Wikipedia:Reference desk/Archives/Computing/2009 September 18

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.

September 18

Where did my question go?

Hi,

I distinctly remember posting my question about asp just a couple of hours ago... How come I can't see it anymore? :( PrinzPH (talk) 01:40, 18 September 2009 (UTC)

Are you talking about the one that a user deleted here? It may have been a completely accidental deletion. -- kainaw™ 02:07, 18 September 2009 (UTC)

Yeah thats the one! :) How do I restore it? (wiki noob) PrinzPH (talk) 02:29, 18 September 2009 (UTC)

I restored your question about ASP from the history. It's the following section. EdJohnston (talk) 02:41, 18 September 2009 (UTC)

Thanks Ed! :D PrinzPH (talk) 02:57, 18 September 2009 (UTC)

And in case anyone's wondering, it's now in the section above. (I had to move it to keep the archiving straight.) —Steve Summit (talk) 01:32, 21 September 2009 (UTC)

Sorry everyone! That seems to have been my fault - I got an edit conflict screen, which are an absolute nightmare on long pages like this one. And I really should stop editing in the middle of the night... - IMSoP (talk) 14:32, 18 September 2009 (UTC)

CAESAR piping software

Hi Anybody can send me CAESAR piping stress analysis software

? You can get it here. http://www.coade.com/

83.100.251.196 (talk) 11:05, 18 September 2009 (UTC)

Reference desk talk page

The talk page refuses to be edited but is not locked -temporary blip?83.100.251.196 (talk) 11:41, 18 September 2009 (UTC)

It's been semi-protected by Ten. Zain Ebrahim (talk) 12:18, 18 September 2009 (UTC)

Ah I see, shouldn't there be an icon (silver padlock) or something on there - I can't spot it.83.100.251.196 (talk) 12:23, 18 September 2009 (UTC)

Me too. I wonder why it's not there. Zain Ebrahim (talk) 13:07, 18 September 2009 (UTC)

Possibly hidden by the "skip to the bottom" button-box at the top-right of the page...83.100.251.196 (talk) 13:23, 18 September 2009 (UTC)

I don't think so. For me, (Beta, IE7) the "skip to the bottom" button goes under the line and the padlock goes above. Zain Ebrahim (talk) 13:34, 18 September 2009 (UTC)

Given how many IPs regularly edit the Reference Desk, isn't protection for four days a little much? It strikes me as a "one day at a time" sort of situation. A look at the history does not make clear to me why it was protected, nor is there any discussion on the page about why it was protected... --98.217.14.211 (talk) 13:28, 18 September 2009 (UTC)

I'm guessing that the edit causing the protection (see link above) was by ex-user"lightcurrent" - who has a "blue touch paper" effect on most admins - as far as I can tell they spend 99% of their time chasing him, his sockpuppets and other his other anonIPs - a behaviour not unlike the game Whac-A-Mole.

This is probably why the protection was immediate, and undiscussed.83.100.251.196 (talk) 13:41, 18 September 2009 (UTC)

iTunes

I have filled both of my hard drives with music and would now like to buy an external hard drive, and move iTunes and all my music over there. So, would this help to speed up my computer as I have heard, and how would I do this with out loosing any of the songs? Any help would be appreciated. Thanks ppl

I don't know to what you are reffering to as there is no link in your question. Please rephrase.

--JD's Web Service 13:52, 18 September 2009 (UTC)

No link should be necessary. If your hard drive is filled, then yes, expanding capacity will improve performance. Generally an internal drive performs better than an external drive, but for data like music and videos, an external drive is fine (and allows for easier portability if needed). Here's one of many articles explaining how to move your iTunes content to a new drive. — Lomn 14:09, 18 September 2009 (UTC)

Telephone/Voice Mail Oddity

A friend of mine recently told me about an odd experience. She was talking to someone else, normally, on the telephone, but when she hung up, there was a new message. It was a message for the person she was talking to. Is this even possible? Thanks, gENIUS101 12:19, 18 September 2009 (UTC)

I don't quite understand how it would be for the person she was talking to - maybe someone had just mis-dialled? Certainly many phone providers (both mobile and fixed line) have free voice mail facilities that kick in when the line is engaged. Often these don't allow you to record an out-going message, so if someone had both numbers and selected the wrong one, they wouldn't have realised they were leaving a message for the wrong person.

If it wasn't even a mutual contact that left the message, then I really don't know. - IMSoP (talk) 15:37, 18 September 2009 (UTC)

I don't think it was a misdial, as it was a work call for the person she was talking to. However, my friend did work there about 10-15 years ago, so that might be it, but it seemed unlikely to me. Thanks, gENIUS101 20:20, 18 September 2009 (UTC)

If I understand correctly, you are saying the other person was using a phone at their workplace and called your friend. After the call, your friend had a new voicemail message but it was meant for the other person.

