Apple → If android has such a huge market share then ..

Well it is not just raw mhz when it comes to things like ps2 emus etc. Android does throttle the cpu back to save battery (and heat etc). Im sure ios does as well we just don't have a easy way to see it like with android. When you root a android or change the rom like i do you can disable any and all cpu throttling. I don't as in real world performance you wont see any difference. In bench mark testing they do things to deliberately let the cpu speed to drop (most androids have a low speed of 300mhz with max of well what ever the cpu cores will push.

They do that to test things like the recovery time which is the time it takes the cpu to go from the low of 300mhz to in my case the max of 1.2 ghz. Back in the day some cpu manus (yeh looking at you intell) and some vid card companies (yeh you best believe im looking at you nvidia and ati) did things like detect when a benchmark program was running and shut off features and turned off throttling to artificially inflate their numbers.

So far i don't thing android has done such a cheat nor apple least no one has found em out. Honestly i never much put faith in to benchmarks as a method to choose a cpu vid card os etc. They are great for comparing your systems pre and post tweaks to find the best settings but to compare say ios vs android na. Dont put stock in them for that. You can mostly trust them for testing drivers etc as well but again some times our friendly hardware manus cheat

Ive caught them cheating my self as have many of my friends and exposed them for it lol

oh as per timing of throttling on android it is measured in milliseconds both to slow down and speed up. Many phones in android when dev options are on have a performance setting that lets you turn off the throttling. If you value battery life don't do it lol.

I don't know, I've been tracking browser user agent strings for quite some time and almost all of the Android browsers, no matter what browser you use on Android, all contain an Android OS and version number in various forms. So unless you choose to specifically modify the user agent string that the browser sends to the server (which I'm sure that most people won't even think about doing), a site administrator will be able to track what kinds of browsers are visiting a web site.

Well i have seen some browsers that never fetched from cache back in the day on pcs. I still think we could be seeing something along those lines to. If lets say for example apple iphones built in browser is ignoring the cached image of »i.dslr.net/sk/bl/logo.gif dslreports logo and loads it fresh from the server each time and all the browsers on android are pulling from cache on the phone most of the time and say only fetching it from the server once per day. Then ios based devices would show a higher statistic.

Now a question to justin. Are those numbers for bandwidth used or visits? And what tracking method is being used? Hits to a image or a pages file (yes i know it is not html but...) say index.html for example. If it is bandwidth i think there might be some merit to my thinking of more fetching from the server for ios devices vs android that is maybe relying more on cache.

In that case, Justin would need to do an analysis of user agents based upon a random set of pages across the site and then add up the statistics.

Those analytics are pre-filtered by google to only reflect pages and sessions, not just random fetches of images.

Only pages have the analytic beacon attached, and google probably uses more than just user-agent to determine device because the beacon pings back all kinds of stuff: window resolution, architecture, stuff the native browser give up as info regardless of rooting and so on.

The only way to escape the reporting or be mis-counted, is if a custom browser somewhere has closed off all those metrics. user agent is probably the least reliable signal.

You probably just have a large user base on apple. Not to surprising really. Just look at how active this forum is vs the android one.

Now i have seen weird stuff though as a web admin for a mmorpg. We had a very high number of visitors on linux based oses. Way out side what we expected based on the fact that we had no linux game client lol.

the world of bots with fake UAs is getting crazy sometimes they have to be observed to spot them.

they even have full javascript based bots now that load and explore sites following all the little buttons and what not that javascript activates. How do you do a robots.txt to block javascript actions? its outdated.

Wowsers i just added up all the droids plus the not set together and the ios you have 3.181 times as many ios users. I know you said you had lots of .x%s in there to not shown so lets just say 3 times as many. Still that is a little weird. I suppose that was what got your attention to begin with.

Not sure you can. But shoot i have some time to burn ill take a look at how the robots.txt is set up and what the specs show. May be a work around that is possible.

It isn't because its an old spec.
When designed it was for URLs.
now URLs are only half what is going on. And everyone wants to index sites like facebook.

Right. Now i did have a thought. Bots work by reading the raw page code to look at text and links. Now if they are looking at say the code for the post button. You might be able to do a no follow on that. the code is
post_form_to('qfn29918652', '/speak/remark,29918652', 'qr29918652'); suspended=0;

So maybe a no follow on /speak/remark/* might tell the bot not to follow the post button obviously no bot will be making a post (well not a legitmate one any how) But it might follow for example any java script related to page 1 2 3 4 next etc. So that should be able to be blocked as far as the bots are concerned.

Allot of what is needed is leg work looking at the various faqs on search engines bots like yahoo google bing etc all they do have faq on how to prevent their bots from indexing this or that.

here's something sense october last year googles bot has been "rendering web pages with javascript and css turned on"

»googlewebmastercentral.b ··· ter.html

Meaning it will follow all javascript etc links and actions or can.

For optimal rendering and indexing, our new guideline specifies that you should allow Googlebot access to the JavaScript, CSS, and image files that your pages use. This provides you optimal rendering and indexing for your site. Disallowing crawling of Javascript or CSS files in your site's robots.txt directly harms how well our algorithms render and index your content and can result in suboptimal rankings.

I would almost bet other engines bots do much the same.

i use imdb a lot too, but i don't use the browser...i use their iPhone app...it is great.

That, plus some large percentage of workers/students spend a large part of their day doing actual work, so they're sitting in front of a desk|laptop.

I can't code on a tablet

I don't know that I would use their app all that much. I use the site say once a day maybe twice a day. I would love to see something along the lines of the xda app for DSLR reports. I keep getting the idea that ehh maybe I'll start developing apps for android. From the looks of what I seen I could learn it pretty easy and I have a huge forum to pump for help xda developers and I'm sure dslreports peeps as well.

use the app instead of the browser...it is quite decent.

Yeh i grabbed it but it does well to much lol.

I literally go to imdb.com type in a show name in search click on the show and see when the next ep is going to air. And that is about it. The news etc stuff i don't bother with and their forums are chock full of iiots and trolls with few exceptions lol

I don't tend to browse the web much on my phone. Can't stand viewing pages on the tiny screen. I switched from an iPhone to Android about 2 years ago.

Most of my non-technical family and co-workers went iPhone/iPad. I think the universal recognition did it.

That's the thing, most phones today have large screens so viewing webpages on them shouldn't be a problem.

Allow me to rephrase in light of a sudden realization. Google Now has been spoon feeding me a massive number of tech related articles I click and read through without thinking about it. Opera shows like 12 tabs...
I can't stand typing on mobile devices would be a better statement.
I have a Galaxy S 4, compared to my HTC Sensation I consider it big. Compared to my last iPhone (4) I consider it massive. But I don't visit any sites I actively post on with it. I tried and I always have trouble finding buttons, or I make an embarrassing number of typos. Dslreports, Slashdot, etc... laptop/desktop only.

I can type very proficiently on my iPhone and with Siri I can just speak what I want to type and Siri puts what I say into words on the screen, complete with things like commas.

I have voice typing, it works well enough, but it isn't the same. Just personal preference. Also, I'm too lazy to turn off auto correct.
I had an iPhone 4, and I had a newer model for work until last October. Siri was interesting to play with.

Lol at typos comment. I feel your pain