korrents

korrents · video

How useful are MacBook Neo benchmarks?

Matt Birchler · 18m · youtube.com ↗

8 korrents from this recording

Matt Birchler did not write this page.

Every claim below is a statement made in this recording, quoted word for word and linked to the second it was said, so you can hear it rather than take our word for it. The wording comes from the transcript published alongside the recording; the sentence above each quote is our reading of the claim, not their wording.

  1. 4:53 · watch on youtube.com↗

    you weren't getting more performance than you were with a Mac Studio, which was significantly less money. So yeah, this product really didn't make sense in the Apple Silicon generation.
  2. 5:13 · watch on youtube.com↗

    I just don't think there's a place for Apple's computers in 2026 for this form factor. They just don't make a computer that remotely makes sense here, right?
  3. 1 min later
  4. 5:54 · watch on youtube.com↗

    I did not want to include benchmarks because I did not think benchmarks were telling you an interesting story or a useful story with what I was trying to tell. I wasn't trying to give you just a spec readout. I was trying to explain what life was like with the laptop.
  5. 1 min later
  6. 7:03 · watch on youtube.com↗

    for the vast majority of use cases i don't think this is going to make a big difference i do think that we stress over disk speed for a lot of things that don't really matter
  7. 2 min later
  8. 9:00 · watch on youtube.com↗

    Snapdragon, the Qualcomm chips have definitely improved their performance from a couple years ago when Apple was smoking everybody in these metrics.
  9. 3 min later
  10. 11:57 · watch on youtube.com↗

    one of the things that kind of gets me about these sorts of benchmarks is to make these benchmarks fair, when I did each of these tests, I stopped everything else on the computer. I just did this one task, and it went as fast as it could.
  11. 4 min later
  12. 15:31 · watch on youtube.com↗

    the limitation there is the context window, because the context window for this model is very small. It's about 4,000 tokens, I believe. And it's simply not enough.
  13. 2 min later
  14. 17:39 · watch on youtube.com↗

    I think it is not an improvement over what they had before. I also think they shouldn't go back. They can't go back. I think they need to figure out how to clean this up and make it feel better.