Talk:DECtape
This is the talk page for discussing improvements to the DECtape article. This is not a forum for general discussion of the article's subject. |
Article policies
|
Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL |
![]() | This article is rated C-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||
|
Patent invalidated?
[edit]I'm skeptical of the claim that DEC's US patent 3,387,293 on DECtape was invalidated.
The Patent was invalidated. A company in Maryland I believe, whose name escapes me, began making a clone and DEC sued. It was discovered DEC had provided several machines for beta testing, which was commonly done. However, they made an arrangement to sell the machine(s) to one of the beta customers. The receipt for the sale was provided as part of the discovery phase of the suit to the company that had cloned the DECtape machines. DEC had not filed for patent protection until after the beta testing was complete. The sale before the patent application invalidated the patent.
The patent covers aspects of DECtape that are nontrivially distinct from LINCtape. For instance, LINCtape was designed for bidirectional search, but DECtape was designed for bidirectional data transfer as well. This required a substantial change to the way the mark track was encoded.
Google doesn't seem to have any other references to this patent being invalidated. Does anyone have a citation to the court case, or at least know who the plaintiff was? If not, I think it should be removed from this page and the LINC page. --Brouhaha 00:44, 19 Oct 2004 (UTC)
- No citation for the patent being invalidated has been forthcoming in almost two years, so I'm removing that claim. I also plan to reword the paragraph comparing LINCtape and DECtape, as it is far too POV at the moment. There were legitimate reasons for some of the major differences between the two, and the patent covers those differences. --Brouhaha 23:33, 25 August 2006 (UTC)
- Apparently the PTO struck a reissue application by DEC on a finding of fraud in the application, and a district court upheld that. The appeals court ruling (653 F.2d 701) vacated the district court ruling and remanded with instructions for the district court to remand to the PTO. I haven't yet found anything regarding the eventual disposition. --Brouhaha (talk) 18:29, 17 April 2011 (UTC)
TU 555 Microtape?
[edit]I found some "TU 555 Microtape"s in the basement that seem to contain a very early version of Tops. I'm not quite sure how these fit into the history, but I assume that they are early DECtapes made before DEC started to private label the tapes. —Preceding unsigned comment added by 66.189.44.116 (talk) 11:58, 18 March 2011 (UTC)
- Microtape was just an earlier name for DECtape. The 555 (dual) and 556 (single) transports (NOT TU555!) were the predecessor of the TU55 and TU56 transports. The controllers were 550 (PDP-1/4/7), 551 (PDP-6), and 552 (PDP-5/8). These early controllers were not software-compatible with the later TC01/TC08 (PDP-8), TD8-E (PDP-8), TC02 (PDP-9), and TD10 (PDP-10).
- Since DECtape (Microtape) has its origins in LINCtape, non-DEC media existed early on. 3M sold it under their own name, and DEC-branded tape was made by 3M. --Brouhaha (talk) 00:32, 19 March 2011 (UTC)
DECtape on PDP-1 and PDP-10
[edit]I'm putting back the reference to DECtapes on PDP-10s. The KA10 system that I grew up on definitely did have DECtapes. Also, the PIP10 utility on OS/8 was specifically designed to allow PDP-8s to read and write PDP-10 DECtapes.
Atlant 18:22, 8 Dec 2004 (UTC)
- Well, but not all PDP-10 models came with DECtapes. Alas, my decsystem10 Reference Manual doesn't explicitly say if all KA-10/KI-10 configurations included DECtapes as a standard, so I don't know about that. (According to the [configuration file], only two of the three ITS KA's at MIT had DECtapes, for what that's worth.) The KL10-B models (TOPS-20) definitely did not include DECtapes. (KL-10A's had them, but they were attached to the console PDP-11 - not sure if any OS's for the KL10-A allowed access to them from the -10.)
- The other thing is that according to the same book, PDP-1's, -4's and -7's had them too. In other words, they were probably an available I/O option for pretty much all PDP's, so to be technically accurate we'd have to list them all, which would look silly.
- In leaving only the 6, 8, and 11 in the list I had tried to list the older, smaller machines for which they would have been a primary mass-storage device - with their small (~300KB) size, once larger disks arrived, they wouldn't even have been useful for backup.
