Jump to content

Wikipedia:Reference desk/Archives/Computing/2012 January 21

From Wikipedia, the free encyclopedia
Computing desk
< January 20 << Dec | January | Feb >> January 22 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


January 21

[edit]

USB Using Processing Power

[edit]

Does USB data transfers use my processing cycles (somehow)? So basically on my windows 7 machine, I make regular backups to my external hard drives using USB 2.0 connections and they typically run into tens of gigs taking maybe tens of minutes. So my question is, if I free up resources such as RAM and the CPUs, does that speed up the transfers? I have heard somebody tell me that such USB transfers go through your processor so they bog down your machine so you should close all programs etc. but I can't find a conclusive answer anywhere really (including wikipedia). Experimenting on my machine I do sometimes (not always) see a speedup if I close all the windows/processes etc. but I am not sure if its just a weird perception error on my part. Does it really make a difference? Is firewire better in that respect? Firewire doesn't go through the processor so it won't bog down the machine? Second, why is cutting and pasting files faster than copying and pasting (again in the context of moving/copying a large number of files)? And third, if I want to move let's say 100GB of data, why do large files (ten files of 10GB each) move much faster than small files (100,000,000 files of 1kB each) of the same total size? Is it because of an overhead of creating new files, naming them, etc.? Thanks! - Looking for Wisdom and Insight! (talk) 01:21, 21 January 2012 (UTC)[reply]

Question 1; try USB 3.0. About the third question: do you mean within the same harddisk? That is because the file is not actually physically moved, only the path is changed. Or do you mean from one disk to another? Von Restorff (talk) 01:29, 21 January 2012 (UTC)[reply]

I mean from one disk to another. And for the first response, but is it true that USB uses your processor even when just moving data? - Looking for Wisdom and Insight! (talk) 03:20, 21 January 2012 (UTC)[reply]

Moving data is all that USB does, so yes. If you are running a(nother) CPU intensive task at the same time as the USB transfer, stopping it might help. RAM won't make a difference for USB specifically, but having more available does tend to make things faster. There is an overhead associated with copying each file because, as you guessed, various metadata has to be written also. This includes the file name and size and creation date and so on, the parent directory index, and the free space bitmap. All of this is written twice on a journaling file system like NTFS. The metadata isn't very large, but it's not contiguous with the file data on the disk, so the read/write heads have to seek back and forth between the metadata areas and the file data, which is slow. -- BenRG (talk) 03:56, 21 January 2012 (UTC)[reply]
USB vs FireWire
USB 1.0 1,5 Mbit/s 0,19 MB/s
USB 1.1 12 Mbit/s 1,5 MB/s
USB 2.0 480 Mbit/s 60 MB/s
USB 3.0 4,8 Gbit/s 600 MB/s
FireWire 400 400 Mbit/s 50 MB/s
FireWire 800 800 Mbit/s 100 MB/s
FireWire 3200 3,2 Gbit/s 400 MB/s

Firewire uses less CPU. USB 3.0 is the fastest. Von Restorff (talk) 04:25, 21 January 2012 (UTC)[reply]

CPUs are extremely fast, compared to the other components of a computer. In a consumer machine, the CPU is rarely going to be fully utilized for even half a second without running out of work to do. There's probably some CPU involved in copying data from one device to another, but the amount is utterly trivial, compared to the time that it takes to physcially spin the disk and physically swivel the head.
Cutting and pasting files is faster because the OS implements it simply by doing a little bit of bookkeeping to change what directory the file is in. To make a duplicate (or move a file from one device to another) requires actually transferring its contents. Moving between physical devices, there should be no significant difference between copying and cutting/pasting.
Your suggested answer regarding lots of small files is correct. There's a certain small overhead associated with each file. For one huge file, the overhead is trivially small. For lots of little files, it might not be.Paul (Stansifer) 06:42, 21 January 2012 (UTC)[reply]
I have no idea how USB 3.0 compares to Firewire in real world performance but AFAIK Firewire, even Firewire 400 usually beats USB 2.0 (which the op has) in performance with mass storage devices (or basically anything with high speed transfers, e.g. camcorders) even on fast modern computers. E.g. [1] (okay 2 years old now but I doubt the speed advancement in CPUs can have made that much of a difference). However the advantage of Firewire 400 is not that significant, in the realm of 30% at most. I'm not that familiar with the technical advancements in USB 3.0 but based on past performance, I would be hesitant to say FireWire 3200 is going to lose to USB 3.0 in real world performance just because it has a higher maximum theoretical performance. Nil Einne (talk) 08:52, 21 January 2012 (UTC)[reply]
P.S. It seems FireWire 3200 is vapourware and with LightPeak and one of the major supporters of Firewire on the PC side and owner of the trademark seemingly coming closer and closer to abandoning it I wonder whether it may just never appear. (On the other hand, the speed may be needed on the non PC side e.g. like by the High-Definition Audio-Video Network Alliance.) Anyway since I doubt USB 3.0 can be that crap, it's probably fair to say USB 3.0 beats Firewire right now. Nil Einne (talk) 09:33, 21 January 2012 (UTC)[reply]

