Differences

This shows you the differences between two versions of the page.

--- haas:spring2024:comporg:projects:pnc2 [2024/04/18 09:22] – [PNC2] wedge
+++ haas:spring2024:comporg:projects:pnc2 [2024/04/18 10:46] (current) – [PNC2] wedge
@@ Line 27: / Line 27: @@
 In order for it to work, we must store all the values we're processing so we can obtain what is left when done– what remains are the prime values.
+Some refined considerations:
+  * include memory management related to the sieve in your timing
+  * for better visibility of performance, take measurements out to FOUR (4) decimal places (instead of the previous 3 we did for pnc0/pnc1)
+  * for graphing purposes, you may want to exclude the brute results, so your graphs do not suffer from needless compression (you want to see how soe performs against the better brute variants; at this point we already know it'll blow baseline brute out of the water)
+  * if the line graph is too noisy (too many things, not enough clear labels), consider doing a bar graph instead.
+  * you can, optionally, choose to bypass running brute once it surpasses a certain threshold; in my example, I set a threshold of 4 seconds, which is why you do not see results provided for 8192, as it'll have exceeded the limit during the 4096 run.
 Here is the wikipedia page: