Not only can you utilize more physical memory, but you can map a larger amount of virtual memory, which is particularly going to be useful once people have 2-3GB of RAM and video cards have aperture sizes into the 512-768MB range, you start running out of 32-bit addressable space.
That's one of the better reasons to make the switch to 64bits. Back when the 32bit stuff came out, 4GB of memory seemed to be outrageous (HD sizes were still in the 100-400MB range). Now we're to the point where it's easy to buy a PC with 2GB of RAM (and not all that expensive either).
What impresses me about the Opteron line is the design choice that AMD made. Making sure that 32bit apps run as fast on 64bit Opertons as they would on today's common 32bit CPUs. Which means that choosing to use Opteron isn't a devil's bargain where you have to choose either future (64bit) or performance. Very smart choice by AMD, leaving Intel to play catch-up for a bit.
A quick Google search shows the Blu-ray 1x spec at 36Mbps... So it's pretty fast. At that speed, the article I read said it would take about 2.5 hours to fill a single sided Blu-ray disk. I'm too lazy to double check the math.:)
OTOH, existing 8x DVD writers can write at 88Mbps. (Assuming DVD 1x is 11Mbps, which is probably a bit high.)
Which makes Blu-Ray's 36Mbps seem slow to me. They need to get the speed up by about 4x in order for it to be considered speedy in my book.
I bought a laptop about 3 years ago for around $2000 and at the time it was an average laptop. Look at what you can get for $2000 today, it's usually top of the line stuff, mine would probably cost around $1000 now.
Back when I bought my first laptop, I paid around $3600 and it wasn't even top of the line at the time. (Usual price for a business laptop was closer to $4000 back in 1992/1994 timeframe.)
Four years ago, a top of the line laptop was around $3000. You could pay more, but for $3k you could easily get a desktop replacement with lots of memory and a decent hard drive.
Prices have now fallen (as you pointed out) to around $2200 for a top-of-the-line laptop. That includes a higher res screen (1400x1050), 1GB of memory instead of 256/512, 60GB HD instead of 30GB, plus a DVD-writer.
Just wish I could get a 127ppi *desktop* LCD. All of the ones I've seen are either 96-100ppi or you have to pay $$$$ for a 200ppi LCD. A 1600x1200 127ppi LCD would be sweet (or even better, a 1920x1080 LCD at that density).
That's because everything resides in DLLs. Go back and look at the size of the old IE4 downloads. IE 4.01 was a 55MB download. IE 5.5sp1 was a 61MB download.
Now figure that IE6 is even more bloated, except that most of it came with the operating system so you don't see it.
how could you post to slashdot without knowing how to type?
You're kidding, right?
Why do you think there are so many spelling nazi's here if there wasn't a dire need?
(I'm a traditional touch typist, learned on a *manual* typewriter. I've learned not to hit the keys so hard over the ages, but I do greatly prefer a full mechanical keyboard over a mushy Dell QuietKey.)
My experience with email is that people use it as an IM system when one is not available, so do yourself a favor, if you can, and add a jabber server and some nice IM clients for internal use
That's an important point, especially if you have users in multiple locations or floors. An internal IM system is pretty important, even if you have a fast e-mail system. IM is for simple stuff like a group discussion or quick "synchro" conversations (do you have X checked out, server Y is being rebooted, let's do lunch at Z today).
We ended up using WiredRed's e/pop software. It's not free, but it does do encryption. There is a java client that you can use on non-Windows machines. But the biggest reason we're sticking with e/pop is the remote control capability which comes in handy for remote support.
I advise using password generating tools to create truly random passwords.
The easiest is to keep a random word generator (or a dictionary) handy. Trying to come up with random words off the top of your head is an iffy proposition. You probably only use a few thousand words regularly, yet a good word file with have a few hundred thousand entries.
For example, here's an output from a very basic generator:
More on-topic now - that's a good point. I can get up to nearly 30MB/s (or VERY roughly 300Mbps) on a 5400RPM drive reading directly (hdparm -t), so assume even under ideal conditions and a 10,000RPM drive you'll get less than 60MB/s or so (or again VERY roughly 600Mbps = 60% theoretical maximum for Gigabit).
