This is a collection of answers to users emails with various questions about the program. If you have a question to ask about, please check if it has been already answered in this list.If you post a message to report a problem you're facing with the program or a possible bug, please specify as much as possible your hardware (especially what soundcard) and software configuration (n-Track version, Windows version etc.) and how to reproduce the problem.
1.a - Q:I can hear jumps in the music when playing back a song. A:If, even with a few tracks you can often hear jumps in the music, try to increase the buffer size and number in the preferences dialog box. Increasing this parameter however will increase the time delay between any action done with the mixer and its audible result.
1.b - Q: I get messages such as "Error opening soundcard" while trying to record the second track, so I’m only able to record the first track. A: Try to enable the option "mono playback while recording" in the preferences dialog. Also, do the “full-duplex test” in the Preferences/Wave devices dialog box. Some soundcards work in full duplex only when recording with certain parameters : if the test fails, try to make some changes in the recording parameters and try again.
1.c - Q: How can MIDI tracks be included in the final mixed down wave file? A: You can record the MIDI files to wave files, activating the MIDI out as a recording source from the soundcard's mixer or Windows volume control. Now just record a new track clicking on the record button. This newly created audio track will contain all the MIDI tracks, and you will be able to use it for the rendering. Not all soundcards can record from their MIDI out so you may have to physically connect the line out of the soundcard to its line input jack, but doing so be aware of the possible feedback problems. Otherwise you'll have to use a software MIDI synth, but the result is not guaranteed to be equal to how you hear the MIDI file through your soundcard synth.
1.d - Q:The program keeps on saying "Creating npk file" as soonas I import or record a wave file. A:The program checks if .npk files (which contain information to speed up the display of waveforms) are up to date checking the file dates. If the computer clock is not set correctly or if for some reason a wave file date is incorrect (for example set in the future) the program will not work. You can adjust the computer internal clock double clicking on the Windows clock on the right side of the taskbar.
1.e - Q:What settings in the Win95->ControlPanel->Multimedia dialog should I use for multiple cards? A:Don't bother with Control Panel/Multimedia settings. They serve only if you choose WAVE MAPPER as wave device. Select directly the desired device from the program preferences.
1.f - Q: Do I need a fast CPU to use effects ? A:Faster CPUswill be able more effects and/or more tracks. If you don't need too many tracks chances are that you'll be able to use some effects even if your processor is not the faster in the market. The best thing is to try: if you can hear jumps in the playback or the system nearly freezes, then it means that it hasn't enough power to do that.
1.g - Q: How can i bring a song off of a regular audio CD into n-Track for editing/mastering? A:You can either record the CD tracks using the soundcard (recording a wave file having selected "CD-in" from the recording view of the volume control), or better directly extract the audio data from the CD using a program that supports audio data extraction. Most CD-R programs, such as Easy CD Creator, do that, or you may use a dedicated "CD Ripper" program.
1.h - Q: How can I create two or more CD tracks that cross-fade into each other, without silence in between? A: You can do this with n-Track using the crossfade function (overlapping the two wave files on the same track and pressing the crossfade button) or alternating the songs to be mixed in two separate tracks and using the volume drawing to apply the fades. Then you would be able to create a final wave file with the whole mix (File/Mixdown command) and use it to burn a CD. Some CD-R programs allows you to create multiple CD tracks from one wave file specifying each track start (index) time. Otherwise you can manually split the wave file in pieces, using n-Track's cut and paste functions. To do this efficiently it's better to start cutting from the end of the song. When you physically create the CD, make sure you use the "Disc At Once" burning mode (which is not supported by some older CD-R device). "Track At Once" mode always introduces 2 seconds (or sometimes less) of silence between CD tracks.
1.i- Q: How is the quality of the processing of n-Track and the plug-ins? A:The quality of the processing is as good as your wildest desires... n-Track and all FASoft plug-ins process the signal with 32 or 64 bit internal precision, while the signal being processed is typically 16 or 24 bit. Any imaginable processing adds noise, but the noise introduced becomes smaller as the internal precision grows. With 32 or 64 bits the noise is absolutely unnoticeable (consider that most hardware digital audio devices, such as guitar effects units or Hi-Fi DSP processors, typically have 16 or 24 bit precision).
