On 6/26/12 9:41 AM, David Brownlee wrote: You may be able to do something clever on the front end with a propeller. I haven't looked at what their performance is like. ST412 data rates are on the order of 10Mhz which isn't going to work with GPIO. Philip is using an FPGA on the Diskferret to capture MFM data (don't know how much time he's been able to spend on this lately).

On Tue, 26 Jun 2012, Dave McGuire wrote: Where exactly can I find those 20 or 30 designs? I've yet to see anything that comes close to what the R-Pi can do for the price they're at. The wait times should be pretty short right now - mine took 3 weeks to get here once I placed my order through RS (Allied). They'll decrease further as production rises to meet the demand. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_! Buying desktop hardware and installing a server OS doesn't make a server-class system any more than sitting in a puddle makes you a duck. [Cipher in a.s.r]

On Tue, 26 Jun 2012 07:13:16 -0700 (PDT) Gene Buckle <geneb at deltasoft.com> wrote: In addition to what Liam postet: Have a look at the FrendlyARMs: http://www.friendlyarm.net/ The Pi is cheap in the bad sense of this word. AFAIK e.g. the Ethernet hangs off USB. The connectors are spread around all four edges of the PCB. This makes it quite hard to put it into a proper enclosure. -- \end{Jochen} \ref{http://www.unixag-kl.fh-kl.de/~jkunz/}

On Wed, 27 Jun 2012 02:48:34 -0700 (PDT) Gene Buckle <geneb at deltasoft.com> wrote: If you need something more basic: GNUBLIN (Sorry, German) http://www.gnublin.org/ No Ethernet, no video, just USB host / device and an assortment of GPIO, SPI, I2C, RS232, ... in 70x70 mm? running Linux for 49,95 EUR. [Enclosure for Raspberry Pi] Doesn't solve the problem, that you need access to all sides of the box to connect this or that. You can't simply shove it into a shleve or the like and connect everything from the front or back. It would have been easy to put all connectors on one side. But this would require more PCB real estate and thus increase cost... -- \end{Jochen} \ref{http://www.unixag-kl.fh-kl.de/~jkunz/}

On 6/26/12 9:49 AM, Dave McGuire wrote: Read write are differential TTL levels, MFM or RLL encoded. Clock recovery on read generally uses a PLL. National app note AN-0413_Disk_Interface_Design_Guide_and_Users_Manual_Jan86 gives a decent overview of the different drive interfaces circa 1986. AN-0414_Precautions_for_Disk_Data_Separator_PLL_Designs_Feb86 AN-0581_Application_Issues_for_the_DP8465_Family_of_Data_Synchronizers_Jan89 AN-0494_Designing_with_the_DP8462.pdf AN-0500_Designing_an_ESDI_Disk_Controller_Subsystem_with_Nationals_DP8466A_Nov87.pdf AN-0415_Designing_with_the_DP8461_Apr85.pdf AN-0501_Interfacing_Nationals_DP8466A_to_the_SMD_Interface_Standard_Mar87.pdf are also of interest

On 26 jun. 2012, at 22:42, Richard <legalize at xmission.com> wrote: I'd like to know as well. Started searching for this but didn't find the '86 version, only newer at National's website. re, Sander

In article <4FEA31AD.9040304 at bitsavers.org>, Al Kossow <aek at bitsavers.org> writes: Looks like the most relevant application notes for this discussion are also included in the 1989 Mass Storge Handbook <http://bitsavers.org/pdf/national/_dataBooks/1989_Mass_Storage_Handbook/> -- "The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline> The Computer Graphics Museum <http://computergraphicsmuseum.org> The Terminals Wiki <http://terminals.classiccmp.org> Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

On Tuesday 26 June 2012, Al Kossow wrote: Make that: http://bitsavers.org/pdf/national/_appNotes/ ;-) Cheers, Lyle -- Lyle Bickley, AF6WS Bickley Consulting West Inc. http://bickleywest.com "Black holes are where God is dividing by zero"

On Tue, Jun 26, 2012 at 1:23 PM, Liam Proven <lproven at gmail.com> wrote: I have yet to put myself on such a list because I hate the waiting game. I do plan to play with them once I can actually *purchase* one without fiddling about. At $25 for no Ethernet or $35 with Ethernet, the price-point is so low that it doesn't take a lot of effort to talk oneself into it, nor does one feel bad about ordering one "just in case" a great idea comes along. When a ready-to-use computing element costs less than a good steak dinner, it's easy to buy one for frivolous reasons. At twice or three-times the cost, I think it's "expensive enough" to have to have a "good reason". I certainly hope to put one to use for digital signage. I see that most of Xibo has been ported to the Raspberry Pi, but it's not quite there yet. I'm still fiddling with 1.0+ GHz boxes with a gig of RAM to get the Ubuntu/Python Xibo client working for technology conventions I work at. I'd *love* to be able to velcro a RP to the back of an LCD and just walk into a hall, hang the LCD on a visible wall, then punch the power button and walk away. Not there yet, but the goal is in sight. Agreed, even here. I haven't gotten into the Beagleboards because they are above my "curiosity" threshold, but I've done a few projects with Arduino because I can roll a finished version on a perfboard or a $3 blank DIY Arduino clone PCB by adding a $3.25 MCU and a handful of cheap parts and by the time I'm done, I've spent less than $10 adding smarts to a project (though I do have one where I repurposed a spare Gen3 MakerBot extruder controller because it had three 12V MOSFETs with screw terminals - it made driving an RGB LED strip a matter of "screw down 4 wires and write 5 lines of code". The hardest part of the project was cutting an Ethernet cable in half to feed power to it via its 8p8c (RJ45) jack). I still plan to play with the Raspberry Pi, but it's IMO a better wee desktop than a physical computing platform precisely because it only has 3.3v current-limited I/O that require external components to do anything with. -ethan

As true as this is, there are far too many places on this planet with respect to which even today's USA is amazingly well off. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On Wed, Jun 27, 2012 at 6:50 AM, Dave McGuire <mcguire at neurotica.com> wrote: [...] I think there is a little bit of history repeating - and in part, it's a deliberate move by the creators. In particular, remember the ZX80/81; technically not very exciting; but cheap enough for anyone to get it. That made it influential. The same argument can be made for the Vic 20. Moreso in Europe than North America, space _is_ a serious consideration. And also the prices for plain old Wintel boxen I see here in France are quite a bit higher than I saw in Montreal - where at points I had trouble giving away P4's that the university was tossing out. Here in Rennes, those still seem to go for about 100 euros. (Judging from advertised prices on the local craigslist equivalent (leboncoin.fr). I haven't needed to sell anything nor any desire to buy such a machine, so I don't know if those people are simply deluding themselves.) Mind you, I have a more powerful machine than the Raspberry Pi in my pocket, with similar CPU, GPU and OS. However, I need that as a tool so I can't mess around with it too much - and it also cost significantly more. It is also less open than the R? (where only the GPU is somewhat closed). That's why I am waiting to get myself a R?. I can play with it and if I break it, it's an inconvenience. If I break my phone, I _have_ to replace it. If I want to mess with a R?, I get it out of a storage bin, do stuff, put it back. A fullsized pentium (or similar) box requires a permanent place on a bench. I do not have that kind of space at this moment. -- Joachim Thiemann :: http://jthiem.bitbucket.org

On 27/06/2012 07:47, Joachim Thiemann wrote: Yes but as its cheap you can treat it like a component. So if I want to build a small IBM1130 emulator I could use the VHDL model and FPGA but that runs at ?200. With the Pi I can afford two or three so I can a Develop, a Test even a Training one if needs be... -- Dave Wade G4UGM Illegitimi Non Carborundum

