Theo Verelst Diary Page

Latest: 26 december 2000

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 familiy, christianity, and othercheats. The simple reason is that it may well by that some people have been ill informed because they've spread illegal 'copies' of mymaterials 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
 

Dec 26, 2000

The text from yesterday isn't finished, but that has time. What about amplifiers? I made some of them during my limited lifetime, and recently my attention was drawn to the subject for two major reasons: I tried a receiver (tuner + amplifier) I found, and apart from the fm reception wondered about the amplifier and speakers effect on the overall sound, and I still think it is a valid idea to make audio amplifiers as a product, because it can be done profitable and with results I myself also have an interest in.

Also, I'll write about some progress in the synthesizer software: the digital filter simulation is getting better. I'll think about wav converting some samples, to put on here, it can be done, but takes some work that doesn't add much usefullness for me, though it may be fun and convincing, a little arpeggio or something.

Stereo listening

Since I was about 12 I think, I was the lucky owner of a stereo cassette deck, and records and radio to record from, and some kind of stereo sound reinforcement equipment. It that important? In the whole of things, probably not that much, but personally, yes. I listened to music easily hours a day. Not that that is very special, but I think for me it was relatively important most of my life, starting before that time already.

I had 2x4 watt EL84 powered (too low voltage..) stereo tube amplifier, and the aforementioned cassette deck with usually well enough recorded tapes. First, two sony speakers were the main source of music, which were not very special, lets say the high and low end not much present, and the rest not clear enough for any satisfactory stereo experience. I think when I was 14 or so, I laid hands on a set of 'obere mittel klasse' and 'spitzenklasse' Philips HiFi speakers, that is the speakers themselves, and I started building encloseres that would last a long time as quality and power enough main speaker systems.

I had a shared small room, so they weren't of extreme size, maybe 60 liters or so I guess, but they were well damped pressure enclosures of great stiffness and strength, and contained speakers good enough to use seriously. The tube amp was not my game. It had lots of switching options, and for the time it was made it didn't hum that much and probably wasn't bad, but I didn't like the transformers, the distortion, the lack of high, the heat up time, and even the lack of power. It took some time to come up with a good alternative, because my allowance would definately not let just go out and buy an amplifier.

The first alternative was a chip based (transistor) amplifier of 5 or two times 6 watts, that made a very clear difference, clearer sound, in fact I liked some of its properties. The power being insufficient, and not up to my speaker standards, the first real improvement was a 2x10 watt transistor amplifier, taken from an electronics book, which performed so well that I was taken by the clarity, openness and stereo picture of the sound. That was something else. It was before I realized it was quite possible to beat the commercial (Pioneer) reciever my dad had, which never made any speakers sound very convincing.

Powerwise, that was different. I went to school parties, and knew that it was needed to have more than a few watts to take my equipment there and make things work, and of course I liked that idea, being music and equipment interested and capable person. In the same book (drama on) the Real Systems were of an other order. We were talking a mono 90 watts quality tube amp, and a solid two time 70 watts transistor amp. Now THAT was something else. The twice 10 watt prototyp I built was satisfactory soniclly, I liked that appearently a quality enough amp made such a listening experience possible, leaving only the power question open. Most home systems were maybe up two a few tens of watts, nice for a party, but of course there were the major disco parties were some more would seem more a desirable deal.

As I though about and was to be anticipated, I saved up, foud ways to find heavy transformers, elco's (capacitors the size of moderate salt containers), and most of all semiconductors, I needed an actual 4 spare power transistors, which were a major budget user in those days, especially the quality ones. After some time I started to get the whole picture together, and would undertake the real major upgrade of going from a twice ten wat decent amp to an abolutely heavy twice 70 wats prototype. Depending on the audience, some may have an idea of these things, but that was like heavy. A commercial amp or receiver of this power rating was heavy (literally) big, expensive enough, and definately not average, not even in the whole of hifi land. So when I say prototype, that was where Major Parts were brought together, this was not childsplay. And seriously, outside teen bravoure, an amp of such rating still takes some consideration.

