Also, more channels wouldn't give a reciever any more reason to clip. Each channel is a seperate amp.
That's not completely true. TFA doesn't really explain the point he's trying to make in this area. Your typical home theater receiver has a power rating for each channel that's usually based on the transistors used. There's also a maximum power rating that comes from how much current the power supply can produce. If you have something with multiple channels being driven at once, there are plenty of receivers where the maximum power you can get is far below the rated power per channel because of the power supply bottleneck.
The much more useful specification then is the maximum power with all channels driven, which some receivers don't even mention because it would show the limitations of their design in an embarrassing light. Obviously a 7.1 receiver is going to be handicapped in that spec when compared to a similar 5.1 receiver because it has more channels to drive.
Regardless, your original point that a typical home theater is usually only blowing out a watt or ten of power under typical operating conditions is still correct.
The ultimate Q-Sound demo disc is also from Roger Waters. On a stereo system that's aligned just right (with you exactly at the center), "Amused to Death" has sounds coming from all over the place; I'll never forget how spooked out I got when first hearing the "The Ballad of Bill Hubbard" for the first time, jumping when the first sound came from behind me.
The Roger Waters album mentioned by the parent post here, "Pros and Cons of Hitchhiking" was recorded with holophonics, a variant of the binaural techniques used to make realistic recordings for headphone listening. That's one of the reasons it sounds particularly good with the low-budget matrix approach.
This approach to generating rear ambience from recordings is generally credited to amplifier designer David Hafler, who popularized it in the 70's. For a clear diagram of how to wire this up that may be easier to understand than the text here see http://sound.westhost.com/project18.htm It's also possible to insert a fixed or variable resistor to adjust the volume of the rear speakers relative to the front; see the "Can I play binaural recordings through loudspeakers?" section of http://www.headwize.com/faqs.htm for a sample there. Finally, you can even derive a passive center channel if you're really hardcore; I don't really like the look of the diagrams at http://kantack.com/surround/surround2.html but it covers all these approaches.
Dynaco sold a little box called the QD-1 that simplified the wiring of these during the original Quad craze. They re-introduced a series 2 version as a cheap Dolby Pro-Logic decoder during the beginning of the home theater craze. The newer version is reviewed at http://kantack.com/surround/surround4.html , which is a pretty spot on commentary about the limitations of this type of decoding. Both models are floating around ebay for not much money.
However, note that the effectiveness of this circuit presumes that your amplifier has a shared ground between the left and right channel circuits. While this is generally true, there are amplifiers (like any balanced design) where the ground of the left channel and that of the right are unrelated. Hooking up this circuit to such an amplifier will either a) shut it down, b) trip a fuse, or c) blow the output transistors, depending on the robustness of the design. Be very careful you know what you're playing with here. If you don't know enough about electronics to check if your amplifier channels have a common ground or not, you probably shouldn't be playing around with this circuit.
Dead Can Dance is close to a best case for this approach; really well produced studio work with lots of hard panned instruments sitting in one channel or the other pop into surround this way, and minimal vocals in the center. The best demo I ever found for how effective this circuit could be was the instrumental title track to the Alan Parsons Project "I Robot", which was downright freaky in pseudo-surround.
Because I can do that pretty easily and more accurately than software.
Yes, but you sitting on your PC typing "confidential" can draw attention to what you're doing that the iPod slurping doesn't. Also, records of what you've did can be left behind in various history files on the PC, especially if you're caught in the middle and don't have a chance to erase them.
Clearly their concerns here are justified, because that's how I spend my time. Just today, I beat down an old lady with a baseball bat, jacked two cars, blew away a cop with a shotgun, and fired rockets at a passing helicopter. Hiring a ho and roughing her up is the only way I've learned that gives a day like that a "happy ending".
I'm reading comments here while also sitting at two tables playing Texas Hold'Em on-line. Why distinguish between Internet and gambling addictions when you can have both?
