Translation for readers:Let's say you have some floating point numbers, like...`````` float A = 42.90F; float B = 00.01F; float C = /* ... */; float D = /* ... */; `````` (A + B + C) will not necessarily produce the same result as (A + C + B).In fact, if you do (A + B + B + B ...) very much, then you will wind up exacerbating the problem --- it can be a source of bugs if you repeatedly increment a counter by small floating point values. For example:`````` const int Iterations = 10; for ( int i = 0; i < Iterations; ++i ) A += B; printf( "%2.4f\n", A ); `````` The output is 43.0000, as you'd expect. But what if we increase Iterations to 100? We get 43.8998.Astute readers probably would note that 00.01F can't be exactly represented in base 2, and wonder whether something like 00.0125F would still suffer from this problem. Alas, yes:`````` A = 42.9000F B = 00.0125F 8 iterations == 43.0000 80 iterations == 43.9001 800 iterations == 52.9006 8000 iterations == 142.8820 /* we would expect 142.9 */ `````` So if you want to parallelize some computations, and you want to do 4000 iterations on core #1 and 4000 iterations on core #2 and then sum the results, well... let's see what happens:`````` float A1 = 42.9000F; float B1 = 00.0125F; float A2 = 42.9000F; float B2 = 00.0125F; const int Iterations = 4000; for ( int i = 0; i < Iterations; ++i ) { A1 += B1; A2 += B2; } float A = A1 + A2 - 42.9000F; printf( "%2.4f\n", A ); `````` The output is 142.8885 -- which is different from our earlier result of 142.8820!You may be tempted to try various things to get around this fact... for example, what about counting from zero rather than from 42.9000? (Answer: the error can be even worse, which is a topic worthy of its own separate post) ... but you will quickly discover that this is a fundamental issue. The order of operations matters.

Search: