Thomas Stromberg

"Great locally brewed beer you won't find anywhere else. The fish & chips were fantastic, as were the sweet potato fries. I'll be back!" - Thomas Stromberg

"One catch with this (and most other) approaches is if you happen to benchmark two nameservers which happen to share cache (8.8.8.8 and 8.8.4.4 - or your ISP vs your home router), you will only be benchmarking the speed of cached responses from the second nameserver. namebench detects this condition and ignores the slower nameserver which is in a cache sharing relationship." - Thomas Stromberg

"That stinks - what does it do? It's tested under WinXP 32-bit here. I'm more than eager to respond to any bug reports posted." - Thomas Stromberg

"Ouch, that would be a bug. It looks like I only find the correct path to the Opera history file when run on Mac OS X. I've opened an issue for this: http://code.google.com/p... - thanks!" - Thomas Stromberg

"By default namebench will generate the final comparison based on whichever nameserver you currently have configured as your primary." - Thomas Stromberg

"Interesting point actually: running a locally caching client can be a detriment to performance due to the extra bounce and very low initial cache hit ratio. It's usually a toss-up, but it can make for a poor experience unless you are using a good set of forwarders." - Thomas Stromberg

"Actually, it does repeat some queries by default (based on a distribution curve of how often you are likely to use it, see selectors.py). It's easy to see which ones hit cache by looking at the response distribution chart. If you want to see something more dramatic, enter "2" for the number of runs. The bar graphs will then show you the performance difference between the first run (somewhat cached) and the second run (entirely cached). namebench used to default to two runs, but I found it to be a poor reflection of real-world performance." - Thomas Stromberg

"Actually, it does repeat some queries by default (based on a distribution curve of how often you are likely to use it, see selectors.py). It's easy to see which ones hit cache by looking at the response distribution chart. If you want to see something more dramatic, enter "2" for the number of runs. The bar graphs will then show you the performance difference between the first run... more... - Thomas Stromberg