USB 3.0 has an upper limit around 5.0Gbps. SATA III has an upper limit of 6.0Gbps. Regardless of overhead these rates are far higher than what a mechanical HDD can sustain for large transfers.
Most mechanical HDDs won't be able to sustain more than about 1.5Gbps (HDD Speed results). So I doubt y...
this answer, post altered after getting quite a bit of rep, failed to acknowlege some of the "issues" of having 2 "controllers" 2 "buffer" sets, and a large dependancy on type of "speed" were talking about.
It did not stop anyone from upvoting it.
While moab says , ya sure about 1/2 that. I do not think that is valid either.
When it is less direct connections, you can get full possible speeds going, with something like a large sequential item. but more latencies can exist, in the whole paths and connections.
@allquixotic Don't particularly like the thought of linking my OS-level account to an online service.
@Psycogeek Depends, though. UASP is quite efficient, and USB 3.1 has alternate modes that could quite literally use plain PCIe or SATA with zero overhead.
UMS is shit, that's for sure. But it's easier and cheaper to implement. And MTP is even slower but has its use case too.
@Bob Depends is right, compare a esata using a marvell controller, to a good 3.0 controller, and the sequential may go UP. provideocoalition.com/a-bunch-of-random-hard-drive-speed-tests has a rack of test (mostly mac style) with results all over the place, but the test itslef looks lameo
A good test should represent some semblance of a reality :-) even have some IwhileO and some random, and some big stuff, and the actual copying of both big and small files. Doing that i would Never put a drive on a usb 3.0 vrses and internal intel controller.
If it is a backup drive, then it really should be at least able to be powered off manually or disconnect manuely . even if that is after you find a virus on your system :-)
It is kinda weird to see people worrying themselves silly (at times) over 1-2% numbers on video cards, and CPU overclocking and all, which can include one card doing 15%+ better in one game and the other the same in a different game. Then the disk, to know there is variations of 10%+ in various operations of controllers and connections, but that All out bottleneck :) not being tested daily.
@Ramhound: Usually it has a bios of its own, and you may need drivers for the OS "pre install". Windows 8 will let you use a USB, can't remember 7, older ones needed a floppy or slipstreaming.
Its tricky, its a Dell, should I even worry about the sata performance on a ssd if my goal is to replace a drive thats been "doing the magic twirl" on certain exploer windows?
ahh. Not sure about those. There's an application for samsung called magician that does a bunch of SSD specific optimsations for you that's worth installing.
Hmm... if I'm using an Dynamic DNS service to provide SRV records to certain ports on my server at home, am I better off looking into doing the same things via something like Apache?
Yea and your a atupid ass machine, if you dont know i can make a cake with ice cream, and that i cant have a shake with a sandwitch , and WTF is that thing in the jello glass.
<--- Stomps feet on floor like 3Year old , If they had said pick the 3 that are most Icecream (or sushi) like , i would get it, but they removed that for some reason?
Sitting at almost 160k and thinking about how to prevent a -2 from downvotes. Even if he had a point, I feel like the suggestion is weird coming from him
...and I also don't see cause for action regardless
Small question regarding pull request, Let's say I fork a branch from X. My branch is Y, I push a lot of commits to Y. When I give a pull request to X, will all the commits from Y be registered on X
@Psycogeek I guess so. I just never thought about changing an old question just because it received a downvote, or 3. That's just how it goes. Some people think an answer is helpful, others think it's stupid. And maybe it is stupid and worthless and deserves to be downvoted to hell
@OliverSalzburg i know some of those , i only downvote old stuff when it was wrong both then and if it applies, now.
But only because i accidentally came across it and it was the usual , not only wrong, but highly accepted for how incorrect it was. Almost expecting if it was more wrong, it would get even more votes :-)
When I'm on Stack Overflow and I see a shitty .NET answer from 2008, I still downvote it. Age is not a deciding factor when I judge the quality of a post.
And maybe the user can't improve the post now, but maybe they should have written a better post back in the day
If they would then delete the post, just because I downvoted it, even though others have upvoted in the past and might find it helpful in the future, that doesn't make sense to me
The one situation where I'd consider deleting an answer would be if it was a software rec where the software is dead/abandoned and/or can't be found. In which case, editing a replacement and how to use it as an alternative sounds better.
I love it when ya try to get a contact, and instead find yourself in an infinate loop on the web pages. . . especially if your reporting an infinate loop bug :-)
So you call them on the phone, and get into an infinate loop , of being passed from office to office "sorry we do not hangle that, you will have to . . ."
