Theo Verelst Diary Page

Latest: Februari 8 2001

I've decided after good example to write some diary pages with toughts and events.

Oh, in case anybody fails to understand, I'd like to remind them that these pages are copyrighted, and that everything found here may not be redistributed in any other way then over this direct link without my prior consent. That includes family, christianity, and other cheats. The simple reason is that it may well be that some people have been ill informed because they've spread illegal 'copies' of my materials even with modifications. Apart from my moral judgement, that is illegal, and will be treated as such by me. Make as many references to these pages as you like, make hardcopies, but only of the whole page, including the html-references, and without changing a iota or tittel...

And if not? I won't hesitate to use legal means to correct wrong that may be done otherwise. And I am serious. I usually am. I'm not sure I could get 'attempt to grave emotional assault' out of it, but infrigement on copyright rules is serious enough. And Jesus called upon us to respect the authorities of state, so christians would of course never do such a thing. Lying, imagine that.

Previous Diary Entries

February 8, 2001

After yesterdays serious what is it, sort of preaching in writing, I think I'll go into some thoughts I've had and some of the subjects I'm in the process of getting results with.

My thoughts about the synth subject

Basically, I like a good instrument to play, and for a long time already wanted a fat sounding analog type of synth, and any variation with rewarding and strong or beautiful sounds. That has been so since I think 15 or 20 years, when in the end I had more than enough equipment, see some pages back for pictures of most of it, I still wanted other instruments as addition, for instance I long considered buying a second hand pro-one, a sequential monophonic analog synth in maybe almost the same area as the minimoog. At the time, they were sometimes offered second hand for about $200, which is a bargain enough, and I should have taken, in retrospect. Also, I bought a TG500 for about $800 at some point, because it was offered as a deal because of some traders deal for about 2/3 of its normal price, and as I wrote, I can't find a picture of it, and it is based on sampling, with a lot of voices, and quite good built in effects, such as reverb, chorussing, etc.

The stuff I'm writing about is not so much to form my thoughts, because they're quite formed enough about the subjects, but to communicate my proceedings maybe, as a way to prepare that what I do makes sense to others as well, and to amuse or edify or in between interessees. For myself the subjects were as I made clear on some older pages, I had an interest in ananlog synths and their digital simulation, and wondered wether I would be able to do essential building blocks myself, especially considering that there are some famous well sounding ones that I until not so long ago would not have tried myself, also because I didn't know what type of circuits were used.

Then I found that mainly the idea were as I had them at the time too about other applications (such as compressors): straight behaviour as much as possible, and that the filter circuit I always wondered about is available in service sheets I found on the web, and could be seen as implementing quite understandable lets say measurement equipment type of electronics. When I could simulate such circuits I found out I'd be able to build and use them right, and already a number of years ago I prototyped some alternative circuits for synth building blocks, such as tdm multiplying dacs, and with fet switches, and switched filters, and the generators I built long ago were of similar quality, and some even more mae to work well in higher frequency range, such as a balanced triange wave generator with switched inverter to generate accurate current controlled sawtooths.

I had tried additive synthesis and computer reproduction of such samples as well as delay lines and such in highschool already, but never tried much digital filtering, and only recently I worked out some examples and tried some straighforward digitial parallels of pleasing analog circuits, after I built and tested them electronically. Is all that fun? In fact a not insignificant part, apart from miserable enough circumstance, was quite work. Making electronics work right with just a multimeter and ears to become agood hum and noiseless and accurately driveable unit is not without work, and getting as many as possible relevant data about the internals of the targetted synth about the table also takes work, thats more fun though.

In the end, when I had the filter as experiment, I did some other things first, and let the subject rest after I tried it in only some test circuits, the same with oscilators, and it wasn't until I made a keyboard, and especially the microcomputer connection with it, that I could try the machines while actually playing sounds, which is fun. That is hardly work, feelingwise, though I have done many experiments before having a better feel about the various sound spectrums and behaviour in practice. That's hours on a row many many times, which works right, I knew my previous synths right too, the same way.

The digital simulation is the part that's pretty new, except that I know what sounds I at least want to be there, and is quite rewarding, as some seem to have made clear, just like some at Yamaha seem to have made clear that it is not for nothing that there are quite some electronics in the profiessional reverb that I had, and screwed open in the end, to see that they didn't indicate for nothing that I was using interesting parts, because some essential ones in my experiments appeared to be the same as they used. I'll get back to the latter when discussing and dealing with quality sampling, in audio and hearing sense of the word.

When doing the digital simulation right, which is pretty much by making a sort of electronics simulation and minding that the digital signal path is uses all its parts right, the sounds can be good, and that was not clear to me from the start. Of course that in principle that is possible I'd acknowledge, a CD player can too reproduce good enough what synths do, but to generate the sounds from pure mathematics and circuit simulations in software, that is another question, that is not trivial. And then to arrive at good enough sounds, and some objectively good ones, and finding that that continues with more sound programming in the same software, that is interesting and makes clear that the efforts are worth it.

