Timothy Miller wrote:
On 6/20/06, James Richard Tyrer <[EMAIL PROTECTED]> wrote:
<SNIP>
65 nm gives you shorter wires. Are they faster?
I'm not sure. I think the capacitance goes up, taking away some of
the advantage.
Yes, just scaling doesn't achieve the expected proportional increase in
speed. That is why there is talk about lower dielectric constants
(using other things to replace Silicon DiOxide) and using Copper
conductors for interconnects. τ = R*C so reducing R does just as much
good as reducing C for the initial charge of the line (if unterminated).
But the initial charge time is determined by the propagation speed as
a transmission line and only reducing C will help. Z = (L/C)^2 but the
speed of the line is determined by C -- larger Coax has a faster
propagation speed because it has lower C/meter. Another problem is that
scaling implies narrower interconnects which means less inductance and a
lower line impedance. Simple solution is to not shrink the interconnect
width but then the chip doesn't shrink as much as you would expect.
Clearly, this isn't simple.
<SNIP>
>> What versions of Unix block I/O to do other things? It is normally
>> I/O that gets preference.
>
> I'm not sure what's being discussed here. Usually, you try to overlap
> CPU and I/O by using DMA. But the process waiting on the I/O is
> blocked. And sometimes, you can't do the I/O via DMA.
The issue that I have is when either keyboard or mouse input is blocked
by other processes.
Linux has a problem that Solaris resolves. Under Linux, the mouse
driver talks to the X server, which talks to the graphics card. If
the X server is heavily loaded, the cursor skips around. Under
Solaris, the mouse and graphics drivers have a standardized interface
that allows the mouse driver to talk directly to the graphics driver,
bypassing X.
A secondary issue is when a redraw has to wait for
another process. Some of the time, I presume that it has to wait for a
swap which can't be fixed except with more main memory.
True, but if it's X11 that's swapped out, at least we can deal with
the mouse cursor issue.
> With a graphics card that uses DMA, you offload the I/O overhead from
> the CPU to the GPU, so the CPU can do other things. This has the
> effect of lowering the load for the X server, so it gets a higher
> process priority.
With a graphics card that runs the X server, you would offload even more
overhead. :-D
True, but is it worth it?
Don't know for sure. This would be a unique product. Although I note
that this wouldn't be a totally new idea. TI was pushing this idea
although I don't think that a commercial product ever made it to market.
DEC used to sell workstations which had a separate processor to run
the X server. IIRC, this was not nearly as powerful processor as the
main CPU.
IAC, it seemed like a way to proved the "Amiga like" performance. Since
the only thing that could slow down the mouse and keyboard input would
be PCI bus contention and the X server would never be swapped out since
it would have its own dedicated memory space.
--
JRT
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
