Thanks for volunteering! I understand if you don't want to do it, though, since you don't have a copy of Adobe Acrobat 10. If you are willing to go through the extra step of installing the trial program, though, that would be great!*Besides the install, it shouldn't take more than about 5 minutes to complete the test.
ADOBE (free trial apparently available)
http://www.adobe.com/products/acrobatpro.html
PROJECT GUTENBERG TEXT (Gibbon's history of the Roman Empire)
http://www.gutenberg.org/cache/epub/25717/pg25717.txt
TEST RESULTS*
(1) Asus 1005ha-pu17 netbook, 2gb memory, Intel Atom N280 (1.66GHz, 512KB L2 Cache, FSB 667MHz). Adobe Acrobat 9.
(2) mba 13"
(3) mbp 13"
(4) mbp 15"
step 1 (preparation):
(1) Asus original conversion = 10.2 seconds (cold start, right click, convert to pdf)
(2) mba 13"
(3) mbp 13"
(4) mbp 15"
step 2 (preparation):*
(1) "save as" tiff file type in a folder= 17.5 seconds (in order to enable conversion. otherwise, i get an error because of "renderable text". also, this will produce 42 images that we can convert into pdf similarly to the way i actually do it with my work.)
(2) mba 13"
(3) mbp 13"
(4) mbp 15"
step 3 (main test):
(1) second conversion = 1:27.3 seconds (combine supported files in Acrobat, larger file size)
(2) mba 13"
(3) mbp 13"
(4) mbp 15"
step 4 (main test):
(1) ocr = 3:07.3 (English (US), Searchable Image output style, lowest (600 dpi) downsample images)
(2) mba 13"
(3) mbp 13"
(4) mbp 15"
step 5 (comments):
The machine didn't seem to be struggling too much. I decided not to start up other programs during the process, because I didn't want to introduce too many variables. However, in past experience, the computer has major problems switching from one application to the next, there is lagging, and the fan is on a lot.
*The methodology isn't perfect. Ideally, this would be done multiple times with 600 dpi scans of printed Chinese or Japanese texts, because those are a lot of the materials I actually work with, and they take far longer to combine and do ocr, but I think this will at least indicate practical performance differences between machines.*