Re: Speed now & then

[Peter Corlett via cctalk]

On Wed, Apr 18, 2018 at 06:32:59PM -0600, ben via cctalk wrote:
> On 4/18/2018 4:47 PM, Eric Smith via cctalk wrote:
>> On Tue, Apr 17, 2018 at 8:18 PM, Fred Cisin via cctalk
>>  wrote:
>>> thousands of movies and TV episodes will fit on a 2TB drive. I am anxiously
>>> awaiting higher capacity thin 2.5" SATA.
>> You can get an 8TB drive in 2.5" form factor, but it doesn't contain
>> spinning rust, and it costs around $6000.

alternate.nl, my local boxshifter, is offering a 4TB 2.5" disk for €164.90.
(Whether this price includes the Dutch "you may now pirate all the things"
copyright levy, I don't know.) Sadly, it's both Seagate and shingled-recording.
Okay for backups and similar streaming write-once workloads, but awful as
general-purpose storage.

A common problem with large-capacity 2.5" disks is that they're 15mm high so
don't fit in laptops or similar consumer electronics. 15mm is more of an
enterprise storage standard that has leaked out.

> At one time you could get a $39 aerial up and get free TV like Dr Who..
> Progress seems to be getting rind of the good old and bringing in the $$$.

There's still plenty of free TV out there. Most of it's not even worth what you
paid for it.

[...]
> As for the BBC and other TV networks, we seem to be getting a lot of high
> priced episodes that have like 3 shows per season with a 2 part Christmas
> special mixed in with 90% reality TV.

Given you're quoting prices in dollars, I guess you're talking about the botch
job found on American cable, and the BBC is a rather different beast on its
home turf, where they manage to broadcast more than endless Top Bloody Gear
repeats. In the UK you can just use said cheap UHF aerial to receive it,
although you are supposed to also pay ~£150/year for a TV Licence.

Re: Speed now & then

[ben via cctalk]

On 4/18/2018 4:47 PM, Eric Smith via cctalk wrote:

On Tue, Apr 17, 2018 at 8:18 PM, Fred Cisin via cctalk <
cctalk@classiccmp.org> wrote:


thousands of movies and TV episodes will fit on a 2TB drive.
I am anxiously awaiting higher capacity thin 2.5" SATA.



You can get an 8TB drive in 2.5" form factor, but it doesn't contain
spinning rust, and it costs around $6000.


At one time you could get a $39 aerial up and get free TV like
Dr Who.. Progress seems to be getting rind of the good old
and bringing in the $$$.
While I admit the new TV's are better res, I can not say much
about what is being broadcast today is better than back then.
Ben.
As for the BBC and other TV networks, we seem to be getting
a lot of high priced episodes that have like 3 shows per season
with a 2 part Christmas special mixed in with 90% reality TV.
Sigh.
BTW the hardest part on my latest computer design is having
a working front panel.

Re: Speed now & then

[Eric Smith via cctalk]

On Tue, Apr 17, 2018 at 8:18 PM, Fred Cisin via cctalk <
cctalk@classiccmp.org> wrote:

> thousands of movies and TV episodes will fit on a 2TB drive.
> I am anxiously awaiting higher capacity thin 2.5" SATA.
>

You can get an 8TB drive in 2.5" form factor, but it doesn't contain
spinning rust, and it costs around $6000.

Re: Speed now & then

[Chuck Guzis via cctalk]

I'm reminded of how fast things have gotten when I use some of my old
media conversion code developed on an 8088 PC, that's been recompiled to
run under 64-bit Linux on a reasonably fast CPU (3GHz quad-core AMD).

I'd sit back for a couple of minutes waiting for the code to churn
through the data and create files and write them to disk.

Now, with essentially the same code, it's done before I can lift my
finger off the RETURN key--150 or 200 files' worth of data.

It strikes me as odd how little time we actually spend today computing
(other than bitcoin mining) in comparison to drawing pictures on a screen.

Back in the day, I often fantasized at what I would do if I had a
processor 10 times faster than the 70s supercomputer I was using.
Little did I suspect that I'd be using the processing power 40-some
years later to watch TV.

--Chuck

Re: Speed now & then

[Boris Gimbarzevsky via cctalk]

TAhanks for that link which fits with my 
measurements (nowhere as detailed) of ones actual 
ability to do things with "modern" hardware.  In 
the 1980's I was used to being able to measure 
events with 0.2 microsecond precision using a 
PDP-11 and my expectation was that the accuracy 
was only going to improve as processors got faster.


I ported a program I wrote on the PDP-11 to a 
Commodore 64 in 1988 and was using it to measure 
finger tapping with a switch array to 1 msec 
accuracy.  This was done through the simple 
expedient of speeding up the sample rate for the 
keyboard to 1 KHz and the adding in my 4 external 
switches as "keys".  Used a 512 K Mac to get the 
serial data and display results.  To do the same 
now would require custom hardware to do the 
timing and a USB link to a "modern" CPU or implimentation on a microprocessor


When I attempted to get this same type of timing 
accuracy from a PC, found out that it was no 
longer easy to get access to interrupts as easily 
as before and keyboard latency was longer as now 
keystrokes were detected by an on board 
microprocessor and sent out as a series of 
packets for each keystroke.  In DOS and W95 where 
one could still easily get at interrupts, then a 
serial port could be used to do msec 
timing.  Once XP and beyond arrived, then the 
best temporal precision one can expect from a 3 
GHz machine is 15 msec.  I suspect the same holds 
for Macs and haven't tried running real time 
Linux as I either pull out my trusty C64 from 
time to time and use it for precision timing 
(unfortunately have only one copy of the code on 
casette tape so when that goes can't do this 
anymore) or I use various microprocessors to do 
the job.  Have a nice microsecond precision timer 
that I wrote for a Propeller chip and feel much 
more comfortable programming for it than the 
latest windoze bloatware system.  The Propeller 
has the same amount of RAM as the PDP-11's I 
started on, runs 20x faster/core and is fun to 
program.  The microsecond timer is attached to a 
geiger counter to generate random bytes for OTP encryption.


Boris Gimbarzevsky

On 29 March 2018 at 19:53, Paul Koning via 
cctalk  wrote:

>
> It would be fun to do a "generalized Moore's 
Law" chart, showing not just transistor count 
growth (Moore's subject) but also the many 
other scaling changes of computing: disk 
capacity, recording density, disk IOPS, disk 
bandwidth, ditto those for tape, CPU MIPS, 
memory size, memory bandwidth, network bandwidth...


This is the most telling I've seen in a long time...

https://danluu.com/input-lag/

--
Liam Proven • Profile: https://about.me/liamproven
Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmaill.com
Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 7002 829 053

Re: Speed now & then (Space and time?)

[Ethan Dicks via cctalk]

On Fri, Apr 13, 2018 at 7:19 PM, Johnny Eriksson via cctalk
 wrote:
> Ethan Dicks  wrote:
>
>> There are times when video instruction makes sense...
>> ... Teaching programming?  I don't want to watch 2
>> hours of someone editing text... g
>
> How about language courses on audio cassettes?

Foreign language?  Sure.  Programming languages?  That would probably
drive me nuts.

> For instance a course on Macro-10, over several of them?

*shudder*

> Yes, it does (or at least did) exist.

I don't doubt that.  Still horrifying to think about sitting through
hours of audio to learn a written languange.

-ethan

Re: Speed now & then (Space and time?)

[Johnny Eriksson via cctalk]

Ethan Dicks  wrote:

> There are times when video instruction makes sense - describing, for
> example, a chemical reaction that produces major visible change in a
> few moments is better to watch than to try to describe.  The vast
> majority of stuff?  Teaching programming?  I don't want to watch 2
> hours of someone editing text... give me words and perhaps a
> screenshot or two if there's something too complicated to simply write
> about.

How about language courses on audio cassettes?

For instance a course on Macro-10, over several of them?

