test-drb

test-drb@ccmp.vtda.org

101090 discussions

Start a nNew thread

R: Re: Mac "Workgroup Server" (or "network server") hardware & AIX

by supervinx＠libero.it

http://www.supervinx.com/OnlineMuseum/Apple/Quadra/950/ Scroll down and you'll find many A/UX pictures

10 years

Fast Unibus Sync Serial?

by seefriek＠gmail.com

From: Ethan Dicks <ethan.dicks at gmail.com> > > DUP-11? Will that do what you need? > Not sure, Ethan. I'm been looking at the doco and it's not clear yet if it's suitable for what I'm trying to do. Thanks for the pointer. KJ

10 years

Mini Map Array Processor - What in tarnation?

by wdonzelli＠gmail.com

Yes, the CSPI box is mounted inside a standard DEC cabinet. I have one of these things in a VAX, originally part of some sort of chemical analysis tool. -- Will On Apr 26, 2016 9:44 PM, "Jon Elson" <elson at pico-systems.com> wrote: On 04/26/2016 07:59 PM, Swift Griggs wrote: > Seriously, > > http://tinyurl.com/j46jg4p > > http://www.computerhistory.org/collections/catalog/102743645 > > What in the world is this thing? Some kind of early parallel or vector > processor? > We got a CSPI 6410 and had it hooked to a VAX 11/780. It was probably a bigger brother to the machine you link to. There was a big math library that came with it. If you had really regular matrix operations, such as FFTs, matrix multiplies and similar classic operations, it could do them quite fast, in the several MFLOP range. The bigger the matrix (as long as it fit in the memory of the unit) the better, as the library just set up all the registers, loaded the data and turned it loose. If you had a bunch of small matrices, it was a lot less efficient, as it had the same setup overhead for every task. Yes, it is a vector processor, with a floating-point multiplier and adder, some address arithmetic logic and a sequencer. Jon

10 years

Mini Map Array Processor - What in tarnation?

by swiftgriggs＠gmail.com

Seriously, http://tinyurl.com/j46jg4p http://www.computerhistory.org/collections/catalog/102743645 What in the world is this thing? Some kind of early parallel or vector processor? External? How delightfully weird, except there is very little online about it. Did any of you work with one of these? What kind of instructions do they use? What is their bus width on the ...uh what? CPUs? What kind of addressing modes & size(s) do these use? Why does the one reference talk about the 64k RAM they had? Were these standalone machines ('cause it looks like it goes to a VAX-11 and "LSI-11") ? This was before my time but fascinating, nonetheless. -Swift

10 years

Using flashable Macintosh ROM SIMM

by dave＠661.org

I have a "MACSIMM" from GGLabs (http://gglabs.us) which I bought from their Ebay store. Unfortunately they haven't followed up on my request for information on how to prepare a ROM image for use with the thing. Has anyone else here used this or a similar device to hack the ROMs of the Mac SE/30, IIsi, IIci, or IIfx? It's composed of four 39SF010A 32-pin PLCC chips the special 64-pin SIMM. -- David Griffith dave at 661.org A: Because it fouls the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail?

10 years

Avionics and amazing gear made by Tatjana (was Re: Data General Nova Star Trek Rockwell Collins vs. Vaisala SPT11A)

by wdonzelli＠gmail.com

There were a few others, as well as some RF devices with the tuned parts inside the bulb. There were also some oddball types made for weather balloon use that had the whole transmitter circuit as one unit. -- Will On Apr 26, 2016 6:30 PM, "Chuck Guzis" <cclist at sydex.com> wrote: On 04/26/2016 12:40 PM, Paul Koning wrote: > This sort of stuff doesn't seem to be all that common; I haven't seen > it elsewhere. Multiple tubes, like dual triodes or triode/heptodes > are pretty common, but those are just the active part. The only "passive in the tube" examples I can think of in US manufacture are simple between-unit resistors, such as the 6N6. --Chuck

10 years

PAL on CDU-700 / 710 Unibus disk controllers

by ian.finder＠gmail.com

Does anyone out there have one of these controllers? There is an unprotected PAL- the address decoder- and I really need a dump of it. Many programmers, particularly the BP technologies ones, can read it

10 years

PAL on CDU-700 / 710 Unibus disk controllers

by glen.slick＠gmail.com

On Apr 26, 2016 3:04 PM, "Ian Finder" <ian.finder at gmail.com> wrote: > > The spec sheet from the PAL mfg. (signetics PLS173) seems to indicate > security fuses aren't an option on the model on my board. I have not > actually tried yet. > Regardless of what the spec sheet does or does not explicitly say, I think when I selected that device in the BPWin device programmer software there was an option to protect the device after programming. I'm mobile now, I can check that later.

