Posts by rubberman

Algorithm:

if remainder < divisor then
    if remainder >= 1/2 divisor then round up
    else round down
endif

What JorgeM said. If you want general information, then visit our friends Google and Wikipedia.

Remember, Google and Wikipedia are your friends!

There are a number of ways to do it. A good way to start is a live usb drive you create with the tool unetbootin: http://unetbootin.sourceforge.net/ That will create a live usb (bootable) drive without touching your computer disc drives. A 4GB thumb drive will do nicely.

The other way is to create a persistent usb image that you can write to just like any disc. Like the live usb, it won't touch your computer discs without you doing so specifically. There are a number of distributions that have such images and tools to install them including Ubuntu, Mint, and others. See their web sites for the tools, images, and instructions. Again, a 4GB or larger thumb drive will work nicely. If I were creating a persistent usb image I would use a 32GB thumb drive as that would have plenty of room for storing documents, installing new software, etc.

Yes it does matter, however most *nix systems these days are running x86 processors so should have the same results, assuming that they have the same size word. That said, most *nix systems are 64-bits any longer, so you might want to specify a long long int (64-bits) instead of a simple int (32-bits).

First rule for Dell computers - always send it in to Dell for repair. Do that now, and demand your $$ back from Fry's. If what they did causes additional costs from Dell, demand them back also. Dell's repair/replacement policies are usually quite reasonable, and you are guaranteed to have a functional system when you get it back.

Since you reset the device, you will need to go into the setup page (using the default admin account "admin" - no password I think, or "admin" if needed) and reconfigure the router again. For most Linksys routers that would be 192.168.1.1

You did ask one question - what are nested classes? So, I will answer that. The other issues I assume are class problems, and we don't do homework...

Nested (inner) classes are classes declared/defined in the context of another (outer) class. They can be public or private to the outer class as desired. Here is an example:

class foo {
public:
    class bar {
    public:
        bar() { std::cout << "instantiating foo::bar instance" << std::endl; }
    };
    foo() { std::cout << "instantiating foo instance" << std::endl; }
    bar& getBarMember();
private:
    bar m_barmember;
};

foo::bar& foo::getBarMember()
{
    return m_barmember;
}

int main(void)
{
    foo aFooInstance;
    foo::bar& bm = aFooInstance.getBarMember();
    return 0;
}

You should get the following output when you run the program:
instantiating foo::bar instance
instantiating foo instance

AND you need to implement all of the methods specified in the Serializable class.

Not having a mac I can't say how to configure Xcode to accept the new compiler; however, my wife does have one and does serious computer systems work for a major physics lab, so I'll ask her. Just be patient - she's getting ready for a family visit tomorrow... :-)

Not a bad starting server, but the power supply is on the meager side. You should have at least a 750W supply (instead of the included 400W one) for a server this size, especially as you add stuff to it. One other thing is that it doesn't seem to come with keyboard, mouse, or display (but does have built-in audio and video w/ DVI and VGA ports). You might be able to configure it to allow console access via the network interface so you can use a laptop or other system to remote in via the network.

The ntpd daemon needs to be able to set the hardware clock, and it can't do that without root permissions.

What kernel are you running? I'm looking at my RHEL 6.2 system (latest 2.6.32 kernel) and there is no such configuration setting. What I have similar to that are as follows:

# CONFIG_SECURITY_DMESG_RESTRICT is not set
CONFIG_SECURITY=y
CONFIG_SECURITYFS=y
CONFIG_SECURITY_NETWORK=y
CONFIG_SECURITY_NETWORK_XFRM=y
# CONFIG_SECURITY_PATH is not set
CONFIG_SECURITY_FILE_CAPABILITIES=y
# CONFIG_SECURITY_ROOTPLUG is not set
CONFIG_SECURITY_SELINUX=y
CONFIG_SECURITY_SELINUX_BOOTPARAM=y
CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=1
CONFIG_SECURITY_SELINUX_DISABLE=y
CONFIG_SECURITY_SELINUX_DEVELOP=y
CONFIG_SECURITY_SELINUX_AVC_STATS=y
CONFIG_SECURITY_SELINUX_CHECKREQPROT_VALUE=1
# CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX is not set
# CONFIG_SECURITY_SMACK is not set
# CONFIG_SECURITY_TOMOYO is not set

They all vary. Some use pattern recognition. Some look at what system api's are accessed and how. Others apply more more sophisticated neural network recognition software. Why do you want to know?

It sounds like a possible firmware problem. Do a complete flash reset on the computer - not sure what the procedure is for this unit (it will be found on the manufacturer's web site most likely), but usually it is something like this:

1. Shut down computer. Unplug from wall power.
2. Remove battery (if a laptop). If not a laptop, remove the flash/bios battery (looks like a coin) from the motherboard.
3. Hold down power key for 10-30 seconds.

If that doesn't work, they download a new flash/bios package from the manufacturer's site and reflash the system. Then, reinstall.

One final possibility is that the system has become infected with a rootkit. I have experienced some that actually infect the recovery partition as well, so doing a reset of the OS to "factory" settings that way will still result in a fubar'd system.

If you are running Windows, there should be a function key combination (look for the blue FN key, and a function key at the top of the keyboard that looks like a display) to switch between monitors. That should work. Another approach is if your system supports multiple monitors, you could plug the display into the other port.

Time to read the x86 assembler manual. You should be able to find a copy on the Intel web site...

The simple explanation of static members:

Static members exist on their own, but are members of the class. They can be accessed by instance objects of the class, or other stuff that has permissions to access it.

Non-static members only exist within an instance object of the class. They can ONLY be accessed within the context of an instance object.

Everything else was said before... :-)

And if you REALLY want to go minimal, try to install Gentoo! You can configure it to ONLY install the components (drivers included) that you need. However, expect a week of sundays to get up to speed on its peculiarities! :-)

It looks like you are trying to initialize an array of integers. From the way you have coded it, the compiler should have at least issued a loud warning. In any case, you don't show what your output is... :-(

Google is your friend! See this wikipedia article: http://en.wikipedia.org/wiki/Virtual_terminal
Or this from Microsoft: http://technet.microsoft.com/en-us/library/cc978134.aspx

Just my opinion - Microsoft Security Essentials? See the following (short version):

1. Uninstall Windows
2. Install Linux
3. Done!

Or walk into traffic and get run over by a bus! :-)
That said, do elaborate on your concerns.

Current Ubuntu distributions by default will require about 2GB or more of RAM. You can install versions of Ubuntu (or Debian/Mint) that use less, or other Linux distributions that are designed to run "lean and mean". That said, your browser itself will take anywhere from 200MB to 2+GB of RAM, assuming you use something like Firefox or Chrome. For example, my Chrome instance, with 4 tabs open, uses about 700MB. The basic OS + GUI (Red Hat 6.4 + gnome 2.x) takes about 600MB. My current RAM usage (running Chrome, VLC to stream some music) and file browser is about 1.3GB.

Myself, I always keep the original source in the original form, and convert that to PDF for publication. If someone needs to do additional processing and PDF isn't "convenient", then if they ask (nicely), I'll send them the original source document. If it's a book, then it may be in Framemaker, which isn't much more useful than the PDF. If it's a simple article or tech doc, then it will probably be in ODF (Open Office) format. It is still their problem to extract the information they need...

I'm drinking a nice glass of Jameson's Irish Whiskey (triple distilled). Going to bed soon... :-)

My current workstation at home is 5 1/2 years old - I had it built by a local white-box vendor to my exact specs. It is still near to state-of-the-art. Intel server/workstation mobo, dual 3GHz quad-core processors, 8GB ECC RAM (I would add more if building one today, but it still is enough for my use), hot-swappable 320GB system drive, 4x500GB data drives (and more now), nVidia 8800GT video w/ dual 1920x1200 HD displays, dual CD/DVD RW drives. I run Scientific Linux 6.4 on it right now (a clone of RHEL 6.4) and run Windows, Solaris, QNX, and FreeDOS in virtual machines (VirtualBox). I also use it for embedded ARM kernel development using a gnueabi tool chain and an Qemu ARM emulator.

It cost me about $5K to build originally - much less than the comparable Apple Mac Pro that I was considering. Not cheap, I know, but consider what I got for that - 8 3GHz cores, 8GB RAM, 2.3TB disc, dual HD displays, dual DVD burners, keyboard, and trackball/mouse, along with the tower enclosure and 1KW power supply. And I haven't even cranked up the GPU floating point processors as yet (I think there are 128 or so of them)! :-)

FWIW, after 30+ years of complex software development, I still us that approach to understanding the problem so I can code a solution for it.

That's why its called home "work". :-) Start by writing out how you might approach the problem, in plain language. That will help you get your mind around the problem, which is 1/2 the way to solving it.

Banfa is quite correct about floating point representations. Trying to get an exact comparison of floats is not simple due to the inability to exactly represent a fractional number like you are trying to do. Integer values are a lot more precise in computer terms since they can be perfectly represented as a set of bits.

In your case, if you used doubles instead of floats, and cast the decimal values accordingly, you might get away with it... :-) Caveat Programmer!

Please elaborate. The term "data representation" can mean a number of things. So, let's be sure we are speaking the same language! :-)

As an aside, if you are inferring "data-driven development", that is a subject I do know something about. However, a fuller explanation of what problem you are trying to solve, and the environment would help focus our efforts on your behalf.

