pete@octopus.UUCP (Pete Holzmann) (02/21/86)
A copy of an internal memo... a proposed Press Release for March 3, 1986:
Long considered a holdout in the PC compatibility ratrace, Cray has finally
joined the pack: The firm has announced its SuperPC I, an MS-DOS portable
that packs many of the features of the reknowned Cray supercomputers.
This computer, however, is not for all: It's exteremely fast -- using GaAs
ICs, tunnel diodes, and Josephson-junction circuits, it has a clock rate
of 2.2 GHz. It costs $4,000,000.00 (US). For many PC users, however, the
cost might be justified. For a Lotus 1-2-3 application that takes 2.3
seconds for recalculation on an AT, for example, the Cray requires
only 7.1 nano-seconds. This time savings, of course, allows the operator
to devote 2.3 seconds (otherwise wasted) to useful pursuits.
The Cray's performance is even more dramatic in other applications. For
programs smaller than 1-2-3, according to the company, the Cray can perform
its operation even before the diskette is inserted. Such power, alas, does
not come without a penalty. Although about the size of a Compaq portable
PC, the Cray portable requires a tractor-trailer support truck for the
liquid helium, the 1 MWatt power generator, and the fire extinguisher; all
are required for normal operation.
(Plagiarized from an internal memo distributed at an IBM subsidiary).
(YES... this is :-) of course!)
--
OOO __| ___ Peter Holzmann, Octopus Enterprises
OOOOOOO___/ _______ USPS: 19611 La Mar Court, Cupertino, CA 95014
OOOOO \___/ UUCP: {hplabs!hpdsd,pyramid}!octopus!pete
___| \_____ Phone: 408/996-7746nather@utastro.UUCP (Ed Nather) (02/22/86)
In article <196@octopus.UUCP>, pete@octopus.UUCP (Pete Holzmann) writes: > > A copy of an internal memo... a proposed Press Release for March 3, 1986: > > Long considered a holdout in the PC compatibility ratrace, Cray has finally > joined the pack: The firm has announced its SuperPC I, an MS-DOS portable > that packs many of the features of the reknowned Cray supercomputers. > > This computer, however, is not for all: It's exteremely fast -- using GaAs > ICs, tunnel diodes, and Josephson-junction circuits, it has a clock rate > of 2.2 GHz. It costs $4,000,000.00 (US). Although about the size of a > Compaq portable > PC, the Cray portable requires a tractor-trailer support truck for the > liquid helium, the 1 MWatt power generator, and the fire extinguisher; all > are required for normal operation. > > (YES... this is :-) of course!) I understand from a local source that Motorola was approached to put the CPU on a chip, but refused unless they could include super-speed graphics sprites and other hardware color graphic assists as well. They also demanded it be called the Cray-ola. -- Ed Nather Astronomy Dept, U of Texas @ Austin {allegra,ihnp4}!{noao,ut-sally}!utastro!nather nather@astro.UTEXAS.EDU
jablow@brahms.BERKELEY.EDU (Eric Robert Jablow) (02/24/86)
In article <196@octopus.UUCP> pete@octopus.UUCP (Pete Holzmann) writes: > >A copy of an internal memo... a proposed Press Release for March 3, 1986: > > >This computer, however, is not for all: It's exteremely fast -- using GaAs >ICs, tunnel diodes, and Josephson-junction circuits, it has a clock rate >of 2.2 GHz. It costs $4,000,000.00 (US). For many PC users, however, the >cost might be justified. For a Lotus 1-2-3 application that takes 2.3 >seconds for recalculation on an AT, for example, the Cray requires >only 7.1 nano-seconds. This time savings, of course, allows the operator >to devote 2.3 seconds (otherwise wasted) to useful pursuits. > >The Cray's performance is even more dramatic in other applications. For >programs smaller than 1-2-3, according to the company, the Cray can perform >its operation even before the diskette is inserted. Such power, alas, does --------------------------------------------------- >not come without a penalty. Although about the size of a Compaq portable >PC, the Cray portable requires a tractor-trailer support truck for the >liquid helium, the 1 MWatt power generator, and the fire extinguisher; all >are required for normal operation. > Actually, it uses thiotimoline-based organic superconducting circuits. Respectfully, Eric Robert Jablow MSRI ucbvax!brahms!jablow I may be a screwy little wabbit, but at least I'm not going to Alcatraz! --E. Fudd--
tuba@ur-tut.UUCP (Jon Krueger) (02/26/86)
Eric Robert Jablow writes: >>This computer, however, is not for all: It's exteremely fast -- using GaAs >>ICs, tunnel diodes, and Josephson-junction circuits... >> >Actually, it uses thiotimoline-based organic superconducting circuits. Making humidity control in the machine room a critical factor in maintaining reliability of operation!
kludge@gitpyr.UUCP (03/05/86)
In article <399@ur-tut.UUCP> tuba@ur-tut.UUCP (Jon Krueger) writes: >Eric Robert Jablow writes: >>>This computer, however, is not for all: It's exteremely fast -- using GaAs >>>ICs, tunnel diodes, and Josephson-junction circuits... >>> >>Actually, it uses thiotimoline-based organic superconducting circuits. > >Making humidity control in the machine room a critical factor in >maintaining reliability of operation! Please note that that use of thiotimoline in switching circuits to produce negative response times is a trade secret of Nonexistant Semiconductors. It is obvious that you have leaked information from the highly secret NoSe Surreal-Time Computation research group. Any production of equipment that could possibly be considered as infringing upon our forthcoming patents will be referred to our lawyers, Drinker, Beerman, Stagger, and Swig. We will sue retroactively. For the past -15+3i years we have been working on these devices, we have seen switching times decrease enormously to the point where our accounts payable will often bill customers before equipment has been ordered. This is progress. ------- Disclaimer: Everything I say is probably a trademark of someone. But don't worry, I probably don't know what I'm talking about. Scott Dorsey Kaptain_kludge ICS Programming Lab, Rich 110, Georgia Institute of Technology, Atlanta, Georgia 30332 ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!kludge USnail: Box 36681, Atlanta GA. 30332
ins_aaaw@jhunix.UUCP (Adlai A. Waksman) (03/07/86)
> >>Actually, it uses thiotimoline-based organic superconducting circuits. > > > >Making humidity control in the machine room a critical factor in > >maintaining reliability of operation! > > Please note that that use of thiotimoline in switching circuits to > produce negative response times is a trade secret of Nonexistant > Semiconductors. It is obvious that you have leaked information from > the highly secret NoSe Surreal-Time Computation research group. . . . Not obvious at all. The endochronic properties of thiotimoline are well documented in the biochemical literature*, and many researchers have been independently working with negative-response-time computation. This is enabling breakthroughs in complexity theory and computability, making possible algorithms and solutions that were hitherto impossible because they cannot be conventionally done in a finite amount of time. * "The Endochronic Properties of Resublimated Thiotimoline", Asimov (widely printed) -- ----------------------------------------------------------------------------- "It is a highly directional ultrasonic beam of pure rock and roll! It KILLS!" ----------------------------------------------------------------------------- Adlai Waksman Mathematical Sciences, The Johns Hopkins University, Baltimore, MD 21218 UUCP: seismo!umcp-cs \ AT&Tnet: (301) 889-8498 ihnp4!whuxcc > !jhunix!ins_aaaw Bitnet: INS_AAAW@JHUNIX allegra!hopkins / INS_AAAW@JHUVMS Arpa: ins_aaaw%jhunix.BITNET@wiscvm.WISC.EDU
ins_ampm@jhunix.UUCP (Michael P McKenna) (03/09/86)
In article <2126@jhunix.UUCP> ins_aaaw@jhunix.UUCP (Adlai A. Waksman) writes: >> >>Actually, it uses thiotimoline-based organic superconducting circuits. >> > >> >Making humidity control in the machine room a critical factor in >> >maintaining reliability of operation! >> >> Please note that that use of thiotimoline in switching circuits to >> produce negative response times is a trade secret of Nonexistant >> Semiconductors. It is obvious that you have leaked information from >> the highly secret NoSe Surreal-Time Computation research group. > . . . > >Not obvious at all. The endochronic properties of thiotimoline are >well documented in the biochemical literature*, and many researchers >have been independently working with negative-response-time computation. >This is enabling breakthroughs in complexity theory and computability, >making possible algorithms and solutions that were hitherto impossible >because they cannot be conventionally done in a finite amount of time. > > >* "The Endochronic Properties of Resublimated Thiotimoline", Asimov > (widely printed) ***** BLATANT PRODUCT ANNOUNCEMENT ***** SOFTWARE FOR YOUR NEW CRAY PC There is one major problem with the Cray PC. It is far too fast. Research has shown that having the computer present answers before the questions are asked causes intense psychological damage to the user, therefore it is important for most tasks to slow your Cray down to human usable speeds. Software from SlowWare Inc. can help you solve this problem. Our first package, available second quarter 1985 (not a typo, the software appeared on our disks, along with a README file explaining that our 1987 selves had written it), is the SLow lANGuage (SLANG) developement system. SLANG is a revolutionary new programming language that enables you to completely negate the Cray PC's incredibly speed. First, SLANG takes your highly