Thank you guys, for all your answers, there were very helpful for me. I decided that I will try to find job as cpp programmer and, if someone will hire me, to survive in a company. Also, as a side project, I'm starting to learn more easier and modern language. Thank you once again. It's great to have opportunity to advise older colleagues.

I think I have breakdown... As you can see on my profile I started to learn C++ in 2006 when second year of my tech high-school class begun. I fell in love with it and I learned it as much as I can, even when the class in school ended. I know C++. I know what's going on under the hood. From the basic types to templates and containers. After I went to University I have break in my relationship with C++, but after all I started to learn Qt to have tool for multi-platform programming. And somewhere in the middle of this I started to looking for some job. And here we have crucial point. As I can see, there is some really huge hype, mostly in US, on start-ups. Today, everything is start-up. Every little idea is becoming start-up. And all that start-ups are mostly web-based applications written in high-level programming languages which (according to the authors of languages) you can learn in 24hours. And if you search for some jobs 90% of companies are looking for web or mobile developer. So, my question is. Is desktop programming still providing employment? Is it worth to learn secrets of such complicated language as C++? Isn't it extinct? I know that there are plenty of problems that can be resolved with using cpp but why I can't find such job offer? Do you know the answer? * Because after High-School I went to Univ. I still don't have professional experience.

Thanks for message, and wishing me luck! :) It's telecommunication company. They are creating embedded software for real-time systems (vxWork) using wireless connections (LTE, CDMA, 1xAdvanced). Now, do you have idea what they can require?

Hi guys! Now I'm in my last years of studies (Computer-Science) so i decided to apply for some company. I send my resume to few companies and in few days I have a test in one of them. Before question here is job description:

excellent, practical knowledge of C/C++
knowledge of UML
practical knowledge of real time systems
good understanding of TCP/IP protocols

Because I never worked as a programmer in company, and I never had such a test, I'm wondering what should I expect. From C++ is more than likely that there will be: classes, templates, pointers (they didn't wrote about some libraries in job description). What's more from C++? What about UML, TCP/IP and real time systems? What should I revise before the test? Best, Luke.

Thank's for replies. It is because I have only 800x600 resolution on VM and it's just pissing me off. When I click Fullscreen the main window is going fullscreen, but the desktop of virtual Kubuntu is still on 800x600. And on that resolution I just can't work. So if you know solution for this problem I will be glad too :)

Hi folks! I’m using Windows 7 and I did disk image via O&O DiskImage. Now I want to install Kubuntu to run some tests on it (I don’t want to use virtual machine). After tests I will not need Kubuntu. As far as I’m concerned GRUB will save some files in MBR to make a ‘choose operating system’ window. Is restoring a disk from image overwriting a MBR and because of that the Kubuntu will be gone? Ofcourse I'm just only asking about GRUB and all stuff that will land on my C:\ partition after installation. Partition with Kubuntu I will delete by some partition editor :)

Hi guys! I'm concerning situation, what will do the job quicker. One server hdd, for example WD VelociRaptor, 3.5'', 300GB, Serial ATA/300, 10000RPM, 16MB cache, or two WD Caviar Black, 3.5'', 1TB, SATA/600, 7200RPM, 64MB cache connected into RAID 0. For me, intuitively raid will be quicker, but some test or personal experience will be useful. Unfortunately I never worked with raid so I only know the theory about it. I know that there is double chance to data loss with raid 0 but I'm planning to buy third external hdd which will be working as a backup storage. I don't wan't to create raid 1+0 because of lack of founds to buy more hdd's. In the price of one 300GB server disk I can get two 1 TB disks. What are you thinking about such situation? The computer will work as a render machine, mostly for 3ds max and autocad as well, so the speed of disk is quite important matter for me. Oh! And I'm planning to use raid controller to relieve processor from counting all that stripe stuff.

So, you are saying that default and maximum size of stack size is 1 MB? I solved this problem with using vector, but still I want to know a C++ mechanisms :)

No, exact size doesn't matter. Just more than approximately 400.000 numbers.

Maybe someone in the future will be searching solution for similar problem so I wrote it here. So the problem is with stack array. Array is so large that it overflows the stack. Solution is to use dynamically sized array. It can be vector class. In this example, it is sufficient to send pointer to function and in the sorting function working on vector. That's all. // Thanks to helios for helping me out (from other cplusplus forum)

hmm, I didn't thought about this, because algorithm works fine, but I did it as you said, and just after debugger started error message came out Process stoped. Process was stopped because it received signal from operating system. Name of signal :SIGSEGV Meaning of singal : Segmentation fault Now it is coming to be more interesting..... ;) I can only click . After click nothing happened. Even debugging is still on, but do nothing. I read about SIGSEGV on wiki, but how to avoid it?

Hi! I have to do a project for my studies. It is a insertion algorithm. I'm reading data from file with using vector. then I'm copying vector to a normal array (the reason is simple: I don't know how to send vector to function, but this is not a main problem). After copying, I'm sending (with using pointers) array to sorting function. Everything is good, algorithm is working but! (yep, there is always but... ;) When I save a lot of data to file, program just after start crashes. The numbers in file are written in that way:

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

So there are 40 numbers in one line. I have 4.000 lines, so after simple math we have 160.000 numbers. The main problem: when I add some new numbers for example 40.000 program crashes I must have more numbers because the task is to compare few sorting algorithms with minimum 15 minutes of sorting. (160.000 digits is sorting in ~1 minute) Here's the code. Maybe it is poor written, but I did my best.

//written with Qt
#include <QtCore/QCoreApplication>
#include <iostream>
#include <time.h>
#include <iomanip>
#include <string>
#include <fstream>
#include <vector>

using namespace std;

//sorting function
void insertion_sort(int *wsk, unsigned long size);

int main(int argc, char *argv[])
{
QCoreApplication a(argc, argv);

//vector to hold data from file
vector <int> array\_vector;

//stream to file
ifstream plik ("numbers.txt",ios::in);

//single number that is being readed from file    
unsigned long single\_number;


while(plik >> single\_number)
{
    array\_vector.push\_back(single\_number);
}

int array\_nonsorted\[array\_vector.size()\];
//copying from vector to array
for (unsigned long a = 0; a < array\_vector.size(); a++)
{
    array\_nonsorted\[a\] = array\_vector\[a\];
}

time\_t start,end; double long dif;

cout << "START\\n";
time(&start);
insertion\_sort(array\_nonsorted,array\_vector.size());
time(&end);
cout << "STOP\\n";
dif = difftime(end,start);
cout << "Array sorted in " << setprecision(5) << dif << "sec"; dif = dif/60; cout << " = " << dif << "minutes";
//cout << "End.";
return a.exec();

}
void insertion_sort(int *wsk, unsigned long size)
{
int tab_sorted[size];

//In final version