. founded march 12, 1995 _| : _____ t r a x w e e k l y # 117 ______________ |___| _ _______/ /\___________________________ / ____________/ /\__\ _ _______/____/_____________________________ / / _________ \/__/ ______\ \_____________________________ / / / `_ . .~ \____\/ _ __ ___ / / / _____ . _ \ __ ___ _/__/\ / / / / /\ _ The Music Scene Newsletter __ __\__\/ _/__/ / ____/ /__\_________________________________ _____ ___ _ / /\/ /___ __________ _ ______ _ ___ \/ /\ / / /____/ \ \ / /\ / __/\ / /\ \ \ / \ /____/ / / \ / \/ /_ \___/___/ \ \_/___/ / \_/ / / \ ___\ / /_/ /______/\/ \ /______/\/ \ /_____/ // \ \ / / / \ / / \ \ \ \_\ \ \ \_\ \ //____/\____\/ / / / / / \______\/ \______\/ \_____\/ \ \ \ \ / / / / \____\/\____\ / / / / _____ _____ _____ _____ _____ _____ / / / /__/ w /\___/ /\___/ e /\___/ /\__ / l /\___/ /\____/ / / __/____/____/____/____/____/____/____/____/____/____/____/________/ / __\ \____\ e \____\ \____\ k \ ___\ \____\ y \__________/ \____\/ \____\/ \____\/ \____\/ \____\/ \____\/WW ------------------------------------------------------------------------------ | TraxWeekly Issue #117 | Release date: 01 Jan 1998 | Subscribers: 1103 | ------------------------------------------------------------------------------ >-[Introduction]-------------------------------------------------------------- As we say goodbye to 1997, we must look at the incredible growth the music scene has seen thanks to the explosion of the information age. As we say goodbye to 1997, we must realize that all of us are approaching the end of another century, and the dawn of a new millennium. Here in 1998, TraxWeekly is still going strong, about to celebrate its third year of reporting and discussing the music scene and the community of people who are responsible for creating it. My heartfelt thanks to each and every one of you for being here, and making all of this possible. For you, the reader, this first 1998 issue of TraxWeekly is here to remind you that no matter the trials of time, no matter the forces that enable and deny us, the ability in all of us to continue with a dream, no matter how worthwhile, no matter how worthless...will forever be part of us. We are happy to present articles on sampling, technology, and Zinc's demotape directory of the year for this issue. Thank you for being a part of the information age. Happy New Year, and welcome, 1998! Gene Wie (Psibelius) TraxWeekly Publishing gwie@csusm.edu >-[Contents]------------------------------------------------------------------ ________ _________________________________________________________________ / ____/_/ __/ \ __/ / _____/ \ __/ __/ ___/_ < \____\ \ \\ \ \\____ __/ __/_\ \ \\____ \_____ \__ \ \ \ \\ \ \ww\ \\ \\ \ \ \ \ \_ _\________\________\\___\____\ \_____\\_______\\___\____\ \_____\_______\ Letters and Feedback 1. Letter from ior 2. Letter from Red Lightning General Articles 3. The Art of Sampling Part Two..................Euji Acha 4. In Tune.......................................Coplan 5. Event Electronics Gina........................Dilvish 6. 1998 Demotape Directory.......................Zinc Closing Distribution Subscription/Contribution Information TraxWeekly Staff Sheet >-[Letters and Feedback]------------------------------------------------------ --[1. Letter from ior]-------------------------------------------------------- From jroth@mailhost2.csusm.edu Date: Mon, 22 Dec 1997 01:35:44 -0800 (PST) From: Jesse Rothenberg To: Gene Wie Subject: Note on MP3 for TW I find it slightly funny that "Blayd Zro" complains about spending months working on perfecting a sample only to have it stolen, and suggests the use of MP3s as a way to stop this 'thievery.' I find this especially funny considering that the primary use of MP3s right now is to rip ENTIRE SONGS of commercial CDs -- songs and CDs that the musicians rely on as their sole method of sustenance and income. It no longer seems so bad that someone is using your sample, does it? Jesse Rothenberg ior jroth@cats.ucsc.edu ------------------------------------------------------------------------------ --[2. Letter from Red Lightning]---------------------------------------------- From redlightning@hotmail.com Date: Fri, 19 Dec 1997 12:06:18 PST From: Red Lightning To: gwie@mailhost1.csusm.edu Subject: TraxWeekly Greetings, As much as I love TraxWeekly and MC5, I think that the debate over the "fairness" of the judgeing is getting a little old. Hell, that was in August! We all know that there are always problems with large contests with popular entries, we just have to try our best. There are always going to be people who liked and disliked the contest. Perhaps they have began to focus more on victory then the music. Anyway, perhaps you could give it a rest? Mike Isaacson Red Lightning ------------------------------------------------------------------------------ >-[General Articles]---------------------------------------------------------- --[3. The Art of Sampling Part Two]------------------------------[Euji Acha]-- Greetings mortals... ...and welcome to the second part of "The Art of Sampling". This part contains a modest introduction to the wonderful pain of noise reduction or "I want this ugly sound to be better", and an explaination of the reverberation and echo functions, that are more powerful than you could think. Before we start The editor used here is Cool Edit. I know you're a lot of Soundforge fans, but Cool Edit is shareware, so anybody can retrieve the shareware version (wich is crippled by the way) and try out what I explain here. And, Cool Edit isn't so bad. And, I don't have Soundforge. :) For the last version of Cool Edit browse to http://www.syntrillium.com For the last version of Fast Tracker browse to http://www.starbreeze.com I - How to remove noise from a sample What I mean by noise is a sound you don't want to have. Cool Edit offers many tools to remove noises, and best results are obtained in combining these tools. You must be awared of the fact that you won't be able to remove every kind of noise... Unless you are using professional hardware and software, this will be out of your range. But don't be discouraged, your simple PC with Cool Edit can do marvellous things. 