HP ZD8110 - 3Ghz P4 HT, Radeon X600 PCI-E 128MB, 2GB Ram, 17inch Widescreen, Ubuntu Breezy - This is a workhorse. It does overheat periodically if it does not get full venting from the three bottom mounted fans
I think you define "workhorse" differently than I do if you can have a machine overheat and still qualify. About a year ago I bought another model from this series of awful HP machines with too many desktop parts in them, and returned it two days later because overheated and crashed regularly during my burn-in testing. I gladly ate the restocking fee rather than put up with it. Whoever is doing the thermal engineering for HP's laptops is either a moron, is hamstrung by some other business process from doing good work, or both.
The T-series Thinkpads have none of these problems, and right now they're the only laptop on the market I consider worth spending money on. I don't quite trust the new Lenovo models yet (while they seem generally fine, I got a little spooked when the first one I saw actually had a Toshiba hard drive in it), luckily there are still plenty of the genuine IBM models floating around for sale.
Finally, someone actually answering the question. It's been months since I had a spyware infection that either Ad-Aware or Spybot were really helpful for; those programs are now obsolete in my opinion. Hijaak This and such are great tools, but with the multi-level spyware infections nowadays (BHO + windows service + constantly reloaded DLL) it's a bear to try and nail everything at once even with it.
I second the recommendation for Ewido for cleaning out nasty infections. The best part is that if your IE still works, you can use their beta free online scanner to try and clean things up.
I've also had success with the somewhat cryptic but powerful Adware Away, which was the only thing I ever found that killed the nastier "about:blank" infections. There used to be a free version of that, but apparently they realized most people ran the program once and never bothered with registering it afterwards. Well worth the $30 if you have one of the infections listed on their site that they kill.
Finally, it's worth mentioning Microsoft's Anti-Spyware package. While it isn't particularly good at killing nasty infections, the proactive tools they include do help at stopping re-infection. For example, when fighting the multi-layer spyware programs, it can stop the service/startup/DLL/BHO sections from re-installing themselves so that you can knock them out one at a time.
Many programs do not run (we use BlueJ and Eclipse, neither work on the Intel).
What an amazing surprise, since Eclipse has always workedperfectly on under Mac OS X before.
As someone who spends a lot of time in Eclipse, the fact that it's never quite worked right under OS X is the only reason I'm still typing this on a PC running Windows. While it's unfortunate for early adopters like yourself, I'm kind of glad it's altogether broken because perhaps this will force Apple to pay more attention to the issues that have been there all along.
Some number of the early sectors on the 1st disk got trashed by the power issues (of course Murphy assured that someone had disconnected the UPS serial cable the week before to move something and not plugged it back in); they showed up as bad blocks when tested later. The MBR started but GRUB died somewhere in the middle of the boot process. Since I wasn't there to do a complete post-mortem and had to go by what I was being read to me over the phone, I can't say for certain that it was the GRUB files, it may have been one of the sectors in the Linux kernel GRUB was running that took a hit.
Once the drive was disconnected the server fired right back up again. The only thing that saved me in this case was that I had every drive in a removable drive bay, all very clearly labeled for such an emergency. Turn the key and the drive doesn't get power.
As you suggested it may have been possible to fix it by adjusting the BIOS boot order. I hesitated to recommend that because even though I know it shouldn't happen, I'm afraid data from the bad drive is going to get mirrored to the good one and corrupt it when I do that.
If you document it thoroughly, any clueless manager should be able to convince the machine to boot from the other device. Or, better yet, don't bother because the odds of the disk failing in exactly the right way to make this necessary are negligible.
"Negligible" in this case meaning that it will only happen when you're on vacation out of the country and can't be reached. I was in Canada the last time I lost a software RAID server this way. Remember the northern East Coast blackout of August 2003? I was trapped in Toronto, unreachable because my hotel's power was out, and the power fluctations at my office in New York caused exactly the scenario you describe.
I am admittedly less lucky than most. Thanks for filling in the details, I didn't remember the trivia beyond recalling why I vowed never to do that again.
Couple of comments on this message, which I generally agree with the theme of but with have some caveats on implementation.
First, one of the benefits of buying cheaper drivers is that you can afford to buy an extra one that sits idle most of the time to use as a hot spare. The expected worst-case scenarios are much less serious if you start getting a rebuild to the spare the minute any one drive fails; you need two failures in the amount of time it takes to copy a disk to be dead, rather than two failures before someone can get to the server to fix the first bad drive. The real worst case scenario is of course a power surge or something that fries all the drives at once.
