Sony to ax 10,000 jobs in turnaround bid: Nikkei - (via Reuters)

nospam writes:

In article , David Dyer-Bennet
wrote:

Guess how long it takes to boot a C64 ...

My knee-jerk reaction is "who cares?". I never had anything to do with
those early generations of home computers because they were so
completely inferior to what I was used to working with at work.

how long does it take to boot a mainframe, especially one that does so
off tape?

Minutes generally; altough I never worked with one that booted off tape
other than during installation.

there were mini computers where you had to toggle in a tiny boot loader
*each time* you booted so that it could read the actual system from
tape or disk.

The "Rim Loader" code is silk-screened on the front panel of the
PDP-8/I, for example -- and that's my screen background, so I'm looking
at it right now. (I also used one some from 1970-1977). Seeing it, the
sequence of instructions is even familiar -- 6032, 6031, 5357, 6036,
..... (That loaded in the BIN loader from paper tape, which then loaded
the actual OS you were going to run.)
--
David Dyer-Bennet, ; http://dd-b.net/
Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/
Photos: http://dd-b.net/photography/gallery/
Dragaera: http://dragaera.info

In article , David Dyer-Bennet
wrote:

there were mini computers where you had to toggle in a tiny boot loader
*each time* you booted so that it could read the actual system from
tape or disk.

The "Rim Loader" code is silk-screened on the front panel of the
PDP-8/I, for example -- and that's my screen background, so I'm looking
at it right now. (I also used one some from 1970-1977). Seeing it, the
sequence of instructions is even familiar -- 6032, 6031, 5357, 6036,
.... (That loaded in the BIN loader from paper tape, which then loaded
the actual OS you were going to run.)

forgot about paper tapes. that high speed paper tape reader on the
pdp-11s (maybe even the pdp-8s) was way cool to watch.

nospam writes:

In article , David Dyer-Bennet
wrote:

there were mini computers where you had to toggle in a tiny boot loader
*each time* you booted so that it could read the actual system from
tape or disk.

The "Rim Loader" code is silk-screened on the front panel of the
PDP-8/I, for example -- and that's my screen background, so I'm looking
at it right now. (I also used one some from 1970-1977). Seeing it, the
sequence of instructions is even familiar -- 6032, 6031, 5357, 6036,
.... (That loaded in the BIN loader from paper tape, which then loaded
the actual OS you were going to run.)

forgot about paper tapes. that high speed paper tape reader on the
pdp-11s (maybe even the pdp-8s) was way cool to watch.

I think they were the same, or very similar. Fan-fold tape moved
through remarkably fast.

But most of our 11s were disk-based, so we very rarely booted from
tape. The 8 was used standalone more often, and hence booted from tape
more often. (We had DECTape on them both. I probably still have my
reel in a box somewhere, but no reason to think anything intereswting is
on it.)
--
David Dyer-Bennet, ; http://dd-b.net/
Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/
Photos: http://dd-b.net/photography/gallery/
Dragaera: http://dragaera.info

In article , David Dyer-Bennet
wrote:

forgot about paper tapes. that high speed paper tape reader on the
pdp-11s (maybe even the pdp-8s) was way cool to watch.

I think they were the same, or very similar. Fan-fold tape moved
through remarkably fast.

it sure did.

"Wolfgang Weisselberg" wrote in message
...
[]
How long does it take to boot and shutdown your computer?
Especially when it's a long used XP with lotsa things
installed, that can take minutes.

Guess how long it takes to boot a C64 ...

-Wolfgang

Perhaps you should try a more recent version of Windows, Win-7 or Win-8?
They boot and shutdown noticeably quicker. XP is over 10 years old now
(but I'm still using it as well).

David

In article , Mxsmanic
wrote:

Nope. I was there. There isn't any disk drive with zero access time or even
anything close to zero.

ssd

As CPU speed has increased, the number of tasks that actually remain CPU-bound
has dwindled. Apart from things like video editing, applying filters to
photos, doing fancy math operations, generating visuals for fancy games, etc.,
very little can keep CPUs busy today. They are idle most of the time.

those are the types of things people do with their computers now.