So I built a frame type of construction on a mahjor heat sink I got somewhere from a another amplifier that was never finished, luckily with 4 power transistors I needed (I had a few myself in my electronics part stock, but I neede dhtese things for other experiment, too, and 5 guilders ($2.5) a piece was steep for me, especially since they could even be blown up...), and I made a construction based on this heatsink, with neatly connected circuit boards, a seperate power supply unit built up on solid 18 mm wood board, and some surrounding circuitry, I built the electronics on the detachable circuit boards, connected the transistors after isolating them from eachother, and fitting temperature sensors on them, and slowly the 10 watt prototype was rebuilt into the desired system.

After quite some neat work, checking, measuring, recheking, and building together, the moment would come to connect the whole thing up to test equipment: dummy loads, to save the speakers from possibly blowing up, a strong signal source that wouldn't be easily damaged if it would get signal fed back into it (my year long well serving mixer), and my voltage meter to see if the whole thing would adjust itself to the right operating point. I don't exactly remember the moment of truth, but certainly that the power on did not in one stroke kill all four power transistors, setting me back for a major investment amount, no sparks were seen, and I assure you that 70 volts on a 2 thousand micro farad capacitor generates more sparks than most flashlight capacitors when short circuited, and even the bias point of the power stage seemed well adjusted.

I waited a few minutes to let the whole thing get its right operating temperature to adjust the rest current. In rest already 7 watts were running through the whole thing, making the heatsink warm. It would take hours to through and through reach operating temperature, which could be noticed when paying attention, but a few minutes were good enough to adjust the major potmeters in the amp, and the required rest current could neatly be set. That was something, that by and large indicated that the whole thing might operate as to be expected. It was time to take the gamble of connecting the speakers, turning the volume control a buit to feed it some signal.. It worked. In fact straight away after I built it together, which was far from trivial, but then again, I had built half a dozen (lesser) power amps already, I was no rookie. No Grounding loops this time, no sir. No humming lines, no noise flimsey soldering connection, not even the need to check some minor circuit errors.

So speakers in place, a favorite cassette running, and a little turn of the mixer and amps volume controls: music! And is sounded wonderfull, even hifi-wise, this needed not to be seen as a raw unrefined power amp. And with small home audience, permission would be granted to play some favorite material at higher volume. Now my room at the time was lets say small, as in 2 by 3.5 small, and this amp powerfull... My gawd, those neighbours were gonna notice this! They must have. Ironically, I had a multimeter set up as a course watt-indicator, and I assure you, this thing did not get not too hot for nothing, it would not be cranked up all the way with people present easily. Ha. That was something at least.

It remained a prototype set up a long time, a few years. It was heavy enough with its mbf ground plate and 18mm surrounded transformer house, the heatsink was attached good enough and though loose wires connected the detachable volume control, all that was missing was a case surrounding the ground plate. It even had a soft start circuit, because the cones of my sensitive, rubber suspended 30 watt woofers in their real airtight (kit!) pressure enclosure would pump a but too much air all too rigorously when this this was turned on with it undamped startup 'doink', realy. They would be seem popping in or out about a centimeter, which would hardly be good for them, so I solved that neatly with a relais taking the offset voltage of the output capacitors during startup, automatically for a few seconds, after all electronics were my serious hobby, and this thing had to work right.

And those capacitors, too, were up to the job: I computeted the required capacitance value to make sure the whole audio range up the the lowest (20 Herz) range would be fed through my amp. Serious test measurements, with sine generator and oscilloscope brought to light that indeed that had paid off, the whole audio range 20 to 20k was no problem, and neither was the measured effective, Root Mean Square power: an actual 75 real watts per channel, over 4 ohm. Period, that was in those days. I had friends with fathers, and family, and so forth, but this was a power rating that would not easily be topped off, this was serious stuff. It was a satisfactory thought that even expensive enough, 'top' equipment it regularly listened to would not be up to these figures.

