For context, the [[IBM 7090]] article in wiki has conflicting sources for the 
machine that played Daisy May. I found it implausible that BTL would still be 
using a 704 in 1961, although I knew that the GM Tech Center in Detroit still 
had one. The 7094 was impossible unless BTL was a test site, and nothing in the 
movie looked like its console.

You're right about reporters; describing the 7090 and 7094 as having 32,768 
bytes is strange unless you qualify it as 36-bit bytes.

I was hoping that somebody had access to one of the old dead trees.

--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
עַם יִשְׂרָאֵל חַי
נֵ֣צַח יִשְׂרָאֵ֔ל לֹ֥א יְשַׁקֵּ֖ר




________________________________________
From: IBM Mainframe Discussion List <[email protected]> on behalf of 
Joel Ewing <[email protected]>
Sent: Saturday, August 30, 2025 12:39 AM
To: [email protected] <[email protected]>
Subject: Re: Daisy Bell Was: Re: Artificial stupidity in action


External Message: Use Caution


The article in Wikipedia on John Larry Kelly Jr, who was one of those
involved in the Speech synthesis part of the 1961 Daisy Bell song demo says:
"In 1961, Kelly and colleagues Carol Lochbaum andLou Gerstman
<https://en.wikipedia.org/wiki/Louis_Gerstman>created one of the most
famous moments in the history of Bell Telephone Laboratories by using
anIBM 7094 <https://en.wikipedia.org/wiki/IBM_7094>computer to
synthesize speech.^[4]
<https://en.wikipedia.org/wiki/John_Larry_Kelly_Jr.#cite_note-4> ^[5]
<https://en.wikipedia.org/wiki/John_Larry_Kelly_Jr.#cite_note-5> A
demonstration by Kelly and Gerstman took place on May 10, 1961, at a
meeting in Philadelphia ...^]
<https://en.wikipedia.org/wiki/John_Larry_Kelly_Jr.#cite_note-6> Their
voice recorder synthesizer/vocoder
<https://en.wikipedia.org/wiki/Vocoder>/recreated the song/Daisy Bell
<https://en.wikipedia.org/wiki/Daisy_Bell>/,^[7]
<https://en.wikipedia.org/wiki/John_Larry_Kelly_Jr.#cite_note-7>"

One of the cited references of the Kelly article is to an article in the
/Sydney Morning Herald/, June 18,1961, p25, written by a reporter for
the Australian Associated Press who witnessed the event.   Without
access to the article, can't prove whether it explicitly referenced a
specific IBM model, but this supports the May 1961 date for the demo
event as reasonable.

I still have the old 1962 Decca vinyl record /Music From Mathematics,/
which includes the Daisy Bell song, and its jacket says an IBM 7090 was
used for all tracks on the record. The record jacket info was written by
members of the Bell Labs project, so if it says IBM 7090 was used for
those specific recordings, that must be the case.  This record is a
public release of  recordings that were internally released in 1961
within Bell Labs.

The first IBM 7094 was supposedly installed in Sept 1962, and there
doesn't seem to be any accounts of Bell Lab acquiring one before general
availability.  Its use for a June 1961 demo seems impossible.

It sounds llike the early research by Bell Labs in this area may have
been done on an IBM 709.  Programs written for the IBM 709 were
supposedly upward compatible with both the IBM 7090
and IBM 7094, and there appears to have been times when the the presence
of these machines overlapped at Bell Labs.  It's possible a 709 and 7090
were both in use in 1961 and both
could have been used in developing the Daisy Bell tapes. it seems likely
that none of Bell Labs computer hardware would have been present for the
1961 Philadelphia  demo.  Instead a recording
of the analog sound generated at Bell Labs would have been played.
Those who saw the demo might not have been clear about what hardware had
been used.

Reporters in those days were pretty ignorant about computers and could
easily  have reported the facts incorrectly, or might never have
mentioned a specific computer model.  It's possible
that those reading about Bell Labs using an IBM 7094 for this research
after 1962 may have just erroneously inferred the 1961 work must have
used the same hardware.

So the correct answer must be 7090, or possibly 709 and 7090 jointly
share the credit of being first; but not the 7094.

JC Ewing

On 8/29/25 11:21 AM, Rupert Reynolds wrote:
 > My memory says 7094, although it was before my time and I can't suggest a
 > source.
 >
 > That said,
 > https://youtu.be/41U78QP8nBk?si=57nEornU6G5VjX7R says 7094 in 1961
 >
 > You've reminded me of The Happy Hacker, which was later and had no
words I
 > think. Bell Labs, maybe?
 >
 > Roops
 > ---
 > "Mundus sine Caesaribus"
 >
 > On Fri, 29 Aug 2025, 14:52 Seymour J Metz, <[email protected]> wrote:
 >
 >> There are conflicting claims as to the first computer to play Daisy
Bell.
 >> Some sources claim 704, with a dater later than the first 7090. Other
 >> sources claim 7094, with a date earlier than the official first customer
 >> ship. Is it possible that BTL had a pre-announcement 7094? Does
anyody have
 >> a definitive reference that wiki could cite?
 >>
 >>
 >> --
 >> Shmuel (Seymour J.) Metz
 >> http://mason.gmu.edu/~smetz3
 >> עַם יִשְׂרָאֵל חַי
 >> נֵ֣צַח יִשְׂרָאֵ֔ל לֹ֥א יְשַׁקֵּ֖ר
 >>

...

--
Joel C Ewing

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN



----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to