SendFaster™ speeds content delivery by up to 500% across platforms and networks, offering you the chance to increase ROI and profits for your website. Studies have shown that a more rapid delivery of content to the end user translates directly into measurable results based on varying metrics for different site categories.
For instance, Google searches drop by 25% with a 500 ms slower page load, a 100ms decrease in load time costs Amazon 1% of sales during the time frame, and Facebook traffic decreases by an amazing 6% with a 1 second slower page delivery.
SendFaster™ isn’t just for those with “bad” bandwidth, it’s the perfect solution for increasing throughput on any bandwidth!
How can SendFaster™ benefit your business? The SendFaster™ protocol specifically targets markets which rely on speed and connectivity for their profits. Slow delivery to end users directly translates into smaller orders, fewer users, and less time spent on site. It offers dramatic improvements to mobile delivery, overcomes higher latencies and packet loss for users connected via Wi-Fi, and greatly improves the HD/streaming experience by minimizing buffering.
The graph illustrates the use of the SendFaster™ protocol versus a standard TCP Reno configuration and explains why sendfaster works. SendFaster sends a fuller stream much more rapidly and the packet loss recovery provides an exponentially better end user experience in terms of page load times, jitter reduction, and overall quality of experience.
PACH Image & Video Compression
When an image is compressed for storage and transmission on the internet, it is decomposed into a series of matrices.
Each matrix is then read out using a "zig-zag" pattern into a list of values, from which the image may be recreated.
The first few numbers in this list represent the "major" features of the block, such as the main color and gradient. Numbers near the end of the list encode "minor" details of the block, such as fine details in an image of grass, or fine patters in an image of human hair.
In order to compress the image and send less data, some of these numbers must be discarded. Most naive algorithms attempt to discard numbers so that the final set of numbers is mathematically most similar to the original image. For instance, in the sum 5 - 4 + 3 - 2 - 1 = 1, an algorithm attempting to minimize the mathematical change might discard "+ 3 - 2 - 1", which would result in 5 - 4 = 1, which is mathematically the same as the original.
However, when the image is reconstructed, these missing values may have been critical to how the image is perceived by the eye. Our algorithm could analyze the image and realize that the most important values to "perception" are 5 - 2 - 1 = 2. While this value is technically "less accurate" than the machine preferred value, it appears more accurate when seen by the human eye.
In addition, most algorithms attempt to evenly remove data from each region of an image. Our algorithm attempts to optimize the image removing data from the areas which have the least detail (such as the solid color of a sidewalk), while preserving the fine detail (such as blades of grass).