Post Reply 
Old RPN w/ discrete components and acoustic memory
03-04-2020, 01:13 PM
Post: #1
Old RPN w/ discrete components and acoustic memory
Maybe it’s been shown before but really cool video.

https://youtu.be/2BIx2x-Q2fE
Find all posts by this user
Quote this message in a reply
03-05-2020, 05:41 AM
Post: #2
RE: Old RPN w/ discrete components and acoustic memory
(03-04-2020 01:13 PM)ndzied1 Wrote:  Maybe it’s been shown before but really cool video.

https://youtu.be/2BIx2x-Q2fE

Amazing. Great find! Thoroughly enjoyed it.

Jaco Mostert, Elec Eng
C47 on DM42, 42S, WP34C&S, 28C, 35S, 32Sii, had 11C; used 67, 85; iOS:42s, Free42, WP31S&34S, HCalc; OSX:WP34C.
Find all posts by this user
Quote this message in a reply
03-05-2020, 05:44 AM
Post: #3
RE: Old RPN w/ discrete components and acoustic memory
(03-04-2020 01:13 PM)ndzied1 Wrote:  Maybe it’s been shown before but really cool video.

https://youtu.be/2BIx2x-Q2fE

I still don't really get acoustic memory. How does it store the information for more than a few milliseconds as the pulse goes around the loop to the end?

Tom L
Cui bono?
Find all posts by this user
Quote this message in a reply
03-05-2020, 06:12 AM
Post: #4
RE: Old RPN w/ discrete components and acoustic memory
(03-05-2020 05:44 AM)toml_12953 Wrote:  
(03-04-2020 01:13 PM)ndzied1 Wrote:  Maybe it’s been shown before but really cool video.

https://youtu.be/2BIx2x-Q2fE

I still don't really get acoustic memory. How does it store the information for more than a few milliseconds as the pulse goes around the loop to the end?

The way I understand it, is that it gets sent around again once picked up by the sensor on the end. This video attempts to show both ends:

https://youtu.be/N9cUbYII5RY

Jaco Mostert, Elec Eng
C47 on DM42, 42S, WP34C&S, 28C, 35S, 32Sii, had 11C; used 67, 85; iOS:42s, Free42, WP31S&34S, HCalc; OSX:WP34C.
Find all posts by this user
Quote this message in a reply
03-06-2020, 01:50 AM
Post: #5
RE: Old RPN w/ discrete components and acoustic memory
If you've ever read a description of how the magnetic bubble memory operated, the design is similar in that the contents of the loop need to be regenerated continuously. At the point (in time) there a specific data bit falls off the end is where you read the value or write a new value.
Find all posts by this user
Quote this message in a reply
03-06-2020, 02:04 PM
Post: #6
RE: Old RPN w/ discrete components and acoustic memory
(03-06-2020 01:50 AM)cruff Wrote:  If you've ever read a description of how the magnetic bubble memory operated, the design is similar in that the contents of the loop need to be regenerated continuously. At the point (in time) there a specific data bit falls off the end is where you read the value or write a new value.

So that's what the fan is for? To blow all of those bits that fell off the end out of the machine? Big Grin

Tom L
Cui bono?
Find all posts by this user
Quote this message in a reply
03-09-2020, 01:04 PM
Post: #7
RE: Old RPN w/ discrete components and acoustic memory
The funny thing is that the program memory in the HP 65 is very conceptually similar, except that it's a very long shift register rather than a mechanical/acoustic device. Technically, the 65 doesn't have "RAM" in that regard, since it isn't "random access", and there are no absolute addresses. It just reads/inserts/swaps bits as they pass by the window. (I imagine there's a bit more to it to allow the calculator to do things like inserting and swapping steps for moving the program pointers around, but that's the general idea.)
Visit this user's website Find all posts by this user
Quote this message in a reply
03-09-2020, 02:41 PM
Post: #8
RE: Old RPN w/ discrete components and acoustic memory
(03-05-2020 05:44 AM)toml_12953 Wrote:  
(03-04-2020 01:13 PM)ndzied1 Wrote:  Maybe it’s been shown before but really cool video.

https://youtu.be/2BIx2x-Q2fE

I still don't really get acoustic memory. How does it store the information for more than a few milliseconds as the pulse goes around the loop to the end?

Many years ago I used to service a proof machine that had delay line memory and that machine had a solid state shift register in between the two ends of the delay line that you would insert and retrieve data from, but since it is a continuous stream of bits that you cannot stop, you need to know when a unit of data is aligned in the solid state register and insert or retrieve data before the shift register advances.

I also took a call on a test scoring machine one time that used delay line memory to hold the correct answers that where compared to the students answer sheets.

Paul.
Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 1 Guest(s)