Did it work ?! Well in my room it worked for many years, at a fractio nof its rating, of course, and though I'm sure it was not up to my current distortion standards, I've at the time compared it with various others, and was not at all dismayed, not at all. The speakers were made with woofers of 'only' 30 wats continuous rating, and 4 tweeters (they happened to be cheaply available quality tweeters) of together 60 official watts, which posed a constraint problem: this amp was actually going to blow away my speakers if I would use it all to enthousiastically. Not that that would happen in my room. Realy, these were 75 real rms watts, not pmp 'music' power with lots of luck and special musical requirements, just because that was so, and that is loud. Quite loud.

Of course the thing had to be tested in some bigger space, and that was possible at some school events, were finally the whole latest system, including speakers, could be tested and used in practice. Major test. In the truck of a borrowed car, set up in the right space, connected up again, rest of the equipment in place, wait for the right time to disrupt the rest of present people (class, school), depending on location... It worked. First, I always constrained the power, simply because it realy was loud, and second, I had no intention doubling the maximum speaker power and ending up with the sorry remains of blown up woofer cones, and I knew what that looked like.

Anyhow, the moment of more truth of course was anticipated, and at some piont the crowd and the trust in the working system would allow volumes to be more than above room volume. Of course the first thing after connecting the equipent up (usually with some friends equipment connected to my mixer, too) was to do a quality check. I was into putting myself roughly at the equal-sided triangle middle optimium listening position, adjusting the speakers or my ears such that the unfortunately not dome speakers would be well alinged, and than playing some favourite music selection from quality tape (TDK SA-X was the definate choice in that time) at a volume at least enough to make some use of the ears dynamic range, knowledgeable of the frequency-wise hearing sensitivity for different sound levels, i.e. the bell curves indicating that outside the midrange the ears perceives relatively more at higher volumes.

And the trick here: no tone controls. No bass boost, no trebble fiddling, no loudness buttons, no nothing. Just freshly cleaned and demagnetized tape heads, well positioned speakers, and nothing else but a straight signal path leading from deck output to the speakers. And as I expected: this was pleasant In my room volume, listening position and 'acoustics' (grrmpfff) were far from optimal. This was a lot better. Basses had space, high volume, and the stereo image was nice enough. And, I repeat, this stuff was something else, the home systems I of course knew from school parties and friends' houses would also be up to some volume, which is nice, and then there would be basses and sounds, and all that, but this was something else. At those volumes, this system was straight, no bass compression, no extra sparkles because those tweeters are moving so fine in their own way at higher volumes, this was worth at least trying to listen to as lets say hifi system in these larger spaces and at some volume. Satisfactory. I wasn't easy to satisfy, but is was of a base and deep enough satisfaction (audio wise that is) that this equipment performed, and good enough, realy I liked listening to it, and it made my favorite loud music come to live enough.

And than of course: the volume! One time I remember the sys was used to do what in a radio show was called the 'hittest'. A synthy background music is started as the candidate takes place, and in maybe half a minute, 5 fragments of hit or classical songs would be mixed in shortly, one at the time, and the candidate would have to guess which they were. We had our own. My mixer, cassette player and deck, tape deck and a friends cassette deck were connected up the mixer, and we made our own hittest on top of a piece of I think it was Vangelis or maybe Jarre fragment. Fun enough to do, of course the hits were taken from radio tapes or LP's, and the challenge I always saw was to arrive at a quality product, which was serious stuff. Limiting the number of generations (overdubs), making sure anything in the signal path is quality enough and well adjusted, and of course making sure the content works was not trivial. The fragments were preferably choosen funny or musically pleasing, and real short, seconds, making it important to do the mixing job accurate and right! And it was not so easy to prepare the whole tape, it requires first that the background synth stuff would be played, and would end up on the master tape, over the mixer and still sound fresh and spacious. Then the fragments would have to be accurately prepared, before the master recording is started, and preferably on two tapes, otherwise tapes would have to be changed during the recording, they would be set 'sharp', and have to be mixed in at the (rythmic) time in the leader music, at the right volume, to let the candidate hear it well and without distortion, and it would have to be stopped again exactly sharp! And then some seconds later, immedeately the next fragment, same recepy. Hard enough, but there was experience, seriously working the pause buttons and my mixer was an experienced act. Making it all sound right, and making no errors, that was the main deal. Because one would not be right, the audience would know it after one test ! So we made many, of various kinds, and every time the last fragment would be messed up, the whole recording would #$% have to be redone, of course.

