Posted on 01/22/2024 8:53:22 PM PST by ProtectOurFreedom
Back in 1978 I (i.e., the narrator Mr. Darling) worked for Cray Research, which made what was at the time the world’s fastest computer, the Cray-1. I thought it would be interesting to compare the Cray-1 with the latest model of iPhone, the iPhone 13. (this is a 2022 video)
| Item | Unis | Cray 1 | iPhone 13 | Factor |
|---|---|---|---|---|
| Weight | Ounces | 176,000 | 6 | 29,300 |
| Price (2022 $) | 2022 | 38,000,000 | 1,000 | 38,000 |
| Speed | MFLOPS | 160 | 15,800,000 | 100,000 |
| Memory | MB | 160 | 512,000 | 60,000 |
| Power consumption | kW | 115 | LiOn 20 hours | |
| Cooling | Freon refr | Air |
All of us computer nerds in junior high and high school in the early 80s loved seeing the specs on those supercomputers compared to our Apple IIs, Commodore 64s, Atari STs, TI 994/As…
That’s what the missus tells me, too. Every year or two.
(TRaSh 80)
It is a lot of change. One of my first jobs after the Navy was working with Maxtor and Seagate. Somewhere I still have a 5MB 5.25” full-size winchester drive. 5MB!
2bad she doesn’t recognize sarc
Cool! I went to a talk at IBM Almaden Valley a long time ago (around 1982) about the history of the IBM 305 RAMAC — the first disk drive. It was a celebration of the 40th anniversary of the tech.
My wife bought me a SuperMac “DataFrame” 20 MB Hard Disk for our Macintosh Plus. It was the first hard disk drive to take advantage of the SCSI port on the Macintosh Plus. She paid $600 for it at the time (almost $1,700 today!). What a sweet present! That was back in the Dual Income, No Kids days. Shortly after that our first was born and that was the end of expensive toys.
I remember BASIC and paper tape to store my programs on.
FORTRAN and punch cards came much later.
I worked in those rooms. I thought they were a crock and when that was done the companies were awash with money. I usually had no trouble visualizing the objectives in 3D but sometimes, for complex geology such as in thrust belts, I would take orthogonal cross sections and make 3D displays or take a map on tops and run rods through it to warp it to the right surface configuration so management could understand.
Oh yeah. Had those in the high school computer lab and even in junior college. Yuck they were.
I worked at Cray Research starting the 3rd year they were in business. I was a group member of the software testing group, then project leader, then department head of software testing. It was a great place to work. There was a lot of money to be spent, and they took very good care of the employees. There were high expectations of performance, and there was a lot of freedom given to figure out how to do your job, no time for anyone to micromanage your work. I really enjoyed my time there.
ping
ping
“The Cray was operating on MUCH tighter code, though.”
No shiite - if one ever gets a chance to see the code behind any windowing GUI, it will gob-smack you. ‘Bloated’ and inefficient only begins to describe it. The linux GUIs are far more efficient but still execute billions of lines of code just to open a window. We have come a long ways from packing multiple flags into a single byte just to save some memory.
I first worked on CDC CYBER mainframes (64, 170, 730, 7400) and the CYBER 203 and 205 Super Computers (designed by Seymour Cray while he worked for Control Data Corporation) before he started Cray Computing. I wrote Fortran and Assembly language for about 10 years. My division was bought out by Siemens Power System Controls in around 1990. I then transitioned to Oracle, SQL and PL/SQL. The 205 I worked on took up a 60,000 sq ft building. It was a 128 bit vector processing unit.
As recently as 1989 I registered for college classes using IBM cards and a checkbook. Now they’re both obsolete.
CC
We’re of the same vintage for college years, and I still have the old slide rule that served me well. Senior year I bought a Craig calculator - 4 functions with a constant key. Fortunately, the cadet store sold products at cost so I was able to afford it, about $90 as I recall.
Part of the education at GA Tech when I was there (late 70s -early 80s) was figuring out how to beat the system to get those Hollerith cards to get a decent schedule. Might have been a more important part of the education than some of the classes.
If you did things the way you were nominally supposed to, you probably weren’t getting a decent schedule.
I was just thinking about coding inefficiency last night. No one seems to care anymore because more memory and processing power is always right around the corner. I think it should still matter.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.