Jump to content

TheAncientOne

Programmer
  • Posts

    726
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by TheAncientOne

  1. Phenomenally inspired. A depressive who could lift my depression. A spinner of wonders. In "SlaughterhouseFive,’’ he drew a headstone with the epitaph: "Everything was beautiful, and nothing hurt.’’ Well, Mr Vonnegut, you've got the answer to one of your questions by now.... If there is a God, I bet you're both having a good laugh. Go in peace. Mike
  2. I bought some as well, even with post to the UK they seemed a deal. They are definitely 24 'clicks' per rev. though I hadn't fired one up yet. I've just had one hitched up to my old Racal counter/timer. I did a 'gold standard' with a Bournes ECW 1J, and I easily read 24PPR. Then I tried the ones from Mr 'gmtcruiser'. They seem to have problems - they produce random counts as if they have very very noisy contacts. The Racal has a reasonable level of deadband available, but only on channel 'B', so I can't debounce the cheap encoders. I'll have to try them as an outboard and test in the near future, but for now I'd recommend trying a longer debounce time. Oh, and the switch common is one of the outside pins, not the centre as in the Bournes ones. I'll report more when I've done some tests. Mike
  3. Strydmulator, the human, (chip), sampler ....... Mike
  4. Just discovered this thread. I'm part way through one of the last of Trevor's 9090's. Waiting for my last batch of resistors from RS before doing the caps. I found a couple of ols Syntom and Synwave PCB's uncased from a previous lot of building - even found the Maplin fron panel stuff. If the diagrams aren't on line. let me know where to put them. They did a 'SynBal', too, a sort of metallic sound gen, got the drawing btu never made one up. My first drum synth was in the 70's. I started with a PE design and kind of went mad from there. It used 7489's for memory, and battery back up, when I did it, was for hours only. I'll try and find the file of analogue module designs. And an idea: there a quite a few telecoms racks going surplus now everyone is upgrading networks. There are a lot of double eurocard rack cases, with backplanes etc. Double euro with a double width front is not a bad drum module size. It would be easy to hack the backplane into power, trigger and audio distribution, and have a set of replaceable drumsounds, say eight to a rack. Rather than use the PCB's, just make a piece of alloy the same size as a double card, and bracket the front panel and rear connectors to it. My later drum sequencer drove this sort of stuff from a BBC micro's user port, (no velocity), then later from a BBC 1Mhz bus card with accent outputs as well, though I got over deep into prgramming it, and never used the messy end result much. I might have another go, now, having got a load of free BBC's. Need to get some pics up. I'll have a trawl through the diagrams box when I find it. Mike
  5. You've got a cave like studio?? Cool! Does everything have big black and white labels, like on the Old Batman TV series? There was a bit of a fun 'Gotcha' here - I've just been using a home made manual tester on my DIN's and I went and double checked before trying two together. I've never seen anything other than R5, so now I know what to watch out for. Thanks Mike
  6. According to Smash's site, the currrent DIN's are designed to "daisy chain", using a single ribbon connector: See http://www.avishowtech.com/mbhp/mbhp_dinR5.html IMHO the pins on the headers are designed to go Out --> In if you use the right connectors. Example: Cure --> J1 on First DIN J2 on First DIN --> J1 on Second DIN J2 on Second DIN --> J1 on 3rd DIN etc Mike
  7. There is a phenomena in software design known as 'Feature creep'. You just get the main code running, when the system specifiers, or more usually the people in marketing, come in with a new feature they've just thought of and is now a vital part of the spec. There are certain well known historical projects that never ran because of this. In fact, I think one well known 'solutions house' extracted an immense amount of money from the UK government by allowing the spec of the software to creep out of range of the hardware capabilities, then getting the project cancelled without ever handing over any hardware or software..... The antidote is 'Design Freeze'. Decide on the spec you want, that is feasible now. Lock yourself in to that, and get it running. You'll get far more feedback on the spec for Version 2 this way, and have something to play with. It's a cool idea, but if it were my project, I'd be looking at a feeze point around now. When you do V2, you can add in your new ideas, and probably do a clean re-write, which will incorporate all you've learned so far. You'll end up a slicker fuller featured machine, but with leaner code. You'll also learn what seemed like a good idea, but never got used, and so can be left out. Believe me I've been there: I wrote a sequencer running on a BBC micro that got too bloated to hold any decent length of sequence. I was a bit like those car customizers that don't know when to stop adding twiddly chrome bits and blue lights. I never used the final thing much. Mike