Word 2007 Error

[edit]
Resolved

Every time I click on a .doc or .docx to open it in Word 2007, I always get an error message saying 'There was a problem sending the command to the program,' after which I am able to open the document by clicking on it again. This only happens when Word is not already open. If I have another document already open, then there is no problem, or if I open the document from Word there is no problem. I have looked around online for a solution, and various people have offered this and that, with the inevitable "Didn't work for me" messages appended to each suggestion. Does anyone know what I should do? It never happens with any other MS Office program. KägeTorä - (影虎) (TALK) 14:02, 21 January 2012 (UTC)[reply]

This is a little anecdotal, in that I had a different problem, but in the solving of that I caused the problem you mention (or something similar). Try Options -> Advanced. Under General, un-tick ‘Ignore other applications that use Dynamic Data Exchange’. - Cucumber Mike (talk) 14:16, 21 January 2012 (UTC)[reply]
Thanks, but I don't have that option there...... KägeTorä - (影虎) (TALK) 14:24, 21 January 2012 (UTC)[reply]
Fair enough. As you were, then! - Cucumber Mike (talk) 14:35, 21 January 2012 (UTC)[reply]
This is a register problem. Von Restorff (talk) 17:52, 21 January 2012 (UTC)[reply]
That's fixed it, cheers! KägeTorä - (影虎) (TALK) 19:34, 21 January 2012 (UTC)[reply]

Recognizing Counterfeit Credit Cards

[edit]

I was running register at work one day and a Hispanic man that couldn't speak English attempted to purchase two gift cards and a candy bar with a suspicious looking Discover card. The card didn't look like any Discover card I'd seen before, the Discover logo on the back was an older style Discover logo, and the hologram didn't change when I tilted the card like they usually do. His ID matched the name on the card. He hit the wrong button on the pin pad (probably because he couldn't read what it was asking), cancelling the transaction and then he left. Of the three other times people tried to use stolen credit cards at my register, two of them were gift card purchases like this time. My question is, do all Discover cards have the same hologram regardless of what country they're issued in, and is it possible there's a legitimate reason that the card looked odd, or were my suspicions correct and the card was likely forged? Additionally, I'm not looking for any legal advice. 184.7.157.90 (talk) 18:46, 21 January 2012 (UTC)[reply]

Your suspicions were 100% correct. Please read this document, that may be useful for someone like yourself. Von Restorff (talk) 18:49, 21 January 2012 (UTC)[reply]

Download an on-line video?

[edit]

I found a really funny Star Wars remake video on-line. Is it somehow possible to download it so I could watch it from my own computer, not from the Internet? JIP | Talk 22:15, 21 January 2012 (UTC)[reply]

RealPlayer allows you to do this (doesn't work for youtube though)--Jac16888 Talk 22:22, 21 January 2012 (UTC)[reply]
Savevid.com Von Restorff (talk) 22:44, 21 January 2012 (UTC)[reply]
netvideohunter for Firefox. -- Finlay McWalterTalk 22:45, 21 January 2012 (UTC)[reply]

If you are using Firefox, this or this are good. AvrillirvA (talk) 00:36, 22 January 2012 (UTC)[reply]

This the best option. Von Restorff (talk) 00:26, 22 January 2012 (UTC)[reply]

AvrillirvA's first link is what I'd use. ¦ Reisio (talk) 02:35, 22 January 2012 (UTC)[reply]