The hittests were successfull enough, and they required the audio system to be clear and loud enough, even in a crowd. In this case (I think one of the first times I tested it at school) it was put in a hallway with stairs, and much open space, so realy not very optimal, this was more a PA situation, but I was confident and not disappointed, those synthy lines and fragments poured out just fine, and even within the bounds I had for myself before I would realy use significant power all the time. Just, but still, even for the test, no problem arose, heatsink lukewarm, woofer behaviour no problem, distortion and 'disco sound' effects minimal. Later on, when no amplifier frying had occured, I would want to try things at a more maximal volume, and prepared some tapes for this, set up the mixer and amp main volume controls to feed the thing some more signal, started the tape, opened the slider, walked to in front of speakers, and pepared for whatever would happen. One test was 'the visitors' the last ABBA album, which progressively had been recorded in a digital studio, and which I had recorded on clean and quality tape the second time the record was played. The song has a nice volume buildup, which in this case had to already start loud, well, it did. And as the drama musically builts up and the (such was my observation at the time) spectrum would be fuller and fuller with distinctly heavy synth lines and rythm tracksm in the end the climax of the song would spank out louder still. Would that happen in these surroundings, at this starting volume, too? Well apart from fear that I now was pushing it in the woofer department, it did, and filled the space with quality loudness I had not produced before. And in this open and not so small space this was like having to shout loud in not so small environment of the system, it was LOUD.

The recent system I described on my other pages, and of course not to small (current) PA systems are of another loudness ballgame, but still, that stuff was getting there were it was loud without too much cracking up. At the louder levels, where the amp would get hot, and then it was loud, it started to indicate its boundaries were reached, and some more distorion would start to become noticable. The recent system was different, probably at least about 4 times as powerfull, but that depends a bit on how that is measured. When I used that in the major hall, that was again another sort of loud, and part of the speakers of such high quality, and the amp of such numerical (figure-wise) quality that comparisons with high end systems can be made. I did do some computations on how it was possible it was as loud as it was, almost without sound degradation except again for the highest levels I tried.

My mos amp chips were fed with a 300 watt officially rated toroid transformer, which under many amps of load would not be far from its rectified voltage of 78 volts (from the top of my head). Lets say it uses 75 volts of this range, and does so until the current limit of at least 10 amps kicks in, then the effective rms power it could generate would be 625 divide by the impedance of the speaker, rated as 8 for black speaker, somewhere around 3 probably for the other, I measured 6, depending on acoustic impedance for the black one, lets say it could than for the highest quality channel have been 150 watts effective RMS continuous power, meaning the heatsick would consume about 50 watts, which could be just ok considering the chip heat transfer specs and the fact that I screwed it straight onto the heatsink.

The transformer and rectifier should then deliver some 400 watts, maybe some more, which I consider no prob, this was one heavy, low impedance torroid with at least 20 amps or so rectifier, and the 2mm2 or so car type wires to connect it all. In principle, with certain (CD compressed) material, the heavy power content could peakwise be almost square wave, which would make it soak a peak power of logically speaking more than 10 amps over the maximum voltage, equalling about 35 volts is 350 watts per channel. Al this depending on where the voltage and current are actually maximum (because there probably are quite some virtual components to it, both because of electrical and acoustical phase behaviour), and what the effective impedances were, which in my experience may vary considerably (hundreds of percents, I measured it on other systems myself).

The car speaker would be up to that, roughly, the little broadband, depending on wether it was on and its relative impedance, of course not, it would fly right out of it airslit, so I kept an eye on it, and I think I shortcircuited it at some point to not blow it up (and make it a bit lower acoustic impedance as hole in the soundboard). A certain tweeters wasn't up to it either, definately not. So it cost me a new tweeter. It sucked anyway. The one that was recommended to me, the Sony silver dome, lived, and indeed I like it.

