935288 - Investigate performance of RIABench benchmark

Status: RESOLVED INCOMPLETE

(Core Graveyard :: Tracking, defect, P4)

Description

[Chris Peterson [:cpeterson]]

RIABench is one of the benchmarks from Tom's Hardware Guide's Web Browser Grand Prix:

http://www.tomshardware.com/reviews/chrome-27-firefox-21-opera-next,3534-6.html

http://www.timo-ernst.net/misc/riabench-start/

The test cases:

 * Primetest
 * Prime factorization
 * 3D-Test
 * JPEG encoding test
 * MD5 hashing test
 * Random key generator test
 * Run-length encoding test (bug 851064)
 * 2D-Test (bug 851062)
 * Memory management test
 * Focus-tests

Comment 1

[Chris Peterson [:cpeterson]]

On my MacBook Pro, Nightly 28 handily beats Chrome 30 on many RIABench tests. \o/

Firefox and Chrome are tied on the Primetest and Prime Factorization tests.

Comment 2

[Hannes Verschore [:h4writer]]

(In reply to Chris Peterson (:cpeterson) from comment #1)
> On my MacBook Pro, Nightly 28 handily beats Chrome 30 on many RIABench
> tests. \o/

When looking to the scores on WBGP with FF19 we were also beating chrome on most tests. The reason we didn't come out as first was the "Run-length encoding test" that was way slower. That's the reason I looked into this and this has been addressed in FF24. So RIABench shouldn't need much attention, right?

Comment 3

[Chris Peterson [:cpeterson]]

Hannes: I agree that RIABench does not need much attention. I filed this bug mostly for completeness with all the benchmarks used by WBGP. <:)

Bug 851062 says we are much slower than Chrome on the Windows drawing tests.

Comment 4

[BMO Automation]

Marking all tracking bugs which haven't been updated since 2014 as INCOMPLETE.
If this bug is still relevant, please reopen it and move it into a bugzilla component related to the work
being tracked. The Core: Tracking component will no longer be used.