alone from the energy efficient X2s pays for the extra CPU cost. The responsiveness of a dual-core system also pays off in improved productivity. Plus a 45W machine tends to be very quiet, even with the OEM heatsink/fan.
We switched to dual-core as soon as dual-core CPUs dropped below $200 (a few years ago now). At that point in time, AMD X2s were the only inexpensive dual-core choice, with the bonus that they were 64bit capable. So we got better performance and a bit of future-proofing.
I have a Canon PowerShot SD550. The 2GB cards work perfectly fine. The 4GB SD cards appear to work, but don't actually work (things go screwy once you exceed 2GB of data on the card).
It's not a new concept... labeling media goes all the way back to cassette tapes. (Eight tracks are before my time, were they writeable?)
Unfortunately regular SD or SDHC cards are at the lower end of useful label size. MicroSD and MiniSD are even worse. So about the best that you can do is some sort of 2-4 character code on each card, and use some other method to keep track of the contents.
Now is the time to make sure that your backup and disaster recovery planning is up to snuff and can deal with malicious intent.
(It doesn't cover data espionage, however... that's a whole different kettle of fish. But you won't have to worry about espionage if all of your systems have been hacked into the ground and your backups aren't any good.)
SVN 1.5 only came out late last year (sometime in the Fall).
Merge tracking and simplification of merges is still very rough around the edges. It's supposed to get better in 1.6 (no ETA). Better then it was by far in 1.4, but still a work in progress.
For a less technical set of users, I still feel that SVN's central repository model is best. For distributed developers with strong technical skills, there's a lot to be said for git. We're staying with SVN because that works better for us.
If anything, I find that the game offers slightly more options and flexibility these days. Mind you, that doesn't necessarily say much. Just that anything is more than zero;)
Pre-Wrath:
- Regular instances in BC almost always required crowd-control. Some required specific forms of CC or else you'd wipe repeatedly. The only way you could get by without CC was to majorly out-gear the instance.
- Aggro was a constant issue that needed to be managed, both by the DPS folks and the tank.
- Healers regularly ran out of mana.
Post-Wrath:
- CC in all instances up until level 80 heroics is purely optional. All trash has such low health compared to the amount of DPS being dished out that everything dies within 10-15 seconds. So everyone just AoEs everything down. Trash mobs become a boring chore on the way to the next boss, because there's no challenge, just a need to beat the trash mobs down so we can get to the boss.
- Aggro is no longer an issue except in really rare cases. DPS classes can simply go all out, never having to watch their aggro. Tank classes do insane amounts of aggro (and damage). All of the various aggro-reduction abilities have become mostly useless. Probably boring as hell for the tanks now that they don't have to work at it.
- Since everything dies so fast, healers don't run out of mana on anything except boss fights.
It was fun for about the first week after 3.0.2 hit and every boss in the game had their health reduced by 30% while all the DPS classes got a huge boost. But since then, it's gotten more and more boring because there is minimal challenge on the way to 80. Everyone says "wait until you do level 80 heroics, then you'll be challenged". My question is why do we have to wait 10 levels in Northrend for a challenge? I could see making the level 70 dungeons easy, but by 75 to 80, there should be a very decent chance of a wipe.
We did Naxx-10 for the first time last night. Out of the 3 wings (9 bosses), only one of them gave us any difficult trouble (Heigan the Unclean). The trash mobs were all a joke. The gargoyles were the only mobs where we had to concentrate and actually have some sort of focus. Other bosses were felled either on the first try or within 3 tries as we learned their special abilities and phases. Nowhere near as interesting as Karazhan was, where you had to adopt different tactics for different areas of trash mobs.
Blizzard has really screwed up if they think "EZ Mode" is healthy over the long term. All EZ-Mode is going to do is ramp up the churn rate on their subscriber base as people get bored with walking into a dungeon, pressing the "I Win" button, and walking back out. The old adage that "things you have to work for are things that you'll value more" comes to mind.
Agreed. Tape makes a lot of sense for high-volume applications.
It's just being crowded out of the low-end market by ever larger and ever cheaper hard drive sizes. Tape costs would have to drop by about a factor of 4 (or more) to compete in the lower end of the market where 100 tapes is a lot.
(If I could backup 800GB for $10, that would be much more of a no-brainer decision. The cost-advantage would be high enough to pay for the expensive tape drive. And $50 LTO-4 tapes are a lot better then back when a lot of large-capacity tapes cost $100 each.)