1 - Noise Reduction Select a part of the wave where you can listen only the sound you don't want to have, or where you can mainly listen it. The most the noise is isolated, the best it is. Select "Noise Reduction" in the menu, and choose the maximum FFT size you can. (when the size is too big, the button "Get Noise Profile" is desactivated), then press "Get Noise Profile". Once the program has gathered the data, close the selection box, and select a part of the wave where you want to remove the noise. Choose "Noise Reduction" and press OK (I advise you to set the precision factor to 9), then play the wave. You will remark that the sound is quite distorded, it is normal since you have certainly not gathered only the noise data. We will try to reduce this distortion. Anyway, before doing this, try to do the same thing, but in gathering the noise data from other parts of the wave. Keep the best result. The noise is generally easy to find when the sound "fades out", it is at the end of the sample. Technical trick : If you still can hear the noise once the filter has been applyed, and if the rest of the sound doesn't seem to be alterated, then you may try to apply the noise reduction function a second time. Also try to change the precision factor and the percentage of noise to remove. The aim is to have a homegenous sound. You should know that the NR (Noise Reduction) function of Cool Edit 96 produces smoother sounds than 95 and 1.52, this is because it "fades" the noise reduction, the algorithm has been upgraded. Read the documentation for further information. Technical trick : When you are working on several different samples from the same source, if you cannot get an isolated noise in one sample, try in the others, and apply the gathered noise on all the others. Because it has been recorded with the same equipement, the noise will be almost the same. 2 - Filtering Filtering a sound is a good way to remove unwanted sounds. The hard part is not creating the filter, as you are going to see, it is quite easy. The hard part is to be able to determine what kind of filter you need... If you don't do this correctly, you will remove too much sound or not enough noise. You have to filter your sample in one of these cases : - you tried the noise reduction function, and even if the noise has been partially or completly removed, the sound has been alterated - you cannot gather the noise - you do not obtain good results with the noise reduction function - you want to play with the filter function :) The noise reduction function you used before is, after all, a kind of advanced filter : it filters out all the frequencies you gather (the noise). If you are familiar with substractive synthesis, this will be very easy for you. It's like playing with the 303's cutoff frequency button. ;) So, if you are such people, you can read this part very fast, just see how we apply the technique to noise reduction. Generally, you will use low-pass filters. A low-pass filter, is a filter that removes the shrill tones, there are also high-pass filters that remove the bass tones. Let's say that your noise is a shrill sound, if you want to remove it, you have to use a low-pass filter. The problem of filtering, is that you will not remove only the noise, but also a small amount of the sound. You certainly know the Dolby B function that all the tape recorders and players have, you remarked that applying this "filter" removes the typical "hissing" noise of cassette tapes, but it also removes a lot of trebble. Even if we loose a bit of the original sound, what matters is the signal/noise ratio. If you loose more noise than signal, then it is fine. That will be the same with filters, we will try to reduce the noise more than the original signal. Technical note : The Dolby B function is more than a simple filter, in fact it increases the high frequencies during recording and weakenes it during playback. Dolby B reduces the noise to about one-third. If you have something better than a lame integrated Hi-Fi system , you may have a Dolby C system, or if you have expensive hardware, a Dolby S system, and per Cthulhu, that's another world. ;) So let me explain you how works the FFT filters, it is (as I said) quite simple. You have a window with two axis, the X-axis (horizontal) is for the frequencies, it goes from 0 (full left) to 22,050 (full right). The Y-axis (vertical) is for the percentage of the frequency you want to keep. It goes from 0 (bottom) to 100 % (top). Keep in mind that the Y-axis can be modified. Ok, but what does it do ? Simple, when you set a point, you tell to the filter : "keep the frequency that corresponds to the X-coordinate of the point at the percentage specified by the Y-coordinate of the point" Look at the schema if you don't understand, it is a mathematic representation of the graph I explained, I hope it will help you. ^ y | a corresponds to the frequency | b corresponds to the percentage to keep b |.......x | . | . | . | . --|--------------------> x '| a x y' So you can set points, and between these points a line will be drawn. Therefore, if you only set a line to the top of the window, the sample will remain unchanged since you asked to keep all the frequencies. If you set a line to the bottom, you will obtain a silence since you asked to remove