Timeline for What does the essence of anything matter, if its effect is the same?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 28 at 11:36 | comment | added | retrospace | "you can implement the same algorithm in C and in Python, and their performance will look nothing alike" This actually gives me more tweaking options to create 2 objects, that have the same input X output X performance set. So, no fallacy concluded. Now, having these 2 objects, how does their difference in essence matter? They do have the same input X output X performance set. Assuming, we can't and won't ever look inside and split it into its compartments. | |
| Apr 28 at 3:02 | comment | added | Kevin |
@retrospace: If you add so much as one nop to any given implementation of an algorithm, it will be ever so slightly slower, despite being exactly the same algorithm (albeit not exactly the same implementation). More pragmatically, you can implement the same algorithm in C and in Python, and their performance will look nothing alike. So this whole idea that performance and "is the same algorithm?" have anything to do with each other is simply fallacious.
|
|
| Apr 27 at 15:16 | comment | added | retrospace | But of course in most cases performance will be different. It is hard to find 2 different algorithms that have exactly the same input X output X performance set. Is it possible at all? And after exact match we could descend into approximate match. But exact match was in question. I like to limit it to exact match. | |
| Apr 27 at 15:09 | comment | added | retrospace | The 2 provided algorithms seem to violate my side-condition, namely that effect is the same. In this particular case, any (input X output X performance) element being identical. That is not the case, if I understood your examples correctly. | |
| Apr 27 at 14:32 | history | answered | Rushi | CC BY-SA 4.0 |