20 years ago...

Message boards : Number crunching : 20 years ago...
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · 4 . . . 6 · Next

AuthorMessage
Profile Mahoujin Tsukai
Volunteer tester
Avatar

Send message
Joined: 21 Jul 07
Posts: 147
Credit: 2,204,402
RAC: 0
Singapore
Message 616291 - Posted: 8 Aug 2007, 11:33:32 UTC

Back in 1987, would distributed computing similar to SETI@HOME (or some other BOINC project) be possible?

Of course, on a smaller scale (fewer people had PCs back then) and less complex workunits (most PCs were 8088/86, 286, 386s back then). The World Wide Web was not yet developed, but workunits could have been sent through some other form of online communication.

ID: 616291 · Report as offensive
peristalsis

Send message
Joined: 23 Jul 99
Posts: 154
Credit: 28,610,163
RAC: 51
United States
Message 616296 - Posted: 8 Aug 2007, 12:37:58 UTC

Let me see...my Commodore C64 with its blindingly fast 300 baud modem would have been up to the task! Tie up the phone line for a day or two d/l'ing a WU and then work on the work unit for a month or three (g)..j
ID: 616296 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 616304 - Posted: 8 Aug 2007, 13:26:07 UTC

In theory it would have been possible, but probably more expensive.

The most logical way to do it would have been to set up a BBS or Bulletin Board System, which were fairly popular back then.

SETI would need multiple phone lines to handle all the clients. WUs would probably have to be 1/16 or 1/32 the size of the original Classic units (pre-Enhanced) just to get downloaded and processed in a timely manor.

WU caching would probably be a bad idea, unless you're going to be away for quite some time. The software would have to be compiled in 16bit to run under DOS (as was most common, though other OSes could of course be supported). A 16bit Windows app would probably be a "new" thing but would slow down processing. They could even use a 32bit extender such as DOS4/GW or DPMI for those "powerful, brand new 386s!"

Programming it under DOS, which has no prioritizing like Windows 9x and newer does, it would have to be programmed as a TSR loaded via CONFIG.SYS or AUTOEXEC.BAT, and would only process when the user is inactive (otherwise the user experience would be horrible).

Basically, the science would have taken forever, but it would have been possible.
ID: 616304 · Report as offensive
Profile KD [SETI.USA]
Avatar

Send message
Joined: 24 Oct 99
Posts: 459
Credit: 2,513,131
RAC: 0
United States
Message 616366 - Posted: 8 Aug 2007, 16:08:35 UTC - in response to Message 616291.  
Last modified: 8 Aug 2007, 16:42:54 UTC

Back in 1987, would distributed computing similar to SETI@HOME (or some other BOINC project) be possible?

Of course, on a smaller scale (fewer people had PCs back then) and less complex workunits (most PCs were 8088/86, 286, 386s back then). The World Wide Web was not yet developed, but workunits could have been sent through some other form of online communication.



It would have been possible. Likely, it would have first started on the internet. I was using the internet even before 1987. (Seriously!) At first, it would have been limited primarily only to the academic communities. Clients would have been developed for UNIX and VMS, to be run by the administrators of their respected machines. WU and results could have been sent by FTP, with each host having their own designated directory. Announcements could have gone out by mailing lists. Perhaps even text files that listed stats could be generated at some interval. Also, the project could have also ran a telnetable BBS on one of their internet nodes.

To bring the "general" public onboard, as already mentioned, dialup BBS's would have been the most efficient approach. Instead of just a single dialup BBS on SETI's end, WU's and results could have been distributed down and upstream over Fidonet. Each participating dialup BBS would be responsible for users who wish to participate in the program. I envision a door program to be used for the distribution of WU's to individual users and the collection of results from individual users, as this approach would be modular so such program could be used with the many dialup BBS programs at the time. The aggregate transfer of WU's and results for all users on the participating dialup BBS would be done during the nightly Fidonet process. For most people, it would have likely taken about a week for their results to swim upstream over Fidonet. Perhaps one a month, statistics could be generated for each dialup BBS (and it's users) to be freq'd out also.

Also, as already mentioned, the TSR approach would have been favorable for the PC. Its likely that applications for other architectures (Mac, Amiga) would have been developed as well. I wouldn't have been surprised if some crafty folks in the Tandy "Coco" and Commodore communities also developed applications. Sure, it would have been s-l-o-w, but...

I think the larger question would be: which would have been more efficient in 1987, a distributed project where you have (at best) only a couple thousand users with the hardware of the time or just a single supercomputer. If the funds were available for a supercomputer, that would have been the big question then as well.

ID: 616366 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65709
Credit: 55,293,173
RAC: 49
United States
Message 616404 - Posted: 8 Aug 2007, 17:17:46 UTC - in response to Message 616304.  