structured code and preprocesses it into S-code (Spaghetti code). On an average program SLANG will increase the size of your program by 117.42% from goto statements alone! This has the added advantage of making it virtually impossible for anyone to reverse engineer your product, EVEN IF THE S-CODE IS READABLE! Other effects of SLANG are to move assignment statements inside loops and introduce several redundant statements. For added slowness the S-code is then interpreted rather than being compiled into machine language. And of course we use the most inefficient methods known to man or machine. The degree of inefficiency introduced is user controllable (within certain limits of course), and is specified by Minimum Acceptable Delay (MAD). If you need extra slowdown you can purchase our VLSS (Very Large Scale Slowness) option. This is an advanced program that scans your code, analyzes the algorithms used and attempts to replace them with less efficient algorithms, many times completely replacing them with a "brute force" approach. VLSS also looks for bottleneck areas in your code, and narrows them even further. SOON TO HAVE BEEN AVAILABLE Network Software form SlowWare Inc. - Slow down your Cray PC enough to communicate with other computers!!! Dwight S. Wilson
ins_akaa@jhunix.UUCP (Ken Arromdee) (03/09/86)
In article <2134@jhunix.UUCP> ins_ampm@jhunix.ARPA (Michael P McKenna) writes: >>> >>Actually, it uses thiotimoline-based organic superconducting circuits. > ***** BLATANT PRODUCT ANNOUNCEMENT ***** This is a violation of all standards of the net, to allow such a blatant advertisement! This has been a public service announcement from Multiple Universe Distributors (MUD). MUD has no need for using illegal substances such as thiotimoline in its computer equipment. We are a conglomeration of software developers in googols of parallel universes. Although travel to parallel universes has been considered far beyond current scientific knowledge, our MUD specialists have found a way. Simply create a set of random ASCII symbols of sufficient length. Anything MUST happen in at least one of the multiversal infinitude. In particular, in at least one parallel universe the random ASCII symbols would just happen to spell out correct, detailed instructions for the con- struction of a machine to send objects across universes. That parallel universe can then send the plans to all the other parallel universes. Furthermore, once this has been done, the same procedure can be used to produce a computer program (with documentation) to do anything desired; just create random letters and the universe that happens to get the result can send it to all the others. This also allows software written to order. Some of our current software products for the Cray PC are: Un-copier: Makes any program uncopyable by Nu-copier or any other copy programs. Nu-copier: Copies any program regardless of any other copy protection (such as Un-copier) on it. IF: Infinite loop finder. Detects programs that will fall into infinite loops when run on specific sets of data. Has been tested with copies of IF modified to go into an infinite loop after printing out the "No loop as currently run" message. BASICBASIC: A subset of Basic for those who have difficulty learning the full language. Currently supports only "PRINT" and "GOTO" (including computed GOTO) which together with our optional "LET" statement interpreter, allows ANY computational task to be performed. Zork XLII: A new adventure game that doesn't just simulate interacting with the real world, it actually does it. Examine and affect anything going on in the world. (Note: MUD's offices are protected by an Improbability Field against invasion by users of this product.) Users must be in the FBI, CIA, NSA, or have top-level government clearance. Coming soon: a new version that will work in space too. Anti-compiler: Un-understandable programs can be made understandable again. Can be used to transform assembly language programs into higher level languages, and can transform any high level language into an even higher level one. Especially good for "impossible to reverse engineer" programs from our competitors. (Disclaimer: manuals are not provided for the infinitely higher levels of languages output. They may be obtained at the low fee of $99.95 each, times the number in a set for a full set of language manuals. This is a low fee if you can afford a Cray PC at the prices charged for one.) Some of our hardware add-ons are: Erase-only Memory: Up to 355 terabytes. Cray PC Unlimiter of Time Limits (CPU Time Limit): With this your Cray PC really CAN do an infinite loop in three seconds. Reality configurators: Changes reality by Infinite Improbability methods. * 4 Dimension Joystick: This device, containing a patented non-thiotimoline integrated circuit, can be used in cases when the computer needs its input before its output. The only joystick with a negative response time! Tardis