Now that is not that average a power anymore, depending a bit on where between 2x150 and 2x350 it was, that's not kilowatt PA, but still, it getting in that range. I was most satisfied with not pushing it to the point of audible sound degradation, being only slight, hitting in. Funny enough that is sort of standing at 10 meter or so distance, not to let my ears distort it for me, and than deciding that the high end audio norm is just starting to fade a little bit, that's sort of the idea here. Apart from some obvious flaws that an expensive high end system wouldn't have, I'd need to balance speaker output levels somehow, and put damping in the left (black) speaker at least.

Anyhow, that is again another level of loud, and I'm sometimes reminded of it, because I have another of the chip amps lying around as prototype (no case) monitor amp, with a bit less solid supply (some halogen lamp transformers), and a only a few quality speakers (the same as in the black box) just lying around (I don't have an enclosure), but oh boy when that stuff, held together by the heavy cardboard woofer box where I put the thing on, upside down to damp the annoying midrange of it, is inadvertedly or purposely exited with enough signal. I made an analog synth with some cmos stuff, that would even clip the whole thing, that's above many rockbands' practicing sound level, in the bass range, realy, when put on a wooden floor that thing is not funny. It makes me suspect that though safely rated at a hundred watts or so (not officially tracible), those 12 inchers I think do not protest all too much about a few hundreds watts of power passing through them. Its still in one piece.

Though frankly, maybe 50 or so is fine to keep sound quality real crisp. I realised yesterday that there may be objective reason these types of speakers distort not muchm comparison wise. The suspension damps the cone action so much it takes in the second range for the rubber foamish suspension to take the cone back to rest when it is released. That in the driving and acoustical impedance game mean it is quite accurately related to the driving voltage. I should write this more scientifically, the model I partwise remember, and think about would be that the current through the coil is linearly causing to the force on the coil/cone (asuming constant magnetic field in the airslit), and that the voltage over the coil, apart from loss of its ohms resistance is the differential of the current change as self induction and as the integral of the travelled distance in the magnets field. During my study I scored a straight A on moderately advanced magnetic field material, I should be able to get this straight, I'l sort it out. The 'virtual' current stuff and all is important, also in the amplifier design area.

Anyhow that system (see my PA and some older diary pages for pictures, circuits, and info) performed quite well, and it is interesting to make it even better, and figure the relevant paramers out. Its a major satisfaction still, for me that is, when such machinery works well producing excellent and loud sound, and is even affordable. And it did. It is was doubtless the best audio system I built, and better than most things in this area, except for expensive high end audiophile equipment (which usuallu will not be loud enough) and quality PA equipment, although I'm not unexperienced in the area, and even the rare enough quality stuff is not often real good in the subtleties of sound reproduction. An interesting area, and it will be nice when I'm able to built some of this equipment agian, maybe in a multi-amped form.

The monitor amp is an attempt to fit the electronics of a 120 watts or so amp in the size of a small matchbox! The electronics, the chip and its surrounding electronics on a small circuit board, see the application notes on the amp page, would fit in a matchbox. Connect the 60 volt or so supply, the speaker, a heatsink preferably and an audio signal, and it works. I considered putting on in one of those small aluminium electronics project cases, using the cse as heatsink, the main point being it still needs a heavy supply, for the rest it could be handy as monitor amp units or so. Anyhow it performs not small, my synth experiments include low filter resonances and oscilator notes, and even at low volumes they make vibrations that are bodily felt, which is better. More rewarding.

Digital sound synthesis, some context

As I'l already mentioned, it is possible to take a good and strong sounding electronic music synthesizer, think about for instance Jean Michel Jarre, or Toto for all I care, and make a computer that imitates the wide and warm sound palette that such a machine can make. The most fundamental way I have described: take the electronic circuit diagram, put it in a circuit simulator (they exist for for instance scientific use), make it simulate the synthesiser circuit with all the knobs and switches adjusted right, make the result into a signal, and voila, a simulated sound.