In theory it would have been possible, but probably more expensive.

The most logical way to do it would have been to set up a BBS or Bulletin Board System, which were fairly popular back then.

SETI would need multiple phone lines to handle all the clients. WUs would probably have to be 1/16 or 1/32 the size of the original Classic units (pre-Enhanced) just to get downloaded and processed in a timely manor.

WU caching would probably be a bad idea, unless you're going to be away for quite some time. The software would have to be compiled in 16bit to run under DOS (as was most common, though other OSes could of course be supported). A 16bit Windows app would probably be a "new" thing but would slow down processing. They could even use a 32bit extender such as DOS4/GW or DPMI for those "powerful, brand new 386s!"

Programming it under DOS, which has no prioritizing like Windows 9x and newer does, it would have to be programmed as a TSR loaded via CONFIG.SYS or AUTOEXEC.BAT, and would only process when the user is inactive (otherwise the user experience would be horrible).

Basically, the science would have taken forever, but it would have been possible.

On Atari computers of the time using the 6502C cpu It would have been 8bits instead of 16bits, In 1990 though 16 or 32bits would have been possible and with a GUI OS too(TOS/GEM/CPM68k/Minix), Initial efforts at creating an external ethernet device for the ST computers had been done, but It was crude and never really got anywhere. The ST computers were Data compatible with PCs back then even though they used Motorolas 68000(68030&68040) family of CPUs. Of course for those that had Amigas(As I did) they also were capable just like the ST computers, I know of one guy who uses Amiga computers in a TV studio that He owns in and among His PCs and I think the Amigas are linked by ethernet possibly(I never asked).
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 616404 · Report as offensive
Profile Dr. C.E.T.I.
Avatar

Send message
Joined: 29 Feb 00
Posts: 16019
Credit: 794,685
RAC: 0
United States
Message 616410 - Posted: 8 Aug 2007, 17:33:13 UTC - in response to Message 616296.  

Let me see...my Commodore C64 with its blindingly fast 300 baud modem would have been up to the task! Tie up the phone line for a day or two d/l'ing a WU and then work on the work unit for a month or three (g)..j


. . . could you imagine what the 'Quorum' would be "dealing with" in that event (Today) . . . sniCkr'n . . .


BOINC Wiki . . .

Science Status Page . . .
ID: 616410 · Report as offensive
Profile KD [SETI.USA]
Avatar

Send message
Joined: 24 Oct 99
Posts: 459
Credit: 2,513,131
RAC: 0
United States
Message 616602 - Posted: 8 Aug 2007, 23:10:06 UTC - in response to Message 616410.  

Let me see...my Commodore C64 with its blindingly fast 300 baud modem would have been up to the task! Tie up the phone line for a day or two d/l'ing a WU and then work on the work unit for a month or three (g)..j


. . . could you imagine what the 'Quorum' would be "dealing with" in that event (Today) . . . sniCkr'n . . .



Well, the WU's would have been a lot smaller to begin with...

It is likely that in 1987 there wouldn't have even been a quorum though. It would have been thought about, but wouldn't have gone past "It would be nice, for science redundancy, to have a quorum if enough resources are available". The likelihood of "cheaters" back then would be negligible as well.
ID: 616602 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 616604 - Posted: 8 Aug 2007, 23:11:48 UTC

According to RFC-1296, there were 28,174 hosts on the internet in December, 1987.
ID: 616604 · Report as offensive
Fred W
Volunteer tester

Send message
Joined: 13 Jun 99
Posts: 2524
Credit: 11,954,210
RAC: 0
United Kingdom
Message 616611 - Posted: 8 Aug 2007, 23:16:09 UTC - in response to Message 616604.  

According to RFC-1296, there were 28,174 hosts on the internet in December, 1987.


I wonder who the other 28,173 were.
ID: 616611 · Report as offensive
Profile KD [SETI.USA]
Avatar

Send message
Joined: 24 Oct 99
Posts: 459
Credit: 2,513,131
RAC: 0
United States
Message 616765 - Posted: 9 Aug 2007, 2:21:59 UTC - in response to Message 616404.  
Last modified: 9 Aug 2007, 2:34:58 UTC

Initial efforts at creating an external ethernet device for the ST computers had been done, but It was crude and never really got anywhere.


I'm going back here and relying on memory, so unfortunately the details have been lost to time.

Back then, in addition to a pee-cee, I also had Tandy TRS-80 Color Computers (Coco's). I got more enjoyment out of the Coco's. They were simple and fun to use, relatively easy to integrate into the "real world" through their A/D converters (joystick ports), could be run under a UNIX-like RTOS called OS-9 by Microware, and had a relatively large community following. Like the Atari and Commodore 8-biters, they were a hobbyist's machine.