If so, then this might have happened if the other person accidentally forwarded their workplace phone to your friend's phone.

Imagine the other person's workplace phone. Perhaps there's a "Call Forward" button right next to the "Outside Line" button. Or maybe the codes are similar, say *8 and 8. So perhaps the first time the other person tried to call your friend, they accidentally used the forwarding code without realizing it. They were expecting a ringing signal, but heard a confirmation beep instead, and so they just hung up and called correctly the second time.

While your friend and the other person were talking, a third person called the work phone, was forwarded to your friend's phone, and went to your friend's voicemail. Perhaps your friend's voicemail greeting is generic, and the third person thought they had the work phone's voicemail and left a message.

Perhaps after the call was over, the other person noticed an indicator on their work phone indicating it was forwarded and turned off the forwarding. --Bavi H (talk) 03:00, 19 September 2009 (UTC)

Why is C++ faster than MATLAB?

Hello, I am not very good at analyzing this stuff. I have to basically do a simulation involving fluid dynamics, Solid Mechanics and Acoustics all in one. My supervisor has asked me to implement it in C++ rather than in MATLAB, where it would be easier. My mesh will contain several thousands (or even millions) of elements. Why did he ask me to do it in C++? Thanks - DSachan (talk) 13:06, 18 September 2009 (UTC)

You should ask him. Perhaps he wants it to be used on a computer that doesn't have MATLAB. Perhaps he wants to share it with people who only know C++. Perhaps he wants it implemented directly instead of slowly parsed by MATLAB. Perhaps he wants you to fail because he is trying to downsize. Perhaps he is a complete idiot and doesn't know what either C++ or MATLAB is. -- kainaw™ 13:24, 18 September 2009 (UTC)

I don't know enough about the speed of MATLAB to give an answer, but the article says MATLAB is implemented in C. But it's dynamically typed which means it may be wasting a LOT of time checking the type of each element in each matrix before doing any calculation. Using C or C++, all the checking is done once at compile time, not repeatedly for each operation on any element in the matrix.

If you're doing it in C or C++ yourself, you'll need to use a software library to implement the matrix algebra. Use a good quality library rather than implementing it all yourself as the libraries are well tested and optimized to make them run as quickly as possible.

One possible reason for using C/C++ is that there are extensions and libraries for performing calculations in parallel over multiple processors. If you don't like C/C++, you could try FORTRAN, which is still very popular in academic circles (I used FORTRAN with LAPACK / NAG libraries Uni). Or even Python with SciPy (this is more similar to MATLAB but may still be a little slow).

I guess the most important bit is to talk to your supervisor and others on campus - if your supervisor only knows C, he can only help you with the problems you get in C. --h2g2bob (talk) 13:55, 18 September 2009 (UTC)

Matlab can be either intepreted or compiled, I'd assume for heavy maths it would be compiled first to avoid the issues you mention with the typing system.83.100.251.196 (talk) 14:08, 18 September 2009 (UTC)

Look at MATLAB#limitations : portable code is one good reason not to use it, and a reason to want it in C or another language.83.100.251.196 (talk) 14:08, 18 September 2009 (UTC)

But MATLAB is more portable than C! And now that GNU Octave exists, MATLAB-compatible code can run on all major platforms, operating systems, processor architectures, without purchasing software from The Mathworks. Nimur (talk) 15:17, 18 September 2009 (UTC)

I admit to being totally out of date.. but more portable? - maybe they see themselves using a non-intel processor in the future - in which case C++ or something would be a better choice, maybe even they have access to cell microprocessor??

I like the sound of C! (programming language) by the way :) 83.100.251.196 (talk) 15:54, 18 September 2009 (UTC)

There is a very prevalent misconception about "performance" in computational research. The idea is that MATLAB, or Java, or Python, are inherently slower because they are interpreted languages. Not only is this untrue (in the case of Matlab or Java, for example, the program can actually be compiled to machine code); it also betrays a 30-year-old understanding of computer processor architecture, and a misunderstanding that "extra instructions" must slow the processor down. In fact, interpreted instructions tend to be more predictable to cache prediction schemes, so executing more instructions may actually increase instruction throughput. A modern CPU or multicore CMP is so sophisticated that a conventional linear analysis of "instruction count" is meaningless when comparing two executables. The only way to determine which language implements a feature faster is comparative benchmarking. To say definitively whether MATLAB or C++ will implemenent your fluid simulation faster, you must code both and test them on the same machine and measure their actual execution time. That being said, it is certainly true that MATLAB has some features which can potentially slow the code down significantly. Dynamic type-checking is a possible issue, but this is probably the least of your worries. What is most troublesome is that MATLAB allows "transparent" activities to be interpreted by the execution environment - e.g., if you have two large matrices A and B, and you add A + B, MATLAB's environment may dynamically choose to swap subsets in and out of main memory or cache at will. Thus, there is a "non-deterministic" aspect to the MATLAB execution pathway (rather, this nondeterminism is really just "beyond the application programmer's control", though totally deterministic to the systems engineer) runtime. As a general rule, by allowing straightforward programming statements (such as "A+B"), even if those matrices were each several gigabytes in size, MATLAB allows the programmer to unwittingly request enormous volumes of computation without really realizing it. So, it breeds a style of programming which is not cognizant of performance and machine limitations. This is either "a wonderful evolution towards higher level abstraction" or a "blight afflicting an entire generation who don't know C++, FORTRAN, and DEC assembler language" depending on your interpretation.