10 years

strangest systems I've sent email from

by blstuart＠bellsouth.net

On Mon, 4/25/16, Swift Griggs <swiftgriggs at gmail.com> wrote: >The idea is somewhat that if students learn a > "good" language that'll teach them some meta-structure that will help them > later. Certainly a lot of people do view it that way, but it's not what I was getting at or how I see it.? Based on my experience, the virtues of any single language are pretty much irrelevant.? What's vitally important is that the student emerge with a deep understanding of how a variety of languages actually work, how they're processed and how the computer executes code written in them.? If you have that deep understanding with a sampling of languages that represent some of the variety of techniques and paradigms (for lack of a better term), then you'll be able to pick up and adapt to most any language that comes along.? To tell you the truth, I'm not very likely to hire anyone who isn't conversant with at least half a dozen different languages.? To summarize, my focus isn't on the skills of any particular lanugage; it's on the understanding of the fundamental concepts, principles, techniques, and mechanisms that make up the world of computing. >Then, let me say that the *idea* that I was attacking Pascal via > Oberon rather than the Ivory Tower Academics is ridiculous. I did understand the point in your first message to be anit-"Ivory Tower Academics."? However, my point it is that viewing the people you have identified as such and dismissing their experience and expertise is a narrow-minded and short-sighted perspective. > Your point might be > logically valid, but ask a 23 year old if they care when they can't get a > job after giving the uni a quarter million bucks and 4-5 years of time > they spent being "educated" rather than "trained". It's interesting that you pick that age as the example.? My daughter is 23.? For her, the undergraduate experience wasn't about a job at all.? It was about exploring the intellectual world and (to borrow >from Thoreau) sucking the marrow out of that life.? In the interest of full disclosure, however, I should point out that she's not typical of most college students (although I wish more were like her). She did grow up in a household that averages more than two degrees per person and she did triple major in her four undergrad years. (A proud daddy can't help but brag a little. :) ) > The underlying point I? was making is that schools don't always > "train" a person ... I apologize if I misinterpret, but I also detect the suggestion that they are supposed to.? I don't disagree that they don't train, but I do disagree that's what their purpose is. I'm not suggesting that some degree of training coming along with the education is a bad thing. However, I'm saying that's not the primary purpose of the university. >? and that's what I wanted and actually needed. There seems to be an implication here of an XOR when I look for an AND. In particular, if I have a candidate sitting across the interview desk from me, I'm not interested unless they have both education and training. I expect the education to come >from a formal environment where people of long experience can help the student understand many perspectives. I expect the training to come from self-directed experience. Unless a candidate shows both the ability to work in a rigorous intellectual manner and the self motivation to go beyond what they've been given, I'm not interested. > all but one of the profs had > turned off their brains in 1986 and it was the 90's. It's certainly true that does happen both in academics and in industry. However, more often than not, the ideas that were seen as "new" in the '80s, '90s, '00s, and '10s, are really ideas that the Computer Science community saw, studied, understood, etc in the '50s, '60s, and '70s. So what appears to be out of touch is often really a broader perspective and one worth understanding and learning from. > True. I wonder though, do you believe that teaching a > language with almost zero commercial value is justified > in the name of education because of it's superior "meta" > qualities ? I'm not sure I can answer the question as you've posed it. As I said, I tend to consider the choice of any single language to be mostly irrelevant. I'm much more interested in the neural pathways that the student builds as a result of the experience of coming to understand a large set of languages. As it turns out, I am currently involved with a restructuring of the introductory programming sequence at one university. Our choice of languages was driven by both pedagogical and vocational considerations. Were our environment different such that we should have looked at only one or the other, then we would have chosen differently. Regardless of what we did pick, we never intended for the freshman languages to be the only languages our students knew before graduation. No one or two languages will give the breadth and depth needed pedagogically. Neither will any one or two languages suffice for building a career as a computer scientist. > Grades aren't meaningless if you have a grant/scholarship > to maintain or need to get into graduate school. There are expected minimums, certainly. Based on my experiences in both academics and industry, I would have my doubts about whether someone is really cut out for a CS major if they can't average Bs in their major classes. On the other hand, I have yet to be in a hiring meeting where one candidate was chosen over another because they had a 3.5 vs the 3.3 the other candidate had. For that matter, I always question candidates that come in front of me who have 4.0 averages. Before I can take them seriously, I need to see some evidence that learning new things is a higher priority to them than the grade. It goes back to the expectation that the candidate show evidence that they can pick things up on their own as well as in the classroom. > They have plenty > of meaning to employers like IBM who might not hire you for > that entry-level position with a low GPA. Actually, it's better to consider companies with unenlightened cutoffs to be simply advertising that they are not where you want to be. > They might not reflect some aspect of education or learning > you think is important, but there is more to going to uni than just > to get a mind expanding education. The purpose of the university is the discovery, dissemination, preservation, and interpretation of knowledge. A necessary, but not sufficient, part of being well-prepared for a professional career is the exposure to that breadth and depth of knowledge. I would rephrase your statement to say that there's more to preparing for a career than just getting a university degree. > I think folks who haven't been to school in 20+ years have a totally > warped view of what is happening nowadays. I might point out that on this mailing list, you will find people who are all across the spectrum with exposure to education. Some of us were students many years ago and haven't had much contact since. Some of us have been students recently. Some of us have siginificant faculty experience. As with the experience of changes in the CS field, I'd suggest that the longer the period of time over which one has experience with education the less warped the view of what's happening. However, each individual must decide for themselves whose opinions and experiences earn their respect. > All our code was in C or C++ and we found that none of the students > from the local universities had those skills. Although there are plenty who disagree with me, I argue that the purpose of the university experience isn't the skills; it's the understanding. I always say I want to hire the preson who I can tell on Friday that we'll be using Intercal for the project and they'll spend the weekend teaching themselves the language and come in Monday morning ready to code in Intercal. I (as well as entrepreneurs I know) have very little interest in how much experience and skill a candidate has in a particular language. The reality is no matter what their experience is coming in, I'm going to have to teach them how to do things the way I want them done. > I could only attribute that to the memorize-and-regurgitate > culture of the schools, but maybe it was just bad luck.? It's a combination of bad luck and the ill effects of misplaced priorities filtering up from the primary and secondary school levels. As I tell my students, there is a continuum of understanding. At one end, there is the mere repeat back to me what I've said to you. If that's all a student can do, then there's not much value in it. I can simply put the same information into a computer and then not bother paying a person for it. At the other end of the continuum is the ability to ask the right next question, the understanding of how our current knowledge came about, and the understanding of how to advance that knowledge. My objective in every class I teach is to help each student move as far along that continuum as they are capable. The farther along that continuum you move, the more you will be able to discover what has never been known before and create what has never existed before. Many would classify that perspective as "Ivory Tower." They might say, I can't put food on the table with that attitude. I would differ with that. I have put food on the table for several decades, and in only a few of those years were the funds coming primarily >from the academy. Indeed I have realized that my level of understanding and my drive to explore, investigate, and create are of significant value to many. Those who recognize that are willing to exchange monatary renumeration for the contributions I can make. Those who don't recognize it self-select themselves out of the pool and I don't have to worry about them. Throughout this, it has not been my intention to in any way dismiss your perspective or to suggest that it is a "wrong" perspective. Indeed, it's a perspective I'm quite familiar with. Instead, my objective has been to suggest there's another perspective whose consideration might lead to deeper understanding. It is a perspective which attempts to temper the immediacy of the question of tomorrow's employment with the longer-term view of how that employment fits into the thousands of years of human civilization. BLS

