| 10:59 pm on Apr 12, 2008 (gmt 0)|
Check your stats, and determine what browsers are being used. There's no point testing browsers that no-one uses. Personally, I take the view that if it works in current versions of Opera and Firefox together with IE6 and 7 that's good enough.
If you have a lot of images on a site it's likely that enormous savings are possible.
Photographic images are usually best served using the jpeg format, but you can certainly mess around with settings and make substantial savings. Adding a slight blur and removing noise before saving can be very effective.
All other (non-transparent) images are usually best served using 256 color png format. If your image editor is rubbish at color reduction (some are) try Ultimate Paint.
| 11:13 pm on Apr 12, 2008 (gmt 0)|
If your site does not work with certain browsers then you will not see them feature highly in the stats - the visitor will be gone.
| 12:28 pm on Apr 13, 2008 (gmt 0)|
You could invest in some VMware and create multiple virtual machines that each run sets of the various browsers.
| 2:53 pm on Apr 13, 2008 (gmt 0)|
which version of VMware? That's the answer I was looking for!
| 8:18 pm on Apr 13, 2008 (gmt 0)|
There's a limit of course, but one solution is to buy three or four secondhand PCs /laptops specifically to test the most common browsers, that's what I do. It's reassuring that I have a few laptops in reserve in case of hardware failure or virus attacks.
| 1:07 am on Apr 14, 2008 (gmt 0)|
Microsoft offers Virtual PC images of windows xp with IE 6,7,8 (separate images). VPC and the images are free, this is the method I use.
| 2:18 pm on Apr 15, 2008 (gmt 0)|
This is something I have been meaning to do also.
I think it is a good idea to still test in IE6 + 7.
Lots of people are still using 6 but the CSS displays a little different in 7.