The Coco's also had a following with the Ham radio community as well. Among many other things, they were used as TTY terminals for packet radio. That, in itself, isn't surprising.

However, I remember reading something at the time that I found very interesting. It is as interesting today as well. I don't remember if I read it in a magazine or online somewhere, but I do remember reading about how NWS (or it may have been NOAA) "storm chasers" were using Coco's as mobile terminals. The interesting part was that they using *TCP/IP* and were a bona fide internet node! They had built, basically, a "black box" which handled the TCP/IP packets to and from their HF radios and translated them into serial data that the computer could handle. Since all the 8-biters at this time had either a DB-9 serial or RS-232 port, this "black box" could turn any of the 8-biters (Coco, Commodore, Atari) into an "internet node".

(In 1987, the way I was accessing the internet was through serial dialup to a local VAX/VMS .edu machine, using just a terminal emulator. PPP, or even SLIP, didn't exist yet. Actually having your own IP address, at home, was a pipe dream!)
ID: 616765 · Report as offensive
Profile Gecko
Volunteer tester
Avatar

Send message
Joined: 17 Nov 99
Posts: 454
Credit: 6,946,910
RAC: 47
United States
Message 616798 - Posted: 9 Aug 2007, 4:22:43 UTC
Last modified: 9 Aug 2007, 4:24:41 UTC

I think the Apple II-Gs I was using at the time at Univ. of TN could have done just fine.
The project just would have scaled to the input/output resources available at the time.

A cool profile of the Seti project on Nova and some articles by Sci American, Discover etc. would have sparked quite a bit of interest even then.
I could envision WUs on floppies that would be crunched and output written back to disks to be returned by mail...kind of like netflix. As a person returned disks, he would simply be sent more to replace them. WUs by mail instead of by internet. Back then, probably a better and more reliable way to reach the general masses who could afford $4K computers anyway. Dealing w/ re-integrating the returned data would have been laborious and tedious, but then again....many things required MUCH more human involvement and resources then, that today is all electronically processed w/ no or few humans required.

All in all, simpler times = simpler solutions, but I think the hardware and software were up to the task in 87' for a more primitive and smaller scale form of the Seti Classic we started to crunch in 99'.

ID: 616798 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 616804 - Posted: 9 Aug 2007, 4:33:59 UTC - in response to Message 616765.  


I'm going back here and relying on memory, so unfortunately the details have been lost to time.

Back then, in addition to a pee-cee, I also had Tandy TRS-80 Color Computers (Coco's). I got more enjoyment out of the Coco's. They were simple and fun to use, relatively easy to integrate into the "real world" through their A/D converters (joystick ports), could be run under a UNIX-like RTOS called OS-9 by Microware, and had a relatively large community following. Like the Atari and Commodore 8-biters, they were a hobbyist's machine.

The Coco's also had a following with the Ham radio community as well. Among many other things, they were used as TTY terminals for packet radio. That, in itself, isn't surprising.

However, I remember reading something at the time that I found very interesting. It is as interesting today as well. I don't remember if I read it in a magazine or online somewhere, but I do remember reading about how NWS (or it may have been NOAA) "storm chasers" were using Coco's as mobile terminals. The interesting part was that they using *TCP/IP* and were a bona fide internet node! They had built, basically, a "black box" which handled the TCP/IP packets to and from their HF radios and translated them into serial data that the computer could handle. Since all the 8-biters at this time had either a DB-9 serial or RS-232 port, this "black box" could turn any of the 8-biters (Coco, Commodore, Atari) into an "internet node".

(In 1987, the way I was accessing the internet was through serial dialup to a local VAX/VMS .edu machine, using just a terminal emulator. PPP, or even SLIP, didn't exist yet. Actually having your own IP address, at home, was a pipe dream!)


LOL...

Yep, the era of stone knives and bearskins. ;-)

And if you tried to tell most people what is was you were trying to do with that collection of gear, they looked at you like you had three heads. :-)

Alinator
ID: 616804 · Report as offensive
zombie67 [MM]
Volunteer tester
Avatar

Send message
Joined: 22 Apr 04
Posts: 758
Credit: 27,771,894
RAC: 0
United States
Message 616809 - Posted: 9 Aug 2007, 5:19:00 UTC - in response to Message 616366.  
Last modified: 9 Aug 2007, 5:25:37 UTC

It would have been possible. Likely, it would have first started on the internet. I was using the internet even before 1987. (Seriously!)

I started working at Stanford University Network in 1987. I quickly discovered USENET, working on my WYSE terminal. There were *plenty* of binaries being exchanged, even then... And it was not limited to just edu and mil.

