Using a USB thumb drive for backups?
in [Mac Systems]
|
From: nospam on 20 Dec 2009 12:38 In article <jollyroger-F98895.10323720122009(a)news.individual.net>, Jolly Roger <jollyroger(a)pobox.com> wrote: > > > USB bus-powered enclosures have a design flaw: the power supplied by the > > > USB bus is barely enough for todays faster, more power-hungry drives. > > That would be your opinion. My opinion is there are still drive being > sold today that require more power. i didn't say there weren't. however, some new drive mechanisms use less power than older ones. check the specs. typically, they've needed 1a to spin up and some have reduced that requirement. that doesn't mean that every drive on the shelf is suddenly the lower power variety. > The fact that such drives come with dual Y-split USB cables lends > credence to my opinion. They require more power than one USB port can > supply. Unfortunately, I suspect some laptops may not have dual USB > buses with separate power supplies, which could result in occasional low > power situations for such drives. most usb drives need more power than a single usb port can provide to spin up the drive, thus the dual cables. once the drive is spun up, the power requirement of a usb drive is usually 500ma or less, which is within the capabilities of a single usb port. > > > To make matters worse, at certain times, the power delivered by USB ports > > > can fluctuate. > > > > only if the host is defective. > > I disagree. Power can fluctuate if devices are aded to or removed from > the same bus. And it is my belief that power can fluctuate at other > times on a laptop where battery drain comes into play. the usb spec requires that the host supply 500ma. if it doesn't, it's not compliant. it's that simple. > > > When a hard drive does not receive enough power during a > > > write, unpredictable things may occur, including data corruption on the > > > drive. Laptops are especially prone to USB bus power fluctuations. > > > > no, they're prone to not being overspeced. desktop computers have a > > much higher power budget and can easily overspec the usb ports. > > Regardless, my opinion is the power delivered USB ports on laptops are > more prone to fluctuation. see above. > > all usb hosts are required to supply 500ma, and most drives need about > > 1a to spin up the drive, but can run fine once spinning on 500ma. since > > usb hard drives are common, some computers provide more than required > > 500ma. > > Unfortunately neither of us has physically monitored the power output of > each USB port on all of Apple's currently-shipping MacBook line through > all of the different phases of startup and under different usage > scenarios. So neither of us can categorically state that 500ma is > consistently delivered to all USB ports on these machines at all times. you haven't measured anything, yet you are so sure it fluctuates?? apple says their computers are usb compliant, and in order to do that, they must meet the requirements of the usb spec. one of those is 500ma output on a port. i'm quite certain that apple has measured it. > I suspect such is not the case. I suspect this is what caused corruption > of my then-USB-connected drive multiple times in a couple months. The > same hard drive, placed into a Firewire enclosure hasn't had a single > instance of corruption since then - for the past 8 months or so. but you never measured the voltage, so it's pure speculation that it was the cause. it could be any number of reasons, especially if it only happened with just one drive.
From: JF Mezei on 20 Dec 2009 13:42 The movie industry found itself with a new problem when movies started to be produced on disk drives instead of plastic film: How do you preserve movies in the long term ? Film degrades gracefully over time. Disks go from being 100% good to 100% bad overnight. For now, what they end up doing is copying drives to new drives every couple of years. This not only get a new "fresh" copy out, but also moves data only drives with current interfaces, ensuring that can still be read. A disk drive with the data intact, but with an interface that no system supports today would be rather worthless. For instance, some old IBM or Digital proprietary drives wouldn't be easy to connect to a current computer system. Even SCSI-1 from the 1980s would start to need additional work to find a system capable of processing it. Does anyone know about long term storage properties of flash drives ? Once data is written (and only dones once to a new drive), would it last "forever" in a vault, unconnected to anything ?
From: JF Mezei on 20 Dec 2009 13:46 nospam wrote: > i didn't say there weren't. however, some new drive mechanisms use less > power than older ones. A 600 meg disk in the mid 1980s required a 10amp 115VAC circuit for each drive, and used a 1/3HP motor to turn the disk. Think about that the next time you turn on your ipod with 30 gigs of storage in it.
