To help my overloaded brain ... when folks refer to "the Prolific problem"
... is that a reference to the usb/serial cable?  or to a chip on the
Windows machine?

Tom M.


On Fri, Apr 5, 2019 at 2:56 PM Brian K. White <[email protected]> wrote:

> I've actually never had a problem with a Prolific based adapter either,
> but I know that it is a real problem anyway.
>
> Pick any random cable purchased off amazon or ebay, and it might or
> might not be garbage, until you prove it by actually using it.
> I've bought probably 20 or more if you count both work and hobby stuff
> over the last 15-20 years, and they've all actually worked ok, but that
> is just luck. The bad ones do actually exist, and they look just like
> the good ones, unless you stick to a few name brands.
>
> And it's also a real thing that even when you have a "good" one where
> the chip isn't counterfeit, the chip and the driver even at their best,
> still lacks certain features that matter.
> John Hogerhuis has explained before (maybe on the fb list instead of
> here) that the Prolific interface doesn't allow you to avoid a certain
> amount of usb buffering, which can screw up the timing of signals, and
> can prevent flow control from actually working, because the flow control
> signal itself is queued up and sent only when the next packet is full
> and ready, and by then you may have already overrun the M100's receive
> buffer etc.
>
> That's the kind of problem that isn't disproved by simply having the
> good luck to not have hit it yet.
> It's still there and will still hit someone else under some other
> condition, or may still hit you tomorrow even if it never has before,
> because it's dependent on variables and variable conditions and usage
> pattern.
>
> In the specific case of TPDD emulation, I think you can actually get
> away with Prolific, and a total lack of flow control, because of the way
> the TPDD protocol happens to work. It works in fixed small size packets,
> where the entire packet fits inside the M100 receive buffer, and, you
> only ever get one packet at a time because the TPDD doesn't send a new
> packet until after the M100 sends an ack after ingesting the previous
> packet. So the tpdd protocol implements a flow control itself, instead
> of relying on the underlying serial channel to be doing that right. So
> it's another way you can think there is no problem when there still is.
> Or, there is no problem *only* as long as you only keep using it for
> exactly that one purpose which happens to work.
>
> --
> bkw
>
>
> On 4/5/19 12:31 PM, Jonathan Yuen wrote:
> > Hello All,
> >
> > I've used a prolific usb-serial cable to connect my laptop to a
> raspberry pi and it didn't have any handshaking at all, just tx, rx, gnd,
> and a constant 5V to power whatever (like the pi itself).  This wasn't real
> rs-232, just the ttl level serial commmunication, but the pi wanted ttl,
> not real rs-232.
> >
> > [email protected]
> > ---
> > När du skickar e-post till SLU så innebär detta att SLU behandlar dina
> personuppgifter. För att läsa mer om hur detta går till, klicka här <
> https://www.slu.se/om-slu/kontakta-slu/personuppgifter/>
> > E-mailing SLU will result in SLU processing your personal data. For more
> information on how this is done, click here <
> https://www.slu.se/en/about-slu/contact-slu/personal-data/>
>
>

Reply via email to