  8. Perhaps I hadn't explained too well: I should have said swap file, (late and tired). For normal storage CF would be fine - and I really like your idea of internal and external CF's. What I was thinking of as a problem was if I ended up with swap, buffer or cache file on the CF. There is a limit to the number of times you can write a location in a CF. It's now in the high tens of thousands, and good CF software will do intelligent writing, only changing the bits it needs to, but a swap file will wear a CF out in months. A friend did a smoothwall using CF, and the first one failed because of this. The optimum solution for him, to get a quiet machine was a laptop drive in a cooler. It might also be possible to use a RAM drive of some kind, I have seen 2G battery backed up ones in some industrial stuff. My initial plan is for a 'test rig' machine I can rack mount, though I wouldn't want a full depth server case, rather one of those rack mount ATX cases Scan near me do new ones, but I'm hoping to get something second hand. Once I know what works, and what I want, then I'll look at doing a custom case. Old Dell desktops often had passive ISA riser cards in them, I've kept a couple somewhere. the later ones had PCI risers. You can often get complete optiplex's (I think they were called) for £20, and the case is narrow enough to adapt for rackmount. In my case I'd consider getting an alloy top cover done, and see what other weight I could shave off. Your rig in a rackbag sounds well cool. One custom Idea I've thought of is to take a disc cutter to an existing case, or drill out the spotwelds, and thus get the 'difficult' backplate metalwork to incorporate into something lighter and smaller. I'm still experimenting. Converter looks good enought to do a stand alone unit from. I'll see how it goes once I've run it for a bit. The 1U idea isn't too easy, but could be done, and would be well cool. On that subject, I'm also looking at some kind of 'heatpipe to fin' idea, to avoid the size and noise of a processor fan. I fancy something more variable in general, though if converter is good, I might do a machine just for it. I do like the concept of an 'appliance' PC, as opposed to general purpose ones. My first appliance was a Smoothwall firewall. My second experiment was with 'FreeNAS', which may go 'final' once I can afford a few more discs. Like you, I'm trying to find a good use for some old motherboards. I've got a couple of the IBM 'Blue Lightning' chips, a sort of souped up 486, which is passive cooled. I'm wondering if one of these might work. I might state here my informal 'Prof's First Rule of RetroTech': "Always have a spare". If you are building a gadget using obsolete technology, make sure you keep some spares, at least enough to cover you whilst you design a mark 2, when they start to run out. I learned this the hard way with my first domestic heating control system, built using an old obsolete PLC. One night it died, and we had a cold week whilst I built a new one. My mum, who I lived with back then, was not happy. Mark 2 used an old BBC micro.I had loads of spares, and of course, true to Murphy's law, it was still running when we sold the house years later. If anyone wants a BBC, by the way, let me know, I've got quite a few, legacy's from when I did repairs for local schools: when they all went to PC's I kept being offered the old stuff. They've got a Texas 76489 sound gen as well as a load of usable IO. Postage outside the UK might be a killer, sadly. Mike
  9. Amazing resource for good basic info on MIDI. http://www.hinton-instruments.co.uk/reference/midi/promidi/index.htm Mike
  10. If you're working on the road, or in a professional environment, it's best to avoid board mounted sockets of all types, if possible. Someone tripping over a lead can trash a board as opposed to just a socket. There is a lot of wisdom on the subject herehttp://www.hinton-instruments.co.uk/reference/midi/promidi/index.htm I put my MIDI in outs on chassis mounted sockets, and use the 3 pin 'molex' style connectors to hitch them to the core board. if a socket gets damaged, I can change it out with a screwdriver, because I've made a couple of spares. Same with the phono's or whatever. If they are on flying leads you could easily change them for 1/4" jacks if the need arose. Mike
  11. THis is my reading of teh diagrams, since I have no experience with the USB card. The card provides: A USB to 2 MIDI in out, (and nothing else at all - the 'stand alone mode to get MIDI on a laptop for instance) A USB to 1 MIDI and 1 Core A USB to 2 Cores. The MIDI on the card is driven from the host of the USB (the computer), and I don't think it is accessible from the MIDIbox core. If you want more MIDI outs driven from a core then you will have to add IIC boards. Mike
  12. Superb work Sasha. I do the hot glue trick too - it works well on the LCD cabling as well, and will peel off without too much trouble if you have to rewire. Mike