all the frequencies. By default, there are two points. (so a line) If you draw a triangle (click in the center of the line, put this point to the top, set one another point to the left bottom, the last to the right bottom), you will remove all the bass and the treeble, and keep the middle. But this will be "progressive", it will keep more and more until it reaches 11,025 Hz (the center), then keep less and less until 22,050 Hz. You may even ehance the middle instead of just kipping it. To do so, modify the Y-axis values, and change the 100 % by 125 % for example. The middle will be intensified at 25 % It is important you check the Y-axis values, if it is set to 200 % don't be surprised if you overdrive your sound... You could of course do something more brutal like a low pass filter. Click on the preset, and see how the graph is... As you can see, there are four points. Two of them are at the top, the two other at the bottom. ^ | | |............. | . | . <------------------ In this section, all the frequencies are | . kept. | . | . | ................. <- In this section, all the frequencies |-----------------------------> are removed. ^ |-------------------- This is the low-pass filter caracteristic. If it is a 16,000 Hz low-pass filter, then the X-coordinate value here will be 16,000 The result, is as expected, a filter that removes high frequencies and let the low frequencies pass. (that's why it's called a low-pass filter) To understand well the filter tool, I advise you to look at the presets, see how is the graph, run the effect on a sample with a large spectrum, and listen. You will see that there are a lot of useful presets. It is important you understand how it works, unless you want to try random things and loose your time. Now, the hard part... How to create a filter wisely ? Hum, again we have to use mathematics. Make a "transform‚e de Fourier" (so exotic :) ) that Cool Edit simply calls "Frequency analysis". A small window will appear with a small graphic inside. It simply shows the amount of frequencies, it goes from 0 to 22,050 Hz (left to right). So at your left you have the bass and at your right the treeble. The schema I made is for a bass sound with a middle and shrill noise. You can see three "pics" that appear. It is probably the noise. So we have to use the filter to attenuate these frequencies. Amount of frequency y ^ |. | . | . | . | . | . . <-----------|----------- This might be the noise | ......... ... . . <--| | .... .. ..... -|----------------------------------> x' | x Frequencies in Hz y' There are two methods for that case. One consist in creating one complex filter that will let all the frequencies pass and attenuate those around the pics. The other is to create one filter to attenuate the frequencies around one pic. In our example that would be 3 filters. I do prefer this method since you are working on a pic one per one, so you can really set the filter fine. Also, you will see if you are removing noise or not. So for the first pic in the middle I would create a filter like this : ^ |.......... ................. | . . | .. | | | | | | -|-------------------------------> | It will attenuate this part of the noise. If it alterates the sound too much I will reduce the attenuation. If it seems to do nothing, I will augment the attenuation. Of course, I will undo between each operation. This will be repeated three times, one time per pic. I should obtain a better sample. DO NOT use filters with a too thin band, you will create another noise. Technical tip : When moving the mouse in the graphic, the coordinates of the points will be displayed. Unless you are a kind of bionic guy, you need this. :) This one just an example of course, the noise can have many different aspect, and it is only a beginning.... :) You may have a nice graph from 0 to 10,000 Hz and then it becomes garbage, it means that the noise is located over the 10,000 Hz. Remember : a noise is generally a random sound created by electric or electronic hardware, therefore it generally looks strange compared to the rest of the sound in the frequency analysis box. I sometimes try to apply very low-pass filters on bass samples. Like a 5,000 Hz low-pass filter. On a bass drum, for example, it works very fine, you will remove almost all, if not all, the noise and won't alterate the sound. Your noise may also be something else than three pics. It may be a regular "hiss" that is not clearly distingishable in the frequency analysis box. (Like the cassette tapes "hiss" I talk about) If you cannot get rid of the hiss with the noise reduction function, and if you cannot see it in the frequency analysis box, you may try the rough method. This method consists in using a low-pass filter and increasing it's value. You start from 4,000 Hz and stop only when you can hear the noise. (You undo between each filter to return to the original sound) Then you do the same with a high pass filter, you start from 20,000 Hz and decreases this value until you can hear the noise. You will have determined the noise "band". If it is not too large (18,000-20,000 Hz for example), then filter this band out. You may also try the reverse of this method. (For the low-pass filter you start from a high value, and reduce until you don't hear the noise anymore, for the high-pass you start from a low value) This is a long work, but if you really want this sample to be better, you have to do it. Desesperate trick : If you can see a lot of garbage everywhere in your sound, except in some places you could use a band-pass filter. This is the reverse method if you prefer. You are not removing the noise, you are removing everything except the clean parts... The sound will certainly be denatured, but the noise will be removed for sure. :) And don't give up ! With Cool Edit and the procedures I described, I succeed in removing almost all the noise from a sound played from a tape wich was a recording of a vinyl. So I removed the classic noise made by used vinyles and also the "hiss" of the tape. (the guy didn't know dolby C obviously ;) ) It just requieres patience and experience. 