Edit: P.S. USENET is still a far superior forum environment. There are many very slick tools to customize the message reading filtering/reading process. *Way* better than these caveman http boards.
Dublin, California
Team: SETI.USA
ID: 616809 · Report as offensive
Profile Mahoujin Tsukai
Volunteer tester
Avatar

Send message
Joined: 21 Jul 07
Posts: 147
Credit: 2,204,402
RAC: 0
Singapore
Message 617080 - Posted: 9 Aug 2007, 17:10:52 UTC

Thanks to everyone for the replies. It seems that back in 1987, distributed computing projects would have been impratical, but possible.

And programming SETI to run on PCs back then would have been more difficult.
ID: 617080 · Report as offensive
Profile Gecko
Volunteer tester
Avatar

Send message
Joined: 17 Nov 99
Posts: 454
Credit: 6,946,910
RAC: 47
United States
Message 617105 - Posted: 9 Aug 2007, 18:07:53 UTC - in response to Message 617080.  

Thanks to everyone for the replies. It seems that back in 1987, distributed computing projects would have been impratical, but possible.

And programming SETI to run on PCs back then would have been more difficult.


We did manage however to land on the moon in 69', still arguably the greatest technological accomplishment of mankind to date w/ even simpler computer tech that still had to coordinate & share data w/ many locations/people....besides actually being a critical technical element of the equipment, navigation and landing. If anything, the moon program is a great example of an incredibly complex integration of computer components w/ the human component. Dependency on either alone would not have gotten us there.

So, I have to absolutely believe this project was possible in 87', but completely agree w/ your point..."not practical". In fairness, the moon program was a "price no object" directive, whereas Seti operates at the far other extreme of budgetary modesty and the generosity of it's supporters & volunteer crunchers.


ID: 617105 · Report as offensive
Profile Philadelphia
Volunteer tester
Avatar

Send message
Joined: 12 Feb 07
Posts: 1590
Credit: 399,688
RAC: 0
United States
Message 617112 - Posted: 9 Aug 2007, 18:26:21 UTC - in response to Message 617105.  

Thanks to everyone for the replies. It seems that back in 1987, distributed computing projects would have been impratical, but possible.

And programming SETI to run on PCs back then would have been more difficult.


We did manage however to land on the moon in 69', still arguably the greatest technological accomplishment of mankind to date w/ even simpler computer tech that still had to coordinate & share data w/ many locations/people....besides actually being a critical technical element of the equipment, navigation and landing. If anything, the moon program is a great example of an incredibly complex integration of computer components w/ the human component. Dependency on either alone would not have gotten us there.

So, I have to absolutely believe this project was possible in 87', but completely agree w/ your point..."not practical". In fairness, the moon program was a "price no object" directive, whereas Seti operates at the far other extreme of budgetary modesty and the generosity of it's supporters & volunteer crunchers.



Anyone know what NASA did use for it's onboard computer for the lunar landing?
ID: 617112 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 617114 - Posted: 9 Aug 2007, 18:28:40 UTC

ID: 617114 · Report as offensive
Profile Gecko
Volunteer tester
Avatar

Send message
Joined: 17 Nov 99
Posts: 454
Credit: 6,946,910
RAC: 47
United States
Message 617122 - Posted: 9 Aug 2007, 18:40:17 UTC - in response to Message 617114.  

Apollo Guidance Computer

Alinator



That was a nifty read. Thanks!
ID: 617122 · Report as offensive
Profile zoom3+1=4
Volunteer tester
Avatar

Send message
Joined: 30 Nov 03
Posts: 65709
Credit: 55,293,173
RAC: 49
United States
Message 617158 - Posted: 9 Aug 2007, 19:55:28 UTC - in response to Message 617122.  

Apollo Guidance Computer

Alinator



That was a nifty read. Thanks!

My Atari 400 was not much slower @ 1.79MHz, But the 6502C cpu sure could address more ram and rom(64k total address space). But It came out about 10 years later.
The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's
ID: 617158 · Report as offensive
Profile Jim-R.
Volunteer tester
Avatar

Send message
Joined: 7 Feb 06
Posts: 1494
Credit: 194,148
RAC: 0
United States
Message 617165 - Posted: 9 Aug 2007, 20:10:25 UTC - in response to Message 617158.  

Apollo Guidance Computer

Alinator



That was a nifty read. Thanks!

My Atari 400 was not much slower @ 1.79MHz, But the 6502C cpu sure could address more ram and rom(64k total address space). But It came out about 10 years later.

And you could program yours just by using basic commands on a regular keyboard. That thing had to have every command, piece of data, etc. entered in either octal or hexadecimal! But it is amazing what it could do for it's time.
Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.
ID: 617165 · Report as offensive
1 · 2 · 3 · 4 . . . 6 · Next

Message boards : Number crunching : 20 years ago...


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.