On 27 June 2012 05:50, Dave McGuire <mcguire at neurotica.com> wrote: It's the first at toy-level prices, perhaps. That is not uncommon in the UK - although oddly it seems slightly better than 2-3Y ago now. Not that this has done me any good at all personally. I can't say what things look like from an American POV. What I can say is 2 things: [1] It's broadly like that here, too, and the improvement, if any, since it all went pear-shaped in 2008 is marginal. [2] I wasn't saying "everyone in America is rich". What I was saying is somewhat different: that many things in the USA are much cheaper than they are in Europe and that this has been true for most of my life, AFAICT. Most USAnians are largely or completely unaware of this. (Additionally, USAnians seem to have more disposable income - something that may have changed in recent times.) Examples: http://www.newyorkstategasprices.com/ I am taking the price of a US gallon of petrol as about $3.50. That is ?2.25. A US gallon is 3.78 litres. That is 92? per litre or ?0.60 (60p) a litre. http://www.petrolprices.com/ ... I am taking the UK price to be about ?1.30 a litre, that is, $2.02. So in Britain, petrol - "gas" - is $7.63 a US gallon. If you think things in the USA are bad now, imagine what would happen if the price of gasoline went from $3.50 to over $7.50. (My estimate of 4-5x was way off; forgive me, I do not own a car, have not ridden a motorcycle in 5y now and have never used a motor vehicle outside the UK. I have not bought petrol in many years; it is a fairly meaningless metric to me personally.) Compare a pair of Levi's 501s on Amazon US: http://www.amazon.com/s/ref=nb_sb_noss_1?url=search-alias%3Daps&field-k… $43. And on Amazon UK: http://www.amazon.co.uk/s/ref=nb_sb_noss_1?url=search-alias%3Daps&field… ?39. ?39 is $60. 40% more in the UK. you pay for something, we will probably pay 40% to 100% more for it. It is not that the average USAnian makes more money or has more money. It's that for all that US$1 = just UK?0.60, that $1 will buy as much or more than that 60p would buy here. Since pay rates are about the same - citations: Average US, $47K http://www.averagesalarysurvey.com/article/average-salary-in-united-states/… Average UK, ?30K http://career-advice.monster.co.uk/salary-benefits/pay-salary-advice/uk-ave… ?30,000 is about ?47,000. So we make about the same, but yours will buy you about 40-100% more goods. *That* is a very big, real difference. For what it's worth, the UK is not a particularly expensive country by European standards. A cup of coffee in a caf? here might be ?1.50; in Norway or Sweden, ?4 or so. A pint of beer in London is about ?3-?3.50; in Norway or Sweden, ?6-?7. There are cheaper countries than here - a pint in the Czech Republic would be about 30-40p - but also ones where everything is twice as expensive. The average USAnian has a quality of life approaching that of the richest, most expensive countries in the world, but the cost of living is that of one of the cheaper countries in the developed West. /That/ is what we get irked about. /That/ is why people get jealous and bitter. Welcome to the C21. The same is happening across the developed world. Ah, but because of its fame, it's appealing to people who don't really know anything much about computers. The R? has a crappy, low-power CPU and not much RAM, but a good powerful GPU which is sadly proprietary, undocumented and with closed-source drivers. However, it's a good stab at a minimum spec which will run Linux with a GUI and a desktop. Yes, but [a] it's not cool or trendy or new, [b] it's big and ugly and boring, [c] it will require some expertise to refurbish and get working, [d] it will be easily damaged by knocking it over or blocking its vents. And it will cost a lot to run. Cost of electricity in the USA: 11? per kWh Cost of electricity in the UK: 22? per kWh Citation: http://en.wikipedia.org/wiki/Electricity_pricing#Global_electricity_price_c… As I said, assume /everything/ here is about twice as expensive. Kids seem to be finding the prospect of a tiny, cheap computer /just for them/ which they can play with and which it doesn't matter if they break appealing and enticing. Also, adults are using them for lots of fun projects: Robot planes (article by me): http://www.theregister.co.uk/2012/06/12/raspberry_pi_drone/ Robot boats: http://www.reghardware.com/2012/06/27/raspberry_pi_skippers_journey_across_… You could not do stuff like this with an old PC. It is a bit, but that is fashion for you. The world of trends and fashion is perhaps one that seldom impacts your average geek. ;?) -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 5 July 2012 13:12, Stefan Skoglund <stefan.skoglund at agj.net> wrote: I am afraid I have never left the capital. :?( But I am glad to hear that it is not as bad as I thought! Owwww! :?( -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 05/07/2012 19:14, Fred Cisin wrote: But it's never been true. A pint here, and anywhere pints are used outside the US, is 20 fluid ounces, and a pound is sixteen ounces... Ah. Well. :-) -- Pete Peter Turnbull Network Manager University of York

On Thu, 5 Jul 2012, Pete Turnbull wrote: I should point out, that I was taught that in school; taught that it was a "British saying", and that "the world around" referred to when "the sun never sets on the British Empire". Q: Does being taught WRONG information have a bad effect? Q: Would charging a pound for a pint (of either size) help the overall world situation? BTW, I just received my "international edition" of 2ed "Art of Electronics". The paper is thin, but the printing is good. It does not have the TITLE PAGE! That, and the obverse have information that I find interesting about the publication history, copyrights, etc. Are those "international editions" AUTHORIZED reprintings?

On 6 Jul 2012 at 2:20, Pete Turnbull wrote: I learned it as "A pint's a pound the world around" for the US and "A pint of water weighs a pound and a quarter" for the British version. --Chuck

On 6 Jul 2012 at 19:49, Tony Duell wrote: I'd be interested to hear of simple rhymes for converting stone to Kg... --Chuck

On Fri, 6 Jul 2012, Tony Duell wrote: Damn! All this time, I've been multiplying by 1.8 and adding 32; no wonder I've been getting it wrong! In elementary school (5th grade), the teacher the book said, PI is about 3.14 or 22/7. The teacher insisted that that meant that it was about 3.14, and exactly 22/7. I got into a LOT of trouble for arguing. -- Grumpy Ol' Fred cisin at xenosoft.com

On Fri, 6 Jul 2012, Chuck Guzis wrote: California. Kensington ("suburb" of Berkeley) The real question is where did the TEACHER go to school. Once people learn that kinda crap, they go on to push it on others. When I was teaching "computer math" (beginning, mostly teaching what floating point is, etc.) I had a couple of students who ALSO had been taught "EXACTLY 22/7" (from other schools)! -- Grumpy Ol' Fred cisin at xenosoft.com

One of my hobbies at school was disrupting lessons by asking nasty qeustions _on the subject of the lesson_. Point being had they kicked me out for doing that, my parents would have exploded with the school Seriously. I once walked out a physics lesson and said I would only return when the teacher had actually learnt some basic physics. That did not go down well. But, you know, I am still waiting for a defintion of a (scientific) measurement that does not involve comparison to a standard. -tony

On 6 Jul 2012 at 19:49, Tony Duell wrote: "A liter of water's a pint and three quarters." I found this interesting tidbit: "The legal standard English spirit measure is 25ml or 35ml. These are the metric conversions of the old measures of 1/6 gill or 1/4 gill (or 1/5 gill in Scotland). The gill is sometimes spelled jill. There is an explanation of the nursery rhyme: Jack and Jill went up the hill to fetch a pail of water. Jack fell down and broke his crown and Jill came tumbling after. When Charles I scaled down the 'jack' (a two-ounce measure) so as to collect higher sales taxes, the jill, by definition twice the size of the jack, was automatically reduced also and 'came tumbling after.' " By what measure does one buy firewood in the UK? Is it still by the cord? --Chuck

On 7 Jul 2012 at 21:20, Pete Turnbull wrote: It does sound a little too neat, doesn't it? But Charles I did have real revenue issues, so maybe there's something to it, although it seems more like something today's politicians would pull. But not all measures make sense. A long tine ago, an English friend referred to "a gallon of potatoes". That one left me scratching my head. --Chuck

potato salad from/for a deli comes in gallons also The other Bob -----Original Message----- From: Pete Turnbull Sent: Saturday, July 07, 2012 7:19 PM To: General Discussion: On-Topic and Off-Topic Posts Subject: Re: Raspberry Pi and America, On 07/07/2012 21:51, Chuck Guzis wrote: Perhaps they were mashed, really soft? ;-) -- Pete Peter Turnbull Network Manager University of York

On Sat, Jul 7, 2012 at 10:51 PM, Chuck Guzis <cclist at sydex.com> wrote: The classic Dutch measure for potatoes is called "mud", from the Latin "modius". The Latin modius was 8.75 liters, or 2.3 gallons. The Dutch mud was around 110 liters, or 29 gallons (it varied according to locality and over time), whereas the Flemish mud was around 640 liters (170 gallons). To measure potatoes, they used a crate that could hold one mud. Eventually, people realised that it made more sense to measure potatoes by weight, so a mud of potatoes became 70 kilograms of potatoes rather than 100 liters of potatoes. I bet the same thing happened in England. Camiel.

