Understanding Benchmark Results

[carloswm85]

I've got this results after running npm run start:benchmark:

AccelerationBands x 332 ops/sec ±0.38% (88 runs sampled)
FasterAccelerationBands x 62,476 ops/sec ±0.29% (90 runs sampled)
AC x 907 ops/sec ±0.43% (93 runs sampled)
FasterAC x 43,950 ops/sec ±0.44% (93 runs sampled)
ADX x 61.82 ops/sec ±0.31% (64 runs sampled)
FasterADX x 11,853 ops/sec ±0.54% (92 runs sampled)
AO x 1,263 ops/sec ±0.46% (95 runs sampled)
FasterAO x 79,166 ops/sec ±0.43% (92 runs sampled)
ATR x 1,256 ops/sec ±1.33% (91 runs sampled)
FasterATR x 34,606 ops/sec ±0.36% (93 runs sampled)
BollingerBands x 153 ops/sec ±0.31% (85 runs sampled)
FasterBollingerBands x 39,396 ops/sec ±0.36% (94 runs sampled)
BollingerBandsWidth x 135 ops/sec ±0.32% (85 runs sampled)
FasterBollingerBandsWidth x 29,354 ops/sec ±1.12% (93 runs sampled)
CCI x 293 ops/sec ±0.55% (86 runs sampled)
FasterCCI x 52,656 ops/sec ±0.53% (93 runs sampled)
CG x 481 ops/sec ±0.49% (91 runs sampled)
FasterCG x 62,563 ops/sec ±0.48% (95 runs sampled)
DEMA x 65.76 ops/sec ±0.38% (68 runs sampled)
FasterDEMA x 55,326 ops/sec ±1.29% (92 runs sampled)
DMA x 1,143 ops/sec ±0.98% (93 runs sampled)
FasterDMA x 61,344 ops/sec ±0.35% (90 runs sampled)
DX x 63.21 ops/sec ±0.40% (65 runs sampled)
FasterDX x 13,782 ops/sec ±0.32% (94 runs sampled)
EMA x 224 ops/sec ±0.53% (86 runs sampled)
FasterEMA x 164,037 ops/sec ±0.29% (96 runs sampled)
IQR x 1,183 ops/sec ±0.71% (93 runs sampled)
FasterIQR x 4,091 ops/sec ±0.48% (94 runs sampled)
LinearRegression x 418 ops/sec ±0.45% (90 runs sampled)
FasterLinearRegression x 155,375 ops/sec ±0.31% (97 runs sampled)
RMA x 4,863 ops/sec ±0.42% (95 runs sampled)
FasterRMA x 164,986 ops/sec ±0.48% (96 runs sampled)
MACD x 76.24 ops/sec ±0.26% (77 runs sampled)
FasterMACD x 48,637 ops/sec ±0.38% (91 runs sampled)
MAD x 560 ops/sec ±0.60% (91 runs sampled)
FasterMAD x 99,571 ops/sec ±0.37% (95 runs sampled)
MOM x 14,225 ops/sec ±0.53% (95 runs sampled)
FasterMOM x 155,145 ops/sec ±0.41% (93 runs sampled)
OBV x 8,698 ops/sec ±0.30% (94 runs sampled)
FasterOBV x 139,736 ops/sec ±0.25% (92 runs sampled)
Period x 3,891 ops/sec ±0.58% (91 runs sampled)
FasterPeriod x 38,286 ops/sec ±1.03% (94 runs sampled)
PSAR x 4,958 ops/sec ±0.35% (92 runs sampled)
FasterPSAR x 140,103 ops/sec ±0.21% (95 runs sampled)
ROC x 1,506 ops/sec ±0.47% (95 runs sampled)
FasterROC x 151,805 ops/sec ±0.54% (94 runs sampled)
REI x 188 ops/sec ±0.76% (85 runs sampled)
FasterREI x 57,005 ops/sec ±0.35% (95 runs sampled)
RSI x 114 ops/sec ±0.25% (83 runs sampled)
FasterRSI x 28,414 ops/sec ±0.29% (96 runs sampled)
SMA x 1,460 ops/sec ±0.32% (96 runs sampled)
FasterSMA x 118,002 ops/sec ±0.53% (93 runs sampled)
StochasticOscillator x 366 ops/sec ±0.33% (91 runs sampled)
FasterStochasticOscillator x 16,841 ops/sec ±0.32% (92 runs sampled)
StochasticRSI x 95.32 ops/sec ±0.38% (80 runs sampled)
FasterStochasticRSI x 13,178 ops/sec ±0.61% (94 runs sampled)
TDS x 25,872 ops/sec ±0.47% (93 runs sampled)
FasterTDS x 315,913 ops/sec ±0.33% (97 runs sampled)
TR x 4,196 ops/sec ±1.15% (95 runs sampled)
FasterTR x 162,758 ops/sec ±0.20% (95 runs sampled)
VWAP x 1,263 ops/sec ±0.54% (93 runs sampled)
WSMA x 1,078 ops/sec ±0.47% (94 runs sampled)
FasterWSMA x 74,301 ops/sec ±0.22% (92 runs sampled)
WMA x 759 ops/sec ±0.30% (93 runs sampled)
FasterWMA x 121,291 ops/sec ±0.30% (96 runs sampled)
getAverage x 26,420 ops/sec ±0.94% (88 runs sampled)
getFasterAverage x 16,615,662 ops/sec ±0.80% (86 runs sampled)
getStandardDeviation x 4,618 ops/sec ±0.29% (93 runs sampled)
getFasterStandardDeviation x 1,143,213 ops/sec ±0.40% (91 runs sampled)
Finished benchmark.

Let's take this sample: FasterWMA x 121,291 ops/sec ±0.30% (96 runs sampled)

Can you help me to read this?

ops/sec is operations per second? What is the percent change indicating? What are the samples from?

How do I tell FasterEMA is faster than EMA?

EMA x 224 ops/sec ±0.53% (86 runs sampled)
FasterEMA x 164,037 ops/sec ±0.29% (96 runs sampled)

[bennycode]

Hi @carloswm85, 164,037 ops/sec means that the algorithm can execute 164K operations per second and thus is much faster than an algorithm which can only handle 200 operations per second. You will notice that the "Faster"-algorithms are by magniture faster than the regular ones which is why I am considering dropping the Big.js variants in v7.

[carloswm85]

Ok. Thanks.