  13. Good point Stryd. I notice that the OP seems to have things fixed, but a couple of powers supply related ideas might not go amiss. Most MIDIbox boards have their regs staning up on end, held on their leads. For a light load, in a fixed installation, this is OK. If the reg runs warm or you're taking it on the road, then its not. As with big wire wound resistors, some of the heat is transmitted down the device legs. If the reg is running quite hot, the pads on the board will heat up, and in the long term the pads will lift, or the solder will seem to 'crystallise', forming what looks like a dry joint. The vibration of moving the things about will add to the effect, or maybe just break the legs by fatigue. If the ground lead goes first you could be in trouble.The reg is quite heavy compared to the legs, and its good practise, especially for mobile stuff, to secure them down. I would always reccommend fixing the tab down somehow. To do it on the board would use quite a bit of extra real-estate and put up Mike and Smash's costs, so I suggest a bit of fin or alloy bracketed to whatever base plate you are building it on. If you are mounting your core module on an alloy base plate, then another old trick can be used. the reg can be put on from the underside, legs going up through the holes, and bolted down to the baseplate. This does need care though. And of course we all do test the supply before inserting the PIC. I'll have a couple of bits done before the end of the month, I hope, so I'll try and get some pictures up to show what I mean.. Pity you're on the other side of the world - I find your posts both illuminating and amusing - a rare mix - you'd be welcome to make it group therapy, It might work for thumbs too. Mike
  14. I think I left the wrong impression here - I was citing the Cray as an example - I have actually been in the machine room at World Weather Centre in Reading, but they were later Crays. I used it an an example of high end ECL, and why they ate so many Amps. I researched the machine very thoroughly for company that wanted a model building. A local retired engineer built it, as a break from his steam engines. The company had a long discussion (they are all tecchies) about whether it would be possible to make a 1/12 model Cray 1 that actually works, using modern parts. The answer seems to be "yes if you can afford about £2K worth of high end FPGA and fast static RAM (the main RAM was 50nS static 64 bits wide plus parity, the main clock ran a real 80MHz - actual cycle time, the registers have to cycle in 6nS!). I'm worried they might decide to ask me to do it, and whilst i have a lot of the system documented, the design of some kind of I/O for the thing would be a killer on it's own. By the way, the Cray used vector processing. The length of a vector is often known as it's ....... 'Stride'. The guy who runs Armari has a couple of rescued Crays http://www.hexus.net/content/item.php?item=903&page=3 and these guys in Germany have real ones you can log on and use http://www.cray-cyber.org/general/start.php My board was a eurocard sized multiply and accumulator, board, for early sort of DSP, part of a project inspired by something in Strawn and Roads' "Foundations of Computer Music". It was connected with one of IRCAM's multivarious projects. We got it running very nicely, then those nice people at TRW came out with a single large chip that did roughly the same job. . . . I think I'd successfully supressed the memory of the thing dropping back to the bench with a thin layer of my skin barbequing on top of the chips. After your comment I'm going to have to do some more therapy. Fortunately I've got a very good Russian therapist called 'Stolichnya'..... Mike
  15. Thoughts on expansions. The multiple sample/hold can be implemented in loads of ways - you could even use a bunch of LM398's and a decoder chip. The old chips were just easy to layout - but there are plently more variants. On the thought of drum triggers. Don't just go wiring those PIC pins to your drum modules etc. There's enough voltage in any analogue synth to scorch a PIC. I know we're all pro's and never accidentally patch a drive line to an output, but but one tired evening it will get done, stopping the session cold. So; put in some buffer chips to protect the PIC or opto isolate the trigger outs. I like the opto version: it helps keep the background hash from the sequencer out of that delicate audio stuff too. As an aside from this: it would be reasonably possible to opto the lines to a DOUT, ot to an A-->D, that way the triggers/ CV's would be on the synth power, not the Box, hopefully keeping the background noise down too. I'm talking about the SO SC and RC lines, by the way. An idea anyway. Mike
  16. Shall we start a club? T shirt or tie? In my case it was complete lack of knowledge of how hot ECL gets. Put my hand on the board just to see it was ok - it stuck to my hand. Whilst swearing over the pain, I was more worried I'd somehow fried the ECL chips (expensive!). It was fine - it just runs that hot! Which is why that cute looking Cray1, ate about 115KW, (correct figure, checked, it had poly phase 400Hz secondary power distribution to the final supplies under the 'love seat' bit.) For those who din't do old style logic, work ECL, Emitter Coupled Logic, was for years the fastest logic chip type you could buy. Some types ran up to 1GHz. The Cray 1, the worlds first real 'supercomputer' was built using 1000's of ECL chips. Cray 1's were like the SR71 Blackbird, (or Bugatti Veyron or whatever), of the computer world: hand made, mind blowingly expensive, and requiring serious maintainance, massive amounts of electricity and cooling, but delivering ultimate performance. See: http://www.nasm.si.edu/imagedetail.cfm?imageID=1118 I think my "favourite" power repair was the faulty transformer that blew the black plastic parts right off a set of 78/79 series regulators. An inter-winding short had produced about 3 times the secondary voltage, and guy at the knitiing factory where it was, had put a bigger fuse in, 'just in case'. I had to replace every chip on two double euro cards. I added a 'crow bar' circuit, 'just in case'. Mike