10 years

SATA SSDs for SGIs using ACARD adapters

by swiftgriggs＠gmail.com

I use Samsung 850 Pro disks exclusively. The main one I tested with was 128GB. 80-pin SCA hot swap Ultra160 SCSI to SATA: This one I've tested in my O2 and Tezro: http://www.acard.com/english/fb01-product.jsp?idno_no=240&prod_no=ARS-2160H… 50-pin ultra-scsi to SATA: This one I've tested in two Indys (R4600 200Mhz and R5k 180Mhz) also a Challenge S with an R4600 200Mhz CPU): http://www.acard.com/english/fb01-product.jsp?idno_no=249&prod_no=ARS-2000S… There is also a 68-pin variant. I've yet to try one because I will probably get rid of most of my Sun gear at some point in the future. However, I'd be interested to know if they work well or not. I still play with older SPARC boxes from time to time. I've upgraded many of my SGI's to use SSD disks. You don't see tremendous improvements in throughput (though it does help some). My results testing with 'fio' showed about 10% throughput increase on most operations overall. However, the latency becomes much lower. This gives the machine a much more snappy feel if you use them as desktops. Web browsing is especially ameliorated (as in, you can almost do it, har har). On my Tezro the results were pretty dramatic. The quad R16k CPUs still keep up pretty well with my "modern" PCs for most operations (including browsing). Using internal SSDs makes the anecdotal experience tangibly improved. I'm curious if anyone has used one of the 50-pin units to upgrade an Amiga 3000 or older mac. I'm thinking of other machines like one of the slimmer and more interesting AlphaStations might also be fun to try (or a newer Multia with the SCSI adapter). Did Atari ST's have SCSI? Without looking it up, I'm guessing so. That'd be another fun one to test. Also before someone pipes up saying "The older SCSI interface is going to limit you too much to feel the difference" let me just say that's NOT been my experience. As I mentioned, you might not get much throughput increase (though it will usually improve some), but the numbers don't lie on the latency which drops outta sight into the sub-1ms range most of the time. You feel that on desktops, bigtime. Also, if you say "that's not stock! It's not original anymore!" Please let me just respond now and also say "Yes, I know that, thank you." Thanks, Swift

10 years

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

test-drb