Finnaly they say , check on microsoft answers LOLOL, wherin you post the info, and they say "this belongs on tech net , not MSA", so you post it on MS TechNet, not really belonging there at that level, and they bitch at you for having said it there.
On your way out, you get the oppertunity to swear with the capslock on to the Survey , which doesnt submit because it gets in an infinate loop.
hm. I have 2 4:3 screens (shockingly) at work . Yet I decided to put the taskbar on my remote desktop *sideways* so I don't waste vertical space from titlebars ._.
The motor controller MOSFET on the AFB1212GHE fan gets very hot. much more so than the motor windings, especially when the motor is stalled or when the rotor is caught between poles oscillating without actually spinning
You can feel the heat from the top side of the label on the fan
On closer examination, the stator seems to be hand-wound
That's probably one of the reasons this fan is so expensive
@DragonLord if you get one of those you could always move your computer out to the garage :-) and you can then get a clickey keyboard and it wont be the most obnoxious thing in the house. I used to use "delta" fans , that is why there was so much push to get to bigger fans (120-240) because rpm isnt just noise it is the kind of noise that drills through you :-)
Going back to delta fans in the larger size would be going "full circle" being right back were we started with 80mm whiners.
if they get the rpm just a wee bit faster, it sounds just like a Dentist drill :-)
The center hub (dead space) is also much larger, very high "pressure" in the minimal fin area, because of the large motor.
wtf. I'm currently calling MS hardware support and they have this stupid "we're sorry for the delay. your call is important to us" bullshit on repetition every 30 seconds
This is really helping with my frustration with their stupid product
@Bob it can also mean they cannot QC. with the large bulk orders you expect 2-3 items out of 100 to be . . . like the potato chips at the bottom of a bag.
Good thing the company motto is "we inspire confidence" (or was it expire)
> When ZFS has direct access to disks, it will bundle multiple read and write requests into transactions; most filesystems can not do this as they only have access to disk blocks.
NO RAID SOLUTION PROVIDES A REPLACEMENT FOR A RELIABLE BACKUP STRATEGY. BAD STUFF CAN STILL HAPPEN AND YOU WILL BE GLAD THAT YOU BACKED UP YOUR DATA WHEN IT DOES.
The commands below are more generic then for kernel version 3.13.0-35 only.
1. Mount the efi partition and copy the kernel files there
$ mount /dev/sda3 /boot/efi
$ mkdir -pv /boot/efi/EFI/ubuntu/
$ cp -uv /boot/vmlinuz-* /boot/initrd.img-* /boot/efi/EFI/ubuntu/
'/boot/vmlinuz-3.13.0-...
And I'm salaried. Of course if I do hours outside of my normal 37.5 yes I get overtime.
At old job I got 'shift pay' because was on a shift pattern. But if they just bumped up the wage I'd be happy too (they'd never do that ,as the base wage is what the progression bonus's were worked out off).
@Bob Well, since you chose software, if you want to utilize all your 3 x disks, you have to use RAID5. Unfortunately, btrfs RAID5 is vulnerable to the write hole: btrfs.wiki.kernel.org/index.php/RAID56
as is linux-md (the traditional software RAID block layer)
Suddenly, a message saying "A problem in the kernel package has been detected" started appearing after I login upon boot. A new message is shown every second, incessantly.
I haven't updated the kernel in the last 12 days (3.19.7-200.fc21.x86_64). I have installed 5 new packages today: subversi...
@allquixotic this has been a nightmare of an installation... deb live image doesn't support efi boot, so I had to grab a copy of lubuntu (I had a 32-bit kubuntu iso, but incompatible with 64-bit efi), then had to juggle loading the deb live image with zfs support and the lubuntu image with efi support, and every reboot required recompiling said support -_-
dropping into lubuntu and installing efibootmgr again now
The commands below are more generic then for kernel version 3.13.0-35 only.
1. Mount the efi partition and copy the kernel files there
$ mount /dev/sda3 /boot/efi
$ mkdir -pv /boot/efi/EFI/ubuntu/
$ cp -uv /boot/vmlinuz-* /boot/initrd.img-* /boot/efi/EFI/ubuntu/
'/boot/vmlinuz-3.13.0-...
Suddenly, a message saying "A problem in the kernel package has been detected" started appearing after I login upon boot. A new message is shown every second, incessantly.
I haven't updated the kernel in the last 12 days (3.19.7-200.fc21.x86_64). I have installed 5 new packages today: subversi...
