On 23 Dec 2004 09:24:48 -0500, Randy Yates <randy.yates@sonyericsson.com> wrote:> "U-CDK_CHARLES\\Charles" <"Charles Krug"@aol.com> writes: > >> On 23 Dec 2004 05:06:29 -0800, absarahmed@gmail.com >> <absarahmed@gmail.com> wrote: >> > Can ne body help me in making the C++ Programme ,,, I will b most >> > thankful to him .. plz help me >> > The Programme is as .: >> > >> >> Sure. >> >> First, decompose the task into the appropriate objects. >> >> Then implement the objects in the language of your choice. >> >> FINALLY, figure out why it doesn't work. >> >> HTH > > That's a good start, Charles, but I think you left a few things > out. > > First, decompose the task into the appropriate objects. > > Then implement the objects in the language of your choice. > > Then, after realizing that this decomposition is suboptimal, > throw away existing implementation and redecompose. > > Re-implement objects of new decomposition. > > Then receive a major requirement that will make the existing > decomposition suboptimal. Throw away existing implementation > and re-decompose. > > Re-implement objects of new decomposition. > > Begin object testing. After several days or weeks of object > testing, realize that the elegant decomposition you've crafted > sucks so many CPU cycles that your application has become > unusable. Throw away existing implementation and testing and > re-decompose. > > Re-implement objects of new decomposition. > > Test objects of new decomposition. > > Integrate objects of new decomposition. Realize that a few > objects were missing. Add those objects to the decomposed > design, implement, test, and integrate. > > Deliver project to customer.Hmmm . . . don't know, Randy, I think you're optimizing prematurly. The "Sucks so many CPU cycles" step usually happens just before "Deliver project to Customer" when you notice the phrase "realtime" in the requirements. Then you add: "Bludgeon the account manager who allowed an unqualified "Realtime" into a requirements contract with his own document" And proceed with the optimization steps you suggest.
How to do video editing through C
Started by ●December 22, 2004
Reply by ●February 2, 20052005-02-02
Reply by ●February 2, 20052005-02-02
absarahmed@gmail.com wrote:> Can ne body help me in making the C++ Programme ,,, I will b most > thankful to him .. plz help me > The Programme is as .:... Most of us try not to do other people's homework. We sometimes help them to get started -- for example, by suggesting how they might organize the work -- and we often help them over a stumbling block when they explain how they got there. Although comp.dsp is frequented by a group of friendly people who write programs and have wide knowledge, it is not about programming per se. To make effective use of the internet, you must learn to start a pertinent thread instead of usurping an existing one with a non sequitur. At the very least, change the subject line. You might find something useful at http://users.erols.com/jyavins/procfaq.htm and http://guide.vsnl.net.in/tcpip/static/usenet/netiquette.html when you find the right group for specific help. While there are many language-specific groups in the comp.lang.xxx hierarchy, and several oriented toward data bases, I know of none that specialize in fundamental program organization. You'll have to learn enough about that on your own to ask specific questions of the form "How can I [some specific task]" or "What is a linked list?" or even "Where would a doubly-linked list be useful?" Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●February 2, 20052005-02-02
> > Then you add: > > "Bludgeon the account manager who allowed an unqualified "Realtime" into > a requirements contract with his own document" > > And proceed with the optimization steps you suggest. > >Real time does not always mean milliseconds; I heard a nice counter example: the monthly salary payment data that must be at the bank at the 22th of the month, and not a day early or late. Wim
Reply by ●February 2, 20052005-02-02
Jerry Avins wrote:> absarahmed@gmail.com wrote:> While there are many language-specific groups in the comp.lang.xxx > hierarchy, and several oriented toward data bases, I know of none that > specialize in fundamental program organization. You'll have to learn > enough about that on your own to ask specific questions of the form "How > can I [some specific task]" or "What is a linked list?" or even "Where > would a doubly-linked list be useful?" > > JerryAsking "what's a good book for..." is rarely frowned upon, and if your question is large (like proper program organization) it may be more help for you than getting a slew of unorganized tidbits from USENET. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com
Reply by ●February 2, 20052005-02-02
Tim Wescott wrote:> Jerry Avins wrote: > >> absarahmed@gmail.com wrote: > >> While there are many language-specific groups in the >> comp.lang.xxx hierarchy, and several oriented toward data >> bases, I know of none that specialize in fundamental program >> organization. You'll have to learn enough about that on your >> own to ask specific questions of the form "How can I [some >> specific task]" or "What is a linked list?" or even "Where >> would a doubly-linked list be useful?" >> >> Jerry > > Asking "what's a good book for..." is rarely frowned upon, and > if your question is large (like proper program organization) it > may be more help for you than getting a slew of unorganized > tidbits from USENET.Here's some good stuff, Ahmed evidently needs it. Merry Christmas! news:comp.programming (No use just to repost the problem ;) Structure and Interpretation of Computer Programs http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-4.html Taligent's Guide to Designing Programs http://pcroot.cern.ch/TaligentDocs/TaligentOnline/DocumentRoot/1.0/Docs/books/WM/WM_3.html How to Design Programs http://www.htdp.org/2003-09-26/ -- Quidquid latine dictum sit, altum viditur.
Reply by ●February 2, 20052005-02-02
On Thu, 23 Dec 2004 21:00:44 GMT, Wim Ton <wimton@blueyonder.co.uk> wrote:>> >> Then you add: >> >> "Bludgeon the account manager who allowed an unqualified "Realtime" into >> a requirements contract with his own document" >> >> And proceed with the optimization steps you suggest. >> >> > Real time does not always mean milliseconds; I heard a nice counter example: > the monthly salary payment data that must be at the bank at the 22th of the > month, and not a day early or late. >You've the right idea. "We can do it in realtime" is an unhelpful buzzword. Sometimes realtime means, "We drop any samples we can't keep up with, as they're not all that important." And sometimes it means, "If we drop any samples, things will blow up where and when they shouldn't . . . sometimes literally." The first is easier than the second.