In summary, MATLAB is not inherently slower than C++, it is not inherently better or easier. Both languages provide different feature sets and different paradigms for interacting with the hardware. If your advisor or research group uses C++ exclusively, it is a cultural preference, rather than a legitimate scientifically grounded analysis based in performance metrics. However, from experience, explaining the details of 2009-computer-hardware-and-software-architecture is rarely effective at uprooting decades of tradition. Certain research communities use FORTRAN. Certain programming teams use g++. Certain teams are dynamic and switch languages at will, based on best tool for a particular task. You may have to learn C++ (which is a worthwhile use of your time), simply to satisfy your professor; and it is very possible that the extra knowledge you gain from this learning experience will make you a better, more hardware-aware programmer. That will likely improve your code performance. Nimur (talk) 15:17, 18 September 2009 (UTC)

It's important to note that when comparing the performance of languages which do not have a "canonical" implementation, you are really comparing the performance of particular implementations of those languages. C++ is defined in terms of a theoretical "abstract machine", and very little information about performance can be inferred from its definition, except that the time complexity of certain standard library operations and algorithms is specified. Given a well-formed C++ program, the C++ standard places no requirements on the executable produced by a compiler, except that it shall exhibit the same observable behaviour as if the program exactly followed a valid execution sequence of the theoretical abstract machine. For example, if one were to write a program that calculates all the prime numbers between 10,006,721 and 0 and prints the lowest, the compiler is free to generate machine code that is equivalent to simply

#include <iostream>
int main() { std::cout << "2\n"; }

My conclusion is similar to that of Nimur:The only way to determine which of two languages is faster, for a given implementation of those languages, is comparative benchmarking. decltype (talk) 17:01, 18 September 2009 (UTC)

Apart from being slow, Matlab is a memory-hog. It amazes me that is manages to run out of memory on a 2GB machine when asked to do a few manipulations with vectors of ~1 million entries. You said your mesh will have that number of entries, in my experience Matlab is a no-go for large scale application.195.128.250.184 (talk) 21:22, 18 September 2009 (UTC)

As I discussed above, MATLAB's syntax allows for much behind-the-scene implicit computation and storage. If you are using excess memory by a factor of 1,000, you are probably storing more than you realize - multiple copies of your million-element array. You might take a look at Mathworks' Memory Managemeny Guide or consider taking one of the free MATLAB courses which are offered to train on how to use MATLAB. A MATLAB array uses exactly the size of the native element array, plus about 60 bytes of overhead for name, length, and dimensionality; if you are finding order-of-magnitude deviation from this, it is because you are doing something else (like copying the array many times, or failing to perform in-place calculation); or you are using cell-arrays instead of true-arrays. These are the same issues you would run into with a FORTRAN array, C++ array, or any other programming language if you performed equivalent calculations. It sounds like you are unaware of how many copies of that million-element array you have actually created. Consider learning the following MATLAB commands:

who
whos
feature('memstats')

As a programmer, only you can be held accountable for understanding the language's memory-management system; ignorance of dynamic memory management is not a justification for blaming the language's performance. Nimur (talk) 16:13, 19 September 2009 (UTC)

Unless they are the languages built-in functions that are not doing their work properly. It hasn't been noted that a good C++ program will be faster than matlab because much more time is devoted the development of C++ compilers than matlab. C++ certainly has more flexible and efficient data structures. Thinking about your program and making an "optimal" design rather than putting together something that works quite well might be a good thing in the supervisor's eyes. --194.197.235.240 (talk) 00:22, 20 September 2009 (UTC)