  17. Definitely worth a least a quarter... I'm with you all the way with this. There is another thing too, control in these devices is very much a dynamical system, (more chaotic in my case). The magnified effects of small variations makes them all sound different, and many patches unique as a result. Want it clinically the same? - write non stochastic stuff in CSound. If you want a real gnarly variable sound, with noise and edges and, dare I say it "Character", then it's got to have analogue parts, and knobs and dials. That being said, I sometimes want glassy, pure impossible textures, so it's out with the CSound, and hopefully, when I can get my head around it Pure:Data will allow me to fuse the two in a way I want to. Getting out there with a hot soldering iron is feeling very good right now, though, (sacrilege!), my first full running project might be not be MIDIbox.... (my tuppence worth as we say in the UK, which makes it about 4 cents at Current rates) Mike
  18. By definition, that is what 'Thru' ports are, simply an output driver connected after the input opto. See: http://www.usyd.edu.au/anaes/rpa/Loadsmanextras/PCmidi.html Which is a useful diagram, showing how to make a box to convert your soundcards' output to proper MIDI without buying expensive cables. This one, notably, has multiple outputs. Hope this helps Mike
  19. Pesonally I'm looking for a 3U industrial case on evilbay, Stryd-One, because that will be far easier to work on and adapt. If the rig runs fine, then is when I'll think about condensing it. Still like the idea of boot from ROM or Flash. I need to see if Converter needs any disc as scratch space, if not it might be possible to run it without a hard disc, given that I can get nice, cheap compact flash discs, which, unlike SD or such like, look just like an ATAPI hard disc to the PC. If the thing needs scratch space, then it's not possible, because the write limit of the CF will soon be exceeded. Mike
  20. Largish board if you're lucky...... However, there may yet be a way: http://bach.ece.jhu.edu/~tim/research/fpaa/fpaa.html An actual product http://www.actel.com/products/fusion/ It makes me wonder about an entire synth on one of these, with DCO's and sequencer implemented in the logic side. This is just me being curious, I've done no more than skim the articles thinking 'looks cool'. I just like the thought of, for instance, being able to reconfgure filters 'on the fly', though emulating that nice Moog or MS20 dstortion might be another problem. It all started with me looking for a substitute for my Electronics Associates TR48 analog computer that I donated to Bletchley Park museum when I moved house. I just wish analogue synths had been made with that level of quality. Presently I've got a tiny analogue computer made of of modern op-amps etc, I thought the array might open the way to a much larger one, and then, of course, if it will build an anlog computer, why not a synth? I'm not sure of the analog/digital balance in the things, so more than one might be needed. Something worth looking at anyway. Mike edited for ypting rerros
  21. I was thinking of speed, and using a direct PC resource rather than one on the soundcard - sometimes you can't use the on card MIDI synth and the on card MIDI port at the same time - depends on the card. Hence, (my thinking, not guaranteed), putting the control surface into a main PC resource will simplify things. So far I've found references to people running two different coundacrds, but no more, on a standard PC. If I can find the right boxes I'm going to dig out a second AWE32 and try it. One thing I did find for DOS, you may need a plug and play tool to get the later soundcards to run up. I think it only runs at boot to initialise the cards. I've got an old Intel tool for this, which may need a little research, as each card will want it's own address, interrupt and DMA settings. Mike
  22. I had to look, hadn't I? <sound of Prof running off and getting a drink> about 700 bloody quid including tax! EEEK! Still damn fine looking though. Mike
  23. Ancient Chinese Curse: "May you live in interesting times" A good friend, and a far better programmer than I will ever, be once called it a 'write only language'. He said he could write things in it well enough, but re-reading the code 6 months later he couldn't make sense of it. On a more salient point, asking because I haven't gone to the mat with C on the MIDIbox yet, does the compiler support nested structures? Mike
  24. Lovely job Wisefire. Fuinnily enough, I used some of the big Ikea bowls for a project last year: light reflectors for a Fortuny style bounce light. I was going to write to Mr Ike or whatever he's called and demand they made bigger ones with a more elliptical profile..... They were for my friend the photographer and of course due to that particular form of Murphy's law, we never got a picture of them..... Looks well cool - a nice job, and definitely deserves it's 'Box of the week' , or even 'hemisphere of the week'. Mike
×
×
  • Create New...