control: Now your Cray PC can be used to control your Tardis! Dimen- sional stabilizers, dematerialization circuits, chameleon circuits can now be directly interfaced with our control panels and your Cray PC. Other interfaces are available through our subsidiary Tardis repairs, Inc. (established 1989) Works with all Tardises made after the Type 39. * Because philosophers have never agreed on what is real, MUD unfortunately cannot guarantee that anything specific is real and thus can be affected by this attachment. -- Kenneth Arromdee | | BITNET: G46I4701 at JHUVM, INS_AKAA at JHUVMS -|------|- CSNET: ins_akaa@jhunix.CSNET -|------|- ARPA: ins_akaa%jhunix@hopkins.ARPA -|------|- UUCP: {allegra!hopkins, seismo!umcp-cs, ihnp4!whuxcc} -|------|- !jhunix!ins_akaa | |
tuba@ur-tut.UUCP (Jon Krueger) (03/12/86)
In article <2141@jhunix.UUCP> ins_akaa@jhunix.ARPA (Ken Arromdee) writes: >We are a conglomeration of software developers in >googols of parallel universes. Although travel to parallel universes has >been considered far beyond current scientific knowledge, our MUD specialists >have found a way. Simply create a set of random ASCII symbols of sufficient >length. Anything MUST happen in at least one of the multiversal infinitude. > . . . Ken, you didn't warn your customers about a limitation of your product: universe incompatibility. For instance, the Blue universe is nearly incapable of communicating with the other universes. Do you sell universe migration tools, or code generation for target universes? If I write my code well, what are its chances of porting to another universe? To a Blue universe? -- Jon Krueger UUCP: ...seismo!rochester!ur-tut!tuba USMAIL: University of Rochester ^^^<<<<dual universe! Pyramid 90x. BITNET: TUBA@UORDBV Taylor Hall Phone: (716) 275-2811 work, 473-4124 home Rochester, NY 14627
henry@utzoo.UUCP (Henry Spencer) (03/13/86)
> There is one major problem with the Cray PC. It is far too fast. > ...Software from SlowWare Inc. can help you solve this problem. Alternatively, you can buy our new RetardaBox product and run your old software unchanged (well, almost). Just add the RetardaBox to your Cray PC (some small additions to the cooling and power-supply systems will be needed) and you'll find it slowing right down to more human speeds. The RetardaBox contains 5 gigabytes of ROM holding the latest 4.7BSD kernel, and also the 56 terabytes of additional RAM needed to run it. This miracle of modern software engineering supplies all the slowness you could ever want! Since only 37% of the system interface has changed since 4.6.2.7BSD, only trivial reprogramming will be needed. Order today! Free with the first 500 orders: a three-volume leather-bound boxed set of the ls(1) manual page!!! (Coming soon: Doug Gwyn's latest compatibility package, faithfully emulating System V Release 573, featuring full compliance with the System V Interface Encyclopedia! Requires AT&T license. Comes with the latest AT&T-approved distributions of the Concatenator's Workbench, the File Renaming Workbench, and the Directory Listing Workbench. Optional libraries available for complete emulation of the latest System V interprocess- communication facilities; please specify the particular facility needed by giving the room number in Columbus, Ohio of its original user(s). No troublesome "cleanup" or "simplification" or "unification" has been done!) For those who need IBM PC compatibility, there is an IBM Emulation switch which divides the clock rate by a factor of five (an additional programmable clock-rate divider controlled by a random-noise generator is available as part of the TSO Emulation Option), adds warm, friendly blur and fuzz to the Cray PC's chillingly sharp and clear display, and reprograms the keyboard encoder so that mundane touch typing becomes an exciting adventure! -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry
ins_akaa@jhunix.UUCP (Ken Arromdee) (03/16/86)
In article <47@ur-tut.UUCP> tuba@ur-tut.UUCP (Jon Krueger) writes: >>We are a conglomeration of software developers in >>googols of parallel universes. Although travel to parallel universes has >>been considered far beyond current scientific knowledge, our MUD specialists >>have found a way. Simply create a set of random ASCII symbols of sufficient >>length. Anything MUST happen in at least one of the multiversal infinitude. >> . . . >Ken, you didn't warn your customers about a limitation of your product: >universe incompatibility. For instance, the Blue universe is nearly >incapable of communicating with the other universes. Do you sell universe >migration tools, or code generation for target universes? If I write my >code well, what are its chances of porting to another universe? To a Blue >universe? You misunderstand. This is only the method used to create our programs (and to create the universe-travelling machine in the first place). Code once produced can work in any parallel universe whose