A Real-World WebAssembly Benchmark

Re:A Real-World WebAssembly Benchmark by markdog12 on Thursday July 05, 2018 @12:52PM (#17464787)

Firefox crushing the competition on this (all on iMac late 2015):

  Firefox Nightly: 2216  
  Chrome Canary: 5078  
  Safari Tech Preview 60: 7216

Brand new iPad 2018: 16653

Moto G4 phone Chrome: 37735

Windows 10 Laptop:

  Firefox Nightly: 2278
  Chrome Canary: 4907
  Edge: 16332

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by titzer on Thursday July 05, 2018 @06:44PM (#17467497)

A big part of the score is compilation time, which improves a lot with Liftoff, our baseline compiler for WebAssembly in V8. We've been in contact with the benchmark authors and have identified other bottlenecks in our engine (e.g. instantiation time), which we are now working on optimizing. Look forward to some increasing scores soon :)

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by reitzensteinm on Thursday July 05, 2018 @09:16PM (#17468223)

Liftoff is still substantially slower at initial compilation than Firefox. Are there any plans to close the gap? A Unity exported game takes over 40 seconds for a new model low end Chromebook to compile. The upcoming 3 fold decrease will be nice, but FF could do it in 2 seconds on the same hardware.

https://lukewagner.github.io/test-tanks-compile-time/

FF 61: 78.4 MB/s

Chrome Canary: 9.7 MB/s

Chrome 67: 3.3 MB/s

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by titzer on Friday July 06, 2018 @01:57AM (#17469237)

We still have a bottleneck at instantation time where we compile many JS<->WASM wrappers. We are working on parallelizing this compilation as well, so we'll get another bump out of that.

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by sitkack on Thursday July 05, 2018 @03:47PM (#17466222)

Pretty soon we will be running Chrome on Firefox.

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by steipete on Friday July 06, 2018 @01:59AM (#17469255)

https://www.destroyallsoftware.com/talks/the-birth-and-death...

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by vanderZwan on Thursday July 05, 2018 @05:59PM (#17467229)

IIRC they've also always had the fastest Typed Arrays. I wonder if there is a connection there - it wouldn't surprise me if memory access is the biggest bottleneck for WASM in other browsers.

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by philipp-spiess on Thursday July 05, 2018 @12:54PM (#17464805)

Have you tried enabling the experimental new WebAssembly baseline compiler in Chrome Canary?

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by markdog12 on Thursday July 05, 2018 @12:57PM (#17464824)

As a matter of fact, I just did! Still getting crushed at 4546

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by grapeli23 on Thursday July 05, 2018 @03:45PM (#17466207)

Check with --js-flags="--no-untrusted-code-mitigations", it should be about 10-13% faster. https://github.com/v8/v8/wiki/Untrusted-code-mitigations

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by symboltoproc on Thursday July 05, 2018 @06:36PM (#17467452)

This does not sound safe.

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by markdog12 on Friday July 06, 2018 @07:45AM (#17470595)

Aw, Snapped! on me

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by nv-vn on Thursday July 05, 2018 @09:10PM (#17468201)

On my phone (OP6) FireFox beats Chrome by two whole seconds

[ Parent | on HN ]

Re:A Real-World WebAssembly Benchmark by vardump on Thursday July 05, 2018 @08:32PM (#17468040)

Samsung Galaxy S9+ (w/ Exynos):

  Firefox: 6900
  Firefox Nightly: 7874
  Google Chrome: 10602

[ Parent | on HN ]