Bit more complicated then that... higher areal densities on the platters will allow for higher sustained transfer rates. So an older 40GB/platter drive will have a slower transfer rate then the 80GB/platter drive with the same rpm. (The 80GB/platter will feed more data past the head per unit of time.)
My 5400rpm 160GB drives and 5400rpm 300GB drives both seem to be right around 30MB/s. That number is based on what was reported while creating the RAID arrays in Linux (also seen when copying files from drive to drive in WindowsXP). With 7200rpm 200/250GB drives, I've seen rates of 45MB/s when transferring between two drives on the same system.
At a guess, we'll need to see 200GB/platter in a 3.5" package in order to hit 100MB/s. IIRC, we're either at 80GB/platter or 100GB/platter today.
On a motherboard that is 2 years old, copying from drive to drive (large video files), I'll see rates of anywhere from 40-45 MB/s. That's with 7200rpm SATA/PATA drives. Copying from a 5400rpm 150GB drive is a solid 30MB/s.
Shooting across the gigabit LAN, those rates drop to 15-20 MB/s. That's with cheap, consumer level hardware on both sides, copying to/from a Windows share.
Still quite acceptable, since the switch could easily be handling multiple streams between different boxes. At least I'm not stuck with the old 8MB/s limit of a 10/100 network.
I'm a die-hard MoO fan, from 1 and 2. Never played 3. Couldn't afford it, tried the demo, didn't like it.
You didn't miss anything.
They took everything that was wrong with the MOO2 interface and made it worse.
(Actually, MOO3 was a great game as designed. The key was the starlanes which re-introduced "terrain" as a key strategic element. Unfortunately, the user-interface was absolutely horrid. Especially comparing the ship design screen between MOO2 and MOO3. MOO2's ship design screen was almost perfect, a nice blend of form and function where all the information you needed was right there.)
One of these day's I'll pickup GalCiv again... played it back in the OS/2 days and it didn't really rock-my-socks as much as MOO2 did.
So what's the take home lesson here? Don't buy stock that will go down? Why didn't I think of that.
The lesson is called "dollar cost averaging". Which means that you put the same dollar amount into a range of stocks/funds once per month or once per week. Sometimes you'll buy low and make out well, other times you'll buy high and not do so well. But over the long run it all averages out and you'll make decent returns.
It also removes the emotional aspect from investing. No jumping on "hot" stocks because you got swayed or staying in a "dying" stock due to an irrational belief that it's about to turn the corner.
Had one when I was a young-un, probably circa 1980 or a bit before. We got ours because our aunt out on the west coast had gotten fed up with it. Ours was dark grey, but without the trailer. The front "laser" was a 6V flashlight bulb and it used a simple plastic membrane keypad on top. Only the middle wheel out of each side's set of 3 was a drive wheel. With a rubber o-ring stretched around a groove in the center to help with traction.
I know I spent *hours* setting up and re-running instructions to make it go from room to room. One of the big problems was the uneven shag carpet that we had (odd-shaped random pattern of long and short fibers). It would really decrease the accuracy of any turn commands that you were lucky to even be +/- 30 degrees on a turn. It didn't do much better on linoleum flooring either. Distance was also tricky.
That's probably why I'm a programmer geek today, too many hours trying to figure out how to program the real-life version of LOGO. Eventually, I took the device apart, probably trying to figure out improvements or fix it. So mine no longer exists in any shape or form.
All-in-all it was a very educational toy as you learned about how plans tended to get trashed by reality and how to improve accuracy (multiple forward commands with small stops were less accurate then a single forward command). Not to mention learning how to devise a plan, see it carried out and then revising it and trying again.
I think it'd be an interesting slashdot poll to see how often people have upgraded their CPU and not their motherboard. I've not yet done it myself other than once replacing a dodgy AMD Duron.
I did it backwards... upgraded my motherboard but not the CPU or RAM. (Switching from the craptastic VIA KT266 chipset to a newer KT400 chipset.) So better have a poll option for that (or is that the Cowboy Neal option?).
One of these days I'll buy something faster then an AthlonXP 1800+ to put in it. At least it has 1.5GB of RAM.
Just for shit and giggles, I highlighted that and used the "search web for" function in Firefox. 839 hits...please don't trash the AMD platform just because of one shoddy chipset manufacturer...