On 6 July 2012 19:49, Tony Duell <ard at p850ug1.demon.co.uk> wrote: How bizarre. We're about the same age, I think, and in the early 1970s I was taught both imperial and metric. I use a mix of both in my head - lengths in feet make more intuitive sense than metres, distances in miles and speeds in mph rather than km or kph, for instance, although I'm quite happy in both. I can't remember how many yards in a mile or ounces in a pound or pounds in a stone or any of that rubbish though, so I use metric for all those sorts of things unless it's a direct comparison - e.g. British male clothes are sized in inches so I just go with the flow. I have a 38" waist and a 36" inside leg because that's what all the shops sell. OTOH I'm 6'2" tall or 1.88m depending on who's asking - either is fine. But Fahrenheit never made any sense to me at all. One end of the scale is entirely arbitrary, the other is an entirely different measure and one that they measured wrongly anyway. It's seemed stupid and random and senseless since I was a small child and I've never ever used it for anything. Celsius is obvious and logical and rational. 0? is freezing, 100? is boiling, bosh, there you are. Fahrenheit is a scale where six apples equals one and a half times the colour of the A three octaves above middle C, so long as it's a Tuesday, and I have never ever been able to fathom (haha) it. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

As I still do. My lathe is calibrated in inches (and 'thou' == 1/1000"), but I use metric twist drills and metric fasteners. For accurate work like that, I use my HP calculator which has the very useful 'unit management'. Incidentally, it is a great annoyance to machinists everywhere that there are 2.5.4mm in an inch. Point being that 254 factorises as 2*127 (127 is prime). SO do do exact metric screwcutting on an imperial lathe or vice versa you need a 127 tooth changewheel. However, there is a little trick... 8000/(3*21) is very close to 127 (in fact 8001/(3*127) _is_ 127). Since 8000 factorises easily into small numbers, you can get way with a 21 tooth changewheel as the only unusual one (all the otehr factsrs are easy to get from normal changewheels) for non-critical work. It's OK for making fasteners, etc, but not for making a micrometer screw :-)). You really didn't want to know that, right :-) I can rememerb all fo them. Mainly becasue a lot of the numbers come up elsewhwere. Thejre are 1760 yeards in a mile. 1760 is a numebr that is etched into my brain. Any idea waht? Well, it's 16*110, and most UARTs (serial chips) need a clock that's 16 times rthe baud rate). So 1760Hz is the clock freqeucny for 110 baud, as needed for taling to ASR33s. But there si a more practical point here. Although miles and yards are both units of length/distance, it's very rare fro the average pserson to need to convert between them. You use yards (feet and inches too) to measure your room for carpet or wallpaper. You use miles to work out how far it is to go to your relaitves (or wahtever). You don't have to relate them. Similarly you use poudns and ounces in the kitchen, stones when figuring out if you've eaten too much from said kitchen, and tons when you have to lift the family car (or whatever). Again, you don't need to relate them. I was once anked on a job applciation form for my 'weight'. I haev no idea why they wanted ot know this, I can't see how it can have any relevance to my abilities as an electronic designer. But I gave it anyway -- in Newtons. If they ask a scientist for a weight, it's going to come in units of weight (or force). If they want mass, they should ask for it :-) It's not totally claear why fixed points were originally used. One version I heard went like this : He wanted all temperatures to be +ve, so he did indded make the coldest ice.salt mixture he could nad used that as a check (but not as the 0 of his scale) Melting ice was 32. Body temperature was 96. Point being there are then 64 degrees betwwen those points which is very easy to divide alogn a thermometer tube (dbisecting is a lot easier than dviding into tenths). Nto to me it isn't. Why pick water anyway? And waht is this insane love of multiples/powers of 10? I much prefer an absolute temperature scale. I normally use kelvin... -tony

On 11 Jul 2012 at 15:45, ben wrote: Here's something odd that I discovered. I needed some M8 flange nuts for my chainsaws (I'm always losing them). US sellers seem to ask about $4-5 each (without shipping) in plain galvanized. Plain M8 galvanized is easy to find here, but flange nuts seem to be a specialty item. The best price by far was from a UK seller, who offered 11 of the things in stainless for $6 shipped. They took less than a week to arrive. So, for some things, the UK has the best deal. I don't know why. --Chuck P.S. For what it's worth, here's the eBay seller, along with his address: Boltbase Iain Sim 6 Taran Alloa, Clackmannan FK101RF UK Why a small town on the Firth of Forth would have the best hardware prices is beyond my reckoning.

It's not the tax that bothers me, it's the 'inspection fee'.. Quite what this fee is supposed to be for is beyond me, many times i've been (correctly, as the value was over the limit) been charged tax on intems shipped ot me from the states aod of course I get this extra fee slapped on. But the parcel has not been opened, they take the value from the green sticker that the sender put on. Hmmm... -tony

I know the problem... That's a very expensive way to buy them. OFten 10 small nuts bought that way will cost more than a box of 1000.... I tend to buy my most-used hardware (M3, M4, etc) in such large boxes, I am going to use them... One thing I use the lathe for is making special threaded parts. I am not goign to start makign standard M3 nutes, but if I need a tapped spacer, or am odd-treaded bolt (M3 * 06.mm pitch, as I mentioend a while back) or somethign like that, I jsut turn it. Finding UNC and UNF parts over here, particularly the 'number' sizes is very difficult. You can sometimes find 4-40 and 6-32 sold for PC assmebling. Some model shops sell them, but at _ridiculous_ prices (like \pounds 2.50 for 4 nuts). I tend to buy them from Digikey. The postage is high, but at least I can get them I guess you haev similar problems finding BA nuts and bolts :-) -tony

'The sun never sets on tnhe British Empire becuase the sun sets in the west and the Birtish Empire is in the east' :-) I think it does. I object ot having to unlearn anything. And as I have said before it is possible ot over-simplify things. I could never understnad the simple books on how telephone exchanges work. Only when I got a copy of 'Telephony' and acutally went through the schematics in there, relay by rely, did it all make sense. Excellent. You have some fun reading ahead :-) -tony

Going through my "junk", unfortunately a bunch of stuff will have to go, fortunately time has corrected my mental picture of much of its value and its now just old stuff. Not on that list; I found what appears to be a new Mattel Aquarius, minus box but still in its original bag with Styrofoam packing. I know I have an Intellivision around someplace too, but in nothing like this condition. Three boxes of Aim65, Sym, and Kym stuff, complete boards, parts, some docs. On the list of things I doubt have any or much value or interest; Box of 9 track tapes, VHS tapes, Hi8, DAT 120, audio cassettes. Big box of printer ribbons, much smaller box of HP plotter pens. Assortment of HP and Compaq 486 and Pentium class desktops, thinking I will pick one type and keep two. Dozen or so IBM PS2 model 25 to I think 95, again thinking I will pick one or two specific models and keep two each. Old big monitors are going to recycle as fast as I dig them out, but some are going to be hard for me to toss, two are brand new a Viking and a Silcon something, they made medical doc systems I recall. Bulk are old, but not real old Macs, maybe two dozen SE or SE/30 I have stacked around, various pizza boxes and LC types. Sorry didn't mean to make this into whats going post, but will need to make one of those soon enough. I'm in City of Orange, Cal. Email me directly at mikeford at socal.rr.com if your are local and interested in something. Things that do have some value I may try selling, but plenty will be free to a good home. Blame it all on the HOA, after 25 years they have decided no parking on the street unless you have two cars in the garage, and no driveway parking if any part reaches the curb.

Hi, I am interested in the Aim65 stuff. Thanks --barrym On Wednesday, June 27, 2012, Mike Ford wrote:

HOA wants to make you park your cars on your garage? Why don't you say to them to (insert lots of filthy expletives here)???? :oO --- Enviado do meu Motorola PT550 Meu site: http://www.tabalabs.com.br ----- Original Message ----- From: "Mike Ford" <mikeford at socal.rr.com> To: "General Discussion: On-Topic Posts Only" <cctech at classiccmp.org> Sent: Thursday, June 28, 2012 4:35 PM Subject: Re: Found a Mattel Aquarius

On Thu, 28 Jun 2012, Cameron Kaiser wrote: Meh. More people just need an old ugly fiberglass boat and other assorted "yard art" for their front lawns :) If even a small percentage of people did this, such HOAs wouldn't be able to do this.

