Re: [quicwg/base-drafts] Remove ack_delay_exponent TP (#2670)

[Kazuho Oku <[e-mail]>]

Unsurprisingly, I prefer _b_ (multiplication), and the reason is as follows.

As @huitema points out, the computational cost of using a multiplier is no worse than using an exponent, and I second @marten-seemann's comment that use of exponent is strictly worse than using a multiplier. I do not think there is a justifiable reason to choose _a_ (exponent).

The network overhead of using no scaling factor would be 2 to 3 bytes per packet that contains an ACK, for the primary use-case of QUIC v1 traffic. This is because we can assume that connections between web browsers and servers would be in most cases greater than 16.383 milliseconds.

Considering the effort we have spent in reducing the number of bytes being transmitted on the wire, I think it might make more sense to have a multiplier than removing the scaling factor.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/2670#issuecomment-501521948