Forgive me for my loud personal opinions, but I strongly disagree with your assertions: more time was devoted to the C++ compiler than to MATLAB? Do you have a source for that? MATLAB has been in continuous development for decades - longer than C++, actually. By the time C++ was conceived, let alone implemented, MATLAB was already a mainstream commercial product available on several architectures and operating systems. Significant work has been done to optimize MATLAB's runtime - for example, automatic copy-on-modify, a feature that (to my knowledge) exists in no commercial or free C++ toolkit, that can reduce memory copy times by orders of magnitude. "Flexible and efficient data structures" begs for a definition of "flexible" and "efficient" - again, I point to MATLAB's ease of portability, compared to C++, as an example of "flexibility". As far as efficiency, again we must revert to the side-by-side benchmarking for a specific feature. I'm a happy C++ and FORTRAN-90 programmer, and I recognize the desire to stick to "native compiled code", but seriously - the vast majority of complaints levied against interpreted languages are buried in fiction. Perhaps you would like to run the benchmarks yourself? You might find that empirical data contradicts "intuitive" preconceptions about "compiled"-vs.-"interpreted" languages. A modern operating system provides so many degrees of separation from the bare metal, anyway - system interrupts, software trap vectors, emulated processors, non-native machine instruction handling, dynamic runtime compiling, fat binaries, virtual memory and paging management - that the notion of "compiled program" is really an antiquated idea. (Try linking without the system library, if you don't believe me! You may seize up at your first "printf"). Unless you run without an operating system, you should be very wary of any claims about programs that run on "raw hardware" because they are "compiled" as opposed to "interpreted." You might find that writing C++ that can actually run on a high-performance Nehalem (without a Linux or a Unix or a Windows Vista to help you stagger along in the platform-specific darkness) is not so very easy after all! But of course, by reputation, we grant C++ and FORTRAN a special place as "native code" because that's historically what they have always been called; and Java and MATLAB are "interpreted". Empirical comparisons are rarely considered when indoctrinating students to prefer one programming methodology over another. I say again - for a specific problem, the only real, relevant benchmark is a side-by-side comparison of equivalent implementations. Nimur (talk) 20:59, 20 September 2009 (UTC)

It feels quite intuitive to me that more time would have been spent in the development of gcc / microsoft visual stuff / whatever, but my feelings are often wrong. I'm not disagreeing with you, but for the sake of indoctrination I must say that if you are smarter and have more time than the matlab developers, you can write a nicer c program than their matlab. Nobody hasn't yet written a self-compiling matlab compiler (?), so I guess any hours spent on c(++) compilers benefit matlab too. --194.197.235.240 (talk) 22:52, 20 September 2009 (UTC)

Appscene

Does anybody know how to download apps from appscene.org? I had some trouble locating a place that will let me browse and download apps.

--Dudforreal (talk) 13:58, 18 September 2009 (UTC)

Contributors sued, assaulted?

Have contributors been sued, or physically assaulted? Perhaps this is not the right place to ask, but I have some concerns about possible retribution, if I were to contribute to certain articles, which I have not done. --DThomsen8 (talk) 15:42, 18 September 2009 (UTC)

People or wikipedia have been sued (or threatened with legal proceedings) in the cases where incorrect (and personally offensive) biographical information has been posted see WP:BIOG for guidelines.

As for physical assault, I'm not aware of this. Excluding the possibility that you live in a heavily censored country where internet activity is monitored and human rights not respected in terms of free speech, etc...

Where you thinking of something in particular? a particular article or set of articles?? Why do you have this concern?83.100.251.196 (talk) 16:01, 18 September 2009 (UTC)

(ec) As far as I'm aware, there haven't been any major newsworthy incidents of wikipedia-based retribution against editors in the form of "assault". That being said, you should not use any Wikimedia projects to break any laws or to commit libel. Wikipedia has policies to preclude you from committing libel; and Wikipedia must comply with the laws (particularly those of the United States), which may involve releasing information to investigators. If you feel that your contribution is important, but must be contributed anonymously to avoid retribution, you might want to read about Wikileaks (an unaffiliated project that is not part of the Wikimedia Foundation or Wikipedia). This will allow you to post sensitive information you have come across while remaining "more" anonymous. However, you should be aware that with sufficient effort, it may still be possible for somebody to locate you. Given your earlier question on the Village Pump, I strongly recommend you read our policies about biographies of living people. This is one of the most strictly enforced policies, as the people who run Wikipedia do not want to propagate libelous material. If you are unsure of the legal implications of your contributions, or the possible ramifications, you need to contact an attorney in your area, because we can not give you legal advice. Nimur (talk) 16:06, 18 September 2009 (UTC)

Probably Wikipedia:Help desk would be better, It is allowed to have a second account to protect your identity in a particular area but you need to be very careful not to mix the two in the same area or use an ip instead. See WP:SOCK#LEGIT Dmcq (talk) 16:16, 18 September 2009 (UTC)

Check Wikipedia:Wikipedia as a court source. ---— Gadget850 (Ed) ^talk 16:39, 18 September 2009 (UTC)

one of the wikipedia admins, i think allison or something, was actually stalked for a while. then the guy was imprisoned. not much information on it though, i think the oversighters have been at work

Problem sending e-mails after Verizon blocked port 25

Maybe there is someone out there who had and solved the same problem? Verizon recently blocked port 25 for outgoing mail and advised customers to change it to port 587 (for outgoing mail). Unfortunately this doesn't seem to work for me after trying with several e-mail clients I have in use. Receiving e-mails works just fine on port 110 but what can I do to be able to send mail again? Help is much appreciated. Thanks, The Magnificent Clean-keeper (talk) 16:33, 18 September 2009 (UTC)

