Try to make it so that all tests were within minutes, allowing the developer who made changes to the code, immediately noticed the error and correct it. If you run all the tests is not possible, usually long-term tests taken separately and run, for example, night. It also contributes to quick detection of errors, at least, the next morning. Manual testing. This is perhaps the final step of any development, but it should not be regarded as a good and reliable methodology. Manual testing must exist, since it is impossible to detect all errors in the automatic mode or code. But count on this method is not particularly worthwhile.
If the program is low quality and a large number of internal defects, its testing and correction may last for a long time and still you can not provide the proper quality of the program. The only method get a quality program – it is quality code. Therefore, we also will not consider manual testing as a complete methodology for the development of large projects. So, what do we have left that deserves the most attention in the development of large software systems? This is a static analysis and unit testing. These approaches can improve the quality and reliability of software code, and they should be given the greatest attention, although, of course, do not forget about the others. We now turn to the issue of testing 64-bit programs, since the use of our chosen methodology is facing some unpleasant difficulties.