3 - Volume enveloppe setting Ok, here we could make the stuff sounds a bit smoother, it is easy and generally increases the sound's quality. But what you will do now depends on how your sample sounds, if the whole sample sounds (relatively) clear, you just have to fade the very beginning and the very end of the sample to zero. Remember what I've said, the first and the last value of a sample MUST be zero. Sometimes, the beginning of the sample, the loud part, is fine, but the rest is... ...strange... Do not panic, you can change this (a bit). Just mute the end (not a too big part of the end), fade out the mute point (to avoid sample clicking) and reverberate the sample. (with the reverb function wich I'm going to explain). That's better isn't it ? This method works generally fine. If your sample will be looped, make sure that the remaining noise (if any) is "regular". It is VERY important, if ever your noise "appears" and "disapears", your sample will be horrible. This is easily noticable in loops. Generally working on a loop is harder. (You may remember that I already said it in the first part of the Art of Sampling) So, that's it. You know enough to remove the dirt in your sounds. Go ahead and clean all your samples now. :) I might explain further how to improve the quality of a sample in the future if I'm asked to. (what a black mailing ;) ) II - How to use the reverb and echo functions 1 - What are the differences between an echo and a reverberation ? In fact there is only one difference : the delay between each repetitions. You all know what's an echo, it's a sound repeated with a certain delay and with a volume variation. When you shout "HELLO !" in a large empty room, you can hear "HELLO ! ...Llo ! ...Lo ! ...o !" because the wave of you voice "bounces", and come back to you. Because it bounces in different places (therefore different distances), it comes back to you with a different delay. The larger the delay is, the more you can feel the echo. When the delay becomes too small, it is called a reverberation. Our ears, like our eyes, are limited. We can only see 24 images per seconds, and we can only distingish sounds over a certain delay. (I'm not sure about the value) Actually, it's not really due to our ears, but more to our brain. (but the kernel v3.0.0 fixes the bug. ;) ) So, a reverberation is an echo with a delay under that fatidic time. That should be as simple as that, but it's not. The algorithmes used by Cool Edit for the echo and reverberation aren't the same, the reverberation algorithm is much more advanced and complex. (much more slow too...) Worst, you can do reverb with the echo function if you set a very small delay (5 ms for example). 2 - Using the echo and reverberation functions with wisdom When you listen to a natural sound, that is, a sound that is not recorded, (a car passing by for example) there is a complex alchemy operating. The wave of the sound goes in every direction and change its direction and its speed according to the matters it encounters and according to the interferences generated by other waves. Hum... The car passing by is not a good example since there is a doppler effect, and that's out of the topic... When you are underwater, haven't you remarked that everything sounds louder ? It's because the sound "goes faster" underwater. But you also remarked that it is quite distorded, like a strange flanged sound... It's because the nature of the wave is altered by the variations of the fluid (the water)... This applies to all the materials, the wave is modified when the environement is modified. I won't explain further these complexes phenomena, many of you hate Physics. Anyhow, if you want to create an empty hangar ambience or a concert hall ambience, you have to know few things. First, to obtain good results, you will have to use both echo and reverberation functions. (Echo chamber is quite different and should be only used with stereo samples) I strongly suggest you to not to use the echo function alone since it can be reproduced easily in a tracker. If you want to create a very complex reverb and echo effect, you may use the echo function, then the reverb or only the reverb. It is also good to know that these functions are very efficient to "spatialise" a sample. (That is converting a mono sample to a stereo one) But unless you want to use real stereo samples in your modules, like I did in "Hydres" (Hydres means Hydras btw), you don't have to worry about this point. If you have an echo chamber or an effect processing unit that allows you to add reverberation and echo to a sound, it could be a good idea to use it (them) instead of a sample editor. Even if it will imply some quality loss (because you will REsample), you will obtain exactly the effect you want thanks to the realtime controls. If you use a sampler (a real one, not a MS-1 :) ), forget what I said about quality loss. (you can do numeric transfers) The answer to "Can my S 3200 XL do it ?" is yes and I hate you. ;) 3 - Few words about the parameters I would like to discuss about the parameters, even if the Cool Edit's help text do it well. a - Echo function This is simple function, it just repeat the sound with a certain delay and reduce the volume according to a falloff ratio. All paramaters can be set differently for both right and left channels. I won't describe all parameters since the the help text of Cool