Yes, it does (or at least did) exist.

> -ethan

--Johnny

Re: Speed now & then

[ben via cctalk]

On 4/12/2018 5:08 AM, Liam Proven via cctalk wrote:

On 11 April 2018 at 17:18, Jay West  wrote:

Liam wrote...
https://danluu.com/input-lag/

Hey thanks for that link... fun read!


High praise indeed! :-) You're very welcome.


But that assumes local character echoing,
I am sure the google "search" box must echo to server for every key press.
Ben.

Re: Speed now & then (Space and time?)

[ben via cctalk]

On 4/11/2018 10:06 PM, Eric Smith via cctalk wrote:

On Wed, Apr 11, 2018 at 6:04 PM, ben  wrote:


On 4/11/2018 5:21 PM, Eric Smith wrote:


On Wed, Apr 11, 2018 at 3:48 PM, ben via cctalk > wrote:

 The FREE fpga development software is only under windows.

Xilinx and Altera (now Intel) FPGA development software, including the
"free" editions, have run under Linux for many years now. I routinely use
them on Fedora and CentOS.






Any PCB designs I build will use through the hole parts, thus 22V10's will
be my biggest programable device.



Sorry, I thought we were talking about FPGAs.


So I buy a pre-made FPGA card, why re-invent the wheel.
Then again with my soldering skill's I better do a valve computer. :)
Ben.

Re: Speed now & then (Space and time?)

[Jon Elson via cctalk]

On 04/12/2018 05:58 AM, Liam Proven via cctalk wrote:

On 11 April 2018 at 23:48, ben  wrote:


The FREE fpga development software is only under windows.


Some of the older Xilinx software (required to create config 
files for their older FPGA chips) will not run on a 64-bit 
platform (either Linux OR Windows) without a lot of fooling 
around.  Their later versions work fine on 64-bit.  That was 
some kind of government ITAR restriction.


Jon

Re: Speed now & then (Space and time?)

[Jon Elson via cctalk]

On 04/12/2018 05:57 AM, Liam Proven via cctalk wrote:

On 11 April 2018 at 20:21, Ethan Dicks  wrote:

I totally agree.  I read faster than most people talk and I retain
more information.  I'd much rather read 1,000 words than watch a 3
minute video.

Strongly agreed.


Yes, me too!  I HATE these terrible videos with shaky 
cameras and unprepared presenters.


Jon

Re: Speed now & then

[Liam Proven via cctalk]

On 11 April 2018 at 17:18, Jay West  wrote:
> Liam wrote...
> https://danluu.com/input-lag/
>
> Hey thanks for that link... fun read!

High praise indeed! :-) You're very welcome.

-- 
Liam Proven • Profile: https://about.me/liamproven
Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmail.com
Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053

Re: Speed now & then (Space and time?)

[Liam Proven via cctalk]

On 11 April 2018 at 23:48, ben  wrote:

> The FREE fpga development software is only under windows.

WINE works well now.

I write in Word 97 under WINE on 64-bit Ubuntu. Works a treat,
blindingly fast, and unlike any Linux tool I can find, it has a
working outliner.

VMs are almost trivially easy, too. Win10 is a free download and works
usably without registration or activation.

> WHAT HAPPEND TO * MINUX VERSION #1 *.

I don't understand the question.

> PS: I do have a better OS kicking around, Oberon but I need a extra
> screen and keyboard for it.

I have that -- well, A2 with Bluebottle -- running on the metal on my
old Thinkpad X200 (installed via a VM). Awaiting more time to learn
and experiment.

-- 
Liam Proven • Profile: https://about.me/liamproven
Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmail.com
Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053

Re: Speed now & then (Space and time?)

[Liam Proven via cctalk]

On 11 April 2018 at 20:21, Ethan Dicks  wrote:
>
> I totally agree.  I read faster than most people talk and I retain
> more information.  I'd much rather read 1,000 words than watch a 3
> minute video.

Strongly agreed.

> Not surprising given how many generations have now grown up watching
> TV as their major input mode.

:-(

> Yes.  People can ramble into a mobile phone and upload to YouTube but
> actually _writing_ the same process... not everyone is a writer.

Worse still, Youtube has commercialised this. The _only_ instructions
I could find anywhere for rooting my smartphone and installing a new
ROM were a Russian language Youtube video, some 30min long. I don't
speak Russian, but thankfully, he had subtitled it in English.

But people make many such vids, with bolted-on title sequences and
rambling "hi guys, welcome, thanks for watching" intros, partly
because that's all they know how to do, and partly because they hope
to get lots of people to watch it all and thus get a kickback from
Google for the revenue from the embedded ads.

99.9% won't of course, but a few get rich doing it: PewDePie etc.

And 99.9% of them are later exposed as racists, anti-semites etc. What
a shock, etc.

Wannabe "models" are doing the same on Instagram -- underweight teen
women making a good living from posting a lot of pics of themselves in
their underwear or swimwear, and so on. A few later "come out" as
suffering from anorexia etc.

https://www.theguardian.com/media/2015/nov/03/instagram-star-essena-oneill-quits-2d-life-to-reveal-true-story-behind-images

Most don't, but they set horribly unrealistic standards of appearance
for young women, etc.

-- 
Liam Proven • Profile: https://about.me/liamproven
Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmail.com
Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053

Re: Speed now & then (Space and time?)

[Peter Corlett via cctalk]

On Wed, Apr 11, 2018 at 03:48:20PM -0600, ben via cctalk wrote:
[...]
> The FREE fpga development software is only under windows.

Altera's Quartus II and Xilinx ISE also have Linux versions. They're as free as
the Windows versions.

Re: Speed now & then (Space and time?)

[Eric Smith via cctalk]

On Wed, Apr 11, 2018 at 6:04 PM, ben  wrote:

> On 4/11/2018 5:21 PM, Eric Smith wrote:
>
>> On Wed, Apr 11, 2018 at 3:48 PM, ben via cctalk > > wrote:
>>
>> The FREE fpga development software is only under windows.
>>
>> Xilinx and Altera (now Intel) FPGA development software, including the
>> "free" editions, have run under Linux for many years now. I routinely use
>> them on Fedora and CentOS.
>>
>

> Any PCB designs I build will use through the hole parts, thus 22V10's will
> be my biggest programable device.


Sorry, I thought we were talking about FPGAs.

Re: Speed now & then (Space and time?)

[Chuck Guzis via cctalk]

On 04/11/2018 02:48 PM, ben via cctalk wrote:

> I have a nice 18 bit cpu here, with only a few hardware bugs.
> Hmm would it work better if I change that around ideas.
> 
> Care to point to a nice 18 bit version of unix or C.
> BTW The cpu has a frame pointer S but no S++ --S operations
> so pushing and popping wild data is not a option.

Well, the Univac 1100/2200 series mainframes ran V7 Unix--and they're
36-bit machines, so probably not far from your 18-bitter--and they're
ones' complement machines.

Univac called it "SX1100", so you have a search term.

--Chuck

Re: Speed now & then (Space and time?)

[ben via cctalk]

On 4/11/2018 5:47 PM, Paul Koning wrote:


I haven't tried pcc, but supposedly that has been ported to the PDP-10, so 
presumably it can be ported to an 18-bit machine too.



Well the original C mostly just and 8 bit bytes and 16 bit ints, with 
floating point for good luck. Now who knows what it needs.
But the good (old) news is you still can get the original C compiler for 
the 11.




You could try gcc; creating a simple back end is not all that hard.  And while 
it makes no attempt to support non-multiple-of-8-bit machines, it can be forced 
to, after a fashion.  One time for grins I banged together a very primitive CDC 
6000 back end.  It wasn't correct but it wasn't horribly wrong...

18 bit Unix, not sure about that one.  It was originally done on a PDP-7 but I 
think that was before C and it's no longer around that I know of.  Still, older 
versions might be somewhat portable.


