2 1 7 Introduction to Dynamic Branch Prediction

Поділитися
Вставка
  • Опубліковано 24 гру 2024

КОМЕНТАРІ • 13

  • @wi0lono
    @wi0lono 4 роки тому +9

    This video series is exactly what I needed

  • @hanboguo3737
    @hanboguo3737 2 роки тому +2

    Love these lectures!!! Well explained with helpful animations and fast-paced. These lectures are for students who have taken an introductory course in architecture.

  • @Mrf.7213
    @Mrf.7213 2 роки тому +3

    Thank you for the lecture! Great stuff!! You could have shown the performance of the 2-bit BHT on the loop example to contrast with the one of the 1-bit BHT

  • @万宇阳
    @万宇阳 4 роки тому +5

    Great explanation

  • @themoviesongofken
    @themoviesongofken 2 місяці тому

    I'm a little confused here. BHT is used to store the history of individual branch. Is there anothere table to record prediction state (prediction wrong or correct)?

  • @antwanwimberly1729
    @antwanwimberly1729 Рік тому

    Why can 2 for
    Loops of equal magnitude say linear time have different run times??

  • @axe863
    @axe863 Рік тому

    Transfer Learning to increase branch predictability ???

  • @haeusi1999
    @haeusi1999 2 роки тому

    Good Video, but is it possible, that the 2-Bit Branch Predictor graph is no correct? If we mispredict from state 10, we should end up in state 01 and not directly in 00. Same goes for 01 taken to 11, shouldnt it change its state to 10? Or am I getting something wrong here?

    • @sellem3
      @sellem3 2 роки тому +1

      isnt that the second graph he showed? the first flips to 00 directly, the other decrements and increments, and in practice they dont show much performance difference

  • @User-pi3nf
    @User-pi3nf 4 місяці тому

    Sehr gut.. Danke

  • @Numinus1
    @Numinus1 Рік тому

    Great video, thank you!

  • @avvarutheja
    @avvarutheja 4 роки тому +5

    This is a good lecture, but for a fresh student (who is just learning micro processors concepts), its too hard to get the concept & basic idea behind the algrithm. You need to take students from a bit low level to high. For example, you should explain how that state machine is formed & what is the basic idea of branch prediction. This makes students to think than just blindly memorizing the state diagram.

    • @mikeoxmaul8678
      @mikeoxmaul8678 2 роки тому +1

      He did a decent job explaining. get gud bruh