ALL parts have different performance. You play this "lottery" every single time you buy ANYTHING.
So you are saying that two Samsung SSDs might have 20% different performance, the same way that a Sandisk and Samsung will have 20% different performance? Prove it. Show me a single benchmark of 20% difference of two properly functioning Samsung SSDs of the same size in an rMBP. You won't find one. Yeah, two Samsung SSDs may literally not have identical performance, but I can guarantee you there won't be a 20% gap. Maybe 2 or 3% from test to test, which is a far cry from 20%.
Abusing Apples, frankly lenient, exchange practices just to get a supposedly more reliable component when there is no actual problem is border line return fraud.
Border line return fraud? Instead of focusing your fury on consumers who don't like the idea of playing the SSD lottery, why don't you focus it on Apple for even putting consumers in the position.
Sometimes 20% is significant, sometimes it's not. If a Honda Civic were expected to get 40 mpg, but you happened to buy one with a different engine component resulting in a max MPG of 32, would you feel like you had gotten what you had paid for? What you feel a bit shafted because, due to luck of the draw, you were missing out on that 20% performance?
I'm not saying it's a perfect example, but if a given consumer has the time and wants to put forth the effort to get a Samsung SSD, which has 20% superior performance to the Sandisk, I don't see how an exchange even comes close to return fraud. You might think it's a waste of time for that consumer to care about the 20% performance difference, but different strokes for different folks, right? Return fraud? No... just no.
There's also the problem of the non-scientific nature of these claims about "20% better performance". You guys are basing these conclusions on a handful of internet DIY he-saids done by different people under different circumstances using potentially different methods.
Not really. It's pretty well settled that the Samsung SSDs pull speeds of about 20% higher than the Sandisk. Yeah, these are synthetic benchmarks that don't necessarily translate to real world use, but they are repeatable and verifiable, as countless benchmark screenshots posted to this very forum have proven.
Keep in mind I'm not the OP, nor am I necessarily saying the OP should return a laptop 50 times to try to get a Samsung SSD. In fact, I tend to agree with many of the posters in this thread that doing multiple returns would be a pretty huge waste of time. But if he has the time and wants to do it, I don't see any problem with that. Again, if there's any problem here, in my view it is that Apple is shipping parts that have such a significant performance difference from others, be it a real-world difference or not... enthusiasts care about benchmarks, right or wrong.