- I would be OK with an edited list which didn't list all the machines that supported them, but just the ones for which they were a primary mass-storage device. Otherwise let's just link to PDP and be done with it... Noel (talk) 15:17, 9 Dec 2004 (UTC)
I've never seen a DECtape on a PDP-1, but if you say they existed, I believe you; this link certainly supports your contention:
http://groups-beta.google.com/group/alt.folklore.computers/browse_thread/thread/ebf63b298ba975b6/1d820bc0c5328407?q=PDP-1+dectape&_done=%2Fgroups%3Fq%3DPDP-1+dectape%26start%3D10%26hl%3Den%26lr%3D%26&_doneTitle=Back+to+Search&&d#1d820bc0c5328407
as does another newsgroup posting from Bob Supnik, who'd certainly know.
On the other hand, *EVERY* photo of a KA10 and every KA10 that I've seen first-hand had DECtapes, right up above and to the right the system console (not necessarily TU56 DECtapes; sometimes the older, relay-controlled TU55 DECtapes). It's a shame I don't have a copy of the Dick Best Options/Modules list any more. :) The PDP-1 apparently had the ancient 555 'tapes.
In any case, I'd be in agreement with omitting the list of all supported platforms and just referring people out to a PDP page and, if people felt it necessary, the LINC page (although LINC is linked from the PDP page).
Atlant 17:19, 9 Dec 2004 (UTC)
The KI10 diagnostics were supplied on DECtape; I suspect that DEC would not offer service on a KI10 without DECtape, just as they later required that KL10 systems have an RP06 to be on service. Early KL10 diagnostics were also on DECtape, but they were supplanted by the KLAD pack.
DECtape was not available until well after the introduction of the PDP-1, but was made available as an option for it. The PDP-1 being restored at the Computer History Museum includes an option to interface to an external Type 550 DECtape control.
--Brouhaha 22:06, 9 Dec 2004 (UTC)
A DECtape manual I have seen includes in the back a compabitily matrix. This lists every PDP family member and indicates what other PDP system's DECtape formats it can handle. So, for example, if you wanted to read a PDP-11 DECtape on a PDP-4, it would tell you whether that is possible (and how). Among other things, this implies that all PDP-n computers had DECtape capability. Paul Koning 20:38, 9 April 2007 (UTC)
- No DECtape for PDP-14 and PDP-16, as far as I can tell. (Some models of PDP-14 had a PDP-8 as a front end, and obviously you could attach a DECtape to the -8.) --Brouhaha (talk) 01:05, 19 April 2011 (UTC)
Tape capacity?
[edit]The blocksize is 129 12-bit words in PDP-8 format, and 256 18-bit words otherwise. The TU55 book on Bitsavers [1] says the tape capacity (260 feet of tape) is 4096 blocks in PDP-8 format, which is 6.34 megabits. The number of blocks in the other format (PDP10/11/15 format) is 578, which translates to 2.66 megabits. This doesn't make sense. I'm positive about the 10/11/15 capacity, but the other comes from a DEC manual... did the length of the tape change? Paul Koning 20:56, 9 April 2007 (UTC)
- More: the TU56 maintenance manual [2] says that the unformatted tape capacity is 2.7 megabits, density is 350 bpi, and length is 260 feet. So the PDP10/11/15 number seems right, which makes me wonder where that other number comes from. Paul Koning 21:02, 9 April 2007 (UTC)
- Ok, the PDP-8 handbook [3] says 1474 blocks, which translates to 2.28 megabits. That fits the other numbers, and it matches the "184k words" mentioned at the top of the article. So the TU55 overview must be in error. Maybe the "4096 blocks" is meant to show the max addressable block number -- 12 bit block numbers -- which could be achieved with a custom format that uses a much smaller blocksize. Paul Koning 21:28, 9 April 2007 (UTC)
It's useful to point out that KL-10 processors came with DECtape drives, but these were connected the PDP-11 front end. They were PDP-11 style DECtapes. The only use I know for them is initial software installation. Once the SW was installed, the front end PDP-11 booted off the system disk. One DECsystem10 model the 1091, came with 8 inch floppies on the front end instead of DECtapes. Same for KL-20 based DECsystem20s. Joe Thursday (talk) 12:56, 1 February 2009 (UTC)
DECtape 2
[edit]I wonder if it's worth describing the lack of relationship between DECtape and "DECtape II". Some of us have long believed that the name "DECtape II" was a marketing ploy to freeload off the good reputation of the original. While DECtape II is block structured, it offers none of the key features of the real thing, in particular that of reliability... It also doesn't come with a reasonable interface, and it was only supported on PDP-11 and VAX. On some at least (RSTS for example) support was as minimal as possible and was dropped as soon as possible. Paul Koning 01:04, 10 April 2007 (UTC)
- I've always assumed that it was mostly the block-addressability that earned it the name, but I also think that the Powers That Be hoped it would become a ubiquitous diagnostic medium much as DECtape had been ubiquitous. That's why we in the hardware world built it into essentially all of the VAX and PDP systems that were released around that time (11/750, 11/730, 11/44, and 11/24), but its obvious performance limitations (especially when compared to the even-cheaper floppy disks, even RX50 floppy disks) doomed it. So by the time of MicroVAX, and MicroPDP, floppies were the universal medium.