And that is were I wouldn't have been all to sure as to what is all needed to make it work right, and a 22kHz or so 9 bit over-all sample reproduction of (at first) gwbasic generated simulation samples was not what I had in mind to make it all tick good enough, but it works good enough. Easy? Straightforward enough, but not easy. One must know about electronical circuits and simulating them, about sampling, and how to use the whole idea of such synths, neither of them easy sibjects. Understandable in part also in general, but not easy, so that was challenge enough.

What I'm not writing about as in 'this is at least how it is' but thinking about myself in the area is the use of fast digital circuits, and about how to achive high quality analog to digital and vv. conversions with parts that are affordable, good, and applicable for real high quality products and efficiently. There start to appear codecs that do 20 and even 24 bit ad and da at 96 kHz with even 6 on a chip or so, but that is not the only thing I'm interested in.

Thinkingwise I'm more into doing the high resolution da at high speed myself, and preferably multiplying, and with analog / digital conditioning, and for instance I've requested an Analog Devices sample of a 14 bit ad converter that runs at 100 mega samples per second or so. At 14 bits, that amazing almost, and then see what else besides a nice osciloscope can be made with such a device and 4(!) nS comparators, and output switchable opamps with high current driving and low distortion capabilities running up to hundreds of megaherz.

That stuff should for instance be applicable to consider multiplexed dealing with various sampled electronics circuits, such as in mixers, eq's, compressor units, and other lets say studio type of equipment, which can then be controlled by such converters and dirvers by a fast dsp without introducing sugnificant delay, and without necessarily sampling every signal to pieces.

That's whats in the call it mildy to quite speculative thinking going on subconsciously and sometimes explicitly.

Not very speculative but more than fun enough is thinking about doing the types of synthesis I've now more than enough experience with on a faster computer again. And drag in bwise, maybe some mpeg and web stuff, that's the free development department in thinking, I know for sure that will work, and know how it will, that is certain stuff, but the results are fun to play around with, and it may make some good products, I;m certain at least it can be competitive, except I don't know how long the competitive edge of the sound generating software can last against competition without dragging in physical modeling and lets say synth programming skills.

I've tried various software synth packages (lots of demos), and not many realy hit hard in the synth sound generating department, they sound often dull, weak, fuzzy, downright bleep-ish, or at best sort of mwaa, which is not a good basis. That can be because of the algorithms, the programmed sounds, or the soundcard (a bit), but not many are realy strong as I've observed.

I would not be all to interested to make a product that is there in hundredfold already, though admittedly that by itself can be interesting enough, and I don't know what kind of technology push I'd be into if I don't regard the reasons for the products that are there, there no reason to become agressive player and get eaten alive by Yamaha or another in no time, unless there is, and I don't think there is.

The string simulator is good enough stuff for some time to come to make a good distinction with, and so can my programming and creativity with known synth types be, but there are some with quite some synth roms on their name that can, too, and that is an area I'd rather see as musician and lets say creative person, and not so much as expertise.

The whole of sound synthesis, mainly for use as an instrument, thats my prefered angle, is rich enough to make very much possible, and the combination of various techniques on a current fast dsp is good enough to make very interesting things with with loads of real variation and freedom for times to come.

Applying the theory I've been delving into in theoretical physics for the physical modeling of instruments, that too is where I'm thinking about, thats a new area enough, and quite advanced, probably directly scientifically interesting.

And lets say some networking question, the electronics end field stuff of putting megabits over kilometers of simple twisted pair wires reliably interests me, as well as the ideas of the wealth of boxes with all kinds of protocols being involved in connections, and how to get to and build parts of exiting and new reliable and fast backbone nets. Interesting.

And of course the brain stuff, as in how does it work, how do my assertions in the area relate to what I read in various literature, and what does my increasing knowledge change about my models. Essential, but long term stuff. A little network with nodes to the power of times of choice possibilities to connect or fire, and faculty related number of ways to make sub graphs of is interesting in another, more software playing kind of way, that sort of generates experience. And of course mastering programming languages and environments that are worth it and staying or becoming fluent in them never hurts.

I soldered a few bandcable wires to the ti DSP sample, which is kind of fun as an idea, but not when its the only thing I currently have at the speed, that I may want to use. Basically, it might work, but as is to be expected, it sucks bad, that 20 wires per inch, times 256, on a poststamp. It might work, I tried conencting a little piece of a broken off hobby knife on the soldering iron (which is ridiculously big), which in itself worked, but the typ wasn't handy enough to use, so I made it simpler, I fyled the tip of a paperclip end into a sort of tiny srewdriver tip shape, twisted the paperclip around the soldering irons' tip, tighened it at a srew, and tried wether it would melt my 1mm 0 tin. With more then a cm paperclip metal in between, heat conduction isn't good, but nearby the tip, all tin bathed, it would work, and then I could solder some wires, making good use of the islands of blobs of tin present on the DSP's connections pads.

