Without seeing the full context of the post in question, let's just say it doesn't sound like it washes.
A 24 bit document in sRGB and a 24 bit document in Adobe RGB have the same encoding. The author may believe that with a smaller color gamut (sRGB), there are more bits to define 'colors' in a smaller gamut than a larger gamut (AdobeRGB)?
A 24 bit image in either sRGB or Adobe RGB (1998) of a gray card will have less colors than an image of a colorful field of flowers. Scene gamut should not be ignored! Gamut and encoding of data should not be ignored! We can talk about color encoding where the math results in billions of color despite the number of colors we can see are nothing approaching this value.
If I take a Granger Rainbow I built in Lab, high bit, convert one to sRGB and one to sRGB
FROM Adobe RGB (1998) in Photoshop, using V2 profiles and no dither, they both have the same number of unique colors according to ColorThink! So I think the original author is discussing something else or is just wrong.
The other issue is, we often
have to convert color spaces. The only reason I can think of converting from Adobe RGB to sRGB is if I needed sRGB! For the web or other emissive non color managed use. There is no such thing as an sRGB printer (or Adobe RGB printer for that matter). While some silly labs may
demand sRGB it is very easy to prove such devices do not follow anything like sRGB in their behavior. Even if someone forced me to send sRGB for output to a printer, I'd still want to start with Adobe RGB (1998) for other output needs.
Actually I'd prefer ProPhoto but that's a story summed up in this long video:
Everything you thought you wanted to know about color gamut
A pretty exhaustive 37 minute video examining the color gamut of RGB working spaces, images and output color spaces. All plotted in 2D and 3D to illustrate color gamut.
High resolution:http: //digitaldog.net/files/ColorGamut.mov
Low Res (YouTube):
http://www.youtube.com/watch?v=n0bxSD-Xx-Q