What JorgeM said. Protocols are the rules by which we communicate. They can be implemented in hardware/firmware, or in software. I have implemented the complete TCP/IP protocol stack in software from the protocol definitions published by the US DOD in the DDN (Defense Department Networking) Protocol Handbooks. I have also worked with implementations in hardware of the same. FWIW, the handbooks take up a good foot of shelf space... Interesting if you like reading compressed BNF (Backus-Naur Form) grammars! :-)

When Bruce gets back from his current vacation in BC, Canada I'll ask him what's what with current Intel designs... :-)

L7Sqr gave a great synopsis of this. It is not a simple subject. If you want more info, see this wikipedia article: https://en.wikipedia.org/wiki/IEEE_754-1985

FWIW, my best college buddy, Bruce Ravenel, was the architect of the Intel 8087 math co-processor back in the late 1970's, the first implementation of the IEEE 754 standard in hardware. It is still part of our Pentium processors today. FWIW, the 8087 did all computations in 80 bits (not 64), resulting in incredibly accurate computations for the time, pretty much eliminating all rounding errors.

Ok. Continued reading indicates that the 8087 design is not part of the current Intel chip sets, though it has to have been a major contributor to the current math capabilities of Intel's processors. I think they are using 128 bits now... :-)

Some newer system will require an unlock key code from the manufacturer. Go to their web site and ask their support staff.

As an addendum, I have observed some viruses that infect dll's in the recovery partition, so even if you run a system recovery operation, you are still infected! I call this "digital whack-a-mole"... This is why I always recommend to my clients that they pay for and get the "optional" installation DVD from their system vendor, and to find another vendor if such is not available.

Often, detection is a 2 (or more) step process. You install a clean system, and take "signatures" (usually checksums) of all files and directories on the system. Then you let the virus infect the system, take it off-line, and using a different operating system, mounting the disc in a "sandbox", you compute the signatures of all the files (and boot sector of the disc), and determine what has been changed. That is the beginning of the analysis. Next you have to determine what changes were made to the files whose signatures were changed. That will lead you to the point where you can determine (usually - sometimes) how to identify this particular virus. Of course, each virus has different characteristics, and modern ones obfuscate their identifying signatures by not infecting any system the same way. Some now even don't bother to modify files, but modify RAM memory, and write themselves to the system virtual memory swapfile so that the next time you boot you are infected again... Very difficult to detect and remove. Sometimes the only solution is a full wipe of the disc and re-installation of the operating systems! :-(

Hey oldman1943 - if that is your BD, you are the same age as my sister! And trust me, she is still hot! :-) Me, you could call rubberman1948...

Your system is running Windows? If so, then you probably are part of a bot-net... :-(

If 3Y is an integer value (up to 64-bits), then to divide by 8 you can use a right-shift operator, shifting 3 places (2^3 == 8). IE,
int64_t result = ((3*Y) >> 3);
This won't work for floating point values, however.

Or just type :lol:, which will bypass most spell checkers, and turn into the appropriate smiley on some web sites! :lol: :-)

That is the speed of the RAM chips and channel. If your RAM modules are 1333 capable, and the channel supports that, then you get about a 1/3 memory I/O speedup, which can be significant - hence my comment about "faster == better". :-)

What Moschops and Banfa said. In any case, the short answer is "it is implementation dependent". This is similar to the order that arguments passed to a function are evaluated. The compiler writer is free to evaluate them right to left, or left to right. So, don't make any assumptions!

Well, we don't do your homework for you. Make an effort and we might be able to help you...

THere is no way, effectively to "dodge" this, other than by manually patching the machine code of the application itself to jump around the "system()" call.

Faster is better. 64-bits is better than 32. ECC memory is better because your computer will continue to run if there is a memory chip problem. As for 8GB of RAM for Win7-32bits? Don't bother - it can only use 4GB of ram, which is why you REALLY want to install the 64-bit version. It can still run 32-bit applications.

In my words, Java == C++ with training wheels. The biggest issue is which components to "include", much like C and C++... :-)

You cannot use static libraries as a "plug-in", though as you surmise you can link it directly to the application code. However, that means it becomes not a plug-in, but a native component of the application. That also may mean that other plug-ins that implement the same functions (signatures) as your bit will not "plug in" because of symbol collision. An example of this is ODBC (database API's). Your user may want to use an Oracle, MySQL, Postgres, or other version of ODBC. If you linked your ODBC library statically, then they would not be able to use those others... Not a good thing probably.

I think it means that the text it is parsing has to be in standard 8-bit ascii or utf-8 format. IE, if your locale (or the user's) is set to Chinese or such (unicode - 2 bytes / character instead of 1), then it cannot parse the text as a number.

SMART are diagnostics and monitoring software built into most modern discs. There are tools to view what it has discovered about the disc. However, from what you are saying, I would say that the disc is fubar and does need to be replaced. Some of the more important information SMART can monitor about the disc:

1. Read Error Rate
2. Reallocated Sector Count (a critical metric)
3. Seek Error Rate
4. Temperature
5. Uncorrectable Sector Count

1. Yes, I know they (argv strings) are modifiable, but IMO not a good idea! It is easy to force a buffer overrun, creating a potential security breach! :-)

2. The getopt() function is a POSIX standard. From the Linux man pages:

   Base Definitions volume of IEEE Std 1003.1-2001, <unistd.h>, the Shell and Utilities volume of IEEE Std 1003.1-2001

3. I agree! :-)