Disk I/O needs be word size aligned so I guess 16 bit unix (if you could 
use it, curse you bell labs) could be ported providing you don't have 
nasty tricks to 18 bit I/O.


Does it have to be Unix?  For a simple character environment, Forth is nice and 
it's very easy to port to pretty much any computer.
Unix was the only thing I can think of that is character I/O , device 
drivers and on 16 bit cpu written

in a high level language.


paul

Re: Speed now & then (Space and time?)

[ben via cctalk]

On 4/11/2018 5:21 PM, Eric Smith wrote:
On Wed, Apr 11, 2018 at 3:48 PM, ben via cctalk > wrote:


The FREE fpga development software is only under windows.


Xilinx and Altera (now Intel) FPGA development software, including the 
"free" editions, have run under Linux for many years now. I routinely 
use them on Fedora and CentOS.


I got version #9 of the Altera software to go with the old DE1 FPGA 
card.I have that card so I can do TTL logic macro's and test before

I do a PCB layout using 74LS chips. More modern software seems to have
dropped that funtionality. I also like programing in AHDL because
I find VHDL and VERLOG too confusing to use.
AHDL also has a similar format to WINCUPL that I use for 22V10
programing. This way I can define logic modules using 22V10's
with the FPGA and test that as well as the 2901 emulation.


The Xilinx Vivado WebPack software supports the entire Artix-7 product 
line, which covers a pretty wide range of device sizes. Xilinx ISE 
WebPack supports Spartan 3, Spartan 6, and CPLD devices.


Any PCB designs I build will use through the hole parts, thus 22V10's 
will be my biggest programable device.

Ben.

Re: Speed now & then (Space and time?)

[Paul Koning via cctalk]

> On Apr 11, 2018, at 5:48 PM, ben via cctalk  wrote:
> 
> ...
>>> and real text screen UNIX is not aviable anymore.
>> Sure it is.
> 
> I have a nice 18 bit cpu here, with only a few hardware bugs.
> Hmm would it work better if I change that around ideas.
> 
> Care to point to a nice 18 bit version of unix or C.
> BTW The cpu has a frame pointer S but no S++ --S operations
> so pushing and popping wild data is not a option.

I haven't tried pcc, but supposedly that has been ported to the PDP-10, so 
presumably it can be ported to an 18-bit machine too.

You could try gcc; creating a simple back end is not all that hard.  And while 
it makes no attempt to support non-multiple-of-8-bit machines, it can be forced 
to, after a fashion.  One time for grins I banged together a very primitive CDC 
6000 back end.  It wasn't correct but it wasn't horribly wrong...

18 bit Unix, not sure about that one.  It was originally done on a PDP-7 but I 
think that was before C and it's no longer around that I know of.  Still, older 
versions might be somewhat portable.

Does it have to be Unix?  For a simple character environment, Forth is nice and 
it's very easy to port to pretty much any computer.

paul

Re: Speed now & then (Space and time?)

[Eric Smith via cctalk]

On Wed, Apr 11, 2018 at 3:48 PM, ben via cctalk 
wrote:

> The FREE fpga development software is only under windows.
>

Xilinx and Altera (now Intel) FPGA development software, including the
"free" editions, have run under Linux for many years now. I routinely use
them on Fedora and CentOS.

The Xilinx Vivado WebPack software supports the entire Artix-7 product
line, which covers a pretty wide range of device sizes. Xilinx ISE WebPack
supports Spartan 3, Spartan 6, and CPLD devices.

Re: Speed now & then (Space and time?)

[Torfinn Ingolfsen via cctalk]

Since you brought it up

On Wed, Apr 11, 2018 at 11:48 PM, ben via cctalk  wrote:
>
>
> The FREE fpga development software is only under windows.
>

Free alternatives exist (and can only get better coverage with time)
http://www.clifford.at/yosys/
https://github.com/cseed/arachne-pnr
http://www.clifford.at/icestorm/

HTH
-- 
Regards,
Torfinn Ingolfsen

Re: Speed now & then (Space and time?)

[ben via cctalk]

On 4/11/2018 9:40 AM, Liam Proven wrote:

On 1 April 2018 at 00:26, ben via cctalk  wrote:


But that is the old fly in the ointment, other software may not be avilable.


It is, you know.


Smart ass response:
Who pirated it for you.



I do run windows


Why?


The FREE fpga development software is only under windows.


and real text screen UNIX is not aviable anymore.


Sure it is.



I have a nice 18 bit cpu here, with only a few hardware bugs.
Hmm would it work better if I change that around ideas.

Care to point to a nice 18 bit version of unix or C.
BTW The cpu has a frame pointer S but no S++ --S operations
so pushing and popping wild data is not a option.




I mean, there are even text-only distros, such as INX:

http://inx.maincontent.net/

... and ADRIANE, a spinoff of Knoppix:

http://www.knopper.net/knoppix-adriane/index-en.html


But my hardware (LOW density SD card) is 256 9 bit charaters per
block for disk I/O.



But you can install Debian, Ubuntu or openSUSE in text-only mode with
no GUI if you wish. All the console-mode tools you could want are
available: web browsers, email and chat clients, twitter clients,
music players, etc.


I still have to finish Tomb Raider 1 under DOS(box) First.


Go the whole hog and install Slackware and go back to installing from
tarballs if you wish.

I'd probably suggest openSUSE as the YaST admin tool works in text
mode, so you don't need to know your way around a hundred config files
-- YaST will do that for you.


WHAT HAPPEND TO * MINUX VERSION #1 *.
Ben.

PS: I do have a better OS kicking around, Oberon but I need a extra
screen and keyboard for it.
PPS:

Software efficiency halves every 18 months, compensating Moore's law

Re: Speed now & then

[Chuck Guzis via cctalk]

On 04/11/2018 11:31 AM, Jecel Assumpcao Jr. via cctalk wrote:

> Besides getting more performance with smaller transistors, we have also
> been increasing performance by taking advantage of more transistors by
> doing more stuff in parallel. So we went from up to dozens of clock
> cycles per instructions to three or four instructions per clock cycle.
> Quite a few of the additional transistors have been use for more and
> more layers of caches.

As one old acquaintance said  "We've given up on reducing the time it
takes one woman to produce a baby.  Our new approach is to use nine
women to get a baby in one month"

...or something to that effect.

--Chuck

Re: Speed now & then

[Bill Gunshannon via cctalk]

On 04/11/2018 03:28 PM, Fred Cisin via cctalk wrote:
>> On Wed, Apr 11, 2018 at 11:09 AM, Chuck Guzis via cctalk <
>> cctalk@classiccmp.org> wrote:
>>> I thought that Moore's "law" dealt only with the number of transistors
>>> on a die.   Did Gordon also say something about performance?
> On Wed, 11 Apr 2018, Eric Korpela via cctalk wrote:
>> You are correct that he only applied it to transistor count.  The 
>> extension
>> to everything else under the sun was made by others.
>
> The number of things under the sun that Moore's "law" gets applied to 
> doubles every 18 months.
>

Isn't that one of the corollaries of Moore's Law?

bill

Re: Speed now & then

[Fred Cisin via cctalk]

On Wed, Apr 11, 2018 at 11:09 AM, Chuck Guzis via cctalk <
cctalk@classiccmp.org> wrote:

I thought that Moore's "law" dealt only with the number of transistors
on a die.   Did Gordon also say something about performance?

On Wed, 11 Apr 2018, Eric Korpela via cctalk wrote:

You are correct that he only applied it to transistor count.  The extension
to everything else under the sun was made by others.


The number of things under the sun that Moore's "law" gets applied to 
doubles every 18 months.

Re: Speed now & then

[Eric Korpela via cctalk]

On Wed, Apr 11, 2018 at 11:09 AM, Chuck Guzis via cctalk <
cctalk@classiccmp.org> wrote:

>
> I thought that Moore's "law" dealt only with the number of transistors
> on a die.   Did Gordon also say something about performance?
>

You are correct that he only applied it to transistor count.  The extension
to everything else under the sun was made by others.