Edit do it. I just want you to pay attention to few things. The "Continue beyond selection" is very important. It means if you select a part to be echoed, it will echo all the rest of the sample. Fortunalty, it won't continue over the window, it will stop at the right side of the screen. (it also applies if you zoomed in) It is a very powerful setting, you can echo only a part of the sound, and it gives incredible effects with digitised sentences or percussion sets. There is also an "Equalizer" that allows you to set what frequencies will be absorbed. Do you want your echo to go in the shrill or in the bass tones ? This is the tool you have to use to set it. The "Delay" function is a kind of "one time echo". It will create ONE echo. In fact it just repeat the sample after the specified delay and with the specified volume. b - Reverb function The reverb function is much more advanced, and it is explained in the help file of Cool Edit why. As they say, using the echo function first, then the reverb function is the best way to obtain a realistic and good reverberation. This function tries to reproduce the complex physics phenomena that occur in the nature, that's why it takes a little bit long to compute on slow computers. (even on faster ones, *grings*) "High frequency absorbtion time" fills a room. The slower is the time, the more the place seems empty. The "perception / timbre" setting is, to me, the magic one. It really changes the nature of the sound. The more this value is important, the more it sounds "wide", "big", "large", etc. Try high values on percussions, it sounds great. (but it's not really original, I admit) The two mixings parameters must be used very carefully. The "mixing - original signal" determines the amount of original signal you want to keep. As you can except, if you set it to zero, you will only hear the reverberation. So, be careful with this one. The other "mixing - reverb" is generally set to 100 %. Unless you want to do special things, you don't have to change it. If you reduce the value, it will sound less natural, but can gives great result. It will give an attenuated reverberation, more a "created" reverberation if you prefer. To be honest with you, I generally use the reverb function not to obtain a realistic reverberation, but to add a particular feedback to a sample. For example, the (wonderful ?) bass drum of "Ioa Petro", another outstanding track by Euji Acha :), was obtained with the reverberation function. The original sample was a simple electronic bass drum sample with nothing particular, believe me. 4 - Creating an echo in Fast Tracker II If you did understand well what I was talking about, you can reproduce an echo and a reverberation in Fast Tracker II. If you repeat the same sample and decrease it's volume, you will create an echo. For example, this will echo the sample 1 : E-6 1 .. 000 ... . .. 000 ... . .. 000 ... . .. 000 E-6 1 20 000 ... . .. 000 ... . .. 000 ... . .. 000 E-6 1 10 000 Yes, it requieres three channels, if you repeat the sample in the same channel, the echo will be different, (if the sample is long enough) but you may not notice it if many other channels are playing. This is why I suggest you to try both and to keep the result you prefer. (as always) A thing you should do to improve your echoes is to play with the panning. Many of you just track in ignoring the panning, and that gives mono modules. You have two stereo channels... Use them ! Here is an example : E-6 1 .. 880 ... . .. 000 ... . .. 000 ... . .. 000 E-6 1 20 820 ... . .. 000 ... . .. 000 ... . .. 000 E-6 1 10 8D0 wich is equivalent to : E-6 1 P8 000 ... . .. 000 ... . .. 000 ... . .. 000 E-6 1 P2 C20 ... . .. 000 ... . .. 000 ... . .. 000 E-6 1 PD C10 Your echo will be certainly deeper. Playing with the stereo is an excellent way to improve the quality of your module noticeably. You can also create a reverberation in looping the end of the sample in a ping-pong loop and in creating an appropriate volume enveloppe. The problem here is to fade the volume out fast enough not to hear the loop. (it won't sound real unless, but again, you may like it) Also, this will create a reverberation : E-6 1 10 000 E-6 20 901 E-6 .. 902 With the sample offset function, we simulate a very short delay between the samples. That delay is short enough to create a reverberation. (if the speed isn't too low) If you prefer, you can use the delay function, like this : E-6 1 .. 000 E-6 20 ED1 E-6 10 ED2 It won't sound exactly the same. I generally prefer to use the sample offset function. You can try this methods and combine them, you will certainly obtain great results and save disk and memory that high quality samples requiere. (87 kb/s for one 44,100 Hz - 16 bits channel) Of course, the effects obtained won't be as great than those you will obtain with Cool Edit, but here you can synchronise them perfectly with the song... So, I would say that you should use Cool Edit to create complex echo and reverberation effect. But a simple echo can be rendered very well by the tracker, so don't waste your time... (and (y)our memory :) ) This concludes the echo & reverberation chapter. And that's (at last ?) the end of The Art of Sampling part two, I hope you enjoyed it... If I am asked to, I might write a third part... If you would like to comment or correct anything in this article, feel free to enter in contact me. May Cthulhu be lenient enough to spare thy insignifant life. Euji Acha / Cloud Nine ------------------------------------------------------------------------------ --[4. In Tune]------------------------------------------------------[Coplan]-- Welcome back for another installment of In Tune. I've made a few changes to the articles format this week. Mainly, I added a