computers are of the appropriate type. Of course, our software may have unintended features in some alternate universes (i.e. those not almost identical to this one; by definition one with incompatible computers is alternate, not parallel. Note: we have not yet found any serial universes). For instance, on Earth-3, the BIOS call on the IBM PC that reads a disk sector instead, due to the different design of the computers, means "format regardless of disk protection." No lawsuits against us have yet succeeded, mainly because we're in a different universe, and there are no extradition or other legal-related treaties between theirs and ours. Furthermore we have encountered another minor problem that should be fixed real soon now. Although random ASCII symbols are guaranteed to produce correct programs in at least one universe, they are also guaranteed to produce every different possible incorrect program in at least one of the infinite universes. Of course, testing the programs will manage to weed out many of these, but we still have some programs which are correct except for a line which says to format the hard disk after the 576th run of the program. These small problems have sometimes caused slight incon- vebiences for our customers, but all in all our customers have been very supportive. Regrettably, however, our company must take a temporary halt in delivering its line of fine software for the Cray PC. As you may recall, our parallel-universe travelling machine was also produced by random ASCII symbols in a parallel universe. Devices produced in such a manner will, at rare intervals, malfunction as the programs do. In theory, the random ASCII instructions may even tell how to build a device that, instead of travelling to parallel universes, destroys them all instantly. There are an infinite number of parallel universes, and anything possible, no matter how improbable, must happen in at least one of them. Some groups, may, of course, be somewhat upset at our building of a machine from random character instructions without any real guarantee that it would actually travel across universes and not do something else on (say) its 167874883'd use, but we have decided that if we don't do it, an alternate universe version of our company will, somewhere. The reason we are making this heretofore proprietary infor- mation available to the public is that our machine has stopped sending to other universes and instead has created a minor disturbance in the space-time continuum. The disturbance is totally under control, and has in fact been growing only about 900 miles and 6 universes per day. So we look forward to your continued patronage, for the short time we all have left. > -- Jon Krueger -- Kenneth Arromdee | | BITNET: G46I4701 at JHUVM, INS_AKAA at JHUVMS -|------|- CSNET: ins_akaa@jhunix.CSNET -|------|- ARPA: ins_akaa%jhunix@hopkins.ARPA -|------|- UUCP: {allegra!hopkins, seismo!umcp-cs, ihnp4!whuxcc} -|------|- !jhunix!ins_akaa | |
kludge@gitpyr.UUCP (Scott Dorsey) (03/17/86)
In article <2126@jhunix.UUCP> ins_aaaw@jhunix.UUCP (Adlai A. Waksman) writes: >> >>Actually, it uses thiotimoline-based organic superconducting circuits. >> > >> >Making humidity control in the machine room a critical factor in >> >maintaining reliability of operation! >> >> Please note that that use of thiotimoline in switching circuits to >> produce negative response times is a trade secret of Nonexistant >> Semiconductors. It is obvious that you have leaked information from >> the highly secret NoSe Surreal-Time Computation research group. > . . . > >Not obvious at all. The endochronic properties of thiotimoline are >well documented in the biochemical literature*, and many researchers >have been independently working with negative-response-time computation. >This is enabling breakthroughs in complexity theory and computability, >making possible algorithms and solutions that were hitherto impossible >because they cannot be conventionally done in a finite amount of time. The paper to which you refer was published in 1938, I believe, and it must be stated that NoSe has held the patent on thiotimoline since almost four years before it was discovered, and the patent on thiotimoline switching devices almost five years before. Thus, since in fact I will not have written this article by the time you read it, legal action will have already been instated against all those discussing this, and my lawyers inform me that a precident exists in the case of Reimann vs. Temporal Processing, inc., 1994. -- ------- Disclaimer: Everything I say is probably a trademark of someone. But don't worry, I probably don't know what I'm talking about. Scott Dorsey " If value corrupts kaptain_kludge then absolute value corrupts absolutely" ICS Programming Lab (Where old terminals go to die), Rich 110, Georgia Institute of Technology, Box 36681, Atlanta, Georgia 30332 ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!kludge