It's called binning, it allows them to test the CPU's and sort them into two bins, one that has all efficiency cores working and another where they can disable two and still utilise them. But they are both the same design. This allows them to keep down cost when there are low yields in the production, most manufacturers do this.
The Pro and Max are
NOT binned .
From Apple's press release
Supercharging MacBook Pro and Mac mini, M2 Pro and M2 Max feature a more powerful CPU and GPU, up to 96GB of unified memory, and power efficiency.
www.apple.com
M2 Pro
M2 Pro
The physical dies are different sizes !!!
Back in the M1 generation :
www.anandtech.com
The M1 Max is about 200mm^2
bigger die. That bigger die area is larger than the whole M1 die !!
There have also been some other posts that the Pro is physically cut from a Max die. There are different widths there above and there is NO die cut line in the middle of the Max. Those reports are just smoke.
Apple did to some very heavy design reuse where they take the layout design of the Max and just make some very narrow adjustments to the design to get a Pro. They design the chip once and they run it through a process to make it smaller without having to do tons of extra work.
If Gurman is right, and there is a chopped down E cluster ( 4 -> 2 ) that doesn't appear at all in the Max , then Apple has taken on the overhead of designing two different dies. There are rumblings about a ginormous iPad Pro (13-14") with a M3 Pro. That might be enoug 'extra unit volume' to help pay for a bigger fork between the Pro/Max. They will drop to something like 70% shared intersection layout from 95+% . It wouldn't be a huge increase in differences.
'Binning' is illustrated by the M2 Pro having 20 GPU cores and only 16 or 19 being active in the delivered SoCS. Same die ( have to be binning the
same physical die) and turn on a variable number of cores.