I am working on an article right now to test download speeds on mobile
devices. I've done some testing like this before, mostly looking at USB
download speeds on a variety of cell phones. What has been intersting
about the testing I've done is that there is such a huge variety in the
amount of time it takes to download content. I had originally figured
that it would be fairly consistent across the different devices, but
was way wrong.
In the new tests I'm going to look at how long it takes to download
content across different. I'm going to test downloading on a
3G network, USB and WiFi across three different phones - the iPhone 3G
and the BlackBerries Storm and Bold. Not only am I going to
look at the amount of time it takes, but also which is easiest.
There is definately something to be said about the
portability of being able to download content on the go with wireless
standards, but are they going to be fast enough to make it worthwhile,
or is the speed benefit from USB going to make it a more attractive
Given the content that is available from the App Store and App World
sites I am going to try to find a program that is offered by both Apple
and RIM, and hopefully something that is relatively useful as an
application so that I can talk about that in a future blog post.
Any suggestions on what to download?
I also want to include a section on future technologies that are going
to improve downloads in the future. The big one that I want
to look at is USB 3.0, but there might be others to consider.
If there are any that you would like me to investigate, let
There are many more variables involved when testing download speeds via 3G wireless networks compared to USB or WiFi connections. To make your 3G tests meaningful, they need to be controlled, repeatable, and reflect real user experience. This necessitates test equipment that emulates a 3G network, preferably portable enough to be easily used on a desk top, and capable of testing multiple versions of 3G technologies.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.