I agree with Robert.
Although I've been in the business of PCs for 20 years or more, I found eBoostr technically challenging. I also found it a worthwhile addition to Win XP even though my speed ratios have never been better that 1.3 (with flashcard).
CONFIGURATION
The need to edit config files and exception lists with notepad will restrict usage to the technically savvy and I suspect that many potential users will not even understand the concept of setting the cache size. Even having a default setting for the flash cache size might be a big help to some (I get better results with a little RAM and 1 GB flashcard than with 2GB or 4GB).
AUTOMATION
Many will not have the enthusiasm or patience of beta testers to fiddle and test the dozens of variables involved. The more automated you can make the eBoostr setup, the more long-term customers you will hold. Maybe your experts can dream up an algorithm which takes into account the speed of the hardisk, the amount of RAM, the pagefile size, the flashcard size etc and come up with a near optimum configuration automatically.
LACK OF INFORMATION
If you scan the internet, there is a fair bit of misinformation about how eBoostr works - even comments that eBoostr cannot possibly work because USB flashcards have longer read times than harddisks. Most PC users and probably some reviewers have never considered the trade-off between seek time and read speed - or even what these words mean.
NEED FOR A TECHNICAL MANUAL
Although most users will have no interest or understanding, it would be well worth your while to spend effort on a comprehensive operations manual.
This might begin with simplified user guide (including basic principles (and how to install eBoostr and how to choose a suitable USB flashcard).
The next section might carry detailed technicalities such as the excellent information emerging from this forum.
There is a need for a hints and tips section covering configuration including which sort of folders might be excluded from the cache and which programs should be given priority.
Information on what the speed test is actually measuring should explain that random testing will naturally give variance beteen results. (In version 3, this particular test drove me silly because I did not understand the nature of the test). I am still uncertain about interpreting the Statviewer and the eBoostrmeasure.exe program initially took a while to work out how it differed from the random speed test.
Such a manual might baffle the non-technically minded but I believe if this information were public, the reviewers and bloggers might understand the product better and pass less misinformation on to potential users.
I don't want to be a wet blanket on such a good product but I have found that even the best engineers and technicians often fail to understand the nature of their users. Maybe a secondary beta test could involve a panel of ordinary PC users who just want their machines to respond faster.
I know this is a tall order but I'm allowed to dream aren't I?
David G
