Silk, iPad, Galaxy comparison(stevesouders.com)
stevesouders.com
Silk, iPad, Galaxy comparison
http://www.stevesouders.com/blog/2011/12/01/silk-ipad-galaxy-comparison/
5 comments
Relevant chart here:
http://loadtimer.org/results.php?data=all
http://loadtimer.org/results.php?data=all
I think it's odd that the author chooses to leave accleration on, despite having just proven that it's detrimental. That, and the fact that you're proxying all of your browsing traffic through a third party, seems like it's all cons and no pros. If I was him, I'd keep it off until there was more of a balance between pros and cons (though the potentially privacy issues alone would probably be nigh-impossible to overcome.)
Why is this confusing? I've seen numerous iPad vs. Android browser comparisons, but yet to see an in-depth Silk test (other than the many reviewers who said it was slower).
Silk (and whether it lives up to it's billing) is the most interesting part.
Silk (and whether it lives up to it's billing) is the most interesting part.
Did you read the article? In it he benchmarks Silk (accel) vs. Silk (no accel). He's proven to himself that the acceleration system doesn't work, using his own testing methodology, which, given that he produced the results, he should be biased into believing.
Nothing is keeping him from turning it on later in the future, when it's objectively proven to be faster. But he keeps it on now, basically because of his optimism in Amazon.
Nothing is keeping him from turning it on later in the future, when it's objectively proven to be faster. But he keeps it on now, basically because of his optimism in Amazon.
Can anyone explain the "acceleration"? I don't understand why a browser would have this option... when would you choose to not use it? Why does amazon give users a choice, if it's supposed to be better (which apparently it's not)?
All traffic is proxied through Amazon's servers. This means that hypothetically they can read you unencrypted HTTP session. Hence the toggle.
>the browser cache was cleared between tests
How is that "real world"?* Not to mention, Silk was designed to speed up as you use it, learning your browsing habits and caching many assets. Using it directly from the box with caching turned off defeats the purpose.
* Ah, I see, the headline was edited before I clicked "submit" on my comment. Point still stands, nearly every one of these artificial benchmarks proves nothing of any real use.
How is that "real world"?* Not to mention, Silk was designed to speed up as you use it, learning your browsing habits and caching many assets. Using it directly from the box with caching turned off defeats the purpose.
* Ah, I see, the headline was edited before I clicked "submit" on my comment. Point still stands, nearly every one of these artificial benchmarks proves nothing of any real use.
It's realworld in the sense he used some of the most popular sites, not some kind of laboratory test with local network.
Caching turned off is for cold-cache test. Storage is relatively small on tablets and it's very plausible to get a cold cache first pageload.
Steve is pretty good at these kinds of tests.
Caching turned off is for cold-cache test. Storage is relatively small on tablets and it's very plausible to get a cold cache first pageload.
Steve is pretty good at these kinds of tests.
Problem is, Silk isn't marketed as just another browser. To see its effects, you have to use it like Amazon intended for it to be used. You don't disconnect the batteries from a Prius then say it's less efficient than a standard car just for the sake of being "fair".
There was nothing wrong with mine, it was a summary that was far more comprehensive than the 4 words that replaced it.