* Posts by DuncanP

4 publicly visible posts • joined 1 Mar 2010

Mozilla man: Firefox 4 will leapfrog JavaScript rivals

DuncanP

... depends on what you want to do with your JavaScript

Significantly faster JavaScript will make available applications as safe web apps with functionality that was previously restricted to desktop applications or to plugins that you have to trust with your security. I'm all for it.

I do however have one concern, and that is that the emphasis is entirely on speed. Rich applications doing client-side manipulation of large volumes of data (e.g. pictures) take up memory. The memory footprint of data in the different JavaScript implementations is vastly different. Opera may be the fastest at the moment (I haven't checked in the last month) but an array of integers takes up more than twice the space of the corresponding array in FF4. The most appalling offender in this regard was IE (all version up to 8) with over 50 bytes per number.

Is this a concern either for implementors or web app developers, or am I alone?

Mozilla stokes Firefox 4 with first beta build

DuncanP
Thumb Up

FF4b is now faster and leaner than Chrome

I have compute intensive JavaScript code that uses large datasets. Chrome has recently slowed down by 120% for my main benchmark and is now 3 times slower than FF4b.

The only browser that is faster is Opera 10.53 (by 15%) but it takes 3 time the memory to hold the same data so for heavy JavaScript performance FF4b is my favourite.

Compared to FF3.6.6 the new beta is more than 10 times faster. They are obviously doing something very right here.

Safari purged of decade-old browser history leak

DuncanP
Thumb Up

Google Chrome dev channel appears to be fixed already

The two sites linked in the article failed to find any history. My version of Chrome is 6.0.422.0.

Mozilla orders Jäger shot for Firefox engine

DuncanP

Interesting confirmation of my observations

This confirms my own observations and I will be interested to see what effect this new technique has. I have some heavily compute-intensive JavaScript which I was seeing running slower in 3.5 and 3.6 with TraceMonkey turned on than in 3.1.

What is interesting to me is that the core of this code, which was absorbing more than 90% of the time, was a fairly simple pair of loops that had no recursion nor anything else fancy at all.