I've owned one of the crappy VIA boards, it was the KT266 chipset (IIRC) that was the real red-headed step child of QA testing. Search for "via pci latency" and the like for some fun horror stories. The motherboard would stutter under load which made it completely unusable (especially since I was trying to do video capture).
I have a KT400 motherboard now (which is what I replaced the old board with). Other then an issue where using the built-in PATA RAID ports causes data corruption (driver issue with Win2000 server) the chipset seems to be stable. (It at least performs well under load.)
My next board will probably be a VIA K8T800 chipset (I've heard iffy things about nForce3 but haven't had time to track down the details or validity).
Of course, there's always dual CPU motherboards...
That's the roadmap that I'm going to be following for any workstation that requires it. Downside is the cost, but from all reports it makes the operating system and everything much more responsive. Still, you know what they say, the only thing faster then X is *two* of X.
I'm hoping for dual-core as well... but unless they break the 90nm issue, are they really going to have room on the die for a 2nd core?
Firewire IS a failure as a general computer interface standard. It ships on what, 10% of all computers?
Out of the (6) boxes in my personal office...
Four of them support firewire. The two that don't are older motherboards from 1999 and 2001 (or thereabouts).
Almost every x86 laptop that I've looked at has FireWire. Sony's laptops have had firewire for quite a while (ever since DVD creation and DV hit the market).
People in the x86 world are just more familiar with USB 2.0 when it comes to external devices. DV cameras are a big driver in the home market.
At 128kbps you're looking at what, 1 meg a minute, roughly? So a CD gives you 600-700 minutes of music, or between 10 and 12 hours. My days are longer than that.:(
I mostly use CD/MP3 in the car. That's probably where it has it's biggest advantage. As an example, I'll have one disc for light pop, another full of my favorite western music, another disc for celtic, one with heavy metal, etc. Changing CDs where you keep the CDs in a visor clip is easy, requires only a brief glance to determine what to put in next (or to line up the CD).
No need to put your entire music collection on CD/MP3, just enough to survive the daily commute with a bit of variation. And it doesn't take long to create new discs for long trips. There's also lots of places to store CDs in a car (I keep mine in the door pockets with frequently used ones in a visor holder).
CD/MP3 is also a good fit for boomboxes (such as taking a trip down to the beach house, or for use in a bedroom) where you need built-in speakers. Discs are cheap and in those situations you'd have to bring along external speakers to hook up to the iPod anyway. Playing an MP3 CD is no more difficult then a regular CD, there's just more tracks to choose from. Boomboxes also solve the battery issue by either plugging them into the AC socket or by being able to pack along extra D cells.
However, for portable use, where your primary method of listening is via headphones... iPod and flash-based players are very tough to beat. Even my mini-CD MP3 player is large and bulk compared to the mini-iPod or regular iPod. The only advantage of the mini-CD player (holds 210MB per disc, discs easily fit in a pocket) was cost. The player was $40, discs are
Note the seemed like a good fit. My next portable player will be either flash based or an iPod. Probably an iPod because by all accounts, they've done it right.
I still can't fit all of my music onto a single iPod... so I'm still going to have to pick and choose.
And the specific case that you cite is one where Sony was forced to follow existing standards/formats instead of making up their own proprietary standards. In those cases, Sony does make good products which are competitive (if usually slightly overpriced).
The problem lies in that Sony likes to lock the user as much as possible into their own weird, twisted, proprietary formats. It's an extremely bad corporate attitude that causes them a lot of self-damage in the market. Even after a few decades (MD, MemoryStick, Beta), Sony management still hasn't seen the error of their ways.
Personally, I use MP3 CDs in the car. Have a RCA boombox that reads MP3 CDs, a mini regular CD/MP3 portable and a mini-CD/MP3 pocket sized device. I've been a big booster of MP3/CD players around here. They're dead simple to use, great for the car when you can change a CD by feel with only a quick glance to see what CD you're going to put in next.
I still plan on purchasing an iPod sometime in the next year. I just haven't decided if I want the mini or the 4th gen regular. The size advantage of the small iPod over even a mini-CD player is huge, and from all accounts, Apple got the user interface right.
The problem was known 4 years ago, but it was marked confidential.