As I write this, my CPU is less than 2%, my GPU utilization is too low to
show, but the systems is still doing disk I/O.

writing has never taxed computers. humans are slow.

now try encoding video and see how idle it is. when i do it, all cores
are pegged and the fans kick in on high.

In article , Mxsmanic
wrote:

how long does it take to boot a mainframe, especially one that does so
off tape?

A couple of minutes. But mainframes are never shut down, so a boot is very,
very rare.

neither are personal computers. booting is a waste of time. sleep it
when you aren't using it and wake it when you want to resume. it wakes
instantly, at exactly where you left off.

On 2012-04-17 14:54 , Mxsmanic wrote:

As I've said, it has gone from around 30 ms to around 8-10 ms. That's an
improvement of roughly 4x over four decades. CPU speeds have increased by many
orders of magnitude over the same period.

The gap between access time and CPU speed is actually worse now than it has
ever been, meaning that access time to disks is more of a bottleneck than it
has ever been before.

You're ignoring many things (as usual).

You're ignoring that file systems write large file blocks sequentially
over many sectors rather than filling in nearest available sectors. This
results in far fewer seeks. (NTFS, HFS, etc). The vast majority of
files on a system lie in a contiguous block (why defrag doesn't pay off
as it used to, esp. under FATx systems).

Since PC memories of 2, 4 , 8 GB and so on are more and more common, the
amount of disk cache allocated to system memory has risen (and is
flexible in size). Recently read/written files are often already in
memory when needed. This reduces seek times dramatically.

As nospam points out, the seek time for SSD's is close to 0 - SSD's are
replacing more and more HD's or are added to HD's to store the most
accessed files in hybrid setups.

--
"I was gratified to be able to answer promptly, and I did.
I said I didn't know."
-Samuel Clemens.

"Alan Browne" wrote:

As nospam points out, the seek time for SSD's is close to 0 - SSD's are
replacing more and more HD's or are added to HD's to store the most
accessed files in hybrid setups.


Yep, SSDs are now mainline. The new peecee here has a 128 GB SSD and a 500
GB HD. I'm being a bit nervous about setting things up, and putting more
program files on the HD than may be necessary. But things like Lightroom and
Photoshop tend to put scratch files on "C", so I need to leave lots of free
space there.

Whatever, it's a new generation. Our old ideas are wrong.

-- David J. Littleboy
Tokyo, Japan

--
"I was gratified to be able to answer promptly, and I did.
I said I didn't know."
-Samuel Clemens.

nospam wrote:
In article , Mxsmanic
wrote:

I have seen it consistently. Something like a browser might do 1000 disk I/Os
as it starts up.

what matters is after it starts up.

Then you would like a steam car ... starting up doesn't
matter to you, does it?


oh yes they have, as have disk speeds. sata is *way* faster than the
crappy ide that existed 10-20 years ago and certainly what existed in
the 1970s.

No, they have not. I'm not talking about capacity or transfer rates, I'm
talking about access time, which has hardly changed at all and is by far the
major bottleneck for disk performance.

access time is *much* faster now, especially on ssd.

CPUs are spending more and more cycles waiting on the same disk operations.


Most I/Os are very small transfers, so they hardly benefit from faster
transfer rates at all. What holds them back is the access time, and it's
terrible.

it's not terrible.

It's terrible for computers.


they've dramatically increased.

No, they haven't. In the 1970s access times were around 30-40 ms. Today they
are around 7-10 ms. That's only a four-fold increase in speed, compared to six
or seven orders of magnitude for CPU speeds.

in the 1980s, drives had 60-80ms access times and today, ssd has
effectively 0 ms access time.

Do a few trillion disk accesses and see how these
"effectively 0 ms" start to add up to quarters and years.

Anyway, CPUs are 1 million times faster since the 80s, and
disk access is only 8,000 times faster since the 80s ...

-Wolfgang