From: Wolfgang Moser (womo_at_news.trikaliotis.net)
Date: 2006-10-08 19:12:19
Hi Marko,
Marko Mäkelä schrieb:
> Wolfgang Moser wrote:
>> Marko Mäkelä schrieb:
>>> A friend of mine recently told me about the MAX3421E from Maxim
>>> <http://www.maxim-ic.com/quick_view2.cfm/qv_pk/3639>.
>>>
>> to me this chip looks like just another USB to any other
>> general availability serial bus. Beside RS232 (ft232bm or
>
> Those chips you mention use their own proprietary USB protocols and
> need special drivers on the PC. The MAX3421E (as well as its precursor
> MAX3420E) allow you to define your own protocol over the USB by
> controlling the chip over the SPI bus. I would guess that the two most
> popular protocol classes to implement would be usb-hid (human interface
> device, i.e., keyboards, mice, joysticks, or low-bandwidth sensors) and
> usb-storage (block device). Both classes are natively supported by
> operating systems.
Oh well, I did not get this point and may have
overlooked the corresponding topic in the datasheet.
Now that I'm rereading I realise that it is similar
to the SL811 in concepts, because it can be used as
host, too.
>> At first there's the Cypress EZ-USB series with the FX
>> and FX2 µCs.
>
> The Keyspan usb-to-serial converters have been built using that chip.
> Do you know if it supports USB 2.0 or USB-on-the-go? Is it available
> in small quantities?
I did not have a detailed look into the Keyspan
adaptors times ago, but from what I can remember and
some conclusions:
* built with EZ-USB FX (not FX2)
* USB 2.0, full speed (12 MBps)
* no USB-on-the-go
Even if the Keyspan USA-19HS claims to be a
High-Speed device it is no USB High Speed device.
The datasheet even more tells that it would be an
USB 1.1 device only (in contrast to my conclusions
based on reports from Linux users on which Cypress
chips were used on newer models).
> That would be understandable, as USB is a very complex standard.
Yup, while on the PC side there is libUSB available
for Linux, FreeBSD, MacOS-X _and_ Win32 (2k, XP),
which did already simplify my first steps, endpoint
programming at peripheral side and more complex
functions is something I did not collect experiences
with yet.
> I think that you are slightly mistaken about the Maxim chip.
> It requires quite a bit of USB programming knowledge,
> although it seems to handle some low-level stuff, such as
> retransmissions. I do not have any experience in low-level
> USB programming yet, though.
Yup, I was. On the other side I had a short look
into the programming manual of the Maxim chip now:
http://pdfserv.maxim-ic.com/en/an/AN3785.pdf
Could not see yet, _how_ easy the USB stuff would
get in the end.
Well, the EZ-USB chips also encapsulate the lowest
level of USB programming and you "only" need to
configure endpoints, buffers, ID strings and such.
Whenever you can save out kernal mode driver
programming, you already made a big improvement.
However as soon as more and more mini USB projects
are arising along with programming samples, this
may teach more and more people. There are already
arising some "Joystick" projects based on the
soft-USB cores from Igor Češko or Objective
Development as well as the SL811 controllers for
6502.
Womo
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.