PS: Verizon's customer service isn't very helpful and I'm kinda tired of being placed on hold etc. The Magnificent Clean-keeper (talk) 16:33, 18 September 2009 (UTC)

Could you explain with as many details as possible what happens when you do try port 587 outgoing? (I had to switch from 25 to 26 when my ISP did a similar thing. Outlook was unperturbed, AFAIK.) Comet Tuttle (talk) 16:44, 18 September 2009 (UTC)

Sure. With "mailwasher pro" I get this error message: Connection timed out (possibly caused by a firewall) while with "Outlook Express" I get: The connection to the server has failed. Account:'[my e-mail here]', Server:'[my server detail here]', Protocol: SMTP, Port: 587, Secure(SSL): No, Socket Error: 10060, Error Number: Ox800CCC0E. While trying, I had only Windows XP's (SP 2) and the Verizon's router firewall enabled as I usually have. BTW, I've also tried now port 26 but without success. If you need more info let me know which and I'll check on it. Thanks for looking into this :) The Magnificent Clean-keeper (talk) 17:18, 18 September 2009 (UTC)

The obvious question that comes to my mind: does your SMTP server actually listen on port 26, 587, or any other port? It sounds like your ISP blocked that port; but your SMTP server has not changed its policy. One possible solution is switching to secure authentication (e.g. SSL or TLS authentication). There may be a configuration setting. You have not told us who operates your SMTP server, but I'm going to guess (a) it is a third-party, not Verizon; and (b) Verizon probably offers its own SMTP server, which you can switch to and use. Most likely, they have changed their configuration to match their new firewall setting, and expect you to use their SMTP server. Nimur (talk) 17:40, 18 September 2009 (UTC)

I was just about to paste the following when I saw your reply: Forgot to mention that I rarely use my Verizon e-mail address so I just checked it and it's still working just fine on port 25. So my problem is just with my non-Verizon accounts. Let me check on authentication again although I'm sure it's set correcly (since my 3rd party e-mail provider requires it).The Magnificent Clean-keeper (talk) 17:55, 18 September 2009 (UTC)

Nope, doesn't work. Authentication is set and I tried with SSL - no difference :( . Any way I can check if my SMTP is actually listening?The Magnificent Clean-keeper (talk) 18:03, 18 September 2009 (UTC)

• Well well, I just tried bouncing some spam from my 3rd party accounts by using Verizon as SMTP and guess what? It doesn't let me because VERIZON refused to send spam. Could it be that Verizon blocked outgoing mail from my other accounts? That would be quite, well you know what I'm trying to say. "Anger is building up".--The Magnificent Clean-keeper (talk) 18:26, 18 September 2009 (UTC)

Yes, Verizon is blocking email relaying, so in order to use their SMTP server, you have to be coming from a Verizon email account. Don't get mad at them for that, if everybody did this it would cut down on spam tremendously. As for your non-verizon accounts, unless they are set up to listen on 587, then you've got no other way to connect to them if VZ is blocking 25. You could try contacting those server providers/owners to see if they can enable 587, otherwise you're kind of stuck. Arakunem^Talk 19:03, 18 September 2009 (UTC)

Yes, the latter response is probably the reason that I had to come here to get it figured out since Verizon doesn't bother to fully inform their customers. Not the first time and not the last.... so I'm still mad at them and will be :P

Anyhow, big thanks to you all for helping me out. Cheers, The Magnificent Clean-keeper (talk) 19:29, 18 September 2009 (UTC)

This is standard for many ISPs. You can collect mail from just about any account on any port - that is fine. You can send mail through their service using your account with them and that's it. For example, my ISP only allows outgoing mail to go through their mail server. So, for my personal email, I check it on my own server. For my student email, I check it on my university account. For my work email, I check it with my work account. When I send mail - no matter which account I have in the "from" field, it is sent through my ISPs mail server. The recipient doesn't know this. They see the from as being from my personal, student, or work account. Only by looking at the email headers will they notice that the actual server doesn't match the email address. -- kainaw™ 05:57, 19 September 2009 (UTC)

Kainaw's description is the "proper" way to use the SMTP server - even if you have multiple email accounts. It's rare to expect that the SMTP server should necessarily match the from: or reply-to: address. The Verizon server should permit you to use its service because you are on a Verizon IP (with or without authentication, based on their policy). Alternatively, you need to authenticate to a third-party SMTP server; this really should be done with SSL to prevent plain-text transmission of your password, etc. The main idea is to separate the processes of outgoing- and incoming- emails. Outgoing mails can be sent from anywhere (like dropping a properly stamped letter in any mailbox) - you are not required to use a particular mailbox for outgoing email. But, to prevent spammer-abuse, most SMTP servers limit who can connect. So, use the Verizon-provided server, even for your non-Verizon email accounts, because it has been specifically configured/firewalled to allow you to use it. Nimur (talk) 06:25, 19 September 2009 (UTC)