> --Chuck
>
>


-- 
Eric Korpela
korp...@ssl.berkeley.edu
AST:7731^29u18e3

Re: Speed now & then

[Paul Koning via cctalk]

> On Apr 11, 2018, at 2:31 PM, Jecel Assumpcao Jr. via cctalk 
>  wrote:
> 
> Chuck Guzis wrote on Wed, 11 Apr 2018 11:09:23 -0700
>> I thought that Moore's "law" dealt only with the number of transistors
>> on a die.   Did Gordon also say something about performance?
> 
> That is correct. The observation that transistors would be faster and
> use less power as they became smaller is called "Dennard scaling" from
> 1974:
> 
> https://en.wikipedia.org/wiki/Dennard_scaling
> 
> This led to the MHz wars of the 1990s. Sadly, as the isolation barriers
> (the "O" in "MOS") became thinner and thinner we could no longer ignore
> leakage currents. In addition, going with lower voltages no longer was
> possible as we got closer and closer to the transistor threashold
> voltages. So we got stuck at 3 GHz or less.

For silicon VLSI logic circuits, yes, give or take a GHz or so.  CMOS can go 
very much higher in RF circuits.  And if you switch to different 
semiconductors, you can go higher still.  People have built oscillators that 
get close to a THz, but those are not logic circuits, let alone computers.

Propagation latencies also become an issue, though you can design fast 
computers with substantial propagation delays if you have to.  Cray did this 
decades ago; the basic stage delay in the CDC 6600 is 1/20th of the clock 
period, and the delay on chassis interconnects is 1/4 of the clock period.  In 
that design, all sorts of things are carefully paced to match the known 
propagation delays.  But most other computers were not designed with such 
complex techniques.

paul

Re: Speed now & then

[Jecel Assumpcao Jr. via cctalk]

Chuck Guzis wrote on Wed, 11 Apr 2018 11:09:23 -0700
> I thought that Moore's "law" dealt only with the number of transistors
> on a die.   Did Gordon also say something about performance?

That is correct. The observation that transistors would be faster and
use less power as they became smaller is called "Dennard scaling" from
1974:

https://en.wikipedia.org/wiki/Dennard_scaling

This led to the MHz wars of the 1990s. Sadly, as the isolation barriers
(the "O" in "MOS") became thinner and thinner we could no longer ignore
leakage currents. In addition, going with lower voltages no longer was
possible as we got closer and closer to the transistor threashold
voltages. So we got stuck at 3 GHz or less.

Besides getting more performance with smaller transistors, we have also
been increasing performance by taking advantage of more transistors by
doing more stuff in parallel. So we went from up to dozens of clock
cycles per instructions to three or four instructions per clock cycle.
Quite a few of the additional transistors have been use for more and
more layers of caches.

-- Jecel

Re: Speed now & then (Space and time?)

[Ethan Dicks via cctalk]

On Wed, Apr 11, 2018 at 11:36 AM, Liam Proven via cctalk
 wrote:
> On 29 March 2018 at 21:35, Fred Cisin via cctalk  
> wrote:
>>
>> MP4s mean that now, not only does it take MUCH longer to create the
>> document, we can now waste MUCH more of the reader's time!
>> I find it very annoying that when GOOGLE'ing to find a simple answer, many
>> of the first hits are YouTube.
>
> I got a room full of very surprised looks a couple of years ago,
> interviewing for a tech-writer position, when I said that I hated
> online videos as instructions, finding them far too slow and
> inefficient.

I totally agree.  I read faster than most people talk and I retain
more information.  I'd much rather read 1,000 words than watch a 3
minute video.

> This concept shocked everyone in the room, AFAICT.

Not surprising given how many generations have now grown up watching
TV as their major input mode.

> Then I realised something -- something connected with my career change
> to tech writing.
>
> A lot of people _can't_ express themselves in text. Some don't care
> and let fly with all their spelling and grammatical errors,
> mostly-missing-and-the-rest-incorrect punctuation etc.

Yes.  People can ramble into a mobile phone and upload to YouTube but
actually _writing_ the same process... not everyone is a writer.

There are times when video instruction makes sense - describing, for
example, a chemical reaction that produces major visible change in a
few moments is better to watch than to try to describe.  The vast
majority of stuff?  Teaching programming?  I don't want to watch 2
hours of someone editing text... give me words and perhaps a
screenshot or two if there's something too complicated to simply write
about.

I've taken several MOOCs.  The only saving grace is the 2X and 4X
buttons on playback.

-ethan

Re: Speed now & then

[Chuck Guzis via cctalk]

On 04/11/2018 09:31 AM, Eric Korpela via cctalk wrote:
> It might break the rules since it only goes back to 1999, but here's
> Moore's law for integer speed, floating point speed, number of processors,
> memory sizes and disk sizes for the machines connected to SETI@home.  Plots
> are averages and medians, unfiltered for errors.   At least one of the
> parameters (total credited ops) is no longer used.expavg credit
> includes GPU work even though it's not included in the processor speed
> numbers.  If someone wants a distillation of the data for comparing to
> earlier machines, let me know.

I thought that Moore's "law" dealt only with the number of transistors
on a die.   Did Gordon also say something about performance?

--Chuck

Re: Speed now & then

[Paul Koning via cctalk]

> On Apr 11, 2018, at 11:05 AM, Liam Proven  wrote:
> 
> On 29 March 2018 at 19:53, Paul Koning via cctalk  
> wrote:
>> 
>> It would be fun to do a "generalized Moore's Law" chart, showing not just 
>> transistor count growth (Moore's subject) but also the many other scaling 
>> changes of computing: disk capacity, recording density, disk IOPS, disk 
>> bandwidth, ditto those for tape, CPU MIPS, memory size, memory bandwidth, 
>> network bandwidth...
> 
> This is the most telling I've seen in a long time...
> 
> https://danluu.com/input-lag/

Nice.  It would be interesting to throw in some even older systems.  One that 
comes to mind is PLATO.  I don't remember exactly what it delivered for 
response time, there were good test results on it.  Somewhere in the 100 ms 
range, I believe, much of which was transmission delays due to low baud rates.  
And that was with 600 users, sharing a quad processor 4 x 10 MHz machine, in 
1976.

paul

RE: Speed now & then (Space and time?)

[Dave Wade via cctalk]

> -Original Message-
> From: cctalk <cctalk-boun...@classiccmp.org> On Behalf Of Liam Proven via
> cctalk
> Sent: 11 April 2018 16:36
> To: General Discussion: On-Topic and Off-Topic Posts
> <cctalk@classiccmp.org>
> Subject: Re: Speed now & then (Space and time?)
> 
> On 29 March 2018 at 21:35, Fred Cisin via cctalk <cctalk@classiccmp.org>
> wrote:
> >
> > MP4s mean that now, not only does it take MUCH longer to create the
> > document, we can now waste MUCH more of the reader's time!
> > I find it very annoying that when GOOGLE'ing to find a simple answer,
> > many of the first hits are YouTube.
> 

Sometimes these are useful, usually not. Some of the slowest and most painful 
come from the biggest players...
Of course when you want a video, like how to remove the casing from my AirCon 
because the drain is blocked
-site:youtube.com gets rid of most


> I got a room full of very surprised looks a couple of years ago, interviewing
> for a tech-writer position, when I said that I hated online videos as
> instructions, finding them far too slow and inefficient.
> 
> This concept shocked everyone in the room, AFAICT.
> 
Why, its obvious...
> 
> --
> Liam Proven • Profile: https://about.me/liamproven
> Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmail.com
> Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
> UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053

Re: Speed now & then

[Eric Korpela via cctalk]

