But that article contradicts itself - it says about it being a silent upgrade, which is true, Microsoft have at no point announced or revealed the "upgrade". Yet a few sentences later it talks about how Microsoft is promising to be sending new ones out? If there source of this comes from a chatlog with a customer service employee then I would take anything they say with a grain of salt.
If the information does not come from Microsoft as an announcement, I would never rely on it as fact. Something a tech support guy/or customer service worker says may not be true - in fact I know a lot of these customer service roles are not even MS employees, but just contracted in.
It's possible that Microsoft initially ordered x number of 4200 processors. As a result of the unexpected initial demand for the SP2, Intel could not keep up with that initial demand and Microsoft bought some available 4300s until Intel could fulfill the original 4200 order.
The article may contradict itself, but I was more referencing my original speculative post on page one. As this wasn't a "planned an announced" change, my speculation is that MS is using either one they can get their hands on to keep up with demand. While I still have no concrete PROOF of this, given the recent release of 4200u equipped units and the relatively low availability/high demand of the 256/8 model, it's not too hard to connect the dots.
That said, I ALSO believe that the main difference in the two chips is psychological more than anything. You may see some slight % changes in synthetic benchmarks, but I doubt you're going to notice much in actual world performance, especially since the SP2 starts throttling itself @ 80c.
Actually CPUBoss is using primarily Passmark itself as the only source of information for their automatic "too close to call" conclusion, because they did not found any other benchmark to scrape.
I don't like CPUBoss either, but I agree with their conclusion this time: the results are inside the margin of error for poorly made benchmarks such as these ones.
Another example is Geekbench, where the differences between the two are so negligible the 4200 actually tops the 4300 in multithreaded performance (though the 4300 tops the 4200 in single threaded).
EDIT: I take it back -- Geekbench's results table sorting seems to be broken. There's probably a 12% difference.
Still, the 1402 is kind of on the edge, it could go either way. I do have an 256GB 1402 with 4300U CPU. Maybe there are multiple manufacturing plants and some of them did not finish their 4200U batch yet ?
Either way, I would complain to Microsoft about this, maybe they can do something about it.
But such (expensive) purchases should be made with great care and prior thought. I, for one, do not expect any kind of support to my advantage after I buy such a product. From my experience, once you buy an item from a big company, they don't really treat you like a customer but more like an "issue" that will be solved in a very standard way. They don't go out of their way at all, no matter what they are lying on the phone. Their support employees probably just wait for their shift to end and go to the pub... However for business-to-business deals, the relationship is different, they respect businesses a lot more, for obvious reasons.
I'm surprised by how "consumer-friendly" and RMA happy MS stores are. Here (no MS stores in this country) I basically had to argue for half an hour in order to get a dead-on-arrival type cover replaced...