Help with binary trees benchmark games example

General Usage
15

I’ve been unable to get a @nexprs based solution to be any faster, and it seems a bit silly to pull in a dependency for something like this. Any suggestions?

@@ -30,7 +30,8 @@
             for _=1:5
                 r, i, t = calc_sum(r, i, cr, ci)
             end
-            all(>(4.0), t) && return 0x00
+            @nexprs 8 k-> t[k] > 4.0 || continue
+            return 0x00
         end
     else
         for _=1:50

Actually, I’m not seeing any difference on my computer from using Baselet.all either:

--- mandelbrot.julia-7.jl	2020-03-03 15:35:30.625960909 -0500
+++ mandelbrot.julia-7.2.jl	2020-04-17 09:51:39.914201367 -0400
@@ -8,7 +8,7 @@
  tweaked for performance by maltezfaria and Adam Beckmeyer
 =#
 
-using Base.Cartesian
+using Base.Cartesian, Baselet
 
 # 0b01111111, 0b10111111, 0b11011111, 0b11101111, etc.
 const masks = (0x7f, 0xbf, 0xdf, 0xef, 0xf7, 0xfb, 0xfd, 0xfe)
@@ -30,7 +30,7 @@
             for _=1:5
                 r, i, t = calc_sum(r, i, cr, ci)
             end
-            all(>(4.0), t) && return 0x00
+            Baselet.all(>(4.0), t) && return 0x00
         end
     else
         for _=1:50

Any thoughts on why you’re seeing a better result? Maybe an AVX2 instruction is being used on your CPU?

julia> using BenchmarkTools

julia> module M7 include("mandelbrot.julia-7.jl") end
Main.M7

julia> module M72 include("mandelbrot.julia-7.2.jl") end
Main.M72

julia> @benchmark M7.mandelbrot($devnull, $16000)
BenchmarkTools.Trial: 
  memory estimate:  42.73 MiB
  allocs estimate:  96031
  --------------
  minimum time:     1.213 s (0.06% GC)
  median time:      1.216 s (0.08% GC)
  mean time:        1.218 s (0.34% GC)
  maximum time:     1.229 s (1.29% GC)
  --------------
  samples:          5
  evals/sample:     1

julia> @benchmark M72.mandelbrot($devnull, $16000)
BenchmarkTools.Trial: 
  memory estimate:  42.73 MiB
  allocs estimate:  96031
  --------------
  minimum time:     1.242 s (0.06% GC)
  median time:      1.246 s (0.08% GC)
  mean time:        1.248 s (0.35% GC)
  maximum time:     1.259 s (1.30% GC)
  --------------
  samples:          5
  evals/sample:     1

(@v1) pkg> st
Status `~/.julia/environments/v1/Project.toml`
  [9718e550] Baselet v0.1.0
  [6e4b80f9] BenchmarkTools v0.5.0

julia> versioninfo()
Julia Version 1.4.1
Commit 381693d3df* (2020-04-14 17:20 UTC)
Platform Info:
  OS: Linux (x86_64-unknown-linux-gnu)
  CPU: Intel(R) Core(TM) i5-3470 CPU @ 3.20GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-8.0.1 (ORCJIT, ivybridge)
Environment:
  JULIA_NUM_THREADS = 4