It might break the rules since it only goes back to 1999, but here's
Moore's law for integer speed, floating point speed, number of processors,
memory sizes and disk sizes for the machines connected to SETI@home.  Plots
are averages and medians, unfiltered for errors.   At least one of the
parameters (total credited ops) is no longer used.expavg credit
includes GPU work even though it's not included in the processor speed
numbers.  If someone wants a distillation of the data for comparing to
earlier machines, let me know.

http://setiathome.berkeley.edu/host_stats/index.html

On Thu, Mar 29, 2018 at 10:53 AM, Paul Koning via cctalk <
cctalk@classiccmp.org> wrote:

>
>
> > On Mar 29, 2018, at 12:20 PM, Murray McCullough via cctalk <
> cctalk@classiccmp.org> wrote:
> >
> > I’m not trying to date myself but have things truly sped up? In 1970’s
> > Toronto I had a classic computer, sorry can’t recall what it was,
> connected
> > to a 300 baud modem; by early 80’s had ‘zoomed’ to 9600 baud. Oh, my! [ A
> > typical file size to download was probably 1 MB. ] Speed indeed! Yet now,
> > here in rural Ontario, Canada, I’m at 5MB/s. Yikes! (Friends in Toronto
> are
> > at 50MB/s.) We can do the math but content, particularly multimedia, has
> > swollen in size.[ 1 GB is not unheard of. ] Were classic computing days
> > that much slower? Happy computing. Murray  -:)
>
> I remember downloading the GCC release kit over a 56k dialup line, in
> 2000.  Took a while.
>
> The ARPAnet in its early days had "high speed backbone" links which were
> 56k bps.  Terminal links presumably 110 bps, that being the speed of ASCII
> teletypes.  And back in the late 1970s you could still find even slower
> links in some places, such as 6 bit links connecting teletype machines for
> newspaper "wire service" feeds.
>
> It would be fun to do a "generalized Moore's Law" chart, showing not just
> transistor count growth (Moore's subject) but also the many other scaling
> changes of computing: disk capacity, recording density, disk IOPS, disk
> bandwidth, ditto those for tape, CPU MIPS, memory size, memory bandwidth,
> network bandwidth...
>
> All these have grown dramatically, but very clearly not in the same
> proportion, for some of these the changes are smooth while others are
> jumps, and the rate of change sometimes varies dramatically over the
> decades.
>
> paul
>
>
>


-- 
Eric Korpela
korp...@ssl.berkeley.edu
AST:7731^29u18e3

Re: Speed now & then (Space and time?)

[Liam Proven via cctalk]

On 1 April 2018 at 00:26, ben via cctalk  wrote:
>
> But that is the old fly in the ointment, other software may not be avilable.

It is, you know.

> I do run windows

Why?

> and real text screen UNIX is not aviable anymore.

Sure it is.

I mean, there are even text-only distros, such as INX:

http://inx.maincontent.net/

... and ADRIANE, a spinoff of Knoppix:

http://www.knopper.net/knoppix-adriane/index-en.html

But you can install Debian, Ubuntu or openSUSE in text-only mode with
no GUI if you wish. All the console-mode tools you could want are
available: web browsers, email and chat clients, twitter clients,
music players, etc.

Go the whole hog and install Slackware and go back to installing from
tarballs if you wish.

I'd probably suggest openSUSE as the YaST admin tool works in text
mode, so you don't need to know your way around a hundred config files
-- YaST will do that for you.

-- 
Liam Proven • Profile: https://about.me/liamproven
Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmail.com
Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053

Re: Speed now & then (Space and time?)

[Liam Proven via cctalk]

On 29 March 2018 at 21:35, Fred Cisin via cctalk  wrote:
>
> MP4s mean that now, not only does it take MUCH longer to create the
> document, we can now waste MUCH more of the reader's time!
> I find it very annoying that when GOOGLE'ing to find a simple answer, many
> of the first hits are YouTube.

I got a room full of very surprised looks a couple of years ago,
interviewing for a tech-writer position, when I said that I hated
online videos as instructions, finding them far too slow and
inefficient.

This concept shocked everyone in the room, AFAICT.

> Dancing kangaroos and yodelling jellyfish has let form triumph over content!
> When will we finally have smell-o-vision?

I used to think this of media such as Snapchat, Tumblr, Pinterest,
Instagram et al.

Then I realised something -- something connected with my career change
to tech writing.

A lot of people _can't_ express themselves in text. Some don't care
and let fly with all their spelling and grammatical errors,
mostly-missing-and-the-rest-incorrect punctuation etc.

At least 1 of my cousins is like that -- I can barely follow when this
woman (who is in her mid 50s) writes to me. It's barely English, with
no punctuation, no capitals, no sentences or paragraphs, just stream
of incoherent severely-misspelled consciousness.

I don't think she realises, and she can't help it.

I suspect many, of all ages, are like her.

And I further suspect that they can't easily consume text, either.
They can't read very fluidly.

Also see:

https://www.joelonsoftware.com/2001/11/16/2006/

... and this from someone who I generally admire. He, as many do now,
see threading as an unnecessary, distracting technical frippery.

Astounding.

But for those who _do_ care about appearances but _can't_ write well
*and know it*, Snapchat etc. are a way out. You can "express yourself"
via pictures. You can take them yourself, use filters and so on to
tart them up a bit, and still communicate with a community of friends,
with no more writing than a few tags at a minimum.

If you can't even do that, Pinterest lets you do it _with someone
else's images_.

-- 
Liam Proven • Profile: https://about.me/liamproven
Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmail.com
Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053

RE: Speed now & then

[Jay West via cctalk]

Liam wrote...
https://danluu.com/input-lag/

Hey thanks for that link... fun read!

J

Re: Speed now & then

[Liam Proven via cctalk]

On 29 March 2018 at 19:53, Paul Koning via cctalk  wrote:
>
> It would be fun to do a "generalized Moore's Law" chart, showing not just 
> transistor count growth (Moore's subject) but also the many other scaling 
> changes of computing: disk capacity, recording density, disk IOPS, disk 
> bandwidth, ditto those for tape, CPU MIPS, memory size, memory bandwidth, 
> network bandwidth...

This is the most telling I've seen in a long time...

https://danluu.com/input-lag/

-- 
Liam Proven • Profile: https://about.me/liamproven
Email: lpro...@cix.co.uk • Google Mail/Hangouts/Plus: lpro...@gmail.com
Twitter/Facebook/Flickr: lproven • Skype/LinkedIn: liamproven
UK: +44 7939-087884 • ČR (+ WhatsApp/Telegram/Signal): +420 702 829 053

Re: Speed now & then (Space and time?)

[allison via cctalk]

On 04/02/2018 01:28 PM, Paul Koning via cctalk wrote:
>
>> On Mar 31, 2018, at 6:26 PM, ben via cctalk  wrote:
>>
>>> ...
>> But that is the old fly in the ointment, other software may not be avilable. 
>> I do run windows and real text screen UNIX is not aviable anymore. 
> Sure it is.  You can perfectly well install Linux without the GUI components, 
> never mind other Unix operating systems like NetBSD which are more oriented 
> towards simple low-overhead deployments.  And my workstations, while they are 
> GUI based, really only serve as text type Unixes with large scroll buffers.
>
>   paul
>
>
That and even the Macbook OS-X will take you to a unix terminal.  IT
runs all the same commands as my Unix-V6 on pdp11.
I'd add the win10/pro has a interface screen that allows a linux/unix
terminal interface.  That made the thing usable!


Allison

Re: Speed now & then (Space and time?)

[Paul Koning via cctalk]

> On Mar 31, 2018, at 6:26 PM, ben via cctalk  wrote:
> 
>> ...
> 
> But that is the old fly in the ointment, other software may not be avilable. 
> I do run windows and real text screen UNIX is not aviable anymore. 

Sure it is.  You can perfectly well install Linux without the GUI components, 
never mind other Unix operating systems like NetBSD which are more oriented 
towards simple low-overhead deployments.  And my workstations, while they are 
GUI based, really only serve as text type Unixes with large scroll buffers.

paul

Re: Speed now & then (Space and time?)