little information section at the end to not only help readers find the song, but it also gives some of the fine details that I don't always discuss within the song (like filesize). I give this idea credit to Mairsil who wrote me earlier in the week with his comments. Apparently, those of you who use Netscape Mail, AOL, Eudora (I assume) can simply click on the link in that section, and the song gets delivered directly to your hard drive. Thanks for the comments Mairsil. All said and done, lets get down to the meat and potatoes. This week, I am reviewing a song by Pyro called "Sad." The song is basically an orchestral peice, though it has quite a bit of influence from modern music styles. This is Pyro's first orchestral tune, and I congratulate him on his success. He definately did his research when composing this song. Pyro brings you into the mood with an introduction consisting of only a few percussion instruments: A distant snare, a tom and a hihat. The repetitive nature of the tom begins as the controlling instrument, and one that succesfully defines the character of the song. However, since it is such a key instrument, I think it could have benefited from an echo, or at least a panning envelope dodging back and forth creating an echo effect. The sample data contains an echo, but playing it static on the panning spectrum doesn't allow the fullness and realism that an echo would add. Shortly after the song begins...the flute lead instrument is introduced, as well as the Atmosphere sample. Those of you GUS owners out there will recognize this sample. Your PAT libraries are wonderful sources for song samples. In this case, Pyro used the Atmosphere sample as a mid-range string section. It is the most effective use of this sample I have ever seen (I have 4 songs on my hard drive using this sample, including one of my own). I imagine this song was written on a GUS, and that brings up something that should be brought to your attention. In the case of this song, the Atmosphere sample would not have the same sound quality if Pyro chose to play it in only one channel at a time. The GUS has a unique tendancy to change the sound quality of two samples played at the same note at the same time...not much, but enough to notice. On a card like the Sound Blaster, this would simply just get louder. I listened to "Sad" on both sound cards, and the doubleing of the notes using the Atmosphere sample sounds nicer on the GUS. Fortunately, it doesn't sound bad on the Sound Blaster AWE32 either. It's just a problem to be aware of. I must compliment Pyro on his percussion. Percussion useing a drum set is a whole different animal than useing a timpany and other orchestral instruments. Pyro had no problems dealing with the Timpany. He even made sure that his Timpany was in tune, which is something many people overlook. Timpanys are considered melodic instruments, so make sure they are tuned. Order 28 starts us with a significantly well done drum solo. Notice in channel 3 there is that little Bubble Blip sample...barely audible, until you take it out, then you know it was there. It's amazing the simple things that can add to your song in general, let alone Pyro's drum solo. The other section of the song I would like to point out starts around order 36. Here, we have a near replica of the beginning, with the addition of that bubble blip sample again. But this time, Pyro cuts to the chase. Suddenly, the flutes come back in a more vibrant manor. They perform their own solo with almost no other instruments in ear shot. This adds to the character of the solo, and upholds the theme of the song. Belive it or not, a base line would make the solo seem happy (I tried, belive me). This style is common in the Star Wars soundtrack (Empire Theme -- violin solo). Eventually, the solo fades back into the style that the rest of the song is written in. Though I think it is a first, I am going to comment on his song text. I am one of those people that reads song text, belive it or not, we do exist. I personally like the big letters made of periods and colons. But the real reason I am commenting is that "quasi-orchestral" statement. As far as I'm concerned, this was a full blown orchestral song. I didn't find any samples within the song that couldn't be taken from a real live orchestra. Its orchestral, and a good one at that. In closing, I feel that I may have come off as being a little hard on Pyro in some spots. I don't intentionally try to seem like I'm being over criticle of his work. Of the orchestral pieces out there, "Sad" is one of the better ones I've heard. I have a lot of respect for the song, and a lot more for a tracker willing to try something new and perhaps not as common. If this week's article seemed silly, keep in mind, it is final project time. Until next time, cheers. --Coplan Song Information: Title: Sad Author: pYro Filename (zipped/unzipped): 237k File Size (Zipped Unzipped): 246k Source: http://www.hornet.org/music/songs/1997/p/py_sad.zip "In Tune" is a regular column dedicated to the review and public awareness of newly released tracked tunes. If you have heard a song you would like to recommend (either your own or another person's), I can be reached at the following address: coplan@thunder.temple.edu Any format playable in either Cubic Player or Impulse Tracker is acceptable. I review single songs only (no musicdisks). Please do not send files over 1MB without first contacting me. ------------------------------------------------------------------------------ --[5. Event Electronics Gina]--------------------------------------[Dilvish]-- Suggested retail price: $499 e-mail dilvish@cyberspace.org for up-to-date pricing and availability. The low down: I've been using the Gina for about a