1P Extension is making Safari slower (tested with SunSpider)
I currently use 1Password 4.2.2 and 1P extension 4.2.0.BETA-15 with Safari 7.0.3 (9537.75.14).
I have a MacBook Pro 13-inch Mid 2013, Processor 2.5 Ghz Intel Core i5, 4GB RAM, Graphics: Intel HD Graphics 4000 1024 MB on System Version: OS X 10.9.2 (13C64), Kernel Version: Darwin 13.1.0.
I noticed whenever I add the 1P extension, Safari goes slower.
I tested my theory with SunSpider 1.0.2 JavaScript Benchmark.
FYI: SunSpider is a JavaScript benchmark. This benchmark tests the core JavaScript language only, not the DOM or other browser APIs. It is designed to compare different versions of the same browser, and different browsers to each other. Unlike many widely available JavaScript benchmarks.
These are the results SunSpider 1.0.2 JavaScript Benchmark:
From: With 1P Extension installed Only
**To: **Without 1P Extension installed
TEST COMPARISON FROM TO DETAILS
===============================================================================
** TOTAL **: 1.29x as fast 224.9ms +/- 10.7% 173.8ms +/- 4.3% significant
===============================================================================
3d: - 32.3ms +/- 16.4% 27.3ms +/- 9.9%
cube: 1.24x as fast 11.5ms +/- 16.6% 9.3ms +/- 13.1% significant
morph: - 9.1ms +/- 20.8% 8.1ms +/- 15.8%
raytrace: - 11.7ms +/- 18.2% 9.9ms +/- 14.2%
access: 1.179x as fast 19.8ms +/- 12.4% 16.8ms +/- 10.0% significant
binary-trees: - 3.1ms +/- 36.8% 2.0ms +/- 29.2%
fannkuch: 1.23x as fast 8.0ms +/- 12.6% 6.5ms +/- 9.3% significant
nbody: - 4.3ms +/- 19.3% 4.1ms +/- 19.2%
nsieve: - 4.4ms +/- 8.4% 4.2ms +/- 7.2%
bitops: - 9.9ms +/- 7.9% 9.5ms +/- 8.1%
3bit-bits-in-byte: ?? 1.6ms +/- 23.1% 1.7ms +/- 20.3% not conclusive: might be 1.063x as slow
bits-in-byte: - 2.3ms +/- 15.0% 2.3ms +/- 15.0%
bitwise-and: ?? 2.0ms +/- 0.0% 2.1ms +/- 10.8% not conclusive: might be 1.050x as slow
nsieve-bits: - 4.0ms +/- 16.8% 3.4ms +/- 10.9%
controlflow: - 2.6ms +/- 19.2% 2.3ms +/- 15.0%
recursive: - 2.6ms +/- 19.2% 2.3ms +/- 15.0%
crypto: - 17.1ms +/- 16.9% 15.0ms +/- 12.3%
aes: - 8.5ms +/- 23.5% 7.7ms +/- 14.5%
md5: - 4.6ms +/- 16.7% 3.9ms +/- 13.5%
sha1: - 4.0ms +/- 22.3% 3.4ms +/- 14.7%
date: 1.63x as fast 35.3ms +/- 28.6% 21.6ms +/- 11.9% significant
format-tofte: 1.78x as fast 23.5ms +/- 40.5% 13.2ms +/- 19.1% significant
format-xparb: 1.40x as fast 11.8ms +/- 18.0% 8.4ms +/- 4.4% significant
math: 1.117x as fast 15.3ms +/- 7.0% 13.7ms +/- 7.8% significant
cordic: - 3.8ms +/- 19.4% 3.2ms +/- 9.4%
partial-sums: - 8.9ms +/- 11.0% 8.2ms +/- 13.5%
spectral-norm: - 2.6ms +/- 19.2% 2.3ms +/- 15.0%
regexp: 1.163x as fast 10.0ms +/- 9.5% 8.6ms +/- 9.8% significant
dna: 1.163x as fast 10.0ms +/- 9.5% 8.6ms +/- 9.8% significant
string: 1.40x as fast 82.6ms +/- 11.8% 59.0ms +/- 6.2% significant
base64: - 6.8ms +/- 32.4% 5.1ms +/- 12.3%
fasta: 1.32x as fast 12.4ms +/- 21.1% 9.4ms +/- 9.6% significant
tagcloud: 1.22x as fast 14.6ms +/- 7.4% 12.0ms +/- 10.1% significant
unpack-code: 1.45x as fast 36.7ms +/- 13.0% 25.3ms +/- 6.3% significant
validate-input: 1.68x as fast 12.1ms +/- 24.2% 7.2ms +/- 14.6% significant
Comments
-
Hi @stamatgeorge,
This can be expected to a certain extent. The 1Password extension uses JavaScript to do its magic, so must inject its scripts in every page.
When we're talking in milliseconds (like measured in the SunSpider report), it's not really going to cause any noticeable difference to users.
With that said, we do hope to make improvements to the extension in the future and hope to make it even lighter. :)
0