On Wed, Jun 27, 2012 at 09:07:28PM +0100, Tony Duell wrote: Price. That's what's exciting about it. But most normal people who might be interested *do* have all of those. If downloading it and putting it onto an SD card is too difficult, then Farnell or RS will be happy to sell you an SD card with an OS installed. If it came with an OS, that would make it more expensive (don't forget, even free software has costs) and it's likely that a lot of users would blow it away and replace it with some other flavour of Linux anyway. Errm, what consumer electronics comes with a schematic these days? The information that normal people will need - people who want to build peripherals, or interface it with other stuff, not just people who want to use it as a cheap media player - is available on the website. They're not hidden. -- David Cantrell | Enforcer, South London Linguistic Massive It's my experience that neither users nor customers can articulate what it is they want, nor can they evaluate it when they see it -- Alan Cooper

I don't talk much on here because most of the time I am in awe of most of you guys fixing things and sharing knowledge at levels I don't have the tools or the talent to aspire to, and hays off it's awesome. Fact is though, this thread is starting to have echoes of that guy at HP that said he could only see a market for 5 computers at most. I have a Raspberry Pi and run it 24/7 as an emulated VAX 3900 using SimH. It does a damn good job of it too. Here's the rub on Raspberry Pi as I see it: It's cheap, magnitudes cheaper than any of that POC Dell or Acer stuff you buy for GBP 299 which has *most* of the same drawbacks (no docs, no schematics, etc). This means you get what you pay for on Raspberry Pi and those systems are exposed for the overpriced hunks of junk they really are. You get a board for your 35 bucks. You can do what you like with it. No frills, no BS. There's even some good community support that's free as in beer. It's real. It's not some imaginary vapourware project that'll never see the light of day. Availability can be sketchy but that'll improve as demand normalises. It runs free software. No Windows license, no Microsoft guff. It's accessible (not everywhere, but it is) - it has GPIO which is open and usable and useful, USB, ethernet etc. Sure, you need a few extras. Most of them are cheap and available. Oh and guess what. Buy a fancy BeagleBoard or an Atom ITX board, something that costs anything from 2x to 6x as much and you... STILL need all the extras. To use a horrid management phrase, it's the Total Cost of Ownership where the impact is. Sure it takes some time fooling about, but that's part of the fun. This is a platform for goofing about with, not running JP Morgan's Oracle stack. Stop being so down on it. It's a great idea and will go far because the sheer variety of stuff you can use it for and the price of it means it's going to be something people don't fear buying because of loss of investment. The hardware itself isn't gonna set the world on fire (at 2W it barely gets warm enough to start one) but that's not the point. The Broadcom SoC is a run-of-the-mill cellphone chip. It isn't fast, the SD card I/O is pretty slow (4-5MB/s which is little more than a DEC RA92 disk) and the power to the USB ports isn't sufficient but you can work around them or just live with it. It cost 35 bucks for pete's sake. So cut em some slack, they are trying to do something rare and enlightened in an industry that increasingly is more interested in turning us into cloud-based service-locked-in drones. -- Mark Benson http://markbenson.org/blog http://twitter.com/MDBenson

I was lucky to get a Pi and I have been playing with it. To me it's something that's interesting and there are already a software you can download to make it play MAME and XMBC. Is it replacing a full computer? No.. Is it giving me some fun for a total of like $65, yea... I have spent more money on less technology I can tell ya. Actually I think I paid more for my arduino

I was wondering if the Raspberry Pi might be used as a very lightweight xbmc fronted?.. George Rachor george at rachors.com On Jun 28, 2012, at 11:38 PM, Tothwolf wrote:

On 6/29/12 1:24 AM, George Rachor wrote: The fans speed up because the system didn't boot, so the power management microcontroller goes into failsafe mode (fans full on). You should see the caps lock key flash once as Open Firmware starts up. If OF doesn't come up, try reseating the DIMMs. The light flash will let you know if USB is working. <cmd><opt>RP resets the parameter ram, and it should loop through the OF light flash sequence if you hold the keys down. That may be enough to get video going. <cmd><opt>OF breaks into the Open Firmware command loop. You should at least be able to tell if the video is working by doing that. Holding the mouse key down will open the CD tray. That is another clue as to how much of the system is working. If the machine was in an unknown environment, you should try to blow the dust out of the power supply in the bottom (which isn't an easy thing to do). Many supply failures occur because of bad air circulation. The problem with batteries occurred mostly on the Q605 systems, because the ADB microcontroller drew too much power. Battery issues got worked out over time as the microcontroller folks got more experience with power management on desktops.

On Fri, 29 Jun 2012, Tothwolf wrote: I got mine from RS/Allied. Total cost with shipping was about $40. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_! Buying desktop hardware and installing a server OS doesn't make a server-class system any more than sitting in a puddle makes you a duck. [Cipher in a.s.r]

On Fri, 29 Jun 2012, Steven Hirsch wrote: I got my "ok to order" email confirmation in May and it was shipped June 16th via DHL from the UK. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_! Buying desktop hardware and installing a server OS doesn't make a server-class system any more than sitting in a puddle makes you a duck. [Cipher in a.s.r]

In the uk you register an interest and get a number. If in the US try registering here:- http://www.alliedelec.com/RaspberryPi/ for updates Dave Wade G4UGM Illegitimi Non Carborundum

On 29 Jun 2012, at 07:38, Tothwolf wrote: It is a nice piece of kit, especially at the price point. For embedded or semi-embedded work it's really nice, my only gripes are the SD IO speed is terrible, and the 5V power input is via a USB Micro connector not a 2-pole barrel connector which would have been more 'standard'. http://www.alliedelec.com/RaspberryPi/ Give Allied Electronics a shot, they are RS's US distributor. The reason they are distributing via Allied and Newark is that Farnell and RS are the *UK's* biggest distributors and offered them a good deal to sort out manufacturing and selling the boards with their Chinese contacts. RS and Farnell in the UK are pretty reasonable to deal with although both have a habit of stinging you on carriage if you aren't paying 100% attention to what you are buying (notably if it's 'cheap'). Ouch. Allied may be better, I can't vouch for them for US customers but RS UK have thus far been speedy and pretty fair with shipping and pricing on the RPi. -- Mark Benson http://DECtec.info Twitter: @DECtecInfo HECnet: STAR69::MARK Online Resource & Mailing List for DEC Enthusiasts.

On Fri, Jun 29, 2012 at 06:38:01PM +0100, Mark Benson wrote: Wrong, the Micro-USB connector is actually a standard (and not just 'standard' as the barrel connector). So, a device gets powered via Micro-USB connector? Fine, I can grab a random wallwart with Micro-USB connector and it will work. I can even order a random dirt cheap wallwart from $RANDOM_CHINESE_SUPPLIER and it will work. I can even just connect it to the nearest computer and it will work. Micro-USB is a fully defined standard: - the connector is standard, no surprises - pin configuration, voltages and power limits are standard Compared to a device powered via a barrel connector: - which diameter? - which polarity? - or maybe it wants AC? - which voltage? - how much power? Chances are, I don't have a suitable wallwart in my pile of wallwarts for that. Kind regards, Alex. -- "Opportunity is missed by most people because it is dressed in overalls and looks like work." -- Thomas A. Edison

FYI: At about 1:30pm EST I received a notice from RS/Newark that they were accepting orders for the RPi in the US. Within 1/2 hour their server was down. And, no, I didn't make it. Got as far as setting up the shopping basket and went bye, bye. Methinks they may have received a lot of responses. Steve --

On Tue, 3 Jul 2012, Mr Ian Primus wrote: I got an email from Newark today with a link to their store. They're limiting the order to 10 each and project late August as the ship date. Might have to order a second one. :) g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_! Buying desktop hardware and installing a server OS doesn't make a server-class system any more than sitting in a puddle makes you a duck. [Cipher in a.s.r]

On Tue, 3 Jul 2012, Steven Hirsch wrote: The site returned to life a bit later in the day and I was able to successfully order (2) units. $35 each, plus state sales tax and about $6 shipping. Not too bad. Delivery schedule is "late August", but I'm not holding my breath. Steve --