All they had to do was pull one of the drives, replace it, and lock up the original off site. In a couple of hours the drives would have been mirrored again.
I can't recommend that method unless you're running triple-active RAID1 (which is simply mirroring across 3 disks instead of just 2 disks). With a triple-mirror, when you pull that 3rd drive to take it offsite, you're at least still running on a 2-drive mirror.
And it's still best if you shut down everything (especially the databases) before you pull that drive and pop in a new one.
Alternately, temporarily hook up a 3rd drive (USB/Firewire), extend the raid array to it, then disconnect it before trimming the array back again.
Even accepting your price that's a cost of about 12.7 cents per gigabyte and you can get 800GB native LTO-4 tapes for about $50, which comes out to about 6.3 cents per gigabyte.
Except that:
- You have to spend $2500-$3500 on the LTO-4 tape drive.
- Oh, better make it two. Just in case the first one breaks. Or make sure that you know two other people with LTO-4 capable tape drives. (On the upside, LTO is pretty stable, with multiple vendors, so getting a replacement drive is probably not hard.)
Until you've bought and used 50 tapes, those individual tapes are costing you $100 each due to the cost of the drive.
2.5" 500GB SATA drives that can be hot-plugged are now about $110. A SATA to USB style miniature case is $15. So for $125 times 3, a small business can get started with an off-site backup for $375. Or go with 1TB 3.5" drives for $100, plus about $30 for a caddy. And they can simply grow the solution by buying additional drives/caddies/enclosures. And until they get to the point where they are juggling a few dozen of these external drives, it's still less expensive then the tape solution.
And restoration is a whole lot easier with hard drives. *Every* computer out there can (assuming the proper software) read from hard drives without the need for expensive tape drives. That ubiquity is worth an awful lot when it comes to disaster recovery.
Yeah, you can spin all sorts of numbers about how tape saves you more money over the long term. But in small / medium business - the ability to avoid a large cash outlay for an expensive tape drive generally wins.
(My preferred backup solution for small/medium companies is to 1. Have a in-house target server where backups get written to. 2. Shove daily updates of key information offsite over the wire every night. 3. Get a hard disk offsite weekly. All of which is generally a lot more then most companies manage to do. You can even script in Linux, with/devfs, that backup scripts will fire when good backup target drives are hot-plugged into the system.)
HTC Touch Pro cell phones are full VGA (640x480), but the principle applies.
Of course, any text size smaller then about 14pt is absolutely unreadable. And people who specify font sizes in pixels (i.e. 6px) deserve their own special level of hell.
Pixels are REALLY tiny on a display of that size... fortunately it runs Opera where you can double-click on the screen area to get it to zoom in.
LCDs interpolate when not running in their native res, which isn't good enough
NVIDIA drivers give you a few options on Windows. You can choose to:
- scale the resolution up to fill the entire screen
- scale the resolution up, but keep the aspect ratio intact (black bars on the sides for 4:3 content)
- don't scale at all (centers the image)
Of course, that 2nd CRT display comes in handy for other reasons...
Slashdot Mirror
Haven't heard the Japanese on it, but I think you can but Princess Mononoke in there too. Didn't the English dub even enjoy a modest Japanese theatrical run, because enough Japanese liked the English voice cast? Not a huge run, but still..
Meh the english dub of Princess Mononoke was pretty bad. They completely change a lot of the subtext in what is going on, who the true villains and who the true heroes are. I suspect it was done to make it more "western" with black/white good/evil rather then shades of grey as in the original version.
The japanese language version (the spoken, not the english subtitles) had a lot more subtlety.
Keeping track of and navigating a few million planes could be done on one server if it was well coded. Which would really cost like 500,000. I'm sure there are a bunch of other things that need doing but i'm so far off of 10billion that i've no idea how they got it that high.
You're living up to your name?
Let's talk about some of the issues:
- Radar is an inexact medium of information. Transponders help a lot, but they only have 4 digits and can be disabled or break. GPS transponders (where the aircraft reports its position) would help, but not every plan has that.
- Communication between the air/ground is not assured.
- Weather, visibility, winds aloft, precipitation, fog are all unpredictable elements that have to be accounted for. You need sensors for all that, or figure out how to do voice interpretation of a pilot's report of bad weather.
- You also need to know the status of the runways. Are they clear of ice, snow, debris or other incursions?
ATC only seems simple on the surface, during good weather, clear skis and everything running on schedules.