Second, I'm not that big of a fan of the Linux software RAID. It's fine in an environment where the server can afford to be down until an admin comes in to deal with failed drives, but the ability of it to automatically repair itself and keep going isn't up to the standard set by the better hardware RAID. The main issue I've had with it in the past is that I've lost the ability to boot far too often. This is not a fault of the Linux software, but of the crappy limitations of the PC BIOS and how it impacts the bootloader. I've found that the real RAID cards are much better at segmenting off a failed drive so that the system boots anyway, and then remapping so the OS doesn't even notice. Too many times I've seen pure software RAID configurations where the boot drive goes, and the PC won't even boot to the point where it should switch to the alternate boot drive because the bad drive makes the BIOS insane. Also, I have my doubts about the feasibility of automatically switching to a hot-spare for the boot disk and having that work transparently. I aim at RAID setups where I can have any random person replace a failed drive while I'm sitting at the beach on my vacation; Linux software RAID doesn't meet that standard in my experience.
Finally, one point I throw out at random while I'm typing here. Hard drives nowadays have gotten pretty good at throwing out SMART errors in advance of actually falling over dead; the last two massive disk failures on recent hardware I suffered from both gave me enough advance warning to save my critical data before the disk became unusable. In order to take advantage of this, you need a solution that lets the drive SMART data flow to the operating system where it can be caught and acted upon. I have found that many cards on the market, both SCSI and SATA, don't do a very good job of passing SMART early failure notices onto the OS so you can act on them.
When I get a new controller in here, I make sure I can use SMART to see details like the drive's temperature (which is also handy to monitor) via the tools in the operating system. If I can't, that means it's unlikely SMART errors are going to make it through as well. 3ware is one of the few companies that seems to recognize the importance of SMART information even in an SATA context. The majority of the SATA cards I've seen (insert Silicon Image and Promise rant here) just throw all that data away. The better SCSI RAID cards just toss the drive out of the array the minute funny smells start coming from SMART, which is a perfectly reasonable approach in an environment where you're always going to have another drive to take over.
The next time I have a long block of spare time, I intend to make a visit to Mr. Gerrold. I plan on lashing him to his word processor until the Chtorr series is complete. I've been watching "Misery" for inspiration and recommended technique.
Talk about stuff you'd do with a time machine: I have a stop to 1984 planned where I track myself down and say "that series you're about to start? You won't find out how it ends for over 20 years; don't read it yet".
> have they downloaded/posted credit card numbers, kiddy porn, terrost plots, maybe post a promise to kill the president, and customized ones for several western and radical countries?
Holy shit, where did you get a copy of my to-do list at? Apparently I need to encrypt my information a bit better myself.
> During my entire Mac using experience (3 or 4 years now), Windows Media Player on my mac would work for about a week. Then it would suddenly stop working. The only thing that would get it working again was a fresh install
Remarkable. I had no idea the Mac port was such a faithful translation of the Windows version.
In most states here, you are "legally drunk" when your BAC reaches 0.08% (this lowers the bar, it used to be 0.10% in some). Since many people start to feel drunk at closer to 0.05%, legally drunk implies a level of alcohol in your bloodstream considerably above just regular old drunk.
Sparco is a interesting company. They're a thin layer on top of massive wholesaler Ingram-Micro. Those warehouses you see? Ingram's, not theirs; they just print the labels that go in the boxes with Sparco's information.
While many resellers do this, Sparco seems to be most transparent about their relationship, which means you can see what's going on in the underlying layer to your benefit. From what I've seen they're getting good prices relative to average for Ingram resellers. By comparison, I've also seen plenty of packages from buy.com that were obviously direct shipped out of Ingram. But while buy.com often has even better prices, they don't let me see the stock information as transparently as Sparco does.
Slashdot Mirror
How many of those years did you spend at the bottom of a hole, trying to levitate out?
I knew the time was approaching where I could finally get some money when I sell the Altair computer in my parent's basement. Ebay, here I come!
Also, more channels wouldn't give a reciever any more reason to clip. Each channel is a seperate amp.
