As someone that has always preferred to carry more smaller capacity cards than one big card Sony's claims of equal performance was why I went with 3 of this 128GB cards after experiencing issues with the ProGrade Gold 128GB (I should have read the specs more carefully as this card could not record 8K Raw).
When I read the PetaPixel testing I was disappointed that the lower capacity Sony cards were not performing as well as their 512GB offering but in real world use this has never been an issue for me so I got over it.
Delkin and Angelbird were outliers for me as the brands were less know² and not nearly as available locally in Canada so I did not research them as much as the other brands. SanDisk feels like a crap shoot due to how often they are counterfeited and I agree Lexar has lost its luster so I went with Sony.
Think I am going to pick up a Angelbird 256GB card and card reader soon for another option.
I have the Sony 256 and can't imagine the differences noted in the article affecting how I shoot. If I were a professional sports photographer I can see wanting the best possible, but for the rest of us, any of these cards look good:
The cards are all so fast that the difference between the top performer in clearing a full buffer and the card that came in second from last is just 1.5 seconds.
Then on how many images at 20 fps you can shoot in a row, for the Sony 256 it's 285 shots versus Sony 512 of 330 shots. That looks like a big deal on a bar chart, but there's no way I'm going to ever do more than 25-30 shots at in a row anyway. At 20fps, 285 shots is 14 seconds, and 330 shots is 16.5 seconds – which is basically shooting 20fps video clips at that point.
The article also points out is they all get throttled once they heat up, which I would think happens quickly on warm days with a lot of shooting. It would be interesting to see the thermal response of all the cards tested.
Something else I just noticed – they don't start all of the bar charts at zero. In my opinion, that makes for a misleading visual impression of the performance differences for those charts.