While I'm sure the recession has something to do with it, the fact of the matter is, unless you're a hard core gamer, or trying to run Vista, any computer bought in the last few years is "good enough."
Pretty much close on the money.
Back in the 1985-2000 period, computer power basically doubled every 12-15 months. So a three year old machine could be 4x-8x slower then a new PC. But by 2000, this rate was definitely down to 15-18 months to double performance. Now it's slowed down to about 24-36 months to double performance, at least if you exclude doubling the cores.
I think a lot of computer makers (and Microsoft) are still stuck in the 90s, thinking that people will upgrade their computers every 3 years. A more realistic view is that any PC bought since about 2002 is probably still a viable machine for light desktop use.
The dual-core PCs that we put in over the last 3 years? Those have a planned lifespan of 6-12 years. Power users will probably get upgraded at the 5-7 year mark (not counting upgrades like boosting the memory), with the older dual-core machines used by less demanding users.
It's definitely a different world since multi-core hit the scene. I don't think manufacturers or Microsoft have quite grasped that yet.
What are your deal breaker applications?
The big one is Microsoft Access. Even the ancient and venerable MSAccess97 doesn't run in WINE. Setting up a WinXP VM inside of a Linux box just for MSAccess doesn't really solve the issue.
For the type of work that we do (projects that only run for a week or two, every project is different, very little reference of old data), a large central database server doesn't make sense for us. We do have a large central database server for the collection of data, but for archival purposes, it gets put into MDBs which are easy to work with.
(On the upside, SubVersion is extremely good at dealing with binary files like MDBs. So that has not been a bottleneck for us.)
I'll probably take another look at alternatives this year.
e-books are, by far, the dumbest invention ever made.
For story-type reading (fiction, historical, basically anything without large images or detailed drawings), they're fantastic. You'll have to pry my Sony PRS 505 out of my cold dead hands.
- 400+ books on my reader (with room for a lot more)
- No DRM on my book files (Baen publishing, Project Gutenberg)
- 2-4 weeks of battery life
- Weight about equal to a large format paperback
- Reasonably sized screen with adjustable text size
- Cost under $300 (still a tad high)
- Readable in bright sunlight
I've used it for a year now. Usually for an hour each night before bed and sometimes longer on the weekends. The next time I move, that's about 100 kg less in book weight that I'll have to deal with.
If I traveled regularly, I would value it even more.
I have two users at the office who are still clinging to their Win98 PCs like its their only hope of salvation.
Everyone else, I managed to get them onto WinXP over the past 3 years.
My evil plan is to switch the internal network over to IPv6. That'll fix those Win98 laggards!
Frankly, I'm hoping that Win7 is approximately Vista SP2. Because, as much as I would love to switch off of Windows desktops and move to Linux w/ WINE. There are a lot of applications that we have which won't run in WINE, and quite a few of those are show-stoppers if we can't run them.
So I'm hoping that MS manages to create a decent OS that we can deploy when we start retiring WinXP desktops in 2011-2014. (All of our desktops are dual-core CPUs with 2GB of RAM, easily expanded to 3.5GB. We're expecting 5-8 years of useful life from them. We may even drop in quad-core CPUs if it seems like we're CPU-limited by 2012.)
Vista was a good generation to skip - there was too much new technology in there (UAC, new driver model, restricted permissions).
I'm still hoping that we can start shifting to Linux desktops, but it will require that WINE is able to run MS Office Professional. Or that ODF actually lives up to its promises (unlikely).
VMWare is too expensive for all but the largest companies. Small companies are not going to pay a few grand per server for VMWare.
Frankly, I see QEMU / Xen / KVM ruling the roost in the small / medium business category.
Hint... 47/300 is still 16%, which is a hell of a lot different then 20%. Even taking that out to 3 digits would not be unusual.
Planning to get rid of it... by putting it on the higher end laptops just released.
I think a lot of the griping about Apple abandoning Firewire comes from the MacBook Air which didn't have a Firewire port (but had USB ports).
Face it, what kind of attention do you get when your servers never fail? When you never lose a database?
One good way to stay proactive is to write up a quarterly report on what happened during the previous quarter and what is planned out for the next year. Works well for keeping your value on the radar.
Corporate work sucks. Rules, HR policies, having to go through dozens of managers to get something changed, and politics out the wazoo. Degrees and certifications and other paper mean more then actual work experience or know-how. Been there, done that, wasted 8 years.
