|
List Info
Thread: Re: -current (4.99.29) on Z50
|
|
| Re: -current (4.99.29) on Z50 |
  United States |
2007-09-13 09:58:00 |
Hi Nick,
>... do you use 48MB ram, or have you done the 64MB
mod?
64MB mod? I've got 48MB. What is the 64MB mod?
>Just wondering as far as usability ...
Usability is fine as long as you don't expect the thing to
win any speed
records
GTK2 apps are probably too much, as is freetype. A WWW
browser with
frames and
Java would be nice but I haven't found one that runs on
mipsel yet.
I've got a 2GB microdrive and a 3c574 ethernet in the system
I'm currently
testing. Your 340MB microdrive will probably be too small
for any X apps
and stuff.
djb
|
|
| Re: -current (4.99.29) on Z50 |
  United States |
2007-09-19 22:11:22 |
Dave,
I will respond to each point as you did, read below..
Dave J. Barnes wrote:
> Hi Nick,
>
> >... do you use 48MB ram, or have you done the 64MB
mod?
>
> 64MB mod? I've got 48MB. What is the 64MB mod?
The 64 MB mod was developed by the main developer of
vr-linux, and as
far as I know of, only a few exist, all of which I believe
are in Japan.
Below is an excerpt from an email from Kawashima:
Just from my memory, this memory chip is 0.5mm fine pitch
TSOP.
So if you have enough soldering experience, it will be
possible.
You need 2 memory chips for 16 MB. (64Mbit DRAM x 2).
But many kind of 64M bit DRAM are there, so you need to
select exactly
compatible
memory chip. This memory chip will be discontinued already,
so you might
need to get
second hand memory board.
I did not document about it, sorry.
Kawashima
>
> >Just wondering as far as usability ...
>
> Usability is fine as long as you don't expect the thing
to win any
> speed records
> GTK2 apps are probably too much, as is freetype. A WWW
browser with
> frames and
> Java would be nice but I haven't found one that runs on
mipsel yet.
>
> I've got a 2GB microdrive and a 3c574 ethernet in the
system I'm
> currently
> testing. Your 340MB microdrive will probably be too
small for any X apps
> and stuff.
>
> djb
>
Yeah, I've been hunting for 2Gb microdrives, and trying to
get a
straight answer on whether microdrives > 2GB will work
with the z50. (I
have a recollection that someone said they were running a
4GB
microdrive, but I can't find the message / email.) I did
manage to chase
down one of the AA battery packs though, that was a nice
addition to my
z50 setup. Anyway, thanks a lot for the info, it's much
appreciated.
Nick / npt
"All unix, all the time."
http://npt.ath.cx
|
|
| Re: -current (4.99.29) on Z50 |
  Canada |
2007-09-19 23:27:37 |
> Yeah, I've been hunting for 2Gb microdrives, and trying
to get a
> straight answer on whether microdrives > 2GB will
work with the z50.
> (I have a recollection that someone said they were
running a 4GB
> microdrive, but I can't find the message / email.)
Well, I'm running mine with a Seagate "8GB" drive
wd0 at atabus0 drive 0: <ST68022CF>
wd0: drive supports 16-sector PIO transfers, LBA addressing
wd0: 7629 MB, 15501 cyl, 16 head, 63 sec, 512 bytes/sect x
15625008 sectors
wd0: drive supports PIO mode 4
and I've had no problems (well, no problems I can ascribe to
drive
size, at least).
/~ The ASCII der Mouse
/ Ribbon Campaign
X Against HTML mouse rodents.montreal.qc.ca
/ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3
27 4B
|
|
| Re: -current (4.99.29) on Z50 |
  United States |
2007-09-26 12:25:19 |
>>>>> "nt" == nick thompson
<nicholas.thompson1 mchsi.com> writes:
nt> Do you notice a difference in how long your
batteries last
nt> with the 8gb vs a smaller microdrive?
The thing that sucks about them is, a long time ago in the
days of
MS-DOS someone wrote some pretentious-sounding academic
paper
convincing everyone that optimal hard disk spindown timeouts
were
extremely short, like one minute or less, because of battery
life or
something. Anyone who's actually used a laptop realizes
that battery
life be damned---the things you really care about w.r.t.
disk spinning
are noise, shock-safety, and access latency. but anyone
who's been
exposed to this awful academic paper has some kind of
hypnotic block
installed inside his mind preventing him from realizing this
obvious
truth from his own experience. It's particularly effective
on
unimaginative electrical engineers.
The microdrive's timer consequently seems to be 5 or 10
_seconds_...!
Making this even worse, NetBSD has some kind of
``trickle-sync'' deal,
or something, so at least for me something hit the drive
about once
every 30 seconds. It would spin up, spin down, spin up,
spin down.
Yes, I know about noatime, didn't help. I found this
hair-clawingly
annoying, much worse than a louder drive, and also sometimes
it would
block something mid-keystroke, so thanks to soft-key-repeat
I would
get ten 'a's instead of one (or ten backspaces, since I was
correcting
things so often), because the software would perceive the
key as held
down while the disk was spinning up and balance its accounts
by
generating a bunch of autorepeated characters. I used to
get so angry
at that thing I'm amazed I never cracked the screen with the
stylus.
I tried, I really tried, thinking if I broke it I wouldn't
ever have
to use it again. god I learned to hate that thing.
anyway I've told this story before so sorry for the spam.
but I would
be very stingy in offering money for a microdrive of any
size.
|
|
| Re: -current (4.99.29) on Z50 |
  Canada |
2007-09-26 12:28:15 |
> The microdrive's timer consequently seems to be 5 or 10
_seconds_...!
> [...]
I haven't noticed any such issues with my z50. Noise - I
have to put
my ear within an inch or so of the microdrive to tell
whether it's
spinning by noise. Shock safety - it's quite rare that the
thing is
running when it's other than on my lap or on some kind of
stable
support like a table, but of course all it takes is once.
Access
latency - now that's the one I notice, though I'm not
noticing the
kinds of problems you describe, like generating false
autorepreat from
the lockup; the sort of thing I notice more is stuff like
typing ls and
getting very slow response the first time. It actually
feels like
stuff getting swapped out, and for all I know it may
actually be; I've
worked with low-RAM machines enough that I know what it
feels like for
the shell to get paged out whenever it's idle for longer
than a handful
of seconds.
The thing I notice most is that, when I'm hitting the disk
hard (eg,
building a kernel), it gets hot. It gets warm enough when
touching the
case outside the CF bay that I wonder if I should be
worrying about the
disk itself....
/~ The ASCII der Mouse
/ Ribbon Campaign
X Against HTML mouse rodents.montreal.qc.ca
/ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3
27 4B
|
|
[1-5]
|
|