Four years ago... The Mozilla Foundation is only what, a year old? So that means that four years ago it would've been under the control of AOL? Who was in control of the group at that point?
What were the defined procedures for handling security-related bugs at that point in time? Did someone just forget about this one? (Dig through Bugzilla sometime... there are lots of bugs outstanding that look like they've simply been lost in the shuffle.)
I have one major problem with FreeBSD - its mascot.
So you only buy your milk from christian cows? (Go listen to some Steve Taylor...) It's a silly cartoon drawing, based on the fact that services in unix have traditionally been called daemons (so if the mascot bothers you, you shouldn't run any flavor of unix, solaris, linux or Mac OS X either).
Not only does it represent the source of evil, but if my family and/or friends were to catch wind of my using anything that uses an image of the Devil to represent itself with, I'd be at the center of controversy, if not out-right ostracized.
You have my pity, you sad sad sad little Christian. If you have to live a perfect life in order to avoid "controversy" in your church, then I think there are some serious issues in that congregation such as gossip, being judgemental, and playing the holier then thou game. I'll give you 50-50 odds, without even meeting your kids, that at least one of them will shake the dust off the hem of their robe as soon as they get out on their own and never come back to that church of their own free will.
(Yeah, I'm one of those statistics. My home church was like that and so were the folks. Everyone so concerned about appearances that you had to cross the street instead of walking past the local bar on the off-chance that someone would think you had gone in. There's a big difference between being religious and actually practicing the faith...)
This feels like a troll duplicate from a year or two ago.
One year ago, SCO had already filed their anti-linux suits and were being widely reviled for it here on Slashdot. In fact, I'm pretty sure it started in April 2003.
7200rpm drives (SATA/IDE) are anywhere from 10W to 13W.
Not all manuf's list all power usages (the three keys are startup, idle, and seeking/active). The only general rule of thumb is that higher rpm drives take more power (in addition to throwing off a lot more waste heat).
Anyone have any ideas on how to back up 1TB in a home environment? i.e., not $3000 tape drives & $200 tapes
I don't have 1TB, but close enough.
Box 1: 250GB RAID1 plus a pair of 250GB scratch disks. Data on the scratch disks doesn't matter if it gets lost or I have to restore it from box #2.
Box 2: 250GB RAID1 for the O/S and critical files, then (4) 300GB 5400rpm drives for backups. Each 300GB drive is it's own partition (not raid'd together or anything). The 300GB drives are used for backing up data for the scratch drives on box #1 and for backing up data for the RAID array on box #2.
I also have a removable drive bay on one of the workstations which backs up the critical data daily, swapping the drive off-site weekly.
And the really critical stuff gets snapshot'd to DVD periodically and stored at the bank. Along with PAR2 recovery data in case the discs start to go bad (see QuickPar).
The key is to separate your data into tiers (critical, replaceable, don't care) and do backups accordingly. It's always possible to re-rip any commercial media (CDs, DVDs), so I don't bother protecting that data against worse case situations.
I I havn't calculated the per MB cost of all the large sizes. someone with more time please do this.
Last I looked, 160GB and 200GB drives are down around $0.50-$0.60/GB. The 250GB and 300GB drives are still up around $0.70/GB.
It still all works out to around $2/GB for even SATA/IDE RAID storage. It's not the drives, it's the box and the controller. There are external enclosures that hold 8 drives and output a SCSI connector, but they're also around $2000.
$400 RAID card, $1200 for the case, CPU, ram, motherboard, DVD, cables, power-supply, misc parts. Then tack on the (8) 200GB drives at $125 each. Cost is around $2600, assuming RAID5 + hot-spare, net drive size is rougly 1.2TB.
$2.12/GB.
IIRC, last I looked, SCSI brought that number up to around $4.50-$5.00 per gigabyte.
Slashdot Mirror
Not only can you utilize more physical memory, but you can map a larger amount of virtual memory, which is particularly going to be useful once people have 2-3GB of RAM and video cards have aperture sizes into the 512-768MB range, you start running out of 32-bit addressable space.
That's one of the better reasons to make the switch to 64bits. Back when the 32bit stuff came out, 4GB of memory seemed to be outrageous (HD sizes were still in the 100-400MB range). Now we're to the point where it's easy to buy a PC with 2GB of RAM (and not all that expensive either).