I'm in the same boat as Gorobei. Jack of all trades means that my job never gets boring. There's always something new to tackle. If I need time off, I get it. (In fact, the less of a fuss you make about time off or "getting what's owed me", the easier it is to get extra time off.)
On the upside, I'm senior enough at the company that I don't get abused. I know what needs to be done, what is urgent, what is nice-to-have, and what is bullpuckey. The CEO listens to me, and we periodically re-arrange priorities.
Even better, I work remotely 3-4 days per week.
I'm hoping more for a quad-core CPU that is a 45W TDP and is under $100. All of the current quad-core CPUs out there are 95-125W (including Intel's) and are $130-$200.
I don't get such comparisons, for really low-end stuff like that wouldn't a P3 1 GHz be enough to?
For the past 2 years, I would consider the purchase of any single-core CPUs to be "penny-wise pound-foolish". Typically, you can get a dual-core CPU for only a few dollars more, and you end up with a much more responsive machine.
Our low-end work desktops have all been dual-core X2s, usually a 45W design where possible. The power savings alone from the energy efficient X2s pays for the extra CPU cost. The responsiveness of a dual-core system also pays off in improved productivity. Plus a 45W machine tends to be very quiet, even with the OEM heatsink/fan.
We switched to dual-core as soon as dual-core CPUs dropped below $200 (a few years ago now). At that point in time, AMD X2s were the only inexpensive dual-core choice, with the bonus that they were 64bit capable. So we got better performance and a bit of future-proofing.
Closer to $1.40 to $1.50 per watt for 24x7 year-round power draw here in the NE (where we pay $0.16 to $0.17 per kWh).
Very hit/miss.
I have a Canon PowerShot SD550. The 2GB cards work perfectly fine. The 4GB SD cards appear to work, but don't actually work (things go screwy once you exceed 2GB of data on the card).
It's not a new concept... labeling media goes all the way back to cassette tapes. (Eight tracks are before my time, were they writeable?)
Unfortunately regular SD or SDHC cards are at the lower end of useful label size. MicroSD and MiniSD are even worse. So about the best that you can do is some sort of 2-4 character code on each card, and use some other method to keep track of the contents.
Some useful card cases are:
CaseAce 3010 Cardsafe SD Card Holder In DVD Case Holds 9
Glossa Aluminum SD Card Case
The aluminum card cases hold 3 SD cards, are slim and sturdy, but are nearly impossible to find anymore (CyberGuys used to sell them).
how much do you trust your fellow employees?
Not very far.
Now is the time to make sure that your backup and disaster recovery planning is up to snuff and can deal with malicious intent.
(It doesn't cover data espionage, however... that's a whole different kettle of fish. But you won't have to worry about espionage if all of your systems have been hacked into the ground and your backups aren't any good.)
SVN 1.5 only came out late last year (sometime in the Fall).
Merge tracking and simplification of merges is still very rough around the edges. It's supposed to get better in 1.6 (no ETA). Better then it was by far in 1.4, but still a work in progress.
For a less technical set of users, I still feel that SVN's central repository model is best. For distributed developers with strong technical skills, there's a lot to be said for git. We're staying with SVN because that works better for us.
If anything, I find that the game offers slightly more options and flexibility these days. Mind you, that doesn't necessarily say much. Just that anything is more than zero ;)
Pre-Wrath:
- Regular instances in BC almost always required crowd-control. Some required specific forms of CC or else you'd wipe repeatedly. The only way you could get by without CC was to majorly out-gear the instance.
- Aggro was a constant issue that needed to be managed, both by the DPS folks and the tank.
- Healers regularly ran out of mana.
Post-Wrath:
- CC in all instances up until level 80 heroics is purely optional. All trash has such low health compared to the amount of DPS being dished out that everything dies within 10-15 seconds. So everyone just AoEs everything down. Trash mobs become a boring chore on the way to the next boss, because there's no challenge, just a need to beat the trash mobs down so we can get to the boss.
- Aggro is no longer an issue except in really rare cases. DPS classes can simply go all out, never having to watch their aggro. Tank classes do insane amounts of aggro (and damage). All of the various aggro-reduction abilities have become mostly useless. Probably boring as hell for the tanks now that they don't have to work at it.
- Since everything dies so fast, healers don't run out of mana on anything except boss fights.
It was fun for about the first week after 3.0.2 hit and every boss in the game had their health reduced by 30% while all the DPS classes got a huge boost. But since then, it's gotten more and more boring because there is minimal challenge on the way to 80. Everyone says "wait until you do level 80 heroics, then you'll be challenged". My question is why do we have to wait 10 levels in Northrend for a challenge? I could see making the level 70 dungeons easy, but by 75 to 80, there should be a very decent chance of a wipe.