I don't know where to begin on this article and the discussion page since both are citing a lot of just plain wrong stuff, partial attempts to refute bad info with other inaccurate or incomplete info. Some is authoritative while some is merely well-intended but muddies the waters.
I'll just add a factoid to clear some other factoids, but I have a lot of raw facts that could go a long way to help retrieve this from chaos.
The original drives were done by an outside vendor for MIT Lincoln Labs for the original LINC. DEC purchased some themselves for the LINC-8. I know even less LINCs were made, but the highest serial number for LINC-8 is #153.
The drive logic uses relays to implement the motor controls, and they are in fact the exact same relay as is used in the LINC and LINC-8 to implement the relays involved with the LINC CPU instruction AC to Relays. Contacts are brought out as DPDT 5-way binding posts for the contacts in that usage, but they are also to be plugged into the LINCtape drives. In fact there is an old software diagnostic to check the contact bounce on the contacts so you can adjust them to work properly as they age and "stick" in use [and can be repaired].
This became the basis of the micro-tape as the old boxes called it, DECtape in the 550 seris controllers almost described properly for the 12-bit machines. The 555 and 552 indeed wind up on a PDP-5, but is not for the PDP-8, since it was completely redesigned. The PDP-5 implementation was worthless for larger memory configurations because there was no separate register for the high 3 bits of the 15-bit address space. They just kludged it to the Data field during the DMA which is preposterous because if interrupts are enabled, when you get an interrupt the DF is zeroed out. Never meant to be, but a bad kludge to try to make something semi-useful.
The PDP-8 got the TC01, a complete redo, 3-cycle data break and requires TU55 and later TU56 using the negative tape buss. The TU55 was known as the "solid-state DECtape" because it COULD be used with logic levels not associated with the grosser relay logic. However, it COULD also be configured to be backward compatible with the older relay logic controllers. [Note: The negative tape buss is also used in the PDP-12 for use with LINCtapes. That the bulk of the machine is TTL is irrelevant to this point despite what an ignoramus might assume.] The TU56 can also be configured as a positive tape buss for some applications such as PDP-11 and PDP-15 as opposed to TC02 of the PDP-9 which is negative tape buss. [The PDP-9 buss itself is negative, and vaguely comparable to the negative PDP-8 buss with bit width variations.] The TC01 got it right about those extra three address bits [and a good thing too!]. Also, the TU55 and TU56 use direct drive motors, not belt-driven as in the earlier drives. I have found a source for a replacment for surgical rubber LINC drives to get them to function today. The TU56 can also be configured without a buss, as in the TD8E configuration with the manchester reader/writer cards populating the drives, so this is not quite like any other usage, but clearly closer to the PDP-11 usage. There was to be but never marketed a PDP-8/e peripheral presumably to be called the TC8E that would have been positive ala the -11 application.