[Peter Corlett via cctalk]

On Sat, Mar 31, 2018 at 04:26:44PM -0600, ben via cctalk wrote:
[...]
> But that is the old fly in the ointment, other software may not be avilable.
> I do run windows and real text screen UNIX is not aviable anymore. All I
> know it is same $$$ cycle as always, BUY the new machine
> for faster software, but you need buy the software that has bug fixes and
> patches for the new system and the software slows down again.
> GUI's gave us 8x bloat and streaming media another 8x bloat.

Decades of being nickel-and-dimed by the Microsoft ecosystem has taught you
that you need to pay for a lot of software. This is far less common on other
platforms.

I'm watching the MNT Reform with some interest as they're much more
user-repairable and -upgradable than typical notebooks, but it's nowhere near
ready yet. The small-run prototype is also an eyewatering €599, so I'll pass on
it for now.

> It is about time NEW notebook computers to come out to let you use them to
> take NOTEs rather than some ap for your phone for notes.

I use a fountain pen and some old file cards for that :)

> PS: Do I need a VALVE computer for the best sounding digital music?

You *could* buy one of those ricer PCs which have a valve amplifier fitted
right there on the PCB, nice and close to all those sharp digital signals and
noisy ground plane. The target market is hipsters who don't understand
electronics.

> PPS: Notice how records are selling again.

There's a lot of talk from the same hipsters about a vinyl resurgence, but the
impressive-sounding percentge issues are against a tiny baselines that's pretty
much an accounting error.

In absolute rather than relative terms, the only growing market for music is
streaming. Spotify is one of those appalling bloated web-wrapper apps I was
railing against. I'm sticking with CDs; they'll even play on my old Amiga.

Re: Speed now & then (Space and time?)

[ben via cctalk]

On 3/30/2018 5:32 AM, Peter Corlett via cctalk wrote:

On Thu, Mar 29, 2018 at 10:07:27PM -0500, Mark Linimon via cctalk wrote:

On Thu, Mar 29, 2018 at 09:00:35PM -0500, Jon Elson via cctalk wrote:

It was an absolute DOG! It took several minutes for Emacs to load.

So, uh, I hate to tell you about the state of the art these days ...


It starts up within half a second or so on my development box, even though my
.emacs pulls in all sorts of weird and wonderful packages. On my remote
servers, it can take a few seconds if the box is overloaded, but is otherwise
instant.

A lot of traditional software has *not* encountered the kind of bloat needed to
counteract Moore's Law, and so performs beautifully. What *is* bloating is
Windows software -- mercifully I don't run Windows -- half-jobbed ports of said
Windows software to other platforms (Firefox, mainly), and web-based stuff. The
latter includes various phoned-it-in apps which are basically written in
Javascript and embed an invisible web browser, which includes Slack, Spotify,
and a few other bits and pieces.

The obvious answer is to not use that bloated crapware and find something else
that does the same job.


But that is the old fly in the ointment, other software may not be 
avilable. I do run windows and real text screen UNIX is not aviable 
anymore. All I know it is same $$$ cycle as always, BUY the new machine
for faster software, but you need buy the software that has bug fixes 
and patches for the new system and the software slows down again.

GUI's gave us 8x bloat and streaming media another 8x bloat.
It is about time NEW notebook computers to come out to let you use them
to take NOTEs rather than some ap for your phone for notes.
Ben.
PS: Do I need a VALVE computer for the best sounding digital music?
PPS: Notice how records are selling again.

Re: Speed now & then (Space and time?)

[Peter Corlett via cctalk]

On Thu, Mar 29, 2018 at 10:07:27PM -0500, Mark Linimon via cctalk wrote:
> On Thu, Mar 29, 2018 at 09:00:35PM -0500, Jon Elson via cctalk wrote:
>> It was an absolute DOG! It took several minutes for Emacs to load.
> So, uh, I hate to tell you about the state of the art these days ...

It starts up within half a second or so on my development box, even though my
.emacs pulls in all sorts of weird and wonderful packages. On my remote
servers, it can take a few seconds if the box is overloaded, but is otherwise
instant.

A lot of traditional software has *not* encountered the kind of bloat needed to
counteract Moore's Law, and so performs beautifully. What *is* bloating is
Windows software -- mercifully I don't run Windows -- half-jobbed ports of said
Windows software to other platforms (Firefox, mainly), and web-based stuff. The
latter includes various phoned-it-in apps which are basically written in
Javascript and embed an invisible web browser, which includes Slack, Spotify,
and a few other bits and pieces.

The obvious answer is to not use that bloated crapware and find something else
that does the same job.

Re: Speed now & then (Space and time?)

[Zane Healy via cctalk]

> On Mar 29, 2018, at 7:00 PM, Jon Elson via cctalk  
> wrote:
> 
> Then, in 1986, I bought a MicroVAX-II CPU board from a broker, and a bunch of 
> 3rd party peripherals, and made a copy of VMS 4.7 (Might have used something 
> earlier for a time.)
> I was in 7th heaven!  A REAL computer at LAST!

And now a Raspberry Pi running SIMH can emulate a faster VAX system than that 
MicroVAX II.  I’m slowly getting my VMS cluster back up and running, but 
through emulation.  My goal is to power down my Alpha, except when I need a 
couple programs that won’t run on a VAX.  My fastest VAX (beating my VAXstation 
4000/60) is SIMH running on an ESXI server.

Zane

Re: Speed now & then (Space and time?)

[Mark Linimon via cctalk]

On Thu, Mar 29, 2018 at 04:05:10PM -0700, Chuck Guzis via cctalk wrote:
> And yet, productive work was performed on it.  Indeed the industrial
> variant, the 1710 was used for early process control.

There were a lot of highway improvements made in the US in the 1950s/
1960s using Bendix G-15s.  That particular branch of civil engineering
was probably that machine's biggest customer.  (OTOH, disclaimer: the
one I used in high school had previously been owned by an oil company.)

mcl

Re: Speed now & then (Space and time?)

[Mark Linimon via cctalk]

On Thu, Mar 29, 2018 at 09:00:35PM -0500, Jon Elson via cctalk wrote:
> It was an absolute DOG!  It took several minutes for Emacs to load.

So, uh, I hate to tell you about the state of the art these days ...

mcl

Re: Speed now & then (Space and time?)

[Jon Elson via cctalk]

On 03/29/2018 04:24 PM, Fred Cisin via cctalk wrote:


I posited that 2 decades ago in a wired article.  My CP/M 
machine booted

in seconds while waiting for
the winders box to decide if it would/could.


"The new machine is so much faster, that it can almost get 
out of its own way!"



From 1978 or so, I had a Z-80 CP/M system running on the 
S-100 bus. In about 1980, I got a Memorex Winchester drive 
on it.  It really made CP/M fly.


In 1982-1984 or so, I was trying to build a 32-bit machine 
very loosely patterned after an IBM 360, but would have used 
a 360 user-level instruction set.  It became obvious that it 
would take years to have an OS, compilers, utilities, etc.  
(I did not know about Unix 360, which I almost certainly 
could have gotten a copy of, we had a PDP-11 Unix license at 
the university.)


So, I cloned a Logical Microcomputer Co. Genix system with 
the Nat. Semi 16032 chip.  It was an absolute DOG!  It took 
several minutes for Emacs to load.  Even vi (which I 
detest!) was horribly slow, like 10X slower than the CP/M 
editor.


Then, in 1986, I bought a MicroVAX-II CPU board from a 
broker, and a bunch of 3rd party peripherals, and made a 
copy of VMS 4.7 (Might have used something earlier for a time.)

I was in 7th heaven!  A REAL computer at LAST!

Jon

Re: Speed now & then (Space and time?)