On Thu, Jun 28, 2012 at 08:57:39PM +0100, Tony Duell wrote: All kinds of reasons. I, for example, want something small and cheap that I can dedicate to my experiments with model railway control. My other computers are either not small and compact, or not cheap, and they're all already in use. Small is obviously good. Cheap is necessary because I will need more than one. And I want something that can run a Unixy OS because of availability of good dev tools that I'm familiar with. I *could* beat my chest manfully and build my own computers out of string and angry weasels, but I have better things to do with my time - such as actually controlling trains. I'm very sorry to hear that. But those don't apply to the vast majority of people who might be interested. The Raspberry Pi website makes it clear. As do the websites selling it. They are not hidden. Let's add up the total extra costs that most people will have: keyboard: 0 mouse: 0 SD card: 0 [have you spotted the theme yet?] power supply: 0 cables: 0 SD card reader: some people might need this GBP 2-ish case: some people might want this GBP 8-ish monitor: this is the only one that might actually be common GBP 50 You are making the common mistake of extrapolating from your rather personal circumstances and assuming that what would be additional costs for you are also additional costs for large numbers of other people. -- David Cantrell | http://www.cantrell.org.uk/david It's my experience that neither users nor customers can articulate what it is they want, nor can they evaluate it when they see it -- Alan Cooper

On 29 June 2012 21:39, Mouse <mouse at rodents-montreal.org> wrote: FWIW, I'm hardly a normal user, and have plenty of spare PC bits lying about, but I have precisely one USB keyboard in the house, which usually hangs out near the lad's PS3. Therefore today I bought a new USB keyboard, and an optical mouse for good measure. Cost me a total of ?7.50 (<?10) from the local supermarket. I'm powering the Pi from the USB media port on the TV that it is using as a display! I can't remember where the micro-usb cable came from, but the HDMI cable was a poundshop jobbie some time back. None of it is going to break the bank and I think that allowing use of such commodity items is a Good Thing - even if people don't have spares, they are cheap enough that the TCO is still not too expensive. Heck, I spent more on a Leapfrog "my first computer", which definitely fits into the "toy" category. Rob

In the UK RS will sell you all the bits needed, and when you order it lists a few but I can't remember if it included everything. I ordered an SD Card with OS installed. I can't remember if I was offered the USB keyboard but I did order the 4gb card with OS installed. Dave Wade G4UGM Illegitimi Non Carborundum

On Sat, Jun 30, 2012 at 07:33:28PM +0100, Tony Duell wrote: There is. RS were *really keen* to sell me all those extra bits when I ordered my board. -- David Cantrell | Enforcer, South London Linguistic Massive What a lovely day! Now watch me spoil it for you.

When I looked (after it was hyped i nthe press, and after I had to navigate past various Rpi pages to get the bits I wanted from Farnell), I didn't see anything to suggest that this was not the final product. Neither in the press nor on the sites I (briefly) looked at. I can think of 3 things wrong with that : 1) It's not open. You yourself said that. Given a piece of oepn-sourve softwre, it is possible (if you are sufficiently technical) to figure out a lot from the source. Given a piece of closed-source software, running on a machine which at the time had no released scheamtics (they are released now, but as you say, things change), you have an alomst impossible job figuring it out without docuemtnation 2) It's not software 3) And therefore it's not free. A peice of software can be easily copied, and thus it cna be made esentially free. A piece of hardware cannot. DOn;t get me wrong, a piece of hardware is a physical thing and it;s obvious you have to pay for it. But even thought the Rpi is cheap, it's still expensive enough that I (for one) would bot buy one unless I knew it coule be useful. Buying things you haev no use for is simply wasting money. AS a result, releasing in a 'beta' version and not stating it was not finished seems like a rather bad idea. -tony

Joachim Thiemann wrote: It is a bit subtle, but in http://www.raspberrypi.org/faqs we have "Will it have a case? Not for the first batch. We'll be making and selling cases by the summer; you'll be able to buy a unit with or without a case, or a case on its own. The education release later in 2012 will have a case by default. There are lots of homebrew case discussions on the forum." The "educational release" (probably out in September or so to coincide with the school year) should be considered the final product and the current version should be considered a developer's preview version. On the parallel thread about "interesting" computers such as Lisp Machines or APL architectures, I design Smalltalk computers. Though the first products will not be in the Raspberry Pi's price range, the plan is to get there as soon as possible. Though the first version will use FPGAs and Verilog, I have a friend who wants to build a TTL implementation as an educational exercise. But that is the future and this list is about the past :-) -- Jecel

On 3 July 2012 19:33, Tony Duell <ard at p850ug1.demon.co.uk> wrote: No, the hardware is not FOSS. It's?not FOS and it's not S(/W). But it is designed and intended to run FOSS OSes. Exclusively, in fact. It is intended to be a cheap Linux machine. The fact that other, non-FOSS OSes are being ported to it - well, one so far, RISC OS - is incidental. As for the unfinishedness and the beta state of this first version - I agree. It's barely been mentioned. It comes up in interviews and things occasionally. I think that they really ought to have emphasized it, not hidden it. But then, whereas I admire the spirit and intent of the project, I don't actually like the execution, myself. I don't like the proprietary GPU, I don't like the closed-source graphics driver, I don't like the crappy old-spec low-powered underperforming ARM core. I don't like the rather underspecced amount of RAM, which isn't really adequate to run Linux with a GUI. I don't like the fact that the Ethernet port is a USB device. I don't like the fact that there's no fast local storage. Now, some of these compromises were doubtless necessary to make it so very cheap, but then again, here is a commercial rival: http://liliputing.com/2012/05/via-apc-a-49-android-computer-with-an-arm11-c… http://apc.io/ Item Description Model APC 8750 Software Android 2.3 (PC System) Chip VIA 800MHz Processor Memory DDR3 512MB Memory 2GB NAND Flash Graphics Built-in 2D/3D Graphic Resolution up to 720p Input and Output HDMI VGA USB 2.0 (x4) Audio out / Mic in microSD Slot Network 10/100 Ethernet Size 170 x 85mm (W x H) Neo-ITX Standard* US$49, or ?31.25. Better CPU, more RAM, more local storage, but inferior graphics and no GPIO. The AllWinner A10 SoC is a 1-1.25GHz ARM SoC with a complete system with GPU onboard for $7 in bulk. It's going to shake up this market and I think soon there will be multiple RasPi rivals at around the same price-point. If this means that an ecosystem of ARM Linux distros for low-end desktops develops, and lots of ARM Linux apps for low-spec machines, then all might be well - if they can all run each other's software, fine. If the kernel and the underlying graphics drivers are different but at the OS API level they're intercompatible, fine. If they're all different iterations of the ARM CPU core and thus are /not/ software compatible, problem. If the latter happens, I think RasPi may lose out, because rivals will be able to offer much more powerful machines for competitive prices. If some of those rivals also offer functionality equivalent to GPIO, and bundle some nice features like optional dual cores (which Linux can exploit well) or on-board SATA, then I think RasPi may end up getting slaughtered. OTOH, if that happens, then perhaps they can just release a RasPi SE or Mark II or something, with an AllWinner CPU and the same form-factor or something. There is a chance, I reckon, that the RasPi may end up like the IBM PC. Quite respectably successful in its own right, but more important for the whole market sector it spawned of enhanced-but-compatible machines. At the end of the day, if it means that the endless march of boring beige x86 boxes is replaced or supplanted or supplemented with little silent, cool-running, all-solid-state no-moving-parts legacy-free ARM machines running Linux, getting kids in the developing world online at prices so cheap that a couple of billion poor people can afford them, then I will be delighted and I don't care who does it - RasPi Foundation, Via, nVidia or whoever. Hell, if the Chinese come up with a rival system based on the Longson/Godson family of MIPS-compatible chips, I'm cool with that as well. Variety is good. Maybe it's time for the RISC resurgence. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 4 July 2012 02:08, Dave McGuire <mcguire at neurotica.com> wrote: Absolutely. Now, tell me, how many of them do you see powering desktops and laptops? Workstations? Servers? Very few. They're in all manner of pocket electronics, and embedded into a billion bits of consumer tech and white goods. They're not sitting driving screens and GUIs. They used to be. At one time, they were the great white hope that was going to vanquish x86. Well, in numbers, they do. But not in money, and not where x86 is strong. That is more of an Intel/Microsoft duopoly than ever. (The competition is stronger than it's been in 15y, but still...) I'd like to see that change. I loved my Archimedes. It was a fantastic desktop computer. I'd like a modern equivalent, some slick fast silent-running monstrously-powerful device with no active cooling and a sleek thin fast OS that you can open up and learn about. I.e., not Windows and not any form of bloody Unix. Bring back BeOS, or Lisp Machines, or something /different./ It's high time. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 4 July 2012 01:56, Liam Proven <lproven at gmail.com> wrote: I think that's actually the point! From http://interviews.slashdot.org/story/11/09/14/1554243/eben-upton-answers-yo… " We?ve said elsewhere that our dream scenario is that someone in China decides to copy our design and start knocking out millions of clones. Remember we?re a not-for-profit organization under English law, and all our trustees have other jobs, so we don?t have the same set of incentives as a regular company." Rob