4 gb ram but showing 3 gb ram

Hello there I have just upgraded my ram to 4 gb. But it's showing only 3 gb. I am using XP SP2 32 bit.Is it faulty ram--119.30.36.40 (talk) 18:28, 18 September 2009 (UTC)

Due to an architectural decision made long ago, if you have 4GB of physical RAM installed, Windows is only able to report a portion of the physical 4GB of RAM (ranges from ~2.75GB to 3.5GB depending on the devices installed, motherboard's chipset & BIOS). From [1] Taggart.BBS (talk) 18:45, 18 September 2009 (UTC)

To remedy this, you can install a 64-bit OS such as Vista 64-bit or Windows 7 b4-bit IF your processor is compatible. If you have a core 2 duo, core 2 quad, an Athlon X2 or other AMD series that's fairly recent or any modern processor you should be able to install a 64-bit OS and take advantage of that extra RAM. Unfortunately this is the only step you can take, and if the processor itself is only 32-bit, you can't even install a 64-bit OS, so you will need to upgrade your hardware. Make sure you look up the exact processor model if it's a Pentium 4, Pentium D, or Celeron to determine if it's 64-bit compatible! Caltsar (talk) 19:39, 18 September 2009 (UTC)

Or the 64-bit version of Windows XP, presumably. Comet Tuttle (talk) 21:27, 18 September 2009 (UTC)

Athlon 64, the first x64 processor, has been out for half a decade now. You don't need a bleeding edge CPU to run a 64-bit OS (and there are some notable recent exceptions, such as Core 1's and Atoms) --antilived^{T | C | G} 03:57, 19 September 2009 (UTC)

Compatible Board(s) for Processor "Q9400"

Hello there, I want to upgrade my mobo to Intel motherboard. I have Core 2 Quad processor which FSB is 1333 MHz. But 3 boards listed in extreme series in intel website are FSB 1600 MHz supported. Will my processor run properly on them? I am bit confused about them.--119.30.36.39 (talk) 19:17, 18 September 2009 (UTC)

If you look on your favorite tech shopping site many motherboards are listed with multiple FSB compatible speeds. In general, you should be able to use a board with a higher FSB listed than your processor has and it will clock down to 1333Mhz to work with your processor. The Intel site is a bit confusing in this regard, but many motherboards will list multiple frequencies in the specs (such as 1600/1333Mhz). Caltsar (talk) 19:34, 18 September 2009 (UTC)

If you open the link you gave, click on a motherboard, you should get somewhere like here [2]

Then click on "compatability", then "supported processors" which should take you to a page like this [3]

This gives a list of compatable processors - as you can see there are various types of core 2 quads with 1333MHz FSB - you need the type eg Q8200 etc.

If it's not on the list it's not supported.

In general there wont be a problem with a 1333Mhz FSB, but you should check first.83.100.251.196 (talk) 21:30, 18 September 2009 (UTC)

Diagrams on TCP/IP and IMPs/TIPs

I am in the final stages of writing a commissioned, non-technical book about the internet and have had 4 diagrams created for the internet history chapter to help explain 'Hub and spoke', 'network', TCP/IP and IMPs/TIPs. These diagrams are to my brief and before I set them in stone I'd like to show them to someone who understands the engineering/computing side. The book as a whole is social-cultural and that's my overall background.

This is also my first time on Wikipedia and I can't seem to upload the .pdf files for you to have a look at, but I have 4 .pdf files which have 'artwork'/B&W line images created specially for me to my brief for this purpose.

Paddy's GirlPaddy's Girl (talk) 19:42, 18 September 2009 (UTC)

You can only upload content to Wikipedia under a free license so anyone can re-use it (including commercial use); or under fair use to illustrate in an article. --h2g2bob (talk) 20:12, 18 September 2009 (UTC)

However, if you host your files on an external site (something like freepdfhosting.com), you can post the links to your files here, and we can take a look at them that way. Indeterminate (talk) 03:52, 19 September 2009 (UTC)

Geejo strikes again!

If you go here [4], this shows what happened to a Yahoo email I was working on. I hadn't touched it for two hours. Yahoo now saves drafts automatically, so I didn't lose anything.

When I started to add something to the email, this is what the cursor did. Then the screen got foggy. Then most of what was on the screen disappeared and at the top it said, "Not responding".

