We put a great deal of focus on performance of the X3 web application, and since the release of v0.9, we have made further improvements. We would like to present performance results to you from the best tools in the business: pingdom, Google Pagespeed, Yahoo Yslow and GTmetrix.
X3 Test Link
How is performance reviewed?
One important factor to note about all these "performance" tests, is that they review based on output of the web page resources, and will not generally rate the raw speed of the server. Therefore, you could have a very slow server and still achieve a high pagespeed score, or have a very fast server and get a low pagespeed score if implementation was weak. Basically these tests cater to "best practices" in the output, but not based on the raw "speed" they are delivered by the server.
As for pure "server speed", this almost entirely depends on power and resources of the specific server. As some of you already know, X3 uses "server caching" to maintain cached results of pages and images, which serves to massively speed up delivery of pages and resources once they are created on your server. In addition to new X3 improvements, the X3 Demo is running on our blazing fast dedicated X3-hosting server, which offers SPDY and SSL combined with Cloudflare and MaxCDN for supersonic delivery of globally distributed image assets.
As a "human" (the ultimate speed test), I think you will find that the X3 demo gallery is incredibly fast. Please also take into consideration that X3 is a modern html5 web application, and generally includes a generous amount of image files and application data. X3 does not require additional html requests after it is loaded, thus optimized for humans at the expense of a slightly bigger initial page load.
http://tools.pingdom.com/fpt/#!/caA75j/ ... gevuex.com
Performance grade 98/100. Your website is faster than 89% of all tested websites
Why not 100?
X3 score is penalized mostly from google fonts and google analytics, which have short cache freshness lifetimes, and also from loading CSS from a cookie-less domain. We cannot control these external resources, and essentially it is reporting false positives that will not affect performance. These resources should technically be set to not cache at all, thus not leading to performance penalization, but a short cache-lifetime is added for beneficial reasons.
https://developers.google.com/speed/pag ... ab=desktop
Why not 100?
Also, pagespeed reports that images in the start page are not optimized for mobile. Although true to some degree, X3 loads high-quality retina-sized images, and the example loads "horizontal" images into a "vertical" orientation, thus leading to a slightly false positive.
Why not 100?
The same reasons as explained in the pingdom and pagespeed reports https://d.pr/i/1g5Rl
(Chrome developer tools plugin)
I challenge you to find a web application of similar depth with better performance results!