I would not say that with such certainty! The only evidence of this is sparse and anecdotal. Right now, there is no evidence that I see that indicates that the performance difference of a physical 1673S and a physical 1693S is any different than the kind of difference that you’ll see between two different physical 1693S drives.
Now, it is possible that a physical 1693S might tend to be better than a physical 1673S because over time, LiteOn may have refined its manufacturing and calibration procedures, but there are always going to be subtle differences between every drive that gets produced simply as the result of normal manufacturing variance. Two drives that came out of the assembly line right next to each other will not necessarily be identical, and I’m sure that a good 1673S drive could very well out-perform a bad 1693S and vice-versa.
Remember, you want to look for differences that are statistically significant, and in this case, you have to show that the inter-1673S/93S variation is greater than the intra-1693S variation.