That's not completely true. TFA doesn't really explain the point he's trying to make in this area. Your typical home theater receiver has a power rating for each channel that's usually based on the transistors used. There's also a maximum power rating that comes from how much current the power supply can produce. If you have something with multiple channels being driven at once, there are plenty of receivers where the maximum power you can get is far below the rated power per channel because of the power supply bottleneck.
The much more useful specification then is the maximum power with all channels driven, which some receivers don't even mention because it would show the limitations of their design in an embarrassing light. Obviously a 7.1 receiver is going to be handicapped in that spec when compared to a similar 5.1 receiver because it has more channels to drive.
Regardless, your original point that a typical home theater is usually only blowing out a watt or ten of power under typical operating conditions is still correct.
The ultimate Q-Sound demo disc is also from Roger Waters. On a stereo system that's aligned just right (with you exactly at the center), "Amused to Death" has sounds coming from all over the place; I'll never forget how spooked out I got when first hearing the "The Ballad of Bill Hubbard" for the first time, jumping when the first sound came from behind me.
The Roger Waters album mentioned by the parent post here, "Pros and Cons of Hitchhiking" was recorded with holophonics, a variant of the binaural techniques used to make realistic recordings for headphone listening. That's one of the reasons it sounds particularly good with the low-budget matrix approach.
This approach to generating rear ambience from recordings is generally credited to amplifier designer David Hafler, who popularized it in the 70's. For a clear diagram of how to wire this up that may be easier to understand than the text here see http://sound.westhost.com/project18.htm It's also possible to insert a fixed or variable resistor to adjust the volume of the rear speakers relative to the front; see the "Can I play binaural recordings through loudspeakers?" section of http://www.headwize.com/faqs.htm for a sample there. Finally, you can even derive a passive center channel if you're really hardcore; I don't really like the look of the diagrams at http://kantack.com/surround/surround2.html but it covers all these approaches.
Dynaco sold a little box called the QD-1 that simplified the wiring of these during the original Quad craze. They re-introduced a series 2 version as a cheap Dolby Pro-Logic decoder during the beginning of the home theater craze. The newer version is reviewed at http://kantack.com/surround/surround4.html , which is a pretty spot on commentary about the limitations of this type of decoding. Both models are floating around ebay for not much money.
However, note that the effectiveness of this circuit presumes that your amplifier has a shared ground between the left and right channel circuits. While this is generally true, there are amplifiers (like any balanced design) where the ground of the left channel and that of the right are unrelated. Hooking up this circuit to such an amplifier will either a) shut it down, b) trip a fuse, or c) blow the output transistors, depending on the robustness of the design. Be very careful you know what you're playing with here. If you don't know enough about electronics to check if your amplifier channels have a common ground or not, you probably shouldn't be playing around with this circuit.
Dead Can Dance is close to a best case for this approach; really well produced studio work with lots of hard panned instruments sitting in one channel or the other pop into surround this way, and minimal vocals in the center. The best demo I ever found for how effective this circuit could be was the instrumental title track to the Alan Parsons Project "I Robot", which was downright freaky in pseudo-surround.
Because I can do that pretty easily and more accurately than software.
Yes, but you sitting on your PC typing "confidential" can draw attention to what you're doing that the iPod slurping doesn't. Also, records of what you've did can be left behind in various history files on the PC, especially if you're caught in the middle and don't have a chance to erase them.
I'd bet that if we expanded the Olympic drug testing, we'd discover traces of performance enhancing h3rbal vi@gra in his system.
Clearly their concerns here are justified, because that's how I spend my time. Just today, I beat down an old lady with a baseball bat, jacked two cars, blew away a cop with a shotgun, and fired rockets at a passing helicopter. Hiring a ho and roughing her up is the only way I've learned that gives a day like that a "happy ending".
I'm reading comments here while also sitting at two tables playing Texas Hold'Em on-line. Why distinguish between Internet and gambling addictions when you can have both?
HP ZD8110 - 3Ghz P4 HT, Radeon X600 PCI-E 128MB, 2GB Ram, 17inch Widescreen, Ubuntu Breezy - This is a workhorse. It does overheat periodically if it does not get full venting from the three bottom mounted fans
I think you define "workhorse" differently than I do if you can have a machine overheat and still qualify. About a year ago I bought another model from this series of awful HP machines with too many desktop parts in them, and returned it two days later because overheated and crashed regularly during my burn-in testing. I gladly ate the restocking fee rather than put up with it. Whoever is doing the thermal engineering for HP's laptops is either a moron, is hamstrung by some other business process from doing good work, or both.