What impresses me about the Opteron line is the design choice that AMD made. Making sure that 32bit apps run as fast on 64bit Opertons as they would on today's common 32bit CPUs. Which means that choosing to use Opteron isn't a devil's bargain where you have to choose either future (64bit) or performance. Very smart choice by AMD, leaving Intel to play catch-up for a bit.
A quick Google search shows the Blu-ray 1x spec at 36Mbps... So it's pretty fast. At that speed, the article I read said it would take about 2.5 hours to fill a single sided Blu-ray disk. I'm too lazy to double check the math. :)
OTOH, existing 8x DVD writers can write at 88Mbps. (Assuming DVD 1x is 11Mbps, which is probably a bit high.)
Which makes Blu-Ray's 36Mbps seem slow to me. They need to get the speed up by about 4x in order for it to be considered speedy in my book.
I bought a laptop about 3 years ago for around $2000 and at the time it was an average laptop. Look at what you can get for $2000 today, it's usually top of the line stuff, mine would probably cost around $1000 now.
Back when I bought my first laptop, I paid around $3600 and it wasn't even top of the line at the time. (Usual price for a business laptop was closer to $4000 back in 1992/1994 timeframe.)
Four years ago, a top of the line laptop was around $3000. You could pay more, but for $3k you could easily get a desktop replacement with lots of memory and a decent hard drive.
Prices have now fallen (as you pointed out) to around $2200 for a top-of-the-line laptop. That includes a higher res screen (1400x1050), 1GB of memory instead of 256/512, 60GB HD instead of 30GB, plus a DVD-writer.
Just wish I could get a 127ppi *desktop* LCD. All of the ones I've seen are either 96-100ppi or you have to pay $$$$ for a 200ppi LCD. A 1600x1200 127ppi LCD would be sweet (or even better, a 1920x1080 LCD at that density).
IE has an executable of a few KB (WinXP).
That's because everything resides in DLLs. Go back and look at the size of the old IE4 downloads. IE 4.01 was a 55MB download. IE 5.5sp1 was a 61MB download.
Now figure that IE6 is even more bloated, except that most of it came with the operating system so you don't see it.
how could you post to slashdot without knowing how to type?
You're kidding, right?
Why do you think there are so many spelling nazi's here if there wasn't a dire need?
(I'm a traditional touch typist, learned on a *manual* typewriter. I've learned not to hit the keys so hard over the ages, but I do greatly prefer a full mechanical keyboard over a mushy Dell QuietKey.)
My experience with email is that people use it as an IM system when one is not available, so do yourself a favor, if you can, and add a jabber server and some nice IM clients for internal use
That's an important point, especially if you have users in multiple locations or floors. An internal IM system is pretty important, even if you have a fast e-mail system. IM is for simple stuff like a group discussion or quick "synchro" conversations (do you have X checked out, server Y is being rebooted, let's do lunch at Z today).
We ended up using WiredRed's e/pop software. It's not free, but it does do encryption. There is a java client that you can use on non-Windows machines. But the biggest reason we're sticking with e/pop is the remote control capability which comes in handy for remote support.
I advise using password generating tools to create truly random passwords.
The easiest is to keep a random word generator (or a dictionary) handy. Trying to come up with random words off the top of your head is an iffy proposition. You probably only use a few thousand words regularly, yet a good word file with have a few hundred thousand entries.
For example, here's an output from a very basic generator:
innatelyagouti0
unpredacious!rah
snowwhiterawly0
betulaceae35fave0
pandani&aerocyst>
schorlous-mondos0
exteriorise!derig0
trophied+unthrust0
essoin+coblenz*
crottels&autumn
ungarter3trudger0
tragicaster75saws=
inemuloussnubber%
synapsed11steichen0
permalloys*bergh0
vakia$czaritza?
upstater*slubs/
predictablefeif0
unbuskkataplexy0
outcrossing!ilona!
And it's kind of amusing that "predictable" is in there... (I swear I didn't put it there on purpose).
Make sure you have jumbo frames enabled on all the machines. Note that you need OS X 10.2.4 or newer on your Mac to use the 9K frames.
What happens if jumbo frames is enabled on the server but not on one of the workstations?
