Normally, you should see difference in the memory tab. Or perhaps you enabled the prefix matching option for this TM? Then CafeTran does not highlight differences in fuzzy matches since it only compares prefixes of the words in the segment (not whole words).
thanks, this did the trick.
Sorry, there is one point concerning fuzzy matches' appearance that nags me.
When I translate the sentence This car is red and then This car is not red CT will show the fuzzy match and point out the word not.
But when I first translate the sentence This car is not red and then This car is red CT will show the fuzzy match and point out – nothing. Only the percentage shows me something must be missing (as there is no text with red background)
This is a pretty simple example, but imagine a longer segment where only one simple "not" (or anything else, a very small, but decisive word) is missing.
I thought this is a mere question of enabling "Compare to target segment", but as far as I see this is not the case. I really suppose I am wrong, but I actually do not find the setting to point this kind of fuzzy match difference out.