Regressions on Performance tests in SunSpider, V8 in Aurora 11

RESOLVED WORKSFORME

Status

()

Core
JavaScript Engine
RESOLVED WORKSFORME
6 years ago
6 years ago

People

(Reporter: lmesa, Unassigned)

Tracking

Firefox Tracking Flags

(Not tracked)

Details

Attachments

(1 attachment)

(Reporter)

Description

6 years ago
Created attachment 591303 [details]
Perf test results 1/24

Hi all,

I always run tests on my windows 7 laptop during the Beta/Auroras per cycle and found a few regressions I wanted to flag in sunspider, sunspider modified and kraken.   

I ran Aurora 11 (whatever build was available today) and Beta 10b4.

Nothing very major--see the attachment and what's highlighted in red.
Thanks. I'm not sure what to make of it: it's hard to tell if it's noise or not. Were all the measurements taken on the same day, or were they over time as those different versions were released?
(Reporter)

Comment 2

6 years ago
(In reply to David Mandelin from comment #1)
> Thanks. I'm not sure what to make of it: it's hard to tell if it's noise or
> not.

Agreed--the differences aren't huge (well, the sunspider ones kind of are).


 Were all the measurements taken on the same day, or were they over time
> as those different versions were released?

I did Firefox 4 and Firefox 8 back in August, Firefox 9 in November, and Firefox 10 & 11 today.
I retested 9, 10, and 11 today, and they all get about 220ms on my machine, so I'm going to chalk this up to random variance.
Status: NEW → RESOLVED
Last Resolved: 6 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.