More on-topic now - that's a good point. I can get up to nearly 30MB/s (or VERY roughly 300Mbps) on a 5400RPM drive reading directly (hdparm -t), so assume even under ideal conditions and a 10,000RPM drive you'll get less than 60MB/s or so (or again VERY roughly 600Mbps = 60% theoretical maximum for Gigabit).
Bit more complicated then that... higher areal densities on the platters will allow for higher sustained transfer rates. So an older 40GB/platter drive will have a slower transfer rate then the 80GB/platter drive with the same rpm. (The 80GB/platter will feed more data past the head per unit of time.)
My 5400rpm 160GB drives and 5400rpm 300GB drives both seem to be right around 30MB/s. That number is based on what was reported while creating the RAID arrays in Linux (also seen when copying files from drive to drive in WindowsXP). With 7200rpm 200/250GB drives, I've seen rates of 45MB/s when transferring between two drives on the same system.
At a guess, we'll need to see 200GB/platter in a 3.5" package in order to hit 100MB/s. IIRC, we're either at 80GB/platter or 100GB/platter today.
On a motherboard that is 2 years old, copying from drive to drive (large video files), I'll see rates of anywhere from 40-45 MB/s. That's with 7200rpm SATA/PATA drives. Copying from a 5400rpm 150GB drive is a solid 30MB/s.
Shooting across the gigabit LAN, those rates drop to 15-20 MB/s. That's with cheap, consumer level hardware on both sides, copying to/from a Windows share.
Still quite acceptable, since the switch could easily be handling multiple streams between different boxes. At least I'm not stuck with the old 8MB/s limit of a 10/100 network.
I'm a die-hard MoO fan, from 1 and 2. Never played 3. Couldn't afford it, tried the demo, didn't like it.
You didn't miss anything.
They took everything that was wrong with the MOO2 interface and made it worse.
(Actually, MOO3 was a great game as designed. The key was the starlanes which re-introduced "terrain" as a key strategic element. Unfortunately, the user-interface was absolutely horrid. Especially comparing the ship design screen between MOO2 and MOO3. MOO2's ship design screen was almost perfect, a nice blend of form and function where all the information you needed was right there.)
One of these day's I'll pickup GalCiv again... played it back in the OS/2 days and it didn't really rock-my-socks as much as MOO2 did.
So what's the take home lesson here? Don't buy stock that will go down? Why didn't I think of that.
The lesson is called "dollar cost averaging". Which means that you put the same dollar amount into a range of stocks/funds once per month or once per week. Sometimes you'll buy low and make out well, other times you'll buy high and not do so well. But over the long run it all averages out and you'll make decent returns.
It also removes the emotional aspect from investing. No jumping on "hot" stocks because you got swayed or staying in a "dying" stock due to an irrational belief that it's about to turn the corner.
Haha, I forgot about that (well kinda).
Had one when I was a young-un, probably circa 1980 or a bit before. We got ours because our aunt out on the west coast had gotten fed up with it. Ours was dark grey, but without the trailer. The front "laser" was a 6V flashlight bulb and it used a simple plastic membrane keypad on top. Only the middle wheel out of each side's set of 3 was a drive wheel. With a rubber o-ring stretched around a groove in the center to help with traction.
I know I spent *hours* setting up and re-running instructions to make it go from room to room. One of the big problems was the uneven shag carpet that we had (odd-shaped random pattern of long and short fibers). It would really decrease the accuracy of any turn commands that you were lucky to even be +/- 30 degrees on a turn. It didn't do much better on linoleum flooring either. Distance was also tricky.
That's probably why I'm a programmer geek today, too many hours trying to figure out how to program the real-life version of LOGO. Eventually, I took the device apart, probably trying to figure out improvements or fix it. So mine no longer exists in any shape or form.
All-in-all it was a very educational toy as you learned about how plans tended to get trashed by reality and how to improve accuracy (multiple forward commands with small stops were less accurate then a single forward command). Not to mention learning how to devise a plan, see it carried out and then revising it and trying again.
I think it'd be an interesting slashdot poll to see how often people have upgraded their CPU and not their motherboard. I've not yet done it myself other than once replacing a dodgy AMD Duron.
I did it backwards... upgraded my motherboard but not the CPU or RAM. (Switching from the craptastic VIA KT266 chipset to a newer KT400 chipset.) So better have a poll option for that (or is that the Cowboy Neal option?).