[Paul Koning via cctalk]

> On Mar 29, 2018, at 7:05 PM, Chuck Guzis via cctalk  
> wrote:
> 
> On 03/29/2018 02:24 PM, Fred Cisin via cctalk wrote:
 HOWEVER, a variant of "Boyle's Law" warns that software and content
 will expand to fit all available space and speed.
>> 
>> On Thu, 29 Mar 2018, allison via cctalk wrote:
>>> We have proof and it is us.
>> 
>> Or, as Walt Kelly ("Pogo") said, "We have met the enemy, and he is us."
>> 
>>> I posited that 2 decades ago in a wired article.  My CP/M machine booted
>>> in seconds while waiting for
>>> the winders box to decide if it would/could.
>> 
>> "The new machine is so much faster, that it can almost get out of its
>> own way!"
> 
> As a real contrast, consider, say, the IBM 1620.  Go look up the cycle
> times on that beast.
> 
> And yet, productive work was performed on it.  Indeed the industrial
> variant, the 1710 was used for early process control.

Or drum machines.  Dutch airplane maker Fokker used one (FERTA) for airplane 
design.  And its successor ARMAC was where the Internet routing algorithm 
(shortest path algorithm) was first run.

paul

Re: Speed now & then (Space and time?)

[Chuck Guzis via cctalk]

On 03/29/2018 02:24 PM, Fred Cisin via cctalk wrote:
>>> HOWEVER, a variant of "Boyle's Law" warns that software and content
>>> will expand to fit all available space and speed.
> 
> On Thu, 29 Mar 2018, allison via cctalk wrote:
>> We have proof and it is us.
> 
> Or, as Walt Kelly ("Pogo") said, "We have met the enemy, and he is us."
> 
>> I posited that 2 decades ago in a wired article.  My CP/M machine booted
>> in seconds while waiting for
>> the winders box to decide if it would/could.
> 
> "The new machine is so much faster, that it can almost get out of its
> own way!"

As a real contrast, consider, say, the IBM 1620.  Go look up the cycle
times on that beast.

And yet, productive work was performed on it.  Indeed the industrial
variant, the 1710 was used for early process control.

Now that I don't need to work on the bleeding edge, I prefer to see how
much can be done with the least.

It's fun.

--Chuck

Re: Speed now & then (Space and time?)

[Fred Cisin via cctalk]

HOWEVER, a variant of "Boyle's Law" warns that software and content
will expand to fit all available space and speed.


On Thu, 29 Mar 2018, allison via cctalk wrote:

We have proof and it is us.


Or, as Walt Kelly ("Pogo") said, "We have met the enemy, and he is us."


I posited that 2 decades ago in a wired article.  My CP/M machine booted
in seconds while waiting for
the winders box to decide if it would/could.


"The new machine is so much faster, that it can almost get out of its own 
way!"




It is hideous.   But you need the picture. 


How should I react to the college administator who told me, "You're not 
seeing the big picture"?   :-)   I was considering defenestration.




HTML has helped that along.

HTML is not nearly so bad its slightly bigger than runoff only wordier.
However that we need HTML for a screen of text is, yes, bad!
I blame WYSISWYG, and Postscript!  WYGINS  (for those that forgot, What
You Get Is No Surprise)
from the days before high resolution printers.


Nothing wrong with a markup language.  It's the application to 
inappropriate uses.
WYSIWYG was touted as being professional approach.  But the typesetter who 
I used to use never needed, nor wanted, a PICTURE of how a line of text 
would look with the fonts he chose.
I did a lot of my early work in YAFIYGI (You Asked For It, You Got IT) 
systems, such as manually embedding Cordata or PCL font commands in my 
text.



One college administrator managed that with ease.  He created the memo
. . . 
But, that was almost a decade ago.  I wonder whether he is now

attaching MP4s?



Eep, the man is batty.


I feel guilty about not having defenestrated the college administration.



Dancing kangaroos and yodelling jellyfish has let form triumph over
content!   When will we finally have smell-o-vision?

Please no, smell-o-vision.  I can see the hackers going for the cross
between skunk, pepperspray,
and some toxic chemical mess.   Obviously a Blacktooth perpiheral.


It will need a Fart virus.



Yes, certainly, the hardware is much faster, and has more storage space.
Yet, the task takes longer, and storage space runs out just as quickly.

Thats the whole sad story.   It is why I still run CP/M, RT-11 and even
a DECMate!  All hail fanfold!
Allison


Allison is wonderful!
Somebody who understands what I've been ranting about!
"So, THIS is 'progress'?"

--
Grumpy Ol' Fred ci...@xenosoft.com

Re: Speed now & then (Space and time?)

[allison via cctalk]

On 03/29/2018 03:35 PM, Fred Cisin via cctalk wrote:
> On Thu, 29 Mar 2018, Murray McCullough via cctalk wrote:
>> I’m not trying to date myself but have things truly sped up? In 1970’s
>> Toronto I had a classic computer, sorry can’t recall what it was,
>> connected
>> to a 300 baud modem; by early 80’s had ‘zoomed’ to 9600 baud. Oh, my!
>> [ A
>> typical file size to download was probably 1 MB. ] Speed indeed! Yet
>> now,
>> here in rural Ontario, Canada, I’m at 5MB/s. Yikes! (Friends in
>> Toronto are
>> at 50MB/s.) We can do the math but content, particularly multimedia, has
>> swollen in size.[ 1 GB is not unheard of. ] Were classic computing days
>> that much slower? Happy computing. Murray  -:)
>
> Application of "Moore's Law" calls for a logarithmic increase in
> speed, such as doubling every 18 months.  Yes, the rate, in terms of
> bits per second has grown a lot.
> Similarly storage capacity has grown.
>
>
> HOWEVER, a variant of "Boyle's Law" warns that software and content
> will expand to fit all available space and speed.
>
We have proof and it is us.

>
> Once, if your handwriting is bad enough, you could type your grocery
> shopping list into Electric Pencil.  Took a few seconds.   later
> WordStar. Scripsit.  WordPervert.  Microsoft Weird. Does Clippy have a
> template for it?
> (PC-Write was a welcome respite in that growing bloat!)
>
I posited that 2 decades ago in a wired article.  My CP/M machine booted
in seconds while waiting for
the winders box to decide if it would/could.


> It's kinda like: the plane flight is half an hour shorter, but the
> airport pre-processing in an hour longer.
>
I fly a Cessna150, cruse speed of 110mph, I could fly to Ohio in six
hours with one fuel stop.
Commercial flight is easily 4x faster and it still takes 6 hours door to
door. 

> Once, the operating system, such as PC-DOS 1.00, fit on a single sided
> MFM 160K floppy disk.  Now, much software comes on DVD, because CD-ROM
> (2/3 GB) isn't large enough!
>
Back when 160k was space, now it's a small entry in a table.

> A memo announcing change of room and time for a meeting is a very
> short paragraph.  That used to be about half a kilobyte.
> Now, it tends to be a few MB.
> It seems that some serious effort has to go into wasting so much
> capacity!
It is hideous.   But you need the picture. 

> HTML has helped that along.
>
HTML is not nearly so bad its slightly bigger than runoff only wordier.
However that we need HTML for a screen of text is, yes, bad!

I blame WYSISWYG, and Postscript!  WYGINS  (for those that forgot, What
You Get Is No Surprise)
from the days before high resolution printers.

> One college administrator managed that with ease.  He created the memo
> in his word processor, printed it on his color printer, signed it,
> SCANNED it, and attached the 24bit-color picture as an attachment to
> an email. The subject line of the email was: "FYI".  The text, other
> than the attachment was: "See attachment".  The attachment was an
> uncompressed picture of a line of text in the middle of a full sheet
> of paper:
> "The curriculum committee has been moved to room D-233 at 2:oo"
> But, in the memo, there was a horizontal rule that was not quite
> horizontal; one end was a few pixels higher than the other! - scanning
> with the paper not quite aligned may well be the easiest way to
> accomplish THAT!
> But, that was almost a decade ago.  I wonder whether he is now
> attaching MP4s?
>
Eep, the man is batty.