The reasons to do this are varied: lots of sounds for and knowledge about such machines are around, and many musicians like them. Some machines go out of tune, all of em are pretty big, not cheap, each has its own character, and it isn't always possible or easy to make analog synthesizer sounds and patches (connections) memorized, that is some machines can't just flip between sounds because they only have knobs which must all be adjusted for each new sound the user wants.

Sample CD's are available to fit certain requests in this area, where lots of well known sounds are recorded and can be loaded in a computer or sampler (a machine that can play back samples of sounds), which is enough sometimes to put a nice synthesizer sound in some music or sound production. That does not make it possible to play around with the sounds, each sound must have its own sample, and loading those samples may not even ne such an easy job. Just twisting a few knobs for some variation isn't possible with this method. Also, the quality is not necessarily even acceptable, especially when a sampler is supposed to respond like a synthesizer, when for instance chords (= more than one sample) are played.

In recent synthesizers therefore not only pianos and other acoustical instruments (brass, reed, strings) are simulated (see physical modeling), but also wel known 'classical' synthesizer architectures are simulated by computers. Some modern instruments (e.g. the Clavia's North Leads (1-3), Waldorf Pulse, Novation, Yamaha's dance machines (to some extend)) are nothing but an exterior like a synthesizer with computer in it that mainly simulates classical type of sythesizers.

Many keyboards and other synths also make a spectrum of sounds, but there is a difference when the actual working, the functioning of the analog electronics plays a role in the machine at hand. Not as scientific in nature as a complete circuit simulator, but more at lets say unit-level: the generators, the filters, envelope generators, mixers, etc. Does this work ?

I've tried a few, and listened to a number, and overall: it certainly gives more then enough nice sounds, with a measure of control over it as in a machine-with-knobs. I mention again here that of course it is not the holy grail of synthesis to have only analog synthesizers, simulated or not: absolutely not. In fact I owned (bought myself) digital synthesizers like the DX-7 / TX802, Kawai Spectrum (well, in fact that was mainly to have another midi keyboard), Yamaha TG500, and a Roland Hp3000 digital piano, that have nothing to do with analog synthesis, minus that the principles can to some extend be found back in a few of em. The TG500 is basically a (fixed) sample playback machine with a lot of ready-to-use onboard sounds of quite decent quality, that does offer (4th order) filters, even with resonance. But dont't try to use too much: nothing much fun. They suck, as in: they don't sound effective enough.

I also had a Poly-800, and DW8000 (both Korg), and a Bit One (I liked it enough), had acess to a Korg MS20, which are basically analog synthesizers with fixed structure, and a microprocessor in it. Each has its own limitations, and NONE of them compare favourably enough to the more classical examples like the Moogs, and Prophets, though I definately used each one with pleasure, and they were worth their money. The bit is the 'fastest' and lets say most direct of the bunch, though its filters weren't good (fat, resonant, warm, tracking) enough, and quite some shortcomings of its circuitry were noticable. It at least was made as a machine with big computer buttons and a metal enclosure, and not expensive.

The poly800 was my first bought synth (not keyboard), and was completely worth the buy, even though it has 'only' 4 notes polyphony with two oscilators per voice, only one (!) 'voltage' controlled filter on board, and no touch sensitive keyboard. It has variation enough, strength enough to be interesting, a for that time heavy digital effect unit, and it 'works' decent, plenty of clean enough sounds. I left mine for a DX7, which was no fair trade, also money wise.. But my my, a touch sensitive fm beast with all these sounds and harmonics and harmonics, and modulation, touch sensitivy, and more modulation, keyboard scaling to change em all, and that for 16 real notes, that was worth the extra thousands, yessir. Even with the generally lets say unbeefy rom sounds to start with. I not so long ago had a DX7 for practicing, I still think I'd want one with those 10k or so internet sounds.