Well, I don't think it should have been hyped as much as it is. There's noting particualrly special avout it. And I can't comment on the design becuase as I said the documentation is inadequate. But form what I've sene it doesn't seen that interesting to me. My worry about anyt of those applications is that from what I ahve heard the supply of Rpis is limited. Maybe large, but limited. And I would not want to sped time developing something only to fidn the Rpi (or whatever) was no longer available. _I_ hae plenyy of 5V suppleis, but nonoe fo them are cheap... Depends on waht sort of composite :-). More seriously, I'd not want ot use TV-rate video if I could avoid it. Not i nthe UK (for the nth time...) OK, it may have improved recently. Certainly when it was anounced I couldn't find any such information (and no, I don't mean on paper, I did poke around web sites, etc). That _is_ what I was told, by somebody who works in the industry. It's quite possible I was misinformed. -tony

On Thu, 28 Jun 2012, David Riley wrote: I don't know about the "leftovers" part. They must have had a LOT of them if so - while the first batch was 10,000 units, they've gotten orders for at least 400,000 and they're well on the way to meeting (or have already met) those orders now. Frankly, I love the stupid little thing. It has huge potential to affect my "main" hobby by driving down the prices of flight simulator avionics for the hobbyist. :) g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_! Buying desktop hardware and installing a server OS doesn't make a server-class system any more than sitting in a puddle makes you a duck. [Cipher in a.s.r]

OK, sounds like they have finally relaesed one. When I looked (at the time the thing was first being hyped), the attidued seemed to be 'we are still deciding if we will release it'. Not the sort of attiitude that encouragees me to buy it. YEs, that's basicially what I was told. The video (in particualr the accelerator) was not going to be documented withotu signing NDAs, etc. Again, not something that (I would consider using. I can rememrbr when you just had to ask for a data sheet and it arrived -- on paper -- the enxt day. No NDAs... That may be so, but it doesn't encourage me to buy the Rpi board... -tony

On 28 Jun 2012, at 21:18, Tony Duell wrote: I imagine when you used to get data sheets delivered next day to your door with no NDAs there probably weren't 100 Chinese companies waiting in the long grass to clone TI or whoever products and beat them over the head with the price difference. NDAs in the electronics industry are a sad but necessary result of the fact that you can't trust anyone with your product data anymore unless you make some kind of allowance to sue them when they steal it. So don't buy one. Non of use that have one (soon to be two and by the year out probably at least five) will mind. Just quit beating us over the head with your opinion. We get it already. -- Mark Benson http://DECtec.info Twitter: @DECtecInfo HECnet: STAR69::MARK Online Resource & Mailing List for DEC Enthusiasts.

Actually, I'd be very likely to approve of it if : It came complete, that is to say I could just plug it in and go without having to pay to download and print manuals, download the OS, etc The docuemtnation let me figure out how to use the thing. Given that, I might actualyl buy one. A couple of serious questions : 1) AS has been confiremd the data sheet on the main 'chip' doesn't include docuemtnation on certain parts of it (video-related?). Are those parts simply not used by the standard linux for the Rpi, or are the drives supplied as binary-only, or what? 2) I understnad the's some kind of GPIO/user port. How many lines, are they individually selectable for direction? Can this be easilly used from C (I assuem there's a C vompiler included with the OS). -tony

An interesting question that may fit on this list There are lots of old ARM devices around. One that is very common in Brazil is the "easybox" internet computer which is an ARM device with an ISA slot (for originally a modem and optionally a NE2000 network board), some peripherals, video output and remote keyboard. These can be had for (very) cheap. How hard is to port e.g.: Angstrom or something like that for a system like this? Maybe I hadn't expressed myself well. I want to UNDERSTAND how the port is done Thanks Alexandre --- Enviado do meu Motorola PT550 Meu site: http://www.tabalabs.com.br

Thanks Chuck, but I want to understand HOW it is done. I have a XYZ device and I know the memory/port map. How do I put it into the code, compile and load on the target?

On 30 Jun 2012 at 9:37, Jochen Kunz wrote: Good point. 2.11BSD is also a good base for very small systems. I see that a version has been created for the PIC32 that's pretty minimal: http://code.google.com/p/retrobsd/ --Chuck

On 06/30/2012 02:56 PM, Tony Duell wrote: I use both regularly. SPI is inferior in some ways, but superior in others. I personally prefer I2C, but there are situation in which SPI is a better choice. It is MUCH faster (I2C tops out at 400KHz, I've seen SPI up in the several-MHz range), and I2C's transfer size is fixed at eight bits, while SPI can transfer arbitrary word widths. (this is handy for, say, ADCs and DACs). I2C, however, uses in-band addressing so it saves board space and I/O pins over SPI. So, while I personally prefer I2C, I believe it is incorrect to consider SPI to be inferior, since it they both have their strengths and weaknesses. The industry agrees, as both have been around for decades and are still considered de-facto standards for local low-speed communications. (I first used I2C on a Philips 87C751 microcontroller on a gov't related project in 1991, and it was well-established even then!) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 06/30/2012 04:27 PM, Chuck Guzis wrote: Very true. :-) But for a real-time clock or a few kB of EEPROM, 400KHz is usually just fine. It's rare for any of my designs for work to not involve both SPI and I2C. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 1 Jul 2012 at 20:45, Tony Duell wrote: Do you want me to send you the Verilog for the CPLD in the middle? --Chuck

On Sun, 1 Jul 2012, Dave McGuire wrote: Don't worry Dave. It'll be mis-spelled and unreadable, so it won't do any damage. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_! Buying desktop hardware and installing a server OS doesn't make a server-class system any more than sitting in a puddle makes you a duck. [Cipher in a.s.r]

On Mon, Jul 2, 2012 at 6:36 AM, Jos Dreesen <jdr_use at bluewin.ch> wrote: /**************************************************************************************** * SPI to I2C interface * January 2007. *****************************************************************************************/ /* Top module */ module SPI_to_I2C (SPI_sclk, SPI_cs, SPI_miso, SPI_mosi, I2C_sda, I2C_scl ); inout I2C_sda; // Bidirectional SDA line inout I2C_scl; // Bidirectional SCL line input SPI_sclk; // SPI clock input SPI_cs; // SPI chip select input SPI_mosi; // Master out slave in output SPI_miso; // Master in slave out //........ endmodule /*********************************************END************************************************/ Didn't want to post the whiole thing here, but you should take a look at: - http://www.altera.com/literature/an/an486.pdf - http://www.altera.com/literature/an/an486_design_example.zip Cheers, Camiel.

Not particualrly, for3 main reasons : 1) From what you say in another message, this came fro ma standard application note, and I can look that up if I ever need it. 2) I don't read Verilog (or any other hardwre description language). Now, if you send me a schematic doing the whole thing with 74xx ICs, I'll make sense of it :-) [1] 3) It's irrelevant anyway. I don;t dispute you can make an itnerfce between SPI and I2C. Heck, I've got an ES232 to I2C (as a master on the I2C) interfce here. But you are stil ltalking I2C to the teletext IC, which is not suprising. You cna't sensibly do anythign else... [1] I wonder what the bneast way to do thsis would be? Off the top of my head, soem kind of microdoed ssytem or complex stat emachine (which at this level are the same thing, of course) would be my first guess. -tony

The only reason I am not of the 'CPLD persuasuion' is that I have no way to use them. Not one of them is totally open. Yes, if I had to do this I would use a microcontroller. But that footnote you quoted specifically refered to a comment I had made about 'doing it with 74xx ICs'. I don't believe there are any such microcontrollers :-) -tony

