Post Reply 
SR-52 was a giant leap. How?
06-17-2017, 12:12 AM (This post was last modified: 06-17-2017 12:13 AM by Matt Agajanian.)
Post: #1
SR-52 was a giant leap. How?
Hi all.

I've wondered about the monumental leaps from the 50A & 51A to the 52/56.

Although they were pricey (perhaps moderately affordable) for their day, what technology and cost issues made it possible to add

over eight parentheses levels and hierarchy stack when the 51 and 50 only had Sum of Products hierarchy?

10 (SR-56) and over 20 (SR-56) data registers,

100 and 224 step program memories,

Single variable statistics for the SR-56,

Well, you know the remaining features sets,

But, what made it possible and how did TI get so far ahead of functionality improvements between the slide rule and the programmable slide rule?

Thanks
Find all posts by this user
Quote this message in a reply
06-17-2017, 12:20 AM
Post: #2
RE: SR-52 was a giant leap. How?
And yes, I could pose the same questions between the HP-65/HP-25/HP-55 and HP-67/29C.
Find all posts by this user
Quote this message in a reply
06-17-2017, 12:21 PM
Post: #3
RE: SR-52 was a giant leap. How?
I think for those closely related Texas model families, the difference was down to a couple of kilo-bits of user RAM space in the form of the TMC0599 device. The slightly older slide-rule models just had no breathing space for more sophisticated functionality. The tiniest increase in RAM/register space completely transformed what was possible.
Find all posts by this user
Quote this message in a reply
06-17-2017, 05:10 PM
Post: #4
RE: SR-52 was a giant leap. How?
This helps clarify the challenge. The way I thought about the transition was by way of economising code. For example, not literal:

In lieu of hyperbolics and inverses, that space of code was used for conditional tests and branching.

Instead of SR-51 US/Metric conversion codes, pending level, parentheses code,

Instead of SR-51 statistics/linear regression, 224 program steps 20 (79 extended) data registers.

In other words, the technologies were there, but it was a matter of coding and quad chips.

Right? Or were there challenges I've not considered?
Find all posts by this user
Quote this message in a reply
06-18-2017, 01:49 AM (This post was last modified: 06-19-2017 03:39 PM by toml_12953.)
Post: #5
RE: SR-52 was a giant leap. How?
(06-17-2017 12:12 AM)Matt Agajanian Wrote:  Hi all.

I've wondered about the monumental leaps from the 50A & 51A to the 52/56.

Although they were pricey (perhaps moderately affordable) for their day, what technology and cost issues made it possible to add

over eight parentheses levels and hierarchy stack when the 51 and 50 only had Sum of Products hierarchy?

10 (SR-56) and over 20 (SR-56) data registers,

100 and 224 step program memories,

Single variable statistics for the SR-56,

Well, you know the remaining features sets,

But, what made it possible and how did TI get so far ahead of functionality improvements between the slide rule and the programmable slide rule?

Thanks

Competition between HP and TI really spurred both companies on to develop products far quicker and with more features than if either was the only company making calculators. Read some of the ads from both companies printed in the mid to late '70's. Many touted the advantages of AOS over RPN (in TI ads) and vice-versa (in HP ads) They were at each others throats and the consumer benefitted. TI claims to have been the inventor of the integrated circuit and had their own chip foundry so they definitely had the technology at their disposal. Market share and ultimately profit drove the companies to hire the best engineers and software people.

Long live competition!

Tom L
Cui bono?
Find all posts by this user
Quote this message in a reply
06-18-2017, 10:23 PM
Post: #6
RE: SR-52 was a giant leap. How?
Yeah. Thanks to Viktor Toth & Joerg's sites as well as numerous memorable visits to Computique, Freeway Stores, Olympic Sales, and Tams, I have most of those brochures.

I recall the illustration on the back pages of the charts in black, yellow, and orange comparing a keystroke approach to this example-- 1+3x[4+5/(7-2/9)]. As an inadequate representation, TI just filled the RPN stack with the first four variables of the sample equation.
Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 2 Guest(s)