This tradition lasted through the early PDP-8 and PDP-11 models, although the PDP-11 CPU light displays were modest compared to earlier CPUs. The earliest PDP-11 disk controllers were very much a match for the earlier machines, though, and included indicator panels which displayed a great deal of information.
Alas, the cost of such displays was high, and they were dropped from later controllers. Their passing was mourned by those who had had the opportunity to work with them, though; in addition to helping with debugging (both hardware and software issues), and giving an extensive insight into how the machine was operating, they were also just a lot of fun. Today, 'das blinkenlitz' (as the are often referred to now) are a focus of the computer heritage movement.
This page documents many of the DEC indicator panels, so that when one is seen, it can be determined what kind of indicator panel it was (the 'inlay' which had the captions for the meaning of the individual lights on it usually did not say what kind of device it was for).
The PDP-8 and PDP-11, although not 36-bit machines, also used the same indicator panel, but in general did not include any 36-bit fields. (The exception is the RP11-C, which was able to read packs written on a PDP-10, and thus included a 36-bit shift register, which was displayed on the RP11-C panel.) The 18-bit PDP-15 also used this indicator panel.
Limited engineering drawings (they do not include the full mechanical drawings) for this indicator panel are included in the RF11 Engineering Drawings set; the indicator panel is shown on pp. 186-190. Of particular interest are the parts lists, which include all the inlays extant as of the date of the drawings (10/1972).
According to those prints, the list of available inlays at that point (other than the ones illustrated below) included:
The tables below list the indicator panels available for each machine type, and include links to larger images of the panel (where we have images - anyone who has an image of one of the missing panel is urged to provide it), and (for some), detailed information about the panels.
|RK08||Like the TC08, only two rows of lights, out of the four available, were used.|
|DX11||The DX11 is an IBM 360/370 Channel interface.|
|RF11||RF11 details||Easily recognized in low-resolution images with the two groups of three single lights on the left-hand side.|
|RP11||RP11 details||Two variants are known for this, known as "RP11" and "RP11-C" in the DEC documentation; they differ only in that one contains two lights (ORDY and IRDY, on the right hand of the second row down) which are not in the other.|
|RK11-C||RK11-C details||The RK11-C engineering drawings show that it was wired to connect up to
an indicator panel, but as far as is known, none were ever actually produced.
None of the extant RK11-C documentation mentions such a panel, and the
indicator panel engineering drawings do not list an RK11-C inlay.
Using the engineering drawings, and the known pinout of the indicator panel, it is possible to determine what such an RK11-C panel would look like; a mockup can be seen here. (Although note that the individual light captions are somewhat hypothetical, since no such inlay was ever produced - although the signals that drive them, and hence their meaning, is known.)
|This panel does not go with a specific device, but includes information from a number of CPU-related systems; the lower rigft quadrant is mostly the KT15 memory management, the upper left is the KA15 priority interrupt system, but there is general CPU/system information throughout (memory parity, power fail, instruction register); the top right line seems to be the paper tape reader input buffer.|
|FP15||As the drawing of the indicator panel connector (pg. 73 of the prints) shows, this panel is full.|
|RF15||Identical to the RF09 indicator panel.|
|RP15||The RP15 had two indicator panels.|
Back to JNC's home page
© Copyright 2015-2016 by J. Noel Chiappa
Last updated: 18/August/2016