The T-series Thinkpads have none of these problems, and right now they're the only laptop on the market I consider worth spending money on. I don't quite trust the new Lenovo models yet (while they seem generally fine, I got a little spooked when the first one I saw actually had a Toshiba hard drive in it), luckily there are still plenty of the genuine IBM models floating around for sale.
After looking at your site for a minute, my laptop is on a hard, elevated surface.
You're obviously unfamiliar with recent versions of Norton's software, which do in fact destroy Windows in short order after installation.
> Maybe you should re-read the book and pay more attention this time?
"re-read" implies that the book was read once already; from its depth, I assumed this review was based on a hard look at the table of contents.
Finally, someone actually answering the question. It's been months since I had a spyware infection that either Ad-Aware or Spybot were really helpful for; those programs are now obsolete in my opinion. Hijaak This and such are great tools, but with the multi-level spyware infections nowadays (BHO + windows service + constantly reloaded DLL) it's a bear to try and nail everything at once even with it.
I second the recommendation for Ewido for cleaning out nasty infections. The best part is that if your IE still works, you can use their beta free online scanner to try and clean things up.
I've also had success with the somewhat cryptic but powerful Adware Away, which was the only thing I ever found that killed the nastier "about:blank" infections. There used to be a free version of that, but apparently they realized most people ran the program once and never bothered with registering it afterwards. Well worth the $30 if you have one of the infections listed on their site that they kill.
Finally, it's worth mentioning Microsoft's Anti-Spyware package. While it isn't particularly good at killing nasty infections, the proactive tools they include do help at stopping re-infection. For example, when fighting the multi-layer spyware programs, it can stop the service/startup/DLL/BHO sections from re-installing themselves so that you can knock them out one at a time.
Many programs do not run (we use BlueJ and Eclipse, neither work on the Intel).
What an amazing surprise, since Eclipse has always worked perfectly on under Mac OS X before.
As someone who spends a lot of time in Eclipse, the fact that it's never quite worked right under OS X is the only reason I'm still typing this on a PC running Windows. While it's unfortunate for early adopters like yourself, I'm kind of glad it's altogether broken because perhaps this will force Apple to pay more attention to the issues that have been there all along.
Some number of the early sectors on the 1st disk got trashed by the power issues (of course Murphy assured that someone had disconnected the UPS serial cable the week before to move something and not plugged it back in); they showed up as bad blocks when tested later. The MBR started but GRUB died somewhere in the middle of the boot process. Since I wasn't there to do a complete post-mortem and had to go by what I was being read to me over the phone, I can't say for certain that it was the GRUB files, it may have been one of the sectors in the Linux kernel GRUB was running that took a hit.
Once the drive was disconnected the server fired right back up again. The only thing that saved me in this case was that I had every drive in a removable drive bay, all very clearly labeled for such an emergency. Turn the key and the drive doesn't get power.
As you suggested it may have been possible to fix it by adjusting the BIOS boot order. I hesitated to recommend that because even though I know it shouldn't happen, I'm afraid data from the bad drive is going to get mirrored to the good one and corrupt it when I do that.
If you document it thoroughly, any clueless manager should be able to convince the machine to boot from the other device. Or, better yet, don't bother because the odds of the disk failing in exactly the right way to make this necessary are negligible.
"Negligible" in this case meaning that it will only happen when you're on vacation out of the country and can't be reached. I was in Canada the last time I lost a software RAID server this way. Remember the northern East Coast blackout of August 2003? I was trapped in Toronto, unreachable because my hotel's power was out, and the power fluctations at my office in New York caused exactly the scenario you describe.
I am admittedly less lucky than most. Thanks for filling in the details, I didn't remember the trivia beyond recalling why I vowed never to do that again.
Couple of comments on this message, which I generally agree with the theme of but with have some caveats on implementation.