One of these days I'll buy something faster then an AthlonXP 1800+ to put in it. At least it has 1.5GB of RAM.
Just for shit and giggles, I highlighted that and used the "search web for" function in Firefox. 839 hits...please don't trash the AMD platform just because of one shoddy chipset manufacturer...
I've owned one of the crappy VIA boards, it was the KT266 chipset (IIRC) that was the real red-headed step child of QA testing. Search for "via pci latency" and the like for some fun horror stories. The motherboard would stutter under load which made it completely unusable (especially since I was trying to do video capture).
I have a KT400 motherboard now (which is what I replaced the old board with). Other then an issue where using the built-in PATA RAID ports causes data corruption (driver issue with Win2000 server) the chipset seems to be stable. (It at least performs well under load.)
My next board will probably be a VIA K8T800 chipset (I've heard iffy things about nForce3 but haven't had time to track down the details or validity).
Of course, there's always dual CPU motherboards...
That's the roadmap that I'm going to be following for any workstation that requires it. Downside is the cost, but from all reports it makes the operating system and everything much more responsive. Still, you know what they say, the only thing faster then X is *two* of X.
I'm hoping for dual-core as well... but unless they break the 90nm issue, are they really going to have room on the die for a 2nd core?
Firewire IS a failure as a general computer interface standard. It ships on what, 10% of all computers?
Out of the (6) boxes in my personal office...
Four of them support firewire. The two that don't are older motherboards from 1999 and 2001 (or thereabouts).
Almost every x86 laptop that I've looked at has FireWire. Sony's laptops have had firewire for quite a while (ever since DVD creation and DV hit the market).
People in the x86 world are just more familiar with USB 2.0 when it comes to external devices. DV cameras are a big driver in the home market.
At 128kbps you're looking at what, 1 meg a minute, roughly? So a CD gives you 600-700 minutes of music, or between 10 and 12 hours. My days are longer than that. :(
I mostly use CD/MP3 in the car. That's probably where it has it's biggest advantage. As an example, I'll have one disc for light pop, another full of my favorite western music, another disc for celtic, one with heavy metal, etc. Changing CDs where you keep the CDs in a visor clip is easy, requires only a brief glance to determine what to put in next (or to line up the CD).
No need to put your entire music collection on CD/MP3, just enough to survive the daily commute with a bit of variation. And it doesn't take long to create new discs for long trips. There's also lots of places to store CDs in a car (I keep mine in the door pockets with frequently used ones in a visor holder).
CD/MP3 is also a good fit for boomboxes (such as taking a trip down to the beach house, or for use in a bedroom) where you need built-in speakers. Discs are cheap and in those situations you'd have to bring along external speakers to hook up to the iPod anyway. Playing an MP3 CD is no more difficult then a regular CD, there's just more tracks to choose from. Boomboxes also solve the battery issue by either plugging them into the AC socket or by being able to pack along extra D cells.
However, for portable use, where your primary method of listening is via headphones... iPod and flash-based players are very tough to beat. Even my mini-CD MP3 player is large and bulk compared to the mini-iPod or regular iPod. The only advantage of the mini-CD player (holds 210MB per disc, discs easily fit in a pocket) was cost. The player was $40, discs are
Note the seemed like a good fit. My next portable player will be either flash based or an iPod. Probably an iPod because by all accounts, they've done it right.
I still can't fit all of my music onto a single iPod... so I'm still going to have to pick and choose.
Not everything Sony produces is outright bad.
And the specific case that you cite is one where Sony was forced to follow existing standards/formats instead of making up their own proprietary standards. In those cases, Sony does make good products which are competitive (if usually slightly overpriced).
The problem lies in that Sony likes to lock the user as much as possible into their own weird, twisted, proprietary formats. It's an extremely bad corporate attitude that causes them a lot of self-damage in the market. Even after a few decades (MD, MemoryStick, Beta), Sony management still hasn't seen the error of their ways.
Personally, I use MP3 CDs in the car. Have a RCA boombox that reads MP3 CDs, a mini regular CD/MP3 portable and a mini-CD/MP3 pocket sized device. I've been a big booster of MP3/CD players around here. They're dead simple to use, great for the car when you can change a CD by feel with only a quick glance to see what CD you're going to put in next.