On 07/01/2012 03:42 PM, Tony Duell wrote: Yes. More often than not, availability is the deciding factor. Yup. They do. SD cards can also be accessed via a four-bit-wide, much faster interface that is proprietary and undocumented, or at least it used to be. I'm sure the specs are floating around in the wild by now, but the SPI mode works just fine, and is fast enough for many applications. You're not convinced that SPI supports multiple word widths?? I can show you the datasheet of the microcontroller in my current design, where it's configurable, if you like. No, that's not at all how it works Tony. Or rather, that's perhaps a suboptimal way to look at it. My particular example is the Philips...erm, "NXP" (grumble) LPC2000-series ARM7 processors, which I'm using all over the place now. They are 32-bit processors, and their SPI controllers can be configured (by setting bits in very well-documented registers) for 8, 9, 10, 11, 12, 13, 14, 15, or 16 bits per transfer. You write to the SPI data register, which is sixteen bits wide, and just treat the other bits as "don't cares". As a digression, normally, assuming programming in C, you'd place a "unsigned short" variable at the address of the SPI data register. (I'll have none of that uncivilized "uint32_t" bullshit that these kids have dreamed up...and, surprise, my code is VERY portable) Yes, of course. But the transfer width of the SPI interface is not at all dependent upon, or even relevant to, the width of the processor it's connected to. Thank heaven. :) I agree 100%. When I have a choice between the two for the same functionality, I err on the side of I2C. :-) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

No, of course I am, convinced of that. I have read the data sheets for SPI decvices. I am not convinced that having arbitrary word lengths (which might not even be a multiple of 4 bits) is an advantage. Sure. I would have expected it to work that way. But what do you do if you, say, have a 20 bit device? 10 bits at a time? That means repackign the 20 bit word. Yo ucan't do 16 bits and then 4 bits (apparently), and if you could you'd still ahve to reprogram the word length between the parts of a single word. I'd prefer to just send an integral number of bytes (or words, or whatever) and have the device ignore the unusued data. That's waht I2C does for 12 bit devices AFAIK. AFAIK sizeof(unsigned short) is not defined anywhere :-). -tony

It was thus said that the Great Tony Duell once stated: I'm not sure I follow you here. In C, you can indeed do a sizeof(unsigned short) and get back the size (in characters) of a short int (with the size in bits of a character defined by CHAR_BIT). Well, that "uint32_t" crap is *very* useful for networking protocols or binary file parsing. -spc (I use uint32_t at work, but then again, I have to deal with network protocols and binary file formats ... )

It was thus said that the Great David Riley once stated: Um .. it *IS* defined in a standard---the ANSI C standard: ... Their implementation defined values shall be equal or greater in magnitude (absolute value) to those shown, with the same sign. CHAR_BIT - number of bits for smallest object that is not a bit-field (byte) CHAR_BIT 8 (C Standard, section 5.2.4.2.1) True. I've been playing around with viola [1] and it *barely* works on a modern system and won't at all on a 64 bit system, due to the "an int is a pointer is a long" mentality of the code. Ick. -spc (It looks like it started out life as K&R C, with some small token attempts towards ANSI C ... ) [1] One of the first graphical web browsers from around 1992. It has a number of concepts which are in use today, including style sheets (not compatible with CSS), scripting (its own language, not JavaScript) and some that are not, such as the ability to traverse the object model/source code (like Smalltalk or Hypercard environments).

On Jul 2, 2012, at 9:18 PM, Sean Conner wrote: I should clarify. Yes, you can get sizeof(whatever) and it will be the size of whatever in bytes; I meant that sizeof(unsigned short) is undefined by a standard. Assuming that it has even a fixed minimum (1 char notwithstanding) is a fool's errand in a lot of cases. 32-bit architectures have dominated computing since I started programming, but I've still seen plenty of breakage (even in new code) caused by such assumptions. I think actually the most obnoxious example I've seen is the assumption that sizeof(int) == sizeof(void *). One of the bog- standard codebases for MMCs on the ATCA architecture, written by PigeonPoint, has some post-processing utilities written for Linux in the source distribution. They made the boneheaded moves of a) typedefing "u32" to "unsigned long" (which made "u32" a 64-bit int on amd64) and b) assuming that "unsigned int" could be used to hold pointer differences (which it cannot on amd64). The resulting code just crashed because the heap was getting stomped, which is hard to debug, and it was so difficult to worm out the errors (especially because some of them just produced slightly-wrong binaries instead of crashing) that I gave up and just forced compilation as 32 bit code. - Dave

On Jul 3, 2012, at 11:24 AM, Sean Conner wrote: OK, yes, the minimums are defined. I've actually run into much more trouble when people assume a maximum, the crypto example I mentioned before being the big one (where you're using integer overflow to perform addition/multiplication/etc modulo 2^32). That's not short, obviously, but a short could be 128 bits if a long were 256. It wouldn't be illegal, and might be totally logical on the host architecture (though it's obviously in the realm of fantasy for the nonce). - Dave

I don't think of it with respect to C very much, but I suspect you're right. C as it's currently defined does impose some constraints; for example, it pretty much requires a binary machine. Even beyond that, it promises a lot less than most people (seem to) think it does; I've often thought about building a portability-test compiler which goes out of its way to break such assumptions. Something like 9-bit chars, 17-bit shorts, 33-bit ints (with irregular endianness, maybe something like BCAD order, and the padding bits scattered around)...and that's just the beginning. :) I've long had similar feelings about POSIX and OS architecture, the worry that anything that doesn't at least mostly fit the POSIX mold kinda can't be done. I spent six months writing glue code to take a research project in distributed, encrypted storage and present a POSIX interface to it; the project was very interesting, but the need they felt to give it a POSIX interface is, to me, an example of this issue. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On 3 Jul 2012 at 10:11, Fred Cisin wrote: Isn't that pretty much a fait accompli? Consider for a moment how the world of computing might appear today if, say, APL was the "standard" computer language taught and used. --Chuck

On 07/03/2012 09:37 PM, Liam Proven wrote: (world the would suck but would be very (elegant)) -- Dave McGuire, AK4HZ New Kensington, PA

On Tue, 3 Jul 2012, Sean Conner wrote: Not all of us are enthusiastic, EVEN NOW, about switching from defacto practices in K&R to arbitrary standards in ANSI-C In K&R C, a short int was often the same as a char That's what K&R originally said, also, and that THAT was ALL that one should assume. They said that an int should be whatever was best handled by the machine. THAT one bit me, and hard, when I was first starting, and using char for byte (contents of 8 bit registers, etc.).

On Tue, Jul 03, 2012 at 11:38:24AM -0700, Fred Cisin wrote: That'd be the Cray compiler. -- David Cantrell | even more awesome than a panda-fur coat It's my experience that neither users nor customers can articulate what it is they want, nor can they evaluate it when they see it -- Alan Cooper

On 07/02/2012 09:18 PM, Sean Conner wrote: Treating pointers as ints has always been dangerous. K&R themselves said that, even in their first edition if I recall correctly. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On Jul 3, 2012, at 2:38 PM, Richard wrote: In fact, I said both (look above). What I did *not* mean to do was imply that MSVC did not have ptrdiff_t, something of which I had no idea and about which I had nothing to say. In retrospect, my statement was worded rather poorly, so apologies for that. I more or less agree; VS2008 was the first one I used that didn't drive me absolutely batty from a standards point of view. 2010 is theoretically much better, but I haven't had much cause to do much Windows programming recently, and the last time I did, some critical components I was using (specifically, CUDA and OpenCL support from ATI) didn't support VS2010 without major patching. I shouldn't imagine it's gotten any worse. My chief issue with MSVC from a standards point of view has almost always been the lack of stdint.h, which I use quite a lot. It's easy enough to drop in a hand-rolled replacement, since Microsoft has gone to great pains to make sure that nothing is changing (even to the point of making "long" 32-bit on 64-bit architectures because there's so much crappy code kicking around blindly assuming that "long" is 32 bits), but I still feel uncomfortable not having that from the compiler vendor. I'm glad to hear that's been rectified; if I ever get the time to do game programming again, I'll be glad to see what else has improved. - Dave

On 07/02/2012 06:53 PM, Sean Conner wrote: I work all day, every day, with network protocols and binary file formats. On every platform I work on, and indeed every platform I've *ever* worked on, unsigned short has been 16 bits, and unsigned int has been 32. That includes the 8-bitters as well. The idea is good, but the syntactic sugar of excessive typedefs where they just aren't needed (and portability to 95% of architectures isn't impacted) is just pointless overcomplexification. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 07/02/2012 07:42 PM, schoedel at kw.igs.net wrote: Yes, so have I. But I'm a commercial embedded systems developer (one who generally doesn't use DSPs!), and I don't work on any of "those" platforms. Literally EVERYTHING I USE has 32-bit ints and 16-bit shorts, and that has been the case for upwards of thirty years. From my point of view here, I see absolutely no reason to add a level of indirection to the native compiler data type when it gives me (ME) no benefit. This has got to be the most corner-case-obsessed group of people I have ever met. ;) (don't get me wrong, I find it fun!) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 07/02/2012 08:09 PM, Dave wrote: Hmm. My two development platforms are x86_64 and UltraSPARC-III+, both of which are 64-bit. An int is 32 bits, and a short is 16 bits on both, by default. My targets are most commonly AVR and ARM7. Ints there are 32 bits and shorts are 16 bits as well. And here's the crux of the matter. I agree 100%. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 07/02/2012 09:09 PM, David Riley wrote: This is gcc v4.4.6. I don't recall it having been a (toolchain) compilation option. I will verify, but I'm porting a few bits of formerly AVR-based C code to ARM7 right now, and not running into any issues. There, my predecessor used madness like "u08" (like a friggin' byte is ever gonna be anything else!) and "u16". -Dave -- Dave McGuire, AK4HZ New Kensington, PA

Honeywell H6000/L66/DSP8 (and several others) have 36 bit integers...