1.j - Q:How many non overlapping wavs can a single track contain? A:The number of non overlapping files is not limited. Adding different wave files to a track is useful if for example you are recording a guitar solo track and there are two solos in the song: instead of recording the two solos together in a track, or recording them in two wave files and associating them to two different tracks, you can record both starting from there actually the solos should start and keeping the two wave files in the same track, so that volume, pan and effects settings will be the same. Just what happens with a "physical" multitrack recorder.
1.k - Q: Does the size of the rendered file depend on the number of tracks of the song? A:The size of the rendered song depends on the length and the sampling frequency, not on the number of tracks. To calculate it: sampling_freq [samples_per_sec] * length [seconds] * 4 = size [bytes]
1.l- Q: Why do commercial CDs have higher volume than the CDs I produce with n-Track? A:Commercial CDs sound louder because they are heavily compressed. You can obtain a similar loudness using a compressor plug-in on the master channel, or even two chained compressors with different settings. For example you may first insert a compressor with actual (soft knee) compression settings, and then a compressor with brick-wall-like limiting settings (with the compression curve knee near 0 dB with very low attack and release times). Experiment with this settings until you obtain a good volume without (excessive) clipping.
1.m- Q: Can I import a wave file directly from a CD-Rom? A:No, you must first copy the wave file to an hard disk. The program needs to write the .npk file (needed for displaying the waveform) in the same location as the wav file, and that's not possible with a CD Rom. Even if it was possible, you would get horrible performance as soon as you'll add more than 1 wave file from a CD rom: the program reads small pieces of data from each wave file, jumping from one file to another, and even very fast CD-Roms are very slow at this because they have poor seek times (I think at least an order of magnitude greater than hard disks).
1.n- Q: How can I line up two tracks? A:You can move a track wave file’s offset simply pressing the four arrows button (or holding the CTRL key) and then dragging the waveform's rectangle sideways. Hold the shift key to move the file only vertically. When the "Grid/Snap to grid" menu command is activated the track's offset will snap to the nearest grid line.
2 - Soundcards & hardware configuration
2.1 - Q: Is there any particular setting for my soundcard? A:Special settings and notes for a few soundcards are listed in the Soundcards notes database
2.a - Q: What is the best soundcard to use with the program? A:The choice of the soundcard can depend on many factors, among which price, sampling quality, number of inputs, on-board preamp, MIDI synth etc., so there isn't a definitive answer to this questions. However here's a few soundcards that have proven to work well with the program:
General purpose cards
Ensoniq AudioPCI (also distributed as "Creative PCI64"): cheap (~25-40$) but has a very good recording quality
Creative PCI 128: the audio part is similar to the PCI64 (good)
Creative Soundblaster Live! is quite good for both audio and MIDI. It has the characteristic that it internally always works at 48000 hz and for different frequencies simply does a sampling frequency conversion. This seems to happen also with digital I/O. This can make the card less attractive if you plan to work at 44100 hz. The only difference between the standard and the Value edition seems to be the digital I/O, so if you don't plan to use it the Value model is probably a better buy.
Creative Labs SB16-32-64 work well with the program apart from the 8 bit playback while recording limitation, which you can extensively read about in the n-Track's help file (Troubleshooting/"Creative Labs's soundcards issues" help topic). A good solution to this is to use a second soundcard for playback during recording.
Audio Recording dedicated cards
Gadgetlabs Wave4 (2 stereo inputs, without preamp) and 8/24 (8 in, 8 outs, 24 bit / 48 Khz)
Event's Gina, Darla and Layla multichannel cards (with 20 bits converters, actually seen by Windows as 24 bits ones, so that they may be exploited using n-Track support for 24 bit soundcards)
Midiman Delta 66 (6 ins - 6 outs, 24 bit, 96 Khz) and Delta 1010 (10 ins - 10 outs, 24 bit, 96 Khz)
2.b - Q:I hear a strong noise during playback that I don't hear with other programs(applies only to Creative Labs soundcards) A: Make sure you turn off the recording vumeter before starting the playback. This must be done with Creative ISA cards (SB16, 32, AWE32 and AWE64) to overcome the limitation of this cards that lets them work in full-duplex only with 8 bit playback. Read the "Troubleshooting/Creative soundcards issues" help topic for more details.
