Speedometer browser benchmark: show me yours and I'll show you mine


So what? A month old browser version is an invalid for running a benchmark? I’m not trying to run it in Windows 98.


I dunno, it’s just not the current released version of Chrome, and the version mentioned in the first post:

Chrome 59 was just released a day or two ago, and it has the new rebuilt Turbofan and Ignition JavaScript engine on by default now! 🎉


One closing comment: Don’t use traditional JavaScript benchmarks to compare phones. It’s really the most useless thing you can do, as the JavaScript performance often depends a lot on the software and not necessarily on the hardware, and Chrome ships a new version every six weeks, so whatever you measure in March maybe irrelevant already in April. And if there’s no way to avoid running something in a browser that assigns a number to a phone, then at least use a recent full browser benchmark that has at least something to do with what people will do with their browsers, i.e. consider Speedometer benchmark.


That’s the whole point of Speedometer. They retired Octane, their old benchmark.

I’m sorry if you are unwilling or unable to understand this. Perhaps try harder? And for God’s sake get on the current version of the damn browser.


I upgraded to Version 59.0.3071.86 but you’re still a condescending asshole who pretends everyone else is dumb instead of reading what they are writing.

You quoted Benedikt Meurer as the authority on how this is so important, yet you FAKE NEWS away his own comments that this is a bad way to compare systems because there is a huge software implementation component to this and the numbers can change massively between each Chrome version (or other software changes). I then show how how there is a literally massive software implementation component happening and you are countering with the idea that I should use a one month newer version of Chrome for a massive performance difference like I am an idiot. That’s my fucking point, and that’s Meurer’s point. Acting like this one single number is a valid comparison between systems is useless because it is massively sensitive to things like browser updates while not a single aspect of the systems themselves change.

So this is not all negative, I will add that I like how you elipsed away if there’s no way to avoid running something in a browser that assigns a number to a phone


And yet, we are all on the same Chrome version, are we not? As cited in the first post:

That’s… the whole point of this exercise. To be on the same, latest version of Chrome. Feel free to scroll up and read it again if you don’t believe me. Hell I can screenshot it if you want!

How so, when we are all on the same browser version? Well, most of us, anyway.


Chrome 58: 145
Chrome 59: 125, nevermind still 145 with ublock turned off.

Office generic Dell Haswell, probably i5 midrange.


Some Browser / OS specific differences, on Windows 10:

Chrome 59        220
Edge 40.15063    130
Firefox 53        92

MacOS on my Macbook (2016)

Chrome 59        101
Safari 10.1.1     69
Firefox 53        39

Poor Firefox.


Why is your MacOS results so much slower than my 3.5 year old iMac?


Did someone say speedometer? My motorcycle’s hasn’t worked in a few years, but I found a nifty app for my android watch that does the trick. Now, obviously I’m not going this fast right now, this was last weekend in a road trip through rural Oregon/Washington.


Macbook 2016 CPU: 1.2GHz Intel Core M5-6Y54 – 2 core, 4 thread, 4.5W TDP.

If your 3.5 year old iMac isn’t kicking this 4.5W TDP CPU’s ass, then it sucks. The thing basically has a tablet CPU.


Ah, in my head I inserted a Pro in there. Still, I wonder why Safari was (barely) faster than Chrome on my iMac compared to the results you got.


New iPad Pro (2017) scores 149 on this.


Don’t know if this is any good but that’s the score on my gaming rig. i5-4690k @ 4Ghz, 16GB Ram, Windows 10


159.1: iPad Pro 10.5 (2017)


We noticed this on Twitter as well. If I reduce the screen to half size (iPad Pro 12.9") I get 156. If I leave it full screen I get 149. Consistently!

Both tablets have the same SoC but the screen is larger on the 12.9" model… So we’re hitting limits of GPU draw speed here 🤣


Surface Pro 5 core i7 gets 110 with Chrome latest. That is … lower than I expected, I need to run it again after full updates.

Ok yeah 136 now that is way more like it.


How do you like it?


I posted some thoughts in the Surface topic. Maybe I’m 😜, but I like topics… on topic.


Ah – there it is. Thanks!