Re: ST developments New graphics card.

Steve Potts wrote:
In message <b10ffd5950.news@xxxxxxxxxxxxxxxxxxxxx>
Richard Porter <dontusethis@xxxxxxxxxxxxxxxxxx> wrote:

[snip]
I'm guessing it could be a timing problem. I wonder if the Vpod was tested
on a 300MHz machine.

Would the processor speed make a difference??? I assumed Vpod ran at its own
speed as far as producing a display is concerned. I don't understand how
these things work to be perfectly honest though.

Hardware wise any podule card will be accessed with the same I/O timings regardless of the processor speed. The RPC originally came with a 30MHz ARM6, and correctly designed peripherals should work with that right up to the StrongARM running at up to 10x the clock speed.

However in practice, a faster processor may deliver events quicker or in a different order compared to slower ones. So with a very complex podule, there may be some unexpected timing consequences, which need to be addressed in firmware.

Or are you saying its a kinetic and therefore its architecture is different? Again, I don't really know what difference that makes - never had my hands on
a kinetic, so not sure about the practical differences.

A Kinetic is a StrongARM card with it's own faster memory on board, which is used in preference but also in addition to the slower memory on the motherboard. This can result in memory intensive operations running quite a bit faster. The drawback is that fast DMA transfers can't be made in to the on board memory, making disc operations from some 3rd party IDE and SCSI cards quite a bit slower.

Podules which work with the StrongARM, should not be affected by the Kinetic apart from the lack of DMA. If they work in a non DMA slot, they should be ok. However, some podules (e.g. older APDL IDE card) if placed in a DMA slot and the DMAManager module was present, would assume they could use DMA and could fail. I got around this by unplugging all instances of the DMAManager module, from the ROM and also any on podule cards, and it then worked perfectly with the Kinetic*

* Until each new Select release where the module numbers in the ROM changed resulting in the DMAManager no longer being unplugged. A bit of fiddling was required each time to sort that out, until ROL released a fix with RO6.

---druck
.

Relevant Pages

  • Re: Testing Sirius RAM card at 8MB?
    ... What SCSI card are you using? ... support DMA within the first 4 MB of a memory expansion card, ... doesn't supply the bank address during a DMA cycle. ...
    (comp.sys.apple2)
  • Re: Tandy 1000HX info
    ... Also, when adding the memory card with the DMA chip on it, how exactly ... video access hardware didn't have to change, just the 'map' of how the ...
    (comp.sys.tandy)
  • Re: DMA transfer speed issues on PCI card with PCI and PCI-X busses
    ... we're using a PCI card with large on-board memory and a PCI interface chip with DMA busmaster engine ... The card is working in PCI and PCI-X busses with 33 MHz and 66 MHz. ... The Scatter-Gather list is located in PC memory ...
    (microsoft.public.development.device.drivers)
  • driver: DMA really really slow
    ... I have a card that can read commands via DMA queue areas from main RAM. ... Usually, DMA speed is around half the speed of main memory, from ... it says DMA speeds are 17-25mb/s Yeuk!!!! ...
    (comp.unix.solaris)
  • [PATCH] spelling fixes: arch/arm/
    ... copy data to/from buffers located outside the DMA region. ... the remainder of memory is at the top and the DMA memory ... Peers on the PCI bus running ... - * devices are controlles by different NPUs on the same bus, ...
    (Linux-Kernel)