2.c - Q:I can hear the signal that I want to record, but the n-Track recording vu-meters don't move A: Hearing the signal coming from a soundcard input and recording it are not related things: for example with most soundcards it is possible to record the signal coming from the line input while monitoring (i.e. hearing from the soundcard's output) the signal coming from the mic input. You set which signal to monitor from the playback view of the Volume Control, while the signal from which to record is selected from the recording view.
2.d - Q: I cannot hear the 'live' sound input while playing back the song I am trying to add a track to. I can add new tracks, but I cannot hear them as they are being recorded. A: You have to activate input monitoring from the soundcard mixer software or Windows Volume control: from the recording sources view make sure that the checkboxes corresponding to the desired recording sources are checked, and in the same manner from the playback view activate the input sources you want to hear. You may choose any combinations: monitor but don't record, record but don't monitor etc..
2.e - Q: I’m having feedback problems (the tracks recorded after the first all contain the preceding tracks) A:Make sure you have selected the correct recording sources, and disabled all the other sources you are not using, in particular the "Wave Out" source. To do so: run the Windows Volume Control (Start Menu/Accessories/Multimedia (or Entertainment in Windows 98)). Chose the Options/Properties menu command, select your soundcard and click on the “Recording” radio button. In the options dialog box make also sure that all the relevant signals are not hidden: the dialog box shows a list of the sources the mixer will show, and sometimes important sources may be hidden by default. After you click on OK the mixer will show the view of all the recording controls. Now remove the checkmark below the level slider of all the sources you don't plan to use, or don't know what they are for. Usually you will only need "Line In" or "Mic in". Some soundcards (such as some Gravis's) always record the wave output on one channel (left or right) so to avoid feedback is necessary to pan all output to the other channel. The "Getting Started/Setting the recording levels" help topic contains a brief explanation of how a soundcard internal mixer works.
2.f - Q: How can I improve my system to be able to handle more tracks? What components should I buy (RAM, Hard Disks, CPUs)? A:Adding RAM above say 32/64 Mb won't help a lot since the program loads the wave files directly from the hard disk, and as soon as the number of tracks grows it would be unlikely that you could have enough RAM so that the system would be able to cache all the files. The things to look for are fast hard disks (UltraDMA or possibly SCSI) and a fast CPU. Regarding to the CPU brand, Intel processors should be preferred because their floating point performances are considerably better than that of their competitors, and the program uses a lot of floating point math. The difference becomes particularly noticeable when applying realtime effects.
2.g - Q: When using two soundcards, the lag value keeps on growing and the resulting tracks are sligthly out of synch. A:Using multiple soundcards, one for recording and one for reproduction or for recording more than one track at a time, will be possible some pitch-shifting effects due to non perfectly matched sampling frequencies of the soundcards. You will be able to notice this while recording watching the lag value. If it keeps to grow and exceeds some thousands, then the problem will be probably hearable. It will be best to try if two soundcards work well toghether before buying them. If the soundcards have a quartz that regulates the sampling frequency, it’s likely that they will work well together. However even if two cards have a crystal oscillator it's not sure that they will sync properly. Even a small difference in the frequency can generate an audible drift after some time. The Intel/Microsoft PC 98 specifications are not too stringent with regards to this aspect: they state that the both the recording and playback sampling frequency should have an accuracy of 0.1%. Even if the uncertainty is +/- 1/10000, at 44100 hz in the worst case the lag can grow by 8.82 samples per second. After 7 minutes the lag grows to almost 1 tenth of a second. The Q of a typical quartz is around 10^4 / 10^5. I don't remember well how the Q relates to the precision of the total oscillator, but the order of magnitude must be that. Now consider that the oscillator will not be made exclusively of the quartz, so cheap resistors and capacitors can further deteriorate the oscillator performance. The moral of the story is that the sampling frequency accuracy is a critical point and it's not too uncommon to incurr in sync problems, even if lately the number of even cheap cards that show this problems has progressively diminished.
2.h - Q:The recording vu-meters seem to show a signal even if all inputs to the soundcard are muted A: The signal that the vu-meters show when all inputs are muted is just the soundcard's internal noise. If you don't wish to see it you may want to decrease the vu-meters range by selecting the desired one from the menu that pop-ups when you right click on the vu-meters window.
