Re: duplicate ops (Re: updated assembler)


"cr88192" wrote:

[HW-drivers]
64-bit drivers: that is the problem we have now,
the OEMs are simply not bothering at this point...

Unfortunately right.
And I add: "if we could have just the needed information"
then we could write the drivers ourself regardless for which OS ever.

Are these new addon-cards designed by robots and nobody knows how they
really work internal ?? (is the former utopic future already present?)

[emulation and speed...]

It will depend on the ratio between CPU/BUS-speed and HW-requirements.

My fast line cameras need IRQ-response within a few nanoSec to keep the
frames stable and in synch with real time. So if I like to use it without
any expensive extra hardware (driver chip and huge,fast buffers) it needs
at least a 2 GHz CPU and a 200 MHZ BUS running unprotected (all ring0).

So a factor *5 wont do it here, but granted this is a special task anyway.

I figured more recently, that since we are only simulating a small piece
of (presumably relocatable) code, we only really need to allow such a
table for "executable" memory, and not a full 4GB address space.
this leads to a much smaller array.

Yes, for not top speed hardware your approach may be fast enough.

But think of IRQ9 driven direct screen (ie: 100Hz/100KHz 1024*768*8/32)
display, a delay of ~13 nanoSec will produce a one dot blurring.
Looks like this IRQ9 method isn't used anymore because windoze's
hardware treating takes too much time even on >2GHz machines.

[side talk..]
I think I need a newer motherboard.
....
dunno, any one else have issues with the NForce3 150 or 250?...

Sorry, due to lack on information I'm bound to S3 and ATI graphics.

Perhaps another problem with BUS-timing or wrong PCI-config on
HW where neiter BIOS nor the OS knows about how to ... ?

....
newer boards seem to demand PCI-E cards afaik.

My newest (AMD64, bought this year) still got an 200 MHz AGP-slot,
but I never buy brand-new brands, I always let others test it first :)

__
wolfgang



.