On Fri, 10 Oct 2003 23:14:45 +0100, Tarquin Mills <[EMAIL PROTECTED]> wrote:



Phoebus Dokos wrote:
<snip>
The following paragraph may be wrong.The hardware is easy, just use a
parallel to USB converter. Parallel is faster than USB 1.1 and below. The
problem is writing the USB stack, other USB writers target keyboards as
they can work with same driver. One route to writing USB might be to
support a device that exists in both USB and ser or par format.
I notice Phoebus mentions getting the solid ink colour printer that I
talked about in the QL Today in Colour argument, which has got a lot cheaper
and is now manufacture by Xerox. I wanted to buy one of these but my mother
stopped so I got a Epson C80, that neither of us are happy with.
I think all retro platforms are suffering from the same problem, we can
either try and keep up, which can help push us forward, or design a retro
printer. A half way house would be an embedded computer converting from
par_ to usb and converting print languages, ESC/P2 -> winprinter. This
could be sold with a printer by dealers.

This is NOT easy although it seems so. So-called USB-to-<something> do not work both ways. They require a USB host which is what PCs (and some other non-PC machines like Apples) have.
In effect that means that you can add a parallel port to a USB capable PC but not a parallel port to a USB-only printer. The USB to Parallel (or USB-to-Serial) have a serial or parallel controller that has to communicate with a USB host in order to work. In effect, no USB host, no functionality.
There is a potential method of adding USB to some QL hardware platforms but would only work on Qx0s. In effect we need an ISA CardBus adapter and then a CardBus controller (Host) for USB -or firewire or whatever else-.


I do not know how feasible that could be given the non-full implementation of the ISA bus on the Qx0 but it is a possibility. Of course three sets of drivers would be needed. One for the Cardbus adapter, one for the whatever interface we would be implementing and finally the printer driver on the program. Is that logical? Without a high-end processor (ie V.5 ColdFire at a decent ie. > 333 MHz) I do not think so.


For the "conversion in printer languages" argument:


The winprinters either are driven directly by the driver (in effect no language exists only a method of transmission over the line) or have a rudimentary language that is so garbled that it's practically impossible to decode for either practical (the documentation on their transmittal methods is non-existent or closely guarded by the manufacturers) or it requires so much processing power that it's simply non realistic to use them.
So in order for the "machine" to work you would need to pay that much that upgrading to a full fledged *DECENT* printer would be probably cheaper!



Inkjets are the real problem, dot matrix printers are still made (why?),

Explanation (btw not only 24 pin are manufactured but also 9pin and even 18pin) is simple.
Many applications require carbon copies. In countries with high tax evasion for example (in my native Greece), there's no way that an invoice can be printed by anything else than an impact printer to ensure "REAL" copies in one go.


Moreover impact printers are relatively simple devices that can produce output in far greater numbers than anything else on the market for batch print jobs like telephone (or other utility) bills for example. (These usually are handled by line printers but for smaller companies, daisy wheel or impact matrix are still required).
One would argue that if speed and mass production is needed we should only keep daisy wheel or line printers (daisy chain). In reality however many printjobs require simple graphics that a line printer or a daisy wheel is incapable of producing.


Finally do not forget the dominance of COBOL in the business market worldwide (before I hear anything weird let me remind you that approx 80% of business computing is still done on mainframes and COBOL) and its default relation to impact printing. Laser printers or other page-based printing solutions just don't cut it there....


they are 24 pin and not warnout. Non cheaper laser are still OK. Linuxs
increasing market share could could change the printer market, though I am
know questioning the market share figure for Windows.


Geoff Wick wrote:
PS: I am having a little trouble displaying the new QL colours on my green
monitor (sob). Can anyone help? (burst of laughter).
I may be able to help.

Hmmm you could use those "coloriser filters" for b&w tv sets... Anyone remembers them? :-D



--
Visit the QL-FAQ at: <http://www.dokos-gr.net/ql/faq/> (Still uploading stuff!)
Visit the uQLX-win32 homepage at: <http://www.dokos-gr.net/ql/uqlx.html>
Visit the uQLX-mac home page at:<http://www.dokos-gr.net/ql/uqlxmac.html>
              • ... Marcel Kilgus
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... Marcel Kilgus
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... desin
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... gwicks
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... Tony Firshman
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... Marcel Kilgus
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... gwicks
              • ... "Phoebus R. Dokos (Φοίβος Ρ. Ντόκος)"
              • ... Robert Newson
              • ... gwicks
  • ... Geogwilt
  • ... merlinch
    • ... Tony Firshman
    • ... Dilwyn Jones

Reply via email to