2.i - Q: Can n-Track record 4 simultaneous tracks? A:The program can handle as many recording sources as you can think. As for the computer, recording a track is quite as much as heavy on the system load as playing back a track: call N the number of tracks you can playback with your computer without clicks or jumps in the music. Now the equation P+R=N must be satisfied for every possible value of P and R, where P is the number of tracks you want to playback and R is the number of tracks you want to record simultaneously. For example if N is 10, if you want to record 4 tracks at a time, you must keep the number of tracks to be played back while recording less than 6. Of course this gives you only an idea of the possible behavior of your computer and actual results may differ slightly. Anyway you can always partially render a group of tracks and playback them as one track only.
2.j - Q: The lag time when recording with two soundcards keeps growing. If I use two of the same kind of sound card, is that guaranteed to work, or not? A:No: if one or both of the soundcards haven't a good clock reference, their sampling rates will differ and the lag will keep on growing. If it stays below some thousand, then the shift will be not hearable. If the soundcards have a crystal oscillator, you may try to change the crystal of both with two exactly matched ones, but this will be quite risky if you aren't fond on this kind of hacking.
2.k - Q: I use a SBAWE64 for input and a SB16 for output. Do you know if and how to "monitor line in/mic" on those? A:If you use 2 soundcards and you record from the mic in input, you can do this way: connect line out of the SB64 to line in of the SB16 (I have connected the output soundcard to a stereo amplifier as a tape deck so that when I activate its channel from the amplifier it routes the rec soundcard output to the input soundcard line in and the makes me hear the output soundcard output). Of course you should activate the mic in monitor from the SB64 mixer and the line in monitor from the SB16 mixer. It may sound complicated but it's worth doing it: with two soundcards it's far better than with a single soundcard (especially if it's a creative one, as the SB32).
2.l - Q: Assuming no effects, how many 44100 mono tracks will I be able to work with before I hit the limitations of my computer? A:At the bottom end, 486 DX4/100s are often able to handle up to 6 tracks, while a Pentium II 333 with a standard UltraDMA hard disk was able to handle up to 26 44100hz mono tracks. Of course, to handle more tracks, you can render a partial version of the song with, say the first 4 tracks, then add another 4, render all and so on. Once you have all the tracks recorded, you can render all the single tracks (which have been left on the hard disk) together in a single final stereo wave file (rendering already rendered tracks theoretically introduces a slight quantity of noise).
2.m - Q: I have two soundcards, one is full duplex while the other not. Is it possible to use the non-full-duplex card as a secondary recording device? A:Yes: only one soundcard in your setup must be full-duplex. However when using multiple soundcards the best thing is to try: check if the lag values keep sufficiently small (they can grow due to small sampling rates mismatches).
2.n - Q: How many soundcards can I add to my system? A:The only limitation in adding a great number of soundcards is the availability of IRQs and DMAs. However even if this resources are sufficient installing more than two soundcards may be a bit tricky, so if you plan to have the need for more than two soundcards probably it's better to look for multiple inputs soundcards, which will also guarantee perfect sync between all the inputs.
2.o - Q: What is the best computer hardware configuration to run the program? A: A fast hard disk is important. Fast UltraDMA disks today offer great performances and are quite cheap. However if you're looking for ultimate performances UltraSCSI is will give you better results, albeit at quite much higher prices. A fast CPU is useful for being able to use many realtime effects. If possible prefer Intel CPUs over other brands as they have better floating point math. For the rest, my suggestions is to buy an overall fast computer without going in to the latest market news which of course will cost, with respect to more current systems, more than what they offer.
2.p - Q: Which operating system n-Track works better with? A: Windows 2000 and Windows NT. With the same hardware setup Windows 9x has slightly worse performances, with particular regards to hard disk input/output. This is directly reflected in the maximum number of tracks the program can handle, which may be up to 30% higher with Windows 2000 or NT. If you can choose among Windows 9x and 2000 or NT, Win 2000 or NT should be preferred, being faster and far more stable. Drivers are a little harder to find for Win 2000/NT so before installing it make sure that all your devices (in particular soundcards) have drivers for the OS. If you're using NT 4.0 and have and IDE hard disk make sure you have installed IDE bus mastering drivers. NT doesn't natively support IDE bus mastering, but there are freely available drivers by Microsoft (contained, but by default not activated, in the SP3 and later) and by Intel.