The DW8000 is a Machine. It looks spacey yet serious, though not enough to beat some more expensive competition, but it has a heavy, big key 5 octave touch sensitive keyboard, unfortunately the Korg joystick instead of wheels, simple but well readable major LED displays, and sounds thick enough. Lush brass and strings, nice enough bass lines, even 8 oscilator uniso (all at the same note, although that effect could have been better, maybe with some more and more refined random detune), nice enough instrument simulations, and later I also gave it a keyboard split extension. Works good on performances, and it has digital waveforms, which make it a bit more varied then instruments with only the classical saws and pulses. I got rid of it because I wanted the piano, and for a long time missed an analog instrument in the setup I had. The warmth and breadh of such a machine compensates for the abundant amount of harmonics from the digital FMs and others, and it can blend together nicely in a mix (for instance doing sequence productions, or multitrack). Live it is fun to have an analog one, but I only found the jp900 (I think it was, a roland machine with maybe 50 knobs on the front panel) up to the actual ANALOG experience in that respect. The others I used, including a jx8p (also roland) are nice enough, and work warm and present enough, but in the end I found out that a bare DX7 produces more fireworks live than most other things, interestingly enough. Just put that thing on the PA, have the sensitive wheels and aftertouch work on some high pitched solo or harmonics rich bass line or brassy stuff, and its presence is very clear. I practiced (and performed) with the JD9000 (or so) on a heavy laney amp, and THAT may defeat the DX7 soloing that way with some work. That made someone ask if I was going for trying Hendrix on the synth. Major fun.

Apart from that, the DX has not much trouble with its studio thinness on stage is my experience, my TX802 (a DX7 II kind of synth with 8 different voices) on stage and in band rehersals fulfilled its position more excellent then I ever anticipated. Then again, I often prefer the roland piano to play, not necessarily a roland, it just happened to be around a lot for the purpose, and it is know for cutting through the band sound well. Piano-wise, the DX wouldn't offer the same, and I played it for at least 4 years often a few times a week with bands, so I have an informed opinion as to its usefulness.

Now what about those analog kinds? Well, I now have a machine on my desk, which makes a noise when a I play a key that goes oomph, or lets say that has a thickness and fatness that is nice, and directly reminds one of some classical synth sounds. That is satisfactory, but also illustrative of something I knew the aforementioned boxes couldn't realy get right, and now my simulator program can ! Those instruments didn't make it thick or distinctly powerfull enough, the DW8000 was nice enough, but not enough compact, a poly800's switched filter simply isn't asked to do a minimoog imitation, and the (In think it is Italian, Crumar) Bit One, well, not in a million years would someone mistake it for making Prophet kind of strings, it lacks the resonance, the complexity, the subtle variations, and simply the basic sound qualities.

Talking about strings, thats lets say sustained sounds on a synth with chords, and nice warm sounding effects, that requires me to go polyphonic, that is, more than one key pressed and sounded at the same time. Thats being prepared, but at least I've experimented with sounds that start to get good enough to notice in that area. Tone-wise I've made it clear enough what is definately true, that a good instrument, and a good sound are important, even essential for satisfactory playing and listening experience. Not that every sound needs solo qualities and high instant attractiveness, but the limitations in certain instruments and the strong points are felt quite objectively, also over time.

I yesterday made the resonance of the simulated filter constant over cutoff-frequency changes, so that filter sweeps have the same resonance, and thus near oscilation stays near oscillation over a frequency range. Suddenly there's heavyness and woofer rattling air being shifted, instead of a sort of pleasing compact ballsiness. If I'd find a synth in some rack that did this sound at some point, not even at high volume (I'll get there), I'd notice for sure. I'm not sure why it is that a minimoog can be expected to do this kind of thing (as it did on thousands of productions), a bit less rattling, and in some yet unsearched ways considerably more profound, and not many others can, but I'm sure that as straight as possible filter with resonance, without compromise, over edgy enough oscilator waveforms has everything to do with it, and that I've heard software, digital and analog equipment, that did NOT do it right. I've modulated the cutoff sweep also with a lf sine wave, which seems also interesting, and I'm sure the rest of the famous synth's sounds have also to do with lfo and modulation at various points, and the simple idea that they have also volume envelopes, but I've got convincing sounds straight from my software, into the synth prototype, and they rock. Well, rock is just an equivalent term.