month now, and I have to say, out of all the audio gear I've used and tested building and selling digital audio workstations, this card is the single most amazing piece of gear that I sell. It's become the backbone of every mid-range workstation I build. Thumbs up. - eric The story so far... About a month ago, a client came to me and asked me to build them a computer based audio workstation. Since that sort of thing is right down my alley, and I had the ability to purchase gear wholesale, I decided to go for it. The first think I had to consider was the computer. I decided on a p233MMX, to minimize waiting for effects to be processed on large audio files, for the same reason, I decided they should have 128 megs of RAM. Since they would need to record and edit large amounts of audio data at once, I set them up with two 6.4 gig hard drives, capable of sustained transfer rates up to 33 MB/sec. I thought they could also use a nice 17" display, and a decent video card so they had a lot of desktop space to work with. I also installed a 120MB UHD IDE Floppy and a CD-R, so they could write backup masters to CD, and exchange large audio files between offices. Now I just had to decide on an audio interface. I started calling my suppliers, and asking for recommendations. One of them told me to check out the Digidesign Session 8. I've got to give Digi a break, because the card is more than a year old, but I wouldn't use one of those if you paid me. It *requires* an external SCSI hard drive to operate, on which, the partitions are limited to 2 gig. It doesn't have native Win95 drivers, and the software support for it is very limited. I considered a number of other options, and decided to try out Event Electronics Gina, because I was very impressed with the same company's 20/20bas near field monitors, so I was counting on their commitment to quality. It was a good bet. I connected it to an Allen & Heath Mix Wizard 20:8:2 mixing console (a very nice board considering the $1500 price tag) plugged in a pair of Genelec 1029A studio reference monitors (winner of this year's Technical Excellence and Creativity award for near field monitor technology) and prepared for the worst. First, I maxed out the volume on the channel 1/2 outs, cranked the speakers to full, and enjoyed the silence. No audible noise, until I cranked the monitor knob on my mixing console a few notches, and /that/ noise came from the console circuitry.. (what can you expect for a $1500 console?). Test number one was passed with flying colors. On to number two. I began listening to a lot of my favorite music. Songs that I was very familiar with. I know how the mixes sound. Or, at least I thought I did. I quickly began to discover little nuances in the mixes that I'd never noticed before... strings, layered percussion effects, and all kinds of textures that I never imagined could have been recorded to a CD all came through crystal clear, even though they were buried in the mix with other sound cards. Next, I did a number of tests, analyzing wav files with a spectral analysis, re-recording them through the Gina, and analyzing again. I carefully noted any differences. I did the same with a diamond monster 3d sound card, and a sbawe32. Both of them were totally blown away by the Gina's score. And the Gina had another one up on them... Usability... The Gina has a stylish little breakout box, connected by a thick cable so all the audio signals are safely away from the interference of your computer. The i/o consists of 2 1/4" unbalanced TRS ins, 8 1/4" analog TRS outs, and an s/pdif digital i/o on the card (RCA terminated coax). I connected all the analog i/o to the mix wizard, being careful to keep them away from power cables and other interference. I connected a Panasonic SV3800 DAT to the digital i/o. Everything worked like a charm the first time. I didn't have to curse, or consult the docs. Installation of the card and software was very easy and straight forward (and quick! It was over in minutes). You don't have to be a recording engineer to use the card, but if you use it long enough, you'll probably want to become one. Selecting output channels is a simple matter in most multi-track software, and if you can only use one i/o channel at a time, switching between the s/pdif digital and analog 1/4" connection is so easy, I assigned the Win95 output selection to a hot key, so all I have to do is hit the hot key and click twice. Event Electronics is becoming one of my favorite audio gear companies. They certainly scored on the echo family of computer sound cards. If you can't afford the Gina, you might want to try the Darla, which is essentially the same thing, minus the breakout box, and digital I/O. If you need more I/O options, wait for the Layla to start shipping. It's the same thing with a rack mount box, *balanced* analog 1/4" i/o (8 in, 10 out), word clock i/o, and midi in/out/through. - eric hamilton dilvish@cyberspace.org Vital stats: Audio Performance (Analog in to analog out) Frequency Response: 10Hz-22kHz, ±0.5dB Dynamic Range: 98dB THD+n: <0.003%, 20Hz-22kHz, A-weighted Hardware + PCI bus master interface + Two analog input channels with precision 20-bit 128x over sampling analog-to-digital converters + Eight analog output channels with high-performance 20-bit 128x over sampling digital-to analog converters + S/PDIF I/O with up to 24-bit resolution + On-board 24-bit Motorola 56301 DSP (66MIPS) + 24-bit data resolution maintained throughout internal signal path + EasyTrimTM automatic input gain adjustment circuitry + Audio interface box with unbalanced 1/4" connectors + Multiple sample rates from 11kHz to 48kHz Event Electronics on the web: http://www.event1.com/ ------------------------------------------------------------------------------ --[6. Demotape Directory]---------------------------------------------[Zinc]-- For the new year, I've included a new listing from the KFMF! It's not reeeeeeally a demo tape, but it's got several audio tracks including one from me, so I might as well list it :) I recommend that everyone orders it if you haven't already done so! Note: You may notice I've slimmed down the intro. I decided it wasn't really all that important. I'll only include brand new text from now on. Also new, a WWW format of the Demo Directory is coming soon! I will keep you posted in this column. -------------------------------------------------------------------------- Demotape Directory - December 1997 -------------------------------------------------------------------------- B00MER - Negative Youth - molotov.bliss CD - $13 US + SH Industrial/Techno September 1996 (re-updated October 1997) b00mer@kosmic.org AND boomer@iglobal.net http://oblique.kosmic.org -------------------------------------------------------------------------- B00MER - anonymous.hate - molotov.bliss CD - $15 US + SH Industrial/Techno/Experimental October 1997 b00mer@kosmic.org AND boomer@iglobal.net http://oblique.kosmic.org Co-Produced by Stein. -------------------------------------------------------------------------- bibby - Subsequence - Seclusion Records CS - $5 includes SH Techno-Rock October 1996 bibby@juno.com Original .ITs unavailable on the internet. -------------------------------------------------------------------------- Electric Keet - version one point zero beta CD - $15 or CS - $10 Everything. Classical to techno January 1997 tracerj@asis.com AND http://asis.com/~tracerj/ek.htm 18 tracks, five exclusive to CD. -------------------------------------------------------------------------- IQ and Maelcum - FTZ "Nothing Is True" CD - $8 US + S&H N/A 1995 maelcum@kosmic.org AND www.kosmic.org/areawww/ -------------------------------------------------------------------------- Kosmic Free Music Foundation - Kosmic Archives CD Volume 2: 1997 CD - $15 + S&H Various February 1998 maelcum@kosmic.org AND http://www.kosmic.org/store/cd.html CD-ROM data PLUS audio tracks! -------------------------------------------------------------------------- Mental Floss - Grey Matter CS - $10 US mixed techno N/A andrewm@io.org AND www.io.org/~andrewm/greymatter -------------------------------------------------------------------------- PeriSoft & SupaMart - Live Inside Your Computer CS - $6 US Ambient/Trance/Techno July 1996 mwiernic@pinion.sl.pitt.edu AND supamart@servtechcom -------------------------------------------------------------------------- Subliminal - Mindscape *COMING SOON* CS (110 min.) - $10 US Various December 1, 1997 (tentative) sub@plazma.net AND www.plazma.net/subliminal -------------------------------------------------------------------------- -------------------------------------------------------------------------- All listings follow this format: Author/Title/[Label] Format/[length]/Price (CS = Cassette, CD = Compact Disc, S&H = Shipping Costs, US = US Funds) Style(s) Used Release Date Contact (email/WWW) Other Think tracked music is commercially viable? Prove it! Support the scene! Suggestions and comments are welcome. Lemons are not. - zinc / rays@direct.ca ------------------------------------------------------------------------------ >-[Closing]------------------------------------------------------------------- TraxWeekly is available via FTP from: ftp.hornet.org /pub/demos/incoming/info/ (new issues) ftp.hornet.org /pub/demos/info/traxweek/1995/ (back issues) /pub/demos/info/traxweek/1996/ /pub/demos/info/traxweek/1997/ /pub/demos/info/traxweek/1998/ TraxWeekly is available via WWW from: www.hornet.org, under section "Information" and subsection "TraxWeekly." To subscribe, send mail to: listserver@unseen.aztec.co.za and put in the message body: subscribe trax-weekly [your *name*, NOT email] To unsubscribe, mail same and: unsubscribe trax-weekly (in the message body) Contributions for TraxWeekly must be formatted for *78* columns, and must have a space preceding each line. Please try to avoid the use of high ascii characters, profanity, and above all, use your common sense. Contributions should be mailed as plain ascii text or filemailed to: gwie@csusm.edu whenever, and it shall be published in the next newsletter at the discretion of the editor. TraxWeekly is usually released over the listserver and ftp.hornet.org every week or so. TraxWeekly does not discriminate based on age, gender, race, or political and religious views, nor does it censor any points of view. The staff can be reached at the following: Editor: Psibelius (Gene Wie)..............gwie@csusm.edu Writers: Atlantic (Barry Freeman)..........as566@torfree.net Behemoth (David Menkes)...........behemoth@mscomm.com Bibby (Andrew Bibby)..............bibby@juno.com Coplan (D. Travis North)..........coplan@thunder.ocis.temple.edu Jeremy Rice.......................jrice@hensel.com Mage (Glen Dwayne Warner).........gdwarner@ricochet.net Nightshade (John Pyper)...........ns@serv.net ascii graphic contributors: Cruel Creator, Stezotehic, Squidgalator2, Thomas Knuppe, White Wizard TraxWeekly is a HORNET affiliation. Copyright (c)1995,1996,1997 - TraxWeekly Publishing, All Rights Reserved. >-[END]----------------------------------------------------------------------- :: ::: : . ..... ..............................:::.................:.... ::: : :::: : .::::. .:::::.:::. ..:::: :::: : :: :: ::: .:: :: :: WW:::: : ::. :: ::: .:: :: .:: :::: : :::.::. ::: .:: .:: .:::::... :: :::.. ... ..: ... ..:::::::::::::::: .:: .::::::: :::::::: ::::::.. ::: ::: ::: : until next week! =) .. ... .. ....... ............... .................:..... .. . :