I still plan on purchasing an iPod sometime in the next year. I just haven't decided if I want the mini or the 4th gen regular. The size advantage of the small iPod over even a mini-CD player is huge, and from all accounts, Apple got the user interface right.
The problem was known 4 years ago, but it was marked confidential.
Four years ago... The Mozilla Foundation is only what, a year old? So that means that four years ago it would've been under the control of AOL? Who was in control of the group at that point?
What were the defined procedures for handling security-related bugs at that point in time? Did someone just forget about this one? (Dig through Bugzilla sometime... there are lots of bugs outstanding that look like they've simply been lost in the shuffle.)
I have one major problem with FreeBSD - its mascot.
So you only buy your milk from christian cows? (Go listen to some Steve Taylor...) It's a silly cartoon drawing, based on the fact that services in unix have traditionally been called daemons (so if the mascot bothers you, you shouldn't run any flavor of unix, solaris, linux or Mac OS X either).
Not only does it represent the source of evil, but if my family and/or friends were to catch wind of my using anything that uses an image of the Devil to represent itself with, I'd be at the center of controversy, if not out-right ostracized.
You have my pity, you sad sad sad little Christian. If you have to live a perfect life in order to avoid "controversy" in your church, then I think there are some serious issues in that congregation such as gossip, being judgemental, and playing the holier then thou game. I'll give you 50-50 odds, without even meeting your kids, that at least one of them will shake the dust off the hem of their robe as soon as they get out on their own and never come back to that church of their own free will.
(Yeah, I'm one of those statistics. My home church was like that and so were the folks. Everyone so concerned about appearances that you had to cross the street instead of walking past the local bar on the off-chance that someone would think you had gone in. There's a big difference between being religious and actually practicing the faith...)
Not a year ago everyone here got hard for SCO.
This feels like a troll duplicate from a year or two ago.
One year ago, SCO had already filed their anti-linux suits and were being widely reviled for it here on Slashdot. In fact, I'm pretty sure it started in April 2003.
5400rpm drives are around 8-11W when active.
7200rpm drives (SATA/IDE) are anywhere from 10W to 13W.
Not all manuf's list all power usages (the three keys are startup, idle, and seeking/active). The only general rule of thumb is that higher rpm drives take more power (in addition to throwing off a lot more waste heat).
Anyone have any ideas on how to back up 1TB in a home environment? i.e., not $3000 tape drives & $200 tapes
I don't have 1TB, but close enough.
Box 1: 250GB RAID1 plus a pair of 250GB scratch disks. Data on the scratch disks doesn't matter if it gets lost or I have to restore it from box #2.
Box 2: 250GB RAID1 for the O/S and critical files, then (4) 300GB 5400rpm drives for backups. Each 300GB drive is it's own partition (not raid'd together or anything). The 300GB drives are used for backing up data for the scratch drives on box #1 and for backing up data for the RAID array on box #2.
I also have a removable drive bay on one of the workstations which backs up the critical data daily, swapping the drive off-site weekly.
And the really critical stuff gets snapshot'd to DVD periodically and stored at the bank. Along with PAR2 recovery data in case the discs start to go bad (see QuickPar).
The key is to separate your data into tiers (critical, replaceable, don't care) and do backups accordingly. It's always possible to re-rip any commercial media (CDs, DVDs), so I don't bother protecting that data against worse case situations.
I I havn't calculated the per MB cost of all the large sizes. someone with more time please do this.
Last I looked, 160GB and 200GB drives are down around $0.50-$0.60/GB. The 250GB and 300GB drives are still up around $0.70/GB.
It still all works out to around $2/GB for even SATA/IDE RAID storage. It's not the drives, it's the box and the controller. There are external enclosures that hold 8 drives and output a SCSI connector, but they're also around $2000.
$400 RAID card, $1200 for the case, CPU, ram, motherboard, DVD, cables, power-supply, misc parts. Then tack on the (8) 200GB drives at $125 each. Cost is around $2600, assuming RAID5 + hot-spare, net drive size is rougly 1.2TB.
$2.12/GB.
IIRC, last I looked, SCSI brought that number up to around $4.50-$5.00 per gigabyte.