2.q - Q: Is it better to use a PCI soundcard with n-Track? A:Speed concerns regarding soundcards have little sense: their data flow is ridiculously small with respect to for example video cards or hard disk controllers. ISA bus can easily handle soundcard's throughput. However the industry thread goes in the direction of eliminating this bus (newer mainboards tend to reduce the number of ISA slots). As far as I know, the only great advantage of PCI soundcards is in using computer memory to store MIDI synth samples, as the new Creative PCI soundcards do.
2.r - Q: My sound card has numerous inputs: spdif, aux, mic, line... Is it possible to record from all these simultaneously on separate tracks? A:The multiple input that you refer to are in practice input to the soundcard's analog input mixer, that mixes all this input in to a single signal that is given as the imput to the soundcard's only analog to digital converter, so you all the inputs can only end up in the same track. I'm not sure about the SPDIF input: being it already digital it may be possible that the soundcard can record simultaneously from it and from the A/D converter, but I doubt that this has been implemented. If you can see only one entry in the Preferences/wave devices dialog box (apart from WAVE_MAPPER), you can only record one track at a time (or two mono tracks, if you can feed two separate analog inputs in the soundcard's line-in, for example connecting two mics to an external mixer and connecting the mixer to the soundcard). Of course there exist soundcards that have multiple A/D, such as the GadgetLabs Wave/4, which for example having four inputs is seen by Windows as 2 stereo standard soundcards.
3 - DirectX Plug-Ins
3.a - Q: Where can I download some DirectX plug-ins? A: Click here for a list of DirectX plug-ins available for downloading.
3.b - Q:I can't no longer save custom presets or I get the error "Can't write to preset file"when exiting the program. A: Using regedit (select Start menu/Run, type regedit.exe and press return) check that this registry key HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\App Paths\ntrack.exe\Presets file points to the old preset file (that usually stays in the program's directory). If you can't find the file delete this registry key. When the program will be restarted i will create another (empty) preset file.
3.c - Q: The program crashes every time I try to insert a plug-in to a track.. A: Please make sure you've installed Microsoft DirectX Media v5.2a or later (that is a different Windows component with respect to the DirectX standard runtime typically needed for games): you can download it at this URL: http://www.fasoft.com/redist/DXMWeb.exe. Also if the problem appears only with a certain plug-in try to uninstall and then reinstall the plug-in that's causing the problem.
3.d - Q: Do DirectX effects work under Windows NT 4.0? A: Yes. Despite the similarities of their names the DirectX Media component needed to use DirectX effects has nothing to do with the DirectX needed for games, and that NT 4.0 supports only up to version 3.0: DirectX Media v5.2a or later runtimes can be installed on Windows 95, 98 and NT.
3.e - Q:After removing some plug-ins from my system they still show up on the available DSP list in n-Track: how do I get them off that list? A: Obviously the uninstall should theoretically do it. It may be possible to directly remove from the registry the plug-in entries but it may be hard (and dangerous) to find them, so the best way is: reinstall the plug-in, run n-Track, browse with the Windows explorer in the plug-in folder and drag its main file (usually with the .dll or .ax suffix) on the n-Track's window. A dialog box should now ask you if you want to run the self registration of the plug-in: if you click on NO, the program will launch the plug-in self-unregistration procedure, thus removing it from the plug-ins list. You can accomplish the same effect from a command prompt typing: "regsvr32.exe /u plugin_filename_with_path". This will launch the self-unregistering routine that should be in every plug-in. You may need to specify the exact path of regsvr32.exe (usually \WINDOWS\SYSTEM).
4 - General questions about recording
4.a - Q: If I was recording a song should I record tracks in mono or stereo, what do most people do, and what are the advantages of doing either... A: It depends on what instrument you are recording. For most purposes a mono recording is most appropriated as for voice, guitars. If you are recording through a multieffect device, maybe you should record in stereo, as such device will enhance a mono sound outputting it in stereo. Also if you record the single tracks in mono, the final result will be always a stereo song: you can pan each track more in one channel or to the other.
4.b - Q: Recording near the computer is quite noisy: how can I get rid of the fan(s) noises? A: The definitive solution to this problem is to put the computer case in another room, make a small hole in the wall and make the cables pass through this hole. Another dirty solution to this problem (that should not be done by anyone who doesn't exactly now what is doing): I've put a switch on the front of the computer and I've truncated one of the power wires of both the CPU and the power supply fan, connecting them to the switch. In this way when I want to get rid of the noises I just turn off the switch. Having it off for 5-10 minutes has never created any problems, but I've found that the power supply heats quite fast so probably longer times can damage it.
4.c - Q: The wave files take up a HUGE amount of space. Can they be saved as mp3 or wma files instead? A: Compressed audio formats, such as mp3, real-audio or wma, are suited for distributing audio files over the internet or in any situation when the size of regular wave files will make their transmission impractical. On the other hand this formats use lossy compression schemes, which degrade the audio quality, and so they aren't suited for storing audio data when the sound quality must be kept as high as possible. For this reason (and also because on-the-fly decompression of compressed files will be way too heavy on the computer CPU) n-Track Studio doesn’t support compressed file formats for recording or importing tracks, but it only allows conversion from finished wave files (i.e. created after mixing down) to mp3 and wma.
4.d- Q: How can I transfer a song from a 4 track analog tape multitrack recorder? A: (posted by Nils) I have found an acceptable, though complicated and not perfect, solution to the '4-synchronous-channels-into-one-soundcard'-problem. In the following discussion, I assume access to a 4-track cassette mixer like a Tascam Portastudio or, like my own trusty machine, a Fostex 280. It should be capable of punch-in and out (preferably automatic) and simultaneous recording on all four tracks. I also assume access to n-Track and some other sound editing program capable of time-streching like, for example, Cool Edit. The imperfection lies in the way tape machines operate. 1) It is impossible to have a tape play back at exactly the same speed (i.e. sample-accurate) twice. 2) It is also impossible to avoid variations in the tape speed (wow and flutter) as it plays. This is how I overcome the first of these problems, as the second one is of a more subtle character and also harder to solve. Here goes... 1) Put some marker on all four tracks of the tape a few seconds before the music itself begins. this can be made by connecting a jack-cable to one input and touching the cable for a split-second while recording. Here is where the automatic punch-in/out comes in handy, punching out of recording mode before the music is reached, so the precious track itself isn't accidentially wiped. Record this signal at a sensible level on all four tracks simultaneously. 2) Repeat the procedure after the take ends, taking care not to wipe the fadeout or the next take on the tape. Now, we have two markers on each of four tracks on the tape, with identical distances between them. 3) Record the take, including the new markers, two tracks at a time, with n-track or another program, and then import the tracks afterwards. Don't split the stereo tracks up yet. 4) Zoom all the way in to the beginning of the song, and offset one of the stereo tracks so that the markers align. Then zoom in to the end of the track and note the difference in samples between the markers on the first and second set of tracks. Use the 'select' feature of n-Track to find the exact numbers at the bottom right. 5) Divide this difference with the sample rate (e.g. 44,100 samples per second for CD quality). This gives the lag time between tracks in seconds. Make a note of this number (use a calculator). 6) Using another program, such as Cool Edit, edit one of the stereo tracks. If the marker on the track is after the reference marker, the track is too long and should be shrunk to fit (i.e. subtracted from the total time) by an amount equal to the number of seconds in step 5). If the marker on the tape is before the reference marker, then the track is too short and should be stretched to fit (i.e. added to the total time) by the same amount. Don't preserve the pitch while stretching, or the result will be out of tune. This will take awhile, so this might be a good idea to 'goof off' for a coffee break. 8) Import the result into n-Track, zoom in on the beginning of the song, and adjust the markers to align with each other. Zoom to the end of the song, and check the difference in time between the markers here. This should be considerably smaller than before. The perfection of the synchronisation between tracks is of the same magnitude as the accuracy of the time stretching.
The advantage of this method lies primarily in the fact that only one sound card is needed, and the inevitable error is suitably small for most types of music. The backdraw is, that variations in tape travel cannot be compensated in this way, and the process is a bit complicated.
The rest of this FAQ, featuring information on auxillery channels, live input processing and registration issues can be be found at http://www.fasoft.com/faq.shtm
