This page is for people 'who are afraid to ask, but need to know'. Including ourselves. The answers we expect from other users. Also we collect questions and answers from the email list. In this new version, some questions and answers have been moved to other pages. So also have a look at the Related Topics bar at the bottom of this page.

The hyperlinked questions have been answered. The questions with a question mark are still waiting for an answer, so feel free to make your contribution!

The signal of a turntable is lower than from a standard Line-connection. Therefore you need to pre-amplify the signal. In jargon: the impedance doesn't match the Line-Input of the EWS. The same goes for the signal of a dynamic microphone.

Possible solutions:

• Buy a stereo pre-amplifier, preferrably one especially designed for record players (they boost the bass a bit). Connect the turntable to it. Connect the pre-amp to the Line In of the EWS.
• In case you already own a mixer: connect the turntable to the mixer, turn the gain knob on the mixer to the max and connect the mixer output to the EWS Line-In. Boost the bass a bit.

Contribution by Jörg Knitter

The situation especially for Win98 is still the same, because nearly nothing has been changed between DirectX5 and DirectX6 (only changes in the Win95 realtime wave mixing algorithm). The next major sound update will be released in autumn 1998 and be called DirectMusic.

Contribution by Philip Koester

If you're recording a digital signal the sample rate of which is 44.1 kHz, you should set the frequency to the same value in order to get a 1:1 copy, otherwise you might get deviations related to sample rate conversion calculations. Besides the algorithm that is used here in order to provide this conversion, it might not be best quality, because it needs to work in real-time.

Same with mixdowns. If you want to record a 44.1 kHz track, again you should set the frequency to this value. OK, what you can do is use 48 kHz in the Control Panel, record a 48 kHz wave, and use a high-quality converter (e.g. WaveLab) to dither the recording back to 44.1 kHz. But there's just no sense in it, and it might sound worse than using 44.1 kHz right from the start.

Terratec Support

The Windows Volume Control is only for the Codec functions, because the Synthesizer functions can not be controlled via the Windows Volume Control. The ControlPanel is for both!

Terratec Support

Search for entries like TerraTec, EWS64, and driver names like ewsaudio.vxd, ews64cfg.dll, cs4232ld.vxd etc.  Normaly you can delete all these entries, but be sure of what you are doing. This is nothing for novices !

Contribution by Harald Riegler

The virtual channels control the AUDIO routing, while the midi clients control the MIDI data flow. Midi does not involve any audio, it is all just about control of synthesizer devices (either the internal Synth/Sampler, or external Keyboards etc.)

The midi stream is something like this:
'Note c5 pressed, note e5 pressed, note e5 released....' and so on (very simplified ;-) )

Obviously the record client is for recording midi data (ex. in Cubasis AV), and the play client is used for the playing of midi files. With the monitor client, a second program - running beside Cubasis for example - can also record, monitor, evaluate etc. the midi data running through the record or play clients.

Contribution by Purple Snail

Set In-2 selector to D. Set the out-1 selector to B.  And make sure that in the FX panel the eq/vspace send button that's right at the bottom is selected. The REC and Input Sel make no difference in this case because we are recording the direct output of the synth.

Contribution by Dave O'Flynn

Full Duplex is recording while playing back at the same sample rate. In other words, if you're playing back a 16bit, 44.1Khz file, anything you record will be recorded at 16bit 44.1Khz.
Enhanced Full Duplex is when you can record at a different sample rate to the rate you're playing at. This means that you can record an 8bit 22Khz file while playing a 16bit 44.1Khz file, or any other combination you feel like, subject to hardware limitations of course.
Practical uses? None. Unless you like that crunchy 8bit sound.

Contribution by Holger Fiergolla

My very personal point of view: because the EWS-Drivers are 16-Bit. One of the past 'EWS64 mailing List' issues tells, that all multimedia Windows drivers are 16-Bit. I do not know if this is right, but the EWSAUDIO.DRV definitely is a 16-Bit driver. Accessing 16-Bit drivers with 16-Bit applications is the easy way. Building 32-Bit Apps doing the same needs a lot of pointer conversions from 16-Bit Segment:Offset notation/allocation/alignment to 32-Bit DWORD pointer notation/allocation (so called "thunking") and some other minor steps, what leads to the second question:

Isn't it more 'crash-proof' when it's 32-bit?

Again my very personal point of view: Nope. Development of 32-bit Software underlies the same rules as 16-Bit development. The only way to get a 'crash-proof' piece of software is intensive testing/debugging until you ironed out all bugs. By the way: You'll never get a 100% bugfree SW since programmers are humans.

Programming 32-bit applications accessing 16-Bit drivers/DLLs is a little more difficult (remember the WOW = WinOnWin32 topics in the mailing list?). I never had produced any 'blue screen' crashes before trying out the EWS-API with a 32-Bit C++ compiler. Today I'm a little older and know a little more about the 'secrets'. The 'blue screen deaths' of my system are reduced to the 'normal' minimum again.

I tried Delphi 2.0 (32-Bit "Pascal") and did not get any connection to the EWS. I tried Delphi 1.0 (16-Bit 'Pascal')and everything works fine. Today I am back to C++ (CBuilder).