We did Naxx-10 for the first time last night. Out of the 3 wings (9 bosses), only one of them gave us any difficult trouble (Heigan the Unclean). The trash mobs were all a joke. The gargoyles were the only mobs where we had to concentrate and actually have some sort of focus. Other bosses were felled either on the first try or within 3 tries as we learned their special abilities and phases. Nowhere near as interesting as Karazhan was, where you had to adopt different tactics for different areas of trash mobs.
Blizzard has really screwed up if they think "EZ Mode" is healthy over the long term. All EZ-Mode is going to do is ramp up the churn rate on their subscriber base as people get bored with walking into a dungeon, pressing the "I Win" button, and walking back out. The old adage that "things you have to work for are things that you'll value more" comes to mind.
Agreed. Tape makes a lot of sense for high-volume applications.
It's just being crowded out of the low-end market by ever larger and ever cheaper hard drive sizes. Tape costs would have to drop by about a factor of 4 (or more) to compete in the lower end of the market where 100 tapes is a lot.
(If I could backup 800GB for $10, that would be much more of a no-brainer decision. The cost-advantage would be high enough to pay for the expensive tape drive. And $50 LTO-4 tapes are a lot better then back when a lot of large-capacity tapes cost $100 each.)
All they had to do was pull one of the drives, replace it, and lock up the original off site. In a couple of hours the drives would have been mirrored again.
I can't recommend that method unless you're running triple-active RAID1 (which is simply mirroring across 3 disks instead of just 2 disks). With a triple-mirror, when you pull that 3rd drive to take it offsite, you're at least still running on a 2-drive mirror.
And it's still best if you shut down everything (especially the databases) before you pull that drive and pop in a new one.
Alternately, temporarily hook up a 3rd drive (USB/Firewire), extend the raid array to it, then disconnect it before trimming the array back again.
Even accepting your price that's a cost of about 12.7 cents per gigabyte and you can get 800GB native LTO-4 tapes for about $50, which comes out to about 6.3 cents per gigabyte.
/devfs, that backup scripts will fire when good backup target drives are hot-plugged into the system.)
Except that:
- You have to spend $2500-$3500 on the LTO-4 tape drive.
- Oh, better make it two. Just in case the first one breaks. Or make sure that you know two other people with LTO-4 capable tape drives. (On the upside, LTO is pretty stable, with multiple vendors, so getting a replacement drive is probably not hard.)
Until you've bought and used 50 tapes, those individual tapes are costing you $100 each due to the cost of the drive.
2.5" 500GB SATA drives that can be hot-plugged are now about $110. A SATA to USB style miniature case is $15. So for $125 times 3, a small business can get started with an off-site backup for $375. Or go with 1TB 3.5" drives for $100, plus about $30 for a caddy. And they can simply grow the solution by buying additional drives/caddies/enclosures. And until they get to the point where they are juggling a few dozen of these external drives, it's still less expensive then the tape solution.
And restoration is a whole lot easier with hard drives. *Every* computer out there can (assuming the proper software) read from hard drives without the need for expensive tape drives. That ubiquity is worth an awful lot when it comes to disaster recovery.
Yeah, you can spin all sorts of numbers about how tape saves you more money over the long term. But in small / medium business - the ability to avoid a large cash outlay for an expensive tape drive generally wins.
(My preferred backup solution for small/medium companies is to 1. Have a in-house target server where backups get written to. 2. Shove daily updates of key information offsite over the wire every night. 3. Get a hard disk offsite weekly. All of which is generally a lot more then most companies manage to do. You can even script in Linux, with
HTC Touch Pro cell phones are full VGA (640x480), but the principle applies.
Of course, any text size smaller then about 14pt is absolutely unreadable. And people who specify font sizes in pixels (i.e. 6px) deserve their own special level of hell.
Pixels are REALLY tiny on a display of that size... fortunately it runs Opera where you can double-click on the screen area to get it to zoom in.
LCDs interpolate when not running in their native res, which isn't good enough
NVIDIA drivers give you a few options on Windows. You can choose to:
- scale the resolution up to fill the entire screen
- scale the resolution up, but keep the aspect ratio intact (black bars on the sides for 4:3 content)
- don't scale at all (centers the image)
Of course, that 2nd CRT display comes in handy for other reasons...