> MP4s mean that now, not only does it take MUCH longer to create the
> document, we can now waste MUCH more of the reader's time!
> I find it very annoying that when GOOGLE'ing to find a simple answer,
> many of the first hits are YouTube.
> A few seconds glance at a text document will likely tell me whether
> the answer to my question is there.  Or a sketch and maybe a
> photograph of somebody's hardware setup.  Instead, sit through minutes
> of talking heads.
> With background music to make it hard to make out what is being said!
> Youtube's "auto-generated CC" is a poor substitute for text.
>
>
> Dancing kangaroos and yodelling jellyfish has let form triumph over
> content!   When will we finally have smell-o-vision?
>
Please no, smell-o-vision.  I can see the hackers going for the cross
between skunk, pepperspray,
and some toxic chemical mess.   Obviously a Blacktooth perpiheral.

 I will nominally run without that peripheral.  Come to think of it I
did that for a decades regarding
sound.    Most of my favorite modern Linux machines can't squawk, peep,
hear, or see me.

> Yes, certainly, the hardware is much faster, and has more storage space.
> Yet, the task takes longer, and storage space runs out just as quickly.

Thats the whole sad story.   It is why I still run CP/M, RT-11 and even
a DECMate!  All hail fanfold!


Allison

Re: Speed now & then (Space and time?)

[Fred Cisin via cctalk]

On Thu, 29 Mar 2018, Murray McCullough via cctalk wrote:

I’m not trying to date myself but have things truly sped up? In 1970’s
Toronto I had a classic computer, sorry can’t recall what it was, connected
to a 300 baud modem; by early 80’s had ‘zoomed’ to 9600 baud. Oh, my! [ A
typical file size to download was probably 1 MB. ] Speed indeed! Yet now,
here in rural Ontario, Canada, I’m at 5MB/s. Yikes! (Friends in Toronto are
at 50MB/s.) We can do the math but content, particularly multimedia, has
swollen in size.[ 1 GB is not unheard of. ] Were classic computing days
that much slower? Happy computing. Murray  -:)


Application of "Moore's Law" calls for a logarithmic increase in speed, 
such as doubling every 18 months.  Yes, the rate, in terms of bits per 
second has grown a lot.

Similarly storage capacity has grown.


HOWEVER, a variant of "Boyle's Law" warns that software and content will 
expand to fit all available space and speed.



Once, if your handwriting is bad enough, you could type your grocery 
shopping list into Electric Pencil.  Took a few seconds.   later WordStar. 
Scripsit.  WordPervert.  Microsoft Weird. 
Does Clippy have a template for it?

(PC-Write was a welcome respite in that growing bloat!)

It's kinda like: the plane flight is half an hour shorter, but the airport 
pre-processing in an hour longer.


Once, the operating system, such as PC-DOS 1.00, fit on a single sided MFM 
160K floppy disk.  Now, much software comes on DVD, because CD-ROM (2/3 
GB) isn't large enough!


A memo announcing change of room and time for a meeting is a very short 
paragraph.  That used to be about half a kilobyte.

Now, it tends to be a few MB.
It seems that some serious effort has to go into wasting so much capacity!
HTML has helped that along.

One college administrator managed that with ease.  He created the memo in 
his word processor, printed it on his color printer, signed it, SCANNED 
it, and attached the 24bit-color picture as an attachment to an email. 
The subject line of the email was: "FYI".  The text, other than the 
attachment was: "See attachment".  The attachment was an uncompressed 
picture of a line of text in the middle of a full sheet of paper:

"The curriculum committee has been moved to room D-233 at 2:oo"
But, in the memo, there was a horizontal rule that was not quite 
horizontal; one end was a few pixels higher than the other! - scanning 
with the paper not quite aligned may well be the easiest way to accomplish 
THAT!
But, that was almost a decade ago.  I wonder whether he is now attaching 
MP4s?


MP4s mean that now, not only does it take MUCH longer to create the 
document, we can now waste MUCH more of the reader's time!
I find it very annoying that when GOOGLE'ing to find a simple answer, many 
of the first hits are YouTube.
A few seconds glance at a text document will likely tell me whether the 
answer to my question is there.  Or a sketch and maybe a photograph of 
somebody's hardware setup.  Instead, sit through minutes of talking heads.

With background music to make it hard to make out what is being said!
Youtube's "auto-generated CC" is a poor substitute for text.


Dancing kangaroos and yodelling jellyfish has let form triumph over 
content!   When will we finally have smell-o-vision?


Yes, certainly, the hardware is much faster, and has more storage space.
Yet, the task takes longer, and storage space runs out just as quickly.

--
Grumpy Ol' Fred ci...@xenosoft.com

Re: Speed now & then

[Ethan Dicks via cctalk]

On Thu, Mar 29, 2018 at 12:20 PM, Murray McCullough via cctalk
 wrote:
> I’m not trying to date myself but have things truly sped up? In 1970’s
> ...  300 baud modem; by early 80’s had ‘zoomed’ to 9600 baud.

What hasn't changed is people.  Back when we had 300 baud, we only had
so many hours a night for eyeballs-on-slow text, so one could only
expect so much content to be transmitted.  File transfers were
hands-off, but even the graphic machines of the time had low-res
screens so a mono-graphic didn't take so long to pull down, unlike
640x480x8bit stuff that was common by the mid-90s.

As telecom speeds (modem then ISDN then DSL, etc, etc) ramped up, what
one could do in 60 seconds or even a 3-second attention span grew.
It's true that webpages have gotten really fat and the amount of media
one pulls down from a single click is staggering, but nobody would
have sat and waited that long for pages to load at dial-up speeds.
More bandwidth enables more to get pulled in the same amount of time,
it doesn't mean that anyone will stand still and load the same
quantity as before but faster.

It's like fixed purchase points for PCs or digital cameras or
whatever.  Each year, the new models cost about the same as the old
models but with more features.  The price points are set and as
technology gets cheaper, they just pack more in the box and leave the
price alone.  Same with telecom - more content, faster lines, same
time in chair.

-ethan

Re: Speed now & then

[Paul Koning via cctalk]

> On Mar 29, 2018, at 12:20 PM, Murray McCullough via cctalk 
>  wrote:
> 
> I’m not trying to date myself but have things truly sped up? In 1970’s
> Toronto I had a classic computer, sorry can’t recall what it was, connected
> to a 300 baud modem; by early 80’s had ‘zoomed’ to 9600 baud. Oh, my! [ A
> typical file size to download was probably 1 MB. ] Speed indeed! Yet now,
> here in rural Ontario, Canada, I’m at 5MB/s. Yikes! (Friends in Toronto are
> at 50MB/s.) We can do the math but content, particularly multimedia, has
> swollen in size.[ 1 GB is not unheard of. ] Were classic computing days
> that much slower? Happy computing. Murray  -:)

I remember downloading the GCC release kit over a 56k dialup line, in 2000.  
Took a while.

The ARPAnet in its early days had "high speed backbone" links which were 56k 
bps.  Terminal links presumably 110 bps, that being the speed of ASCII 
teletypes.  And back in the late 1970s you could still find even slower links 
in some places, such as 6 bit links connecting teletype machines for newspaper 
"wire service" feeds.

It would be fun to do a "generalized Moore's Law" chart, showing not just 
transistor count growth (Moore's subject) but also the many other scaling 
changes of computing: disk capacity, recording density, disk IOPS, disk 
bandwidth, ditto those for tape, CPU MIPS, memory size, memory bandwidth, 
network bandwidth...

All these have grown dramatically, but very clearly not in the same proportion, 
for some of these the changes are smooth while others are jumps, and the rate 
of change sometimes varies dramatically over the decades.

paul