But the though of doing it for more then a few, and then starting the second row at the same smaller than 1 mm distance of the first, and knowing that every wire broken means doing a square of 16 at least or so again looked very unapealing. I think it can be done with trial and error and building up of experience and skill, but I'm not sure I want that. Lets hope soon enough a good enuogh job opportunity (maybe a hundred and hour, that would be something at least to get things going again) falls through, that a prefered way, and than generate a little work in the printed circuit board producer area. Or use the TI evaluator board straight from the box, srew it in an existing enclosure with keyboard, include some keyboard decoder and maybe display unit, and fool em all with the product.

I have the distinct, general intelligence insulting impression, that some don't like that idea, and that there are forces (what a word) in this world that are against such peaceble, fun, and completely un-negative activity going on. Dispicable.

Essential Greek

I was sort of taken still by James 1, I read through verses I think until 9 or so, about the temptations, and who brings them about, the flower, that stuf. The flower is supposed to be in an enclosed space, and the rich let no water near, so it withers and dies. Temping or testing by God (as in an an extreme case "he chastises his son" or if you are free from being corrected by him, you may not be his, as an indication that the principle at least is know in other bible books) seems to make ones desires or longing release one as if from slavery, and get away, and should not be delusive and if brought into the world may lead to a wrong idea in it and produce death.

Quite different stuff then in the translation. I'm not finished, and consider doing a text on also the beginnings of aeschilus and maybe demostenes (didn't check that one out yet) to bring forward a good idea as to why I think about most main relevant greek words in also James the way I started to, and why I think that at the time these texts may well have been deliberately aluded to, because of their subjects.

Completely interesting, this is lets say hard core stuff, christian wise, wisdom wise, background wise, refutation and correction wise, and when I'm more finished, doctrine wise.

This stuff is about the essence of life, and in essential ways, and the lies that are about the same subjects are the reasons for many, many, wrong doctrines and ways of dealing with life, and that being to expected as a result of evil in general, I think it of complete imperativeness to at least aim for a good basis for those who want to know, and religionwise there is no other way for me.

No Windows?

I'm aware of various operating systems, and quite skilled in the use and reasonably to quite exert on the buildup of a few at least, and have a few years ago taken a vanilla PC, whiped the compacs' hardisk, messed about a bit trying various partion sizes, and trying again, and installed redhat, I forgot which one, which worked fine enuogh, minus a lot of not-well-done things such as the installer not working easy enough, and not allowing easy wrapping up and reinstalling of packages (seems to have been solved shortly after), postscript readers that can work but need lots of configuration, slow, though operative, netscape, hard to compile opengl libraries, and such.

Fine for a free operating system, and after a little experimentation: a working and efficient and robust enough real multi user, multi tasking operating system, and not some NT kind of (as it seems to me) tricked cooked up system, with maybe components of age such as X windows, which however are optimizable for a range of display kinds with work, and then can perform, but then again, don't we all end up with bigxbig with 3 or 4x8 or 12 bits standard screens. Apart from certain little or 19' machines..

I've never been particularly fond of X windows, I did program at least two applications in it years ago, curvedrawing sort of things, and previewing wireframe pictures with for the time big models of hundreds of thousands and more wires, but the ideas (client server approach for instance), the sort of implicit object oriented C approach, and the whole idea of those windows in the first place in a general and portable way were capable of maybe 8 years ago already make me look at a life captured and transmitted over the internet american TV channel, just opening an X window connected to some server. Fun, and definately not trivial at the time.

The linux system worked fine enough, and at least is capable of doing multi processing without pitfalls I found in the not so long time I used it, making cgi's and parallel processes work fine and not with the windows delay as far as I could see.

Yesterday I checked the Linux magazine, I don't remember the exact title, and they are at least making a jsut profi enough impression to take serious in the whole of os things, I think, and various interesting sujects could be read about, amoung others the idea of web-interfacing over forms with some computation engine, sort of what I did a few years ago with tcl/tk on a image database, interestingly sort of the basic materials needed that I then figured out, too.

They didn't use a server with built in efficient cgi functions, no connections with a fixed computation server process, and the nothing special was done database or presentation wise, but they did mention a few html keywords I wasn't aware of, or that didn't exist yet to deal with the streaming of large amounts of data over the browser server link, I look it up to read.

I saw there is new mag on embeded linux, let's see if I can check it out somehow, they suggested hand helds with that os running, could be interesting, but would it be possible to have enough linux running to do development?

I forgot

With the revelation stuff, I forgot the idea that some must have to do certain things, because some beast seems to receive authority (why the hell would one want that), and it seems that it is of interest to some that such a construction has the power to speak, and it seems to exhibit certain signs taken for 'life'.

Well, well.

I remember a 20 year old commodore 64 of a friend at the time that spoke, and was lively enough, music, graphics, the whole thing. What would win? And who gives a * ?

I guess it has to do with the idea that the holy spirit can give life, gifts, change of heart all these things and that late after such display of real power, some suckers following the one that still has this preoccupation, a sort of compulsion, an idee fix, that it is better to present itself as realy, realy at least just like God. Isn't that something. And when the holy spirit will be gone for 3.5 (x2?) years, it will indeed fool quite some without direct correction. Amazing. Incredible.