From: Jolly Roger on 20 Dec 2009 15:06 In article <201220091238269346%nospam(a)nospam.invalid>, nospam <nospam(a)nospam.invalid> wrote: > In article <jollyroger-F98895.10323720122009(a)news.individual.net>, > Jolly Roger <jollyroger(a)pobox.com> wrote: > > > > > USB bus-powered enclosures have a design flaw: the power supplied by > > > > the > > > > USB bus is barely enough for todays faster, more power-hungry drives. > > > > That would be your opinion. My opinion is there are still drive being > > sold today that require more power. > > i didn't say there weren't. however, some new drive mechanisms use less > power than older ones. check the specs. typically, they've needed 1a to > spin up and some have reduced that requirement. that doesn't mean that > every drive on the shelf is suddenly the lower power variety. > > > The fact that such drives come with dual Y-split USB cables lends > > credence to my opinion. They require more power than one USB port can > > supply. Unfortunately, I suspect some laptops may not have dual USB > > buses with separate power supplies, which could result in occasional low > > power situations for such drives. > > most usb drives need more power than a single usb port can provide to > spin up the drive, thus the dual cables. > > once the drive is spun up, the power requirement of a usb drive is > usually 500ma or less, which is within the capabilities of a single usb > port. > > > > > To make matters worse, at certain times, the power delivered by USB > > > > ports > > > > can fluctuate. > > > > > > only if the host is defective. > > > > I disagree. Power can fluctuate if devices are aded to or removed from > > the same bus. And it is my belief that power can fluctuate at other > > times on a laptop where battery drain comes into play. > > the usb spec requires that the host supply 500ma. if it doesn't, it's > not compliant. it's that simple. Yes, and it seems neither you nor I are in a position to say whether or not all USB ports on every Mac are compliant to that requirement all of the time. > > > > When a hard drive does not receive enough power during a > > > > write, unpredictable things may occur, including data corruption on the > > > > drive. Laptops are especially prone to USB bus power fluctuations. > > > > > > no, they're prone to not being overspeced. desktop computers have a > > > much higher power budget and can easily overspec the usb ports. > > > > Regardless, my opinion is the power delivered USB ports on laptops are > > more prone to fluctuation. > > see above. > > > > all usb hosts are required to supply 500ma, and most drives need about > > > 1a to spin up the drive, but can run fine once spinning on 500ma. since > > > usb hard drives are common, some computers provide more than required > > > 500ma. > > > > Unfortunately neither of us has physically monitored the power output of > > each USB port on all of Apple's currently-shipping MacBook line through > > all of the different phases of startup and under different usage > > scenarios. So neither of us can categorically state that 500ma is > > consistently delivered to all USB ports on these machines at all times. > > you haven't measured anything, yet you are so sure it fluctuates?? Neither have you. Are you sure it doesn't? Common sense tells me if the battery power level in a laptop happens to fluctuate, so will the power supplied to USB ports. Are you saying that cannot ever happen? > apple says their computers are usb compliant, and in order to do that, > they must meet the requirements of the usb spec. one of those is 500ma > output on a port. i'm quite certain that apple has measured it. Are you certain they have measured it through all possible usage states, not just at idle? > > I suspect such is not the case. I suspect this is what caused corruption > > of my then-USB-connected drive multiple times in a couple months. The > > same hard drive, placed into a Firewire enclosure hasn't had a single > > instance of corruption since then - for the past 8 months or so. > > but you never measured the voltage, so it's pure speculation that it > was the cause. it could be any number of reasons, especially if it only > happened with just one drive. Of course it's speculation. And without the necessary equipment, that's all it is going to be. -- Send responses to the relevant news group rather than email to me. E-mail sent to this address may be devoured by my very hungry SPAM filter. Due to Google's refusal to prevent spammers from posting messages through their servers, I often ignore posts from Google Groups. Use a real news client if you want me to see your posts. JR
From: Richard Maine on 20 Dec 2009 16:02
nospam <nospam(a)nospam.invalid> wrote: .... > the usb spec requires that the host supply 500ma. if it doesn't, it's > not compliant. it's that simple. .... > apple says their computers are usb compliant, and in order to do that, > they must meet the requirements of the usb spec. one of those is 500ma > output on a port. i'm quite certain that apple has measured it. Are you sure that Apple says without qualification that all their computers are USB compliant? That sounds like speculation to me - and speculation that doesn't match my recollections. Let's see... Well, it isn't directly from Apple, and I'm too lazy to go chase down a first-hand citation, but lowendmac is usually pretty reliable (certainly far more so than random posters on usenet) and I'd suspect they would not just invent the following. See <http://www.lowendmac.com/macbookpro/15in-macbook-pro-june-2007.html> which is a profile of June 2007 15" MacBook Pro models. Note in particular the paragraph: "Unlike earlier models, where every USB port could provide 500 mA of power, only a single high-powered device can be attached to the USB ports, and software will enable one of its downstream ports to supply 500 mA of power. If a second high-powered device is attached, it will behave like a normal bus-powered hub and only provide 100 mA per downstream port." -- Richard Maine | Good judgment comes from experience; email: last name at domain . net | experience comes from bad judgment. domain: summertriangle | -- Mark Twain |