Pi Approximation using Monte Carlo
Learning and science
interactive math visualization
SHOW HN2014
How HN took it
3 positive, 19 neutral or asking, 0 mixed, 4 critical, out of 26 comments about the project. Thread score -0.07.
Idea: 1 for, 2 againstPerformance: 0 for, 2 againstDesign: 1 for, 0 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Valuable for teaching and education
From 1 comment
Do not underestimate the value of these visualizations. They reach 1000s of kids, students, and who-knows, adults fooling around on Sundays. So, perhaps you can turn this around,…
MrQuincle on HN
Objections
Converges slowly; estimate often far from pi at low iteration counts
From 2 comments
Nice visualization. After playing with it for a while, the number never really converged to Pi. The value even reached 3.135 after 10000 iterations. Do you have any measure how…
clemsen on HNI got 3.1544 after 10000 iterations, which also didn't match my intuition of how quickly it "should" converge. So I decided to knock up a quick deterministic version in C that…
Karellen on HN
Monte Carlo is an inefficient way to compute pi
From 2 comments
I will never understand why people think Monte Carlo is such a neat solution. There is nothing random about pi. Why didn't you just pick the center of the square, use that to…
wfunction on HNMonte Carlo methods are never "seriously" used to approximate pi – there are algorithms which converge much, much faster. This here, however, is not a "serious" attempt at…
adwn on HN
Requests
Graph of pi approximation versus n
From 1 comment
Nice visualization! Some UI comments I think the reset button should be smaller (you usually don't want to reset the simulation). I expected that the number in the upper right…
gus_massa on HN
Error estimates or bounds
From 1 comment
Nice visualization! Some UI comments I think the reset button should be smaller (you usually don't want to reset the simulation). I expected that the number in the upper right…
gus_massa on HN
Advanced mode and smaller reset button
From 1 comment
Nice visualization! Some UI comments I think the reset button should be smaller (you usually don't want to reset the simulation). I expected that the number in the upper right…
gus_massa on HN
Questions
Why does the estimate overshoot more than undershoot?
From 1 comment
I notice that it seems to overshoot more than it undershoots. Any idea why this is?
skykooler on HN
How many iterations are needed for stable digits?
From 1 comment
Nice visualization. After playing with it for a while, the number never really converged to Pi. The value even reached 3.135 after 10000 iterations. Do you have any measure how…
clemsen on HN
Alternatives named
Deterministic grid sampling
From 2 comments
I got 3.1544 after 10000 iterations, which also didn't match my intuition of how quickly it "should" converge. So I decided to knock up a quick deterministic version in C that…
Karellen on HNI will never understand why people think Monte Carlo is such a neat solution. There is nothing random about pi. Why didn't you just pick the center of the square, use that to…
wfunction on HN
Quasi-Monte Carlo with Sobol sequences
From 2 comments
It would be interesting to see how quickly a quasi-random, low-discrepancy sequence converges to PI (e.g. a Sobol Sequence).
brotchie on HNThe comment has a point. It is possible that the particular deterministic rule that chooses points could interact with the shape (quarter circle). In this case, the error would…
mturmon on HN