Updated benchmarks in readme

README.md

@@ -16,23 +16,38 @@ develop native C modules for CPU-intensive tasks. Enjoy the result!
 __Benchmarks:__
 
 ```
-node v0.10, 10mb sample:
+node v0.10.26, 1mb sample:
 
-   deflate-gildas x 4.62 ops/sec ±1.93% (15 runs sampled)
-   deflate-imaya x 3.82 ops/sec ±1.31% (13 runs sampled)
- ! deflate-pako x 9.09 ops/sec ±0.42% (26 runs sampled)
-   deflate-zlib x 14.16 ops/sec ±3.33% (63 runs sampled)
+   deflate-dankogai x 4.74 ops/sec ±0.68% (15 runs sampled)
+   deflate-gildas x 4.61 ops/sec ±1.73% (15 runs sampled)
+   deflate-imaya x 3.10 ops/sec ±3.73% (11 runs sampled)
+ ! deflate-pako x 7.11 ops/sec ±0.26% (21 runs sampled)
+   deflate-pako-untyped x 4.34 ops/sec ±1.35% (14 runs sampled)
+   deflate-zlib x 14.34 ops/sec ±2.90% (68 runs sampled)
+   inflate-dankogai x 31.29 ops/sec ±0.72% (56 runs sampled)
+   inflate-imaya x 30.49 ops/sec ±0.84% (53 runs sampled)
+ ! inflate-pako x 70.00 ops/sec ±1.60% (71 runs sampled)
+   inflate-pako-untyped x 17.67 ops/sec ±1.27% (33 runs sampled)
+   inflate-zlib x 70.82 ops/sec ±1.69% (81 runs sampled)
 
-node v0.11, 10mb sample:
+node v0.11.11, 1mb sample:
 
-   deflate-gildas x 5.10 ops/sec ±4.94% (16 runs sampled)
-   deflate-imaya x 3.42 ops/sec ±4.11% (12 runs sampled)
- ! deflate-pako x 11.28 ops/sec ±0.42% (32 runs sampled)
-   deflate-zlib x 14.17 ops/sec ±3.34% (64 runs sampled)
+   deflate-dankogai x 5.61 ops/sec ±0.30% (17 runs sampled)
+   deflate-gildas x 4.97 ops/sec ±5.68% (16 runs sampled)
+   deflate-imaya x 3.53 ops/sec ±4.19% (12 runs sampled)
+ ! deflate-pako x 11.52 ops/sec ±0.23% (32 runs sampled)
+   deflate-pako-untyped x 5.12 ops/sec ±1.44% (17 runs sampled)
+   deflate-zlib x 14.33 ops/sec ±3.34% (63 runs sampled)
+   inflate-dankogai x 42.96 ops/sec ±0.19% (57 runs sampled)
+   inflate-imaya x 85.05 ops/sec ±1.07% (71 runs sampled)
+ ! inflate-pako x 97.58 ops/sec ±0.69% (80 runs sampled)
+   inflate-pako-untyped x 18.06 ops/sec ±0.65% (56 runs sampled)
+   inflate-zlib x 60.60 ops/sec ±2.04% (67 runs sampled)
 ```
 
-If you doubt, that zlib is slow because of marshalling, try benchmark with `level 0`.
-You will see, that node bindings don't add noticeable slowdown.
+zlib's test is partialy afferted by marshling (that make sense for inflate only).
+You can change deflate level to 0 in benchmark source, to investigate details.
+For deflate level 6 results can be considered as correct.
 
 __Install:__
 

benchmark/benchmark.js

@@ -156,4 +156,4 @@ module.exports.run               = run;
 
 run(process.argv.slice(2).map(function (source) {
   return new RegExp(source, 'i');
-}));
+}));