And now the rectangles at the bottom of the screen have switched places. I'm able to do everything fine here. It's just there.Vchimpanzee · talk · contributions · 20:29, 18 September 2009 (UTC)

Wait, now the screen is back but it's much smaller and the "Not responding" message is still there.Vchimpanzee · talk · contributions · 20:31, 18 September 2009 (UTC)

Weird. The border of the screen with the problem appeared, with a foggy image of the top, on top of what I was doing, and when I scrolled down, what was in the area inside the border didn't scroll.Vchimpanzee · talk · contributions · 20:37, 18 September 2009 (UTC)

Now the text and everything are extremely large, although the screen is small.Vchimpanzee · talk · contributions · 20:39, 18 September 2009 (UTC)

I tried clicking on the red X and it gave me the choice of restarting or closing. I tried restarting. That small screen is still there, but it's black and I can't do anything to it.Vchimpanzee · talk · contributions · 20:42, 18 September 2009 (UTC)

Restarting the computer has solved the problem. I'd still like to know what happened. The information about my computer is on here somewhere.Vchimpanzee · talk · contributions · 20:51, 18 September 2009 (UTC)

1. When your computer's mouse cursor changes into a spinning wheel or an hourglass, this is the "wait..." cursor. This usually happens (a) when the application that "owns" the window (the one that your mouse cursor is currently over) has told Windows to display the "wait..." cursor; it can also happen if (b) the application stalls and is not responding to Windows's queries; or (c) if a humorist like Geejo creates a "mouseover" effect causing this.

2. In your 2nd paragraph, above, you got the "wait..." cursor and then a "not responding" message because of 1(b) in the above paragraph.

3. Have you tried the previous suggestion of adding RAM to the computer? I think you are using a computer that is old and needs more RAM, and probably a hard disk wipe. Comet Tuttle (talk) 21:26, 18 September 2009 (UTC)

I don't need to add RAM. It's only a year old. If I could figure out where my computer information was, I'd put it here. It's somewhere in the reference desk archives now.Vchimpanzee · talk · contributions · 18:25, 19 September 2009 (UTC)

Yahoo textareas are sensitive to time-outs. If you leave something open but untouched for a while, you are better off closing that page/tab and re-opening it from the Drafts folder. If a reply or forwarding fails to work, same thing...save it as a draft, open it from that folder and send it immediately. Once a mail page goes "not responding" it never will, so close it and open a new one. - KoolerStill (talk) 16:52, 20 September 2009 (UTC)

I tend not to save drafts. It just causes problems (for example, I might never go back to the draft folder). At least Yahoo saves drafts automatically. All I'm really doing is what "Notepad" software at the library does. I don't know whether I have such a thing, but it's my way of saving information I'd like to keep. I know, since it's my own computer I can use bookmarks. If I go to a library and want to see the same thing, email is the best way.

I wonder if I should complain to Yahoo? It never happened before.

Now, regarding my RAM and my disk. Here's what I got when I looked to see if I needed to do anything.

C: 360 GB free of 454 GB. D:1.04 GB free of 11.0 GB. I'm not allowed to delete anything on D because there's only one item there and it says I need that for System Restore. There is a disk cleanup function I've never used, and it scares me to do anything right now.

I have an HP Pavilion KT369AA-ABA a6512p Intel Pentium Dual CPU E2200 @2.20 Ghz, 4 GB RAM, 64 bits. Windows is 89583-OEM-7332157-00061. I have Internet Explorer 8, though that's new as of July 13.

I defragment every week. I used to delete my history (after my uncle had to delete a bunch of stuff I had left on his, not knowing it didn't go away) until I found out that after two days it did me no good to have saved something (apparently the default is to save only the current and previous days's work). I did that so if the Internet went out, I'd have long pages to read. Didn't work, because I don't use my computer for two or three days at a time.Vchimpanzee · talk · contributions · 20:08, 21 September 2009 (UTC)

Okay, Disk Cleanup will gain me 3.66 MB. One of the checked items was Temporary Internet Files, which are gone anywya after I've been off the computer for two days. It apparently won't do anything for me.Vchimpanzee · talk · contributions · 20:14, 21 September 2009 (UTC)

Thanks for the extra information. It supports the conclusion that neither RAM nor hard disk space is the problem; you have 4GB of RAM and lots of free hard disk space. (By the way, you can cut down the defragging to like once a year; once a week is not going to speed anything up.) I'm going to continue to go with my item 1(b) above; it's a glitch in the Yahoo software or IE crashed (possibly due to the glitch?). Was this a one time event or does it occur regularly? Comet Tuttle (talk) 16:43, 22 September 2009 (UTC)

One time. IE was still running because I was doing other stuff.Vchimpanzee · talk · contributions · 16:27, 23 September 2009 (UTC)

Yahoo is messed up. I went to an email I was working on, which had been saved as a draft 20 minutes earlier. I couldn't copy and paste. The cursor wasn't blinking. I went to the draft folder. It let me in to the folder but wouldn't let me in to the email. It won't let me click on anything, even the red X in the upper right corner.