First, one of the benefits of buying cheaper drivers is that you can afford to buy an extra one that sits idle most of the time to use as a hot spare. The expected worst-case scenarios are much less serious if you start getting a rebuild to the spare the minute any one drive fails; you need two failures in the amount of time it takes to copy a disk to be dead, rather than two failures before someone can get to the server to fix the first bad drive. The real worst case scenario is of course a power surge or something that fries all the drives at once.
Second, I'm not that big of a fan of the Linux software RAID. It's fine in an environment where the server can afford to be down until an admin comes in to deal with failed drives, but the ability of it to automatically repair itself and keep going isn't up to the standard set by the better hardware RAID. The main issue I've had with it in the past is that I've lost the ability to boot far too often. This is not a fault of the Linux software, but of the crappy limitations of the PC BIOS and how it impacts the bootloader. I've found that the real RAID cards are much better at segmenting off a failed drive so that the system boots anyway, and then remapping so the OS doesn't even notice. Too many times I've seen pure software RAID configurations where the boot drive goes, and the PC won't even boot to the point where it should switch to the alternate boot drive because the bad drive makes the BIOS insane. Also, I have my doubts about the feasibility of automatically switching to a hot-spare for the boot disk and having that work transparently. I aim at RAID setups where I can have any random person replace a failed drive while I'm sitting at the beach on my vacation; Linux software RAID doesn't meet that standard in my experience.
Finally, one point I throw out at random while I'm typing here. Hard drives nowadays have gotten pretty good at throwing out SMART errors in advance of actually falling over dead; the last two massive disk failures on recent hardware I suffered from both gave me enough advance warning to save my critical data before the disk became unusable. In order to take advantage of this, you need a solution that lets the drive SMART data flow to the operating system where it can be caught and acted upon. I have found that many cards on the market, both SCSI and SATA, don't do a very good job of passing SMART early failure notices onto the OS so you can act on them.
When I get a new controller in here, I make sure I can use SMART to see details like the drive's temperature (which is also handy to monitor) via the tools in the operating system. If I can't, that means it's unlikely SMART errors are going to make it through as well. 3ware is one of the few companies that seems to recognize the importance of SMART information even in an SATA context. The majority of the SATA cards I've seen (insert Silicon Image and Promise rant here) just throw all that data away. The better SCSI RAID cards just toss the drive out of the array the minute funny smells start coming from SMART, which is a perfectly reasonable approach in an environment where you're always going to have another drive to take over.
The next time I have a long block of spare time, I intend to make a visit to Mr. Gerrold. I plan on lashing him to his word processor until the Chtorr series is complete. I've been watching "Misery" for inspiration and recommended technique.
Talk about stuff you'd do with a time machine: I have a stop to 1984 planned where I track myself down and say "that series you're about to start? You won't find out how it ends for over 20 years; don't read it yet".
> have they downloaded/posted credit card numbers, kiddy porn, terrost plots, maybe post a promise to kill the president, and customized ones for several western and radical countries?
Holy shit, where did you get a copy of my to-do list at? Apparently I need to encrypt my information a bit better myself.
If you think this article is stupid now, just wait until it shows up again as a dupe.
> During my entire Mac using experience (3 or 4 years now), Windows Media Player on my mac would work for about a week. Then it would suddenly stop working. The only thing that would get it working again was a fresh install
Remarkable. I had no idea the Mac port was such a faithful translation of the Windows version.
In most states here, you are "legally drunk" when your BAC reaches 0.08% (this lowers the bar, it used to be 0.10% in some). Since many people start to feel drunk at closer to 0.05%, legally drunk implies a level of alcohol in your bloodstream considerably above just regular old drunk.
Sparco is a interesting company. They're a thin layer on top of massive wholesaler Ingram-Micro. Those warehouses you see? Ingram's, not theirs; they just print the labels that go in the boxes with Sparco's information.
While many resellers do this, Sparco seems to be most transparent about their relationship, which means you can see what's going on in the underlying layer to your benefit. From what I've seen they're getting good prices relative to average for Ingram resellers. By comparison, I've also seen plenty of packages from buy.com that were obviously direct shipped out of Ingram. But while buy.com often has even better prices, they don't let me see the stock information as transparently as Sparco does.
Rather then reject submissions that are misleading, the "editors" here instead reward the good articles by posting them at least twice.