The downside of that otherwise rational argument is soon they'll be saying that about minority architectures we actually care about (PowerPC comes to mind). -- ------------------------------------ personal: http://www.cameronkaiser.com/ -- Cameron Kaiser * Floodgap Systems * www.floodgap.com * ckaiser at floodgap.com -- Simplicity is the ultimate sophistication. -- Leonardo da Vinci -----------

On 2 Jul 2012 at 20:25, Dave McGuire wrote: No, but I suspect a resurgence might occur in some of the old ideas. One such is that as word length increases beyond a certain limit (say, 64 bits), lareg *integers* are less important, particularly if they exceed the width of an address. In that case, it makes sense to treat integers as a special case of floating point. Thus, in a 64-bit machine, an integer may have a range of only +/-2^47-1 (48 bits); a short, only 24 bits; a long 96 bits. It was thus on machines as late as the ETA-10 (I don't recall if Cray ever adopted that convention) and I wouldn't be surprised to find a modern system that operated that way. And one of these days, I expect to see a variable-word length CPU again. Variable-length BCD arithmetic might be considered a boon by the financial people. --Chuck

On 2 Jul 2012 at 22:28, Dave McGuire wrote: Well, the NEC V-series did have instructions for variable-length packed BCD add/subtract/compare up to 254 digits, but sadly, no multiply/divide. So it's not that far from current architectures. At least one company (http://www.silminds.com/) specializes in decimal add-ons to existing architectures and decimal IP cores. Must be some money there... The STAR-100 had BCD add/subtract/multiply/divide of packed decimal numbers up to 65Kbytes in length (that's 2 digits per byte). To date, it's the only hardware that I've seen that carried BCD arithmetic to such lengths. Enough even for the most ambitious hedge fund manager... --Chuck

It was thus said that the Great Dave McGuire once stated: Well, at work, we develop code that has to run on both 32 and 64 bit systems (Pentium and SPARC, so I get the fun of byte swapping too!). I don't even know what the size of a 'long' is on the 64 bit system (but I don't really have to care when I can select uint32_t or uint64_t as needed). I learned C during the large 16-bit/32-bit switch (MS-DOS) and there, the sizes of nearly *everything* changed on what memory model was used ... -spc (If I recall correctly, there were several different compilers for 68k based systems where an 'int' could be 16 bits or 32 bits ... )

It was thus said that the Great Dave McGuire once stated: It hasn't been that bad. In fact, the testing tools I wrote (that's my job---to test the code the other developers write) were more portable than the code I was testing! (our lead developer has this insane embedded mindset where I swear he thinks we're writing our code for 1MHz 6502s instead of 64bit multicore SPARC but that's a rant for another time; he didn't bother with byte swapping (because every cycle is sacred) on the network protocols because, well, hey, we're using only SPARCs, until that wasn't the case). I seem to recall the Latice C compiler for the Amiga (I still own it) spent a chapter or two on linking with the output from other compilers that may use different sized ints (68000, internal 32 bit, external 16 bit bus, so arguments could be made for using either size for an int). -spc

On Tue, Jul 3, 2012 at 1:10 AM, Dave McGuire <mcguire at neurotica.com> wrote: ISTR there was a difference between the default behavior of the Manx C compiler and Lattice C, but I skirted most of it because when I plunked down the big bucks, I got Lattice. IIRC, Manx had early popularity because of lower acquisition cost bolstered by the fact that it made "smaller binaries" (because it did a lot of 16 bit operations in places where Lattice C for the 68K (or even UNIX C compilers on VAXen) would default to 32 bits. I also recall some turbulence from source that would compile on Manx but not on Lattice C, but most of the examples to be found would probably be in the oldest 5%-10% of the Fred Fish library. -ethan

On Mon, 02 Jul 2012 19:26:37 -0400 Dave McGuire <mcguire at neurotica.com> wrote: No. avr-gcc uses 16 bits / 2 bytes for int / unsigned int. I am looking at this very moment to a LCD connected to a ATmega168 that printed sizeof( unsigned int) on said LCD. Your confusion about sizeof( int) is the exact reason why uint32_t and its friends where introduced with C99. IIRC a int is 16 bits on 2.11BSD too, as the PDP-11 is a 16 bit machine. But it is way to long ago since I hacked C on my 2.11BSD machine. So I may be wrong on this. The point is: An int is usuly the largest intish type a machine can handle "comfortly". This is true for the AVR, as at least some AVRs have instructions that work on 16 bit values. Also: sizeof( int) is expected to be > sizeof( char). (> not >=) So if we imply char to be a 8 bit byte an int must be at least 16 bits. Thus sizeof( int) == 2 for the AVR, since sizeof( int) == 4 would be to much overhead. The 16 bitnes of the PDP-11 is well knowen. Regarding uint32_t and its friends: They are in the C99 standard. (inttypes.h) I use them all over the place. Especially on AVRs. I wrote C code that I compiled without change on an AVR and on a PeeCee running NetBSD/i386. I used (u)int(16|32)_t all over there to avoid the sizeof( int) missmatch. This was important, as I did some fixed piont arthmetic. -- \end{Jochen} \ref{http://www.unixag-kl.fh-kl.de/~jkunz/}

On Jul 3, 2012, at 3:27 PM, Dave McGuire wrote: The bright side is that going *up* in bit size (like you're doing from AVR to ARM) usually works. The big counter-example is when someone has done something too-clever-by-half like using integer overflow to perform an automatic modulo in order to save cycles. In cases like that, you absolutely need to know the bit width of your variables, and the not-knowing can produce sneaky bugs that are very difficult to track down. I absolutely agree that it's important to know the salient features of your architecture, such as native bit-widths. "Salient" for me for the AVR architecture includes the fact that it's an 8-bit architecture and that 16-bit math is actually not so cheap on it (the only 16-bit native ALU instructions are for moving and adding/subtracting immediates, and I would hazard a guess that you only get them because they are leftover "bonuses" in the ALU for the X, Y and Z regs). So when I'm writing code I know will be targeted for the AVR, I tend to use uint8_t a lot more (and yes, I use it because the name carries the semantics for the reader that I am not necessarily using this as a character string). I guess that brings up another reason I happen to like those stdint.h typedefs: they give semantics to the reader ("I REALLY WANT this to be 16 bits") that "unsigned short" does not. It's proven useful to me when porting other people's code. - Dave

On 07/04/2012 05:50 PM, arcarlini at iee.org wrote: Very true; and if K&R had originally specified C's native data types to explicitly include their size, I'd have no problem with it. My issue is that I started with C long before these typedefs came into being, and I just don't like 'em. ;) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

I don't know whether there has been any C implementation (gcc or not) for the PDP-10. I'd expect to see 9-bit char, 18-bit short, and 36-bit int. long could be 36 or 72; I could argue for either. I think a good case could be made that short and int should each be 36 bits, but I'm not sure whether I agree with it. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

...until you want to port to an architecture that doesn't have a 32-bit data type. I would much prefer to use uint_atleast_32_t (or however it's spelled) and pack/unpack the octet stream explicitly. Significantly more portable, clear, and future-proof. Or at least that's how I see it. For hardware drivers, which IIRC is where this came from, I see little to no real difference between (say) unsigned short and uint16_t for these purposes; the portability demands for hardware drivers are rather different. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B