A few weeks ago I couldn't sign in because they refused to let me go any further until I provided a cell phone number for a lost password. I don't have a cell phone. Going to the help screen somehow solved the problem--until it happened again, and I got in only by pretending I had lost my password. They're running out of second chances.Vchimpanzee · talk · contributions · 19:22, 23 September 2009 (UTC)

Now I can't even click on the rectangle at the bottom of the screen.Vchimpanzee · talk · contributions · 19:23, 23 September 2009 (UTC)

Everything works on another screen, but it's inconvenient.Vchimpanzee · talk · contributions · 19:25, 23 September 2009 (UTC)

I just checked the drafts folder. What I was working on when I asked the question did NOT get saved, and I have no idea what it was or how to get it back.Vchimpanzee · talk · contributions · 19:27, 23 September 2009 (UTC)

Now I can't use the back button. I can, but it doesn't do anything. The list of folders disappeared, then the font changed. All I'm trying to do is copy part of a stupid email so I can delete the rest of it.

I found the email. There was an error message but I was somehow able to see the subject, and I found it using a search. But I'm getting a LOT of error messages. They need to get their act together.Vchimpanzee · talk · contributions · 19:36, 23 September 2009 (UTC)

Okay, now the font is different, the folders are gone, and I can't get anything to happen if I click on "Mail". I keep going back to the same place.Vchimpanzee · talk · contributions · 19:38, 23 September 2009 (UTC)

I got everything done that I need to do. I also got back the screen where I couldn't do anything, not even with the red X. I got the Geejo cursor again, but then it went away and now I have a hand. I can't click on the red X. I know who is to blame. There's an AT&T ad, and I complained about them on the reference desk earlier. I still have problems with them on Firefox with their orange ads. This one is blue.Vchimpanzee · talk · contributions · 19:43, 23 September 2009 (UTC)

One more detail: the Norton and Yahoo tool bars disappeared. The hand is now an arrow, but it won't do anything. The AT&T ad has changed.Vchimpanzee · talk · contributions · 20:19, 23 September 2009 (UTC)

I sent a report to Yahoo. As with the pasword issue, they're not bothering to do anything about this, although the password problem was an emergency. If all I do is store stuff and find another avenue for keeping a notepad while I work, I suppose they've fulfilled their responsibilities to me. Nevertheless, today I've had the back button go bad and put me in an endless loop, with an error message because one of the emails in the history got moved and another one got deleted. One of the moved emails came back up on the screen without the list of folders. I couldn't even click on "mial" and get results, though the Yahoo toolbar at the top of the screen does work.

I referred the Yahoo people to this thread. I didn't bother to tell them what the problem was because there are so many.

The forward button blinks and turns gray if the back button actually works. That's one clue.Vchimpanzee · talk · contributions · 17:18, 25 September 2009 (UTC)

At long last, a clue. A yellow triangle on the lower left part of the screen, and if I double-click on the triangle I get Line 61896993 and character 111.Vchimpanzee · talk · contributions · 21:18, 25 September 2009 (UTC)

Continued [5].Vchimpanzee · talk · contributions · 19:49, 28 September 2009 (UTC)

Downloading fonts

When I go to a free font website, I can find fonts I like, and when I download the fonts, I get a .ttf file. Do I need to do something further to make these fonts accessible to programs like Word, or is that all that I need to do? Who then was a gentleman? (talk) 22:13, 18 September 2009 (UTC)

Quit Word, then go to Control Panel -> Fonts and drag the .ttf file into the resulting window. Upon launching Word, the new font will be available. Comet Tuttle (talk) 22:54, 18 September 2009 (UTC)

Drag it from where? When I go to the page where the .ttf file is, in another Window, and try to drag it into the Control Panel pane, it doesn't work. And the files are located in C:\Documents and Settings\Compaq_Administrator\Local Settings, which is not accessible from the Control Panel. Who then was a gentleman? (talk) 23:03, 18 September 2009 (UTC)

What OS version? If you're using Windows XP, try Control Panel -> Fonts and then look in the File menu of the Fonts window for "Install New Font...". You may have to be logged in with an account that has administrator rights; I'm unsure on that point. Comet Tuttle (talk) 23:18, 18 September 2009 (UTC)

Download them to your desktop (you can put them anywhere you want, but I find having them on my desktop is 'tidier'). Assuming you're using Windows, click on Control Panel, open Appearance and Themes and click Fonts. Then click 'File', 'Install New Font' and under 'Folders' click 'Documents and Settings' and choose your username and 'Desktop'. The font(s) will then appear and you can click 'OK' and it will upload them to the font list. If you have a program that uses fonts open when you do this, you'll have to restart the program before the font can be used. HalfShadow 23:31, 18 September 2009 (UTC)

Great, getting them to the desktop worked. Thanks. Who then was a gentleman? (talk) 00:07, 19 September 2009 (UTC)