Timeline for Why were chips socketed in early computers?
Current License: CC BY-SA 4.0
21 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 9, 2018 at 15:41 | answer | added | David Lovering | timeline score: 2 | |
| Sep 9, 2018 at 11:42 | comment | added | David Tonhofer | The good old Trash-80. "1980: US$399 (equivalent to $1,185 in 2017)". Oh yeah, the joys of money printing. | |
| Sep 9, 2018 at 11:42 | review | Suggested edits | |||
| Sep 9, 2018 at 12:06 | |||||
| Sep 9, 2018 at 11:34 | comment | added | David Tonhofer | @MichaelKay You know you are getting old when "early computers" means the machines in use when you were using lego. Integrated Circuits were used in the 60s, but let's consider the PDP-11/03 as a cutoff. Wikipedia says: "The LSI-11 (PDP-11/03), introduced in February 1975 is the first PDP-11 model produced using large-scale integration; the entire CPU is contained on four LSI chips made by Western Digital". There is a nice photo of the four socketed chips, with soldered support logic. | |
| Sep 7, 2018 at 13:48 | comment | added | TextGeek | I upgraded an original Mac's memory, and it took almost 8 hours to remove the old soldered RAM (4-layer board, 16 16-pin DIPs). You can bet I socketed the new stuff.... | |
| Sep 6, 2018 at 19:51 | comment | added | Michael Kay | Early computers did not use chips. | |
| Sep 6, 2018 at 18:30 | answer | added | rgoers | timeline score: -3 | |
| Sep 6, 2018 at 15:20 | comment | added | JoL | I wonder if in the future all car parts will be welded together, and complete car replacement when any piece breaks will be the norm, and then we'll wonder, why did we join the pieces with nuts and bolts before? | |
| Sep 6, 2018 at 10:51 | comment | added | Tobia Tesan | You know how people rage on internet forums when Apple or HP ships a product with soldered RAM? In the '70s soldered anything would have elicited the same reaction. Only they didn't have forums, so they would have picked up actual torches and pitchforks. | |
| Sep 6, 2018 at 2:09 | answer | added | hotpaw2 | timeline score: 12 | |
| Sep 6, 2018 at 0:30 | answer | added | Artelius | timeline score: 6 | |
| Sep 5, 2018 at 19:51 | answer | added | cmm | timeline score: 20 | |
| Sep 5, 2018 at 18:34 | answer | added | StessenJ | timeline score: 8 | |
| Sep 5, 2018 at 17:49 | vote | accept | rwallace | ||
| Sep 5, 2018 at 17:16 | comment | added | nbloqs | This was the general common practice with DIP (and other non SMD) IC packages. It allowed for quick fixes of the boards without using desoldering stations, etc. And the manufacturing techniques were different too. | |
| Sep 5, 2018 at 16:57 | answer | added | Raffzahn | timeline score: 67 | |
| Sep 5, 2018 at 15:50 | comment | added | tofro | Customers of very early computers expected sockets for maintainability (at least in the "nerd market". | |
| Sep 5, 2018 at 14:53 | history | edited | fadden | CC BY-SA 4.0 | Clarify title |
| Sep 5, 2018 at 14:33 | answer | added | Stephen Kitt | timeline score: 9 | |
| Sep 5, 2018 at 14:30 | comment | added | manassehkatz-Moving 2 Codidact | I'll let someone who has more extensive experience give an official answer. But I suspect that it has to do with complexity. Your most complex chips - and the CPU is generally at the top of the list - are most likely to (a) have problems requiring replacement (e.g., design bugs found after manufacture, which has happened a few times to Intel) and (b) are most important to keep protected from any damage during assembly - wave soldering being pretty safe but why risk the most important chip when you can put in a socket and insert the chip after the soldering is all done. | |
| Sep 5, 2018 at 14:20 | history | asked | rwallace | CC BY-SA 4.0 |