The differences between the formats may or may not be a patent issue, but in any case, Computer Operations Inc. of Beltsville, MD sold a drive that was physically more like the LINC-8 drives [or an odd one-drive layout] and they called it LINCtape, but in fact the intent was DECtape depending on what controller you hooked to it. What I heard at the time was DEC attempted to get it suppressed on the usual patent/copyright issues, but that DEC lost because it had sold to eager customers some early units before the patent application so DEC's objection was overturned. Computer Operations LINCtape was definiely successful, so something like that must have happened. It definitely got hooked to PDP-11-compatible controllers if nothing else in terms of DECtape format. However, an obscure application was to a PDP-8 positive buss peripheral that mimicked just enough of the LINC-8 tape interface that it was software compatible with the LINC-8 original even if the CPU was for example a PDP-8/L. I only have the name of the developer, Kurt Metzger of Cooley Labs in Michigan. Clearly the drive doesn't actually determine the tape format, that's the controller's job. In fact, a standard option of the PDP-12 allows software to run to read and write [but NOT format] DECtapes on the PDP-12. [Note: Metzger is one of the authors of custom modifications to the LINC-8 to accomplish the same thing on that machine. I used it in conjuntion with other hardware mods I helped design and the relevant software to take advantage of it; the net result is a superset of both Metzger's original stuff and the stuff I participated in to make it more suitable for running certain software systems on the LINC-8 such as OS/8 [and a proprietary system copyright myself which I would detail here].
OS/8 does NOT take advantage of a second tape drive per se, but a clever user can arrange for certain files to be on the other tape to save time depending on specifics of the moment. In point of fact, all swapping is ALWAYS on the first tape for all operations such as binary loading or system function overlay swapping, etc. To make overall improvement, files should be placed on other drive[s]. If you have a third drive, this allows still more functions in less time depending. [For example, you could run an editor on the system device, swapping and all, and an input file on a second tape and an output file on a third tape. Further, OS/8 has a restriction that you cannot open more than one file for output on any device, thus sophisticated operations might need to have yet another tape for some auxiliary writing such as splitting up source files, etc. Of course disk users had far less to care about in this and similar regard, etc.]
Just a smattering of some factoids. Straightening out this mess is not something I really want to do myself, but I would be glad to proof-read any attempts at reordering stuff. Contact me at [email protected] for any followup.
cjl
External links modified
[edit]Hello fellow Wikipedians,
I have just modified one external link on DECtape. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
- Added archive https://web.archive.org/web/20100807150030/http://www.bitsavers.org/pdf/dec/dectape/EK-0TU58-TM-001_TU58tech.pdf to http://bitsavers.org/pdf/dec/dectape/EK-0TU58-TM-001_TU58tech.pdf
When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}
).
This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}}
(last update: 5 June 2024).
- If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
- If you found an error with any archives or the URLs themselves, you can fix them with this tool.
Cheers.—InternetArchiveBot (Report bug) 17:18, 4 December 2016 (UTC)
Read/write speed?
[edit]There are many references in the article to the format being reasonably quick (certainly compared to paper tape, anyway) and usable like a slow disc drive in practice. But actually how fast was it? There's no mention anywhere. Presumably it may take a few seconds to seek through the whole tape looking for a particular block out of the ~512 (for 18-bit records/blocks) total (using simpler, broader marks on one of the tracks that could be counted off at high speed), and then another half second or so to read that actual block? (=512 words/sec raw read rate, a little less sustained when reading blocks sequentially without any further seeking, merely waiting out the small gaps between them, and rather less for full random access). That would be a significant improvement over the ~60cps (or ~20wps) of paper tape, but still grant an worthwhile advantage to a more costly drum memory or extremely expensive early hard disc system (a few dozen kwords/sec)... However, that's just an educated guess based off how tape drives appear to behave in footage of their use (rapid, jerky movements when performing actual reads/writes with maybe a couple of spurts per second, and very rapid seeking back/forth in between any sequential R/W), and their market position between the better known speeds of those other devices, so does anyone have actual information? 146.199.0.175 (talk) 15:37, 20 March 2018 (UTC)
DECtape SIZE is not mentioned by "inches"
[edit]I saw the sizes in "inches" information for the incompatible LINC tapes, but not for this tape. Did it come in 10" (10 inches) and/or other sizes? Misty MH (talk) 19:18, 1 December 2022 (UTC)