they are absolutely not.
-
-
Replying to @whitequark
I'm perplexed how it works then. Non differential seems to poop out in the 100-200 Mbps rates in my experience without power hungry termination.
1 reply 0 retweets 3 likes -
Replying to @bml_khubbard
it's pseudo-differential: it looks an awful lot like differential most of the time (though the signals are specified relative to ground anyway), but instead of a comma with a proper line code, they use a condition of both D+ and D- being low to delimit packets
1 reply 1 retweet 14 likes -
Replying to @whitequark @bml_khubbard
that means you can't actually use a comparator to receive USB. you basically have to sample D+ and D- separately referenced to a voltage you generate locally, and then detect D+>D-, D+<D- and D+=D-.
2 replies 1 retweet 22 likes -
-
Replying to @JodiTheTigger @bml_khubbard
> I'm perplexed how it works then. this is generally what happens when you open the USB spec and read it. also, it doesn't work very well.
1 reply 5 retweets 38 likes -
AFAIK you do use a differential receiver for HS data, it's just that you *also* need single ended comparators to detect EOP. Also this design is almost the worst possible design for audio devices. USB audio whine is now a plague, even in professional productions.
1 reply 1 retweet 20 likes -
if I remember correctly, the data lines are referenced to ground, so while you certainly -could- use a differential comparator, it doesn't give you a lot of advantage
1 reply 0 retweets 4 likes -
The termination is through GND (half the differential impedance on each leg), but you still need to use a differential receiver. There's no way you can get the required noise immunity without one.
2 replies 1 retweet 17 likes -
Replying to @qfp64 @whitequark and
It's shitty cursed differential.
12:13 AM - 25 Jun 2019
0 replies
0 retweets
9 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.