Contribution by Dave O'Flynn

Yes, there is a big performance hit if you run your card without any IRQ. Running the card without an IRQ should only be done as a last resort. If your card uses an interrupt (IRQ = Interrupt Request Line) then whenever it needs attention, it can interrupt the CPU and do what ever is necessary. However, if you run the card without an IRQ the driver has to constantly check the card to see if it needs anything, as there's no way for the card to tell the system that it needs attention. This is called 'polling', and wastes a lot of time, because most of the time the card won't need attention, and you'll have wasted some CPU time. This slows your system down quite a lot.

Contribution by Dimitris Siatiras

The AXDIST.EXE, which is bundled with the EWS64 drivers, includes Microsoft oriented ActiveX distribution DLL required by some applications in order to register their components on the system. This distribution code is available for example with the VC++ SDK. The contents of the AXDist.exe are the following:

• WinInet.dll
• InLoader.dll
• SChannel.dll
• URLMon.dll
• HLink.dll
• HLinkPrx.dll
• OLEAut32.dll
• StdOLE2.tlb

Most programs have this AXDist bundled with their installation files (e.g. Netshow). In most circumstances there is no need for the user to run this file at all.

Contribution by Philip Koester and Paul Heijmann

Philip: If you have a SCSI drive you can just copy the tracks digitally to a wav-file. I think there are programs that can do that with IDE drives as well, check out some pieces of CD recording software.

Paul: Indeed there are such programs, I'm using DIDO 4.0, which works with IDE drives. You'll have to install MSCDEX though.

Contribution by Dave O'Flynn:

All SCSI (as far as I know) and 90% of IDE drives can read audio directly from the drive. This is known as RAW mode - the drive treats the audio data as if it were standard CD-ROM data. For this to work, the drive must support the "Read Long" command. Most drives do, although some don't - Mitsumi are probably the highest profile company, though there are reports of some newer Mitsumi's supporting CD Audio extraction. Programs such as CDDA and CD Worx will read the data directly. You can find more detailed info and drive compatibility lists at the homepages of both the programs. These programs do not require MSCDEX (MicroSoft CD EXtensions - the DOS CD-ROM driver standard), as they work directly with the Win '95 (and NT) driver standards.

Contribution by Dave O'Flynn

We estimate that it won't be soon. But apart from that, people sometimes only ask for a PCI-version because they have heard that PCI is faster. Technically that's true, but for your information we will sum up a few additional facts.

• When Terratec started development work on the card, the PCI interfaces available for digital audio work were very immature and did not offer the features that were available from ISA chipsets such as the DREAM set, so they had no real choice - they had to develop for ISA.
• Other important soundcard manufacturers are also reluctant in releasing PCI-versions of their cards. So it's not reasonable to expect a PCI release from Terratec at short notice.
• PCI soundcards can (and do) use an interrupt to provide legacy compatibility for DOS games. The difference between PCI and ISA cards is that (for reasonable performance) all ISA cards have to use interrupts, while PCI cards don't.

Contribution by Kristofer Maad

This explanation was originally an integrated part of his answer to the Signal Routing question. So it helps if you read that first.

Now over to recording things:

• When recording with the Codec (Codec Record), you set the input with the Input Selection switch in the Control Panel. You can select MIX, IN-1, MIC, or SYNTH. Nothing fancy about that, just remember that you hear the whole mix, but you only record one particular signal. The Monitor slider in the Control Panel can be used to give a feedback on what is recorded. This is especially useful when recording from the MIC-IN, since it doesn't go to the mixer otherwise. The recording level can be set with the slider on the far right in the control panel, when the small slider under it is in the A position. You can get a VU meter by pressing VU.
• When recording with the Synthesizer (Wave Record), the recording level is set by the same slider on the right, when the small slider is in the B position. You can not get a VU meter in the mixer when using the Wave Record, unfortunately.

The interesting thing about the Synthesizer (Wave Record) is, that you don't have to record the input signal at all. If you take a look at the FX panel, you see a small slider in the upper right corner. If this is in the Audio-In position, the input signal gets recorded. However, if it's in the Mix position, only those inputs routed to the EQ/V-Space effect are recorded - before the EQ/V-Space effect, however. So, you can mix the Wave, MIDI, and Input signals digitally and record it to your hard drive. Or, skip the Input signals and get an absolutely noise-free mixdown of your song (if you're only using Wave and MIDI sounds, that is).

Something else, if you don't need the internal effects for one particular input of the digital mixer, you can use four-channel mode for that particular input. That way, you can have:

• four Wave Play devices, each one routed to one mono output,
• or two Wave Play devices, each routed to one stereo output. These pannings are controlled from the Virtual Channels Panel.

Note from Kristofer: Well, that's about what I've figured out - and it turned out to be quite a lot, actually. I guess there might be some errors in this description, so I take no responsibility for correctness here! :)

Official answer by Terratec as stated in their FAQ of 11/11/97:
(Translated from German by Andrew Morgan)

(Question: What is the difference between Codec and Synthesizer?)

The EWS 64 CODEC is equivalent to what you would expect to find on a standard hi-end soundcard. In other words, you can record audio files, listen to and sample audio CD´s, record using a microphone and connect external devices (line-in 1) using 16-bit converters.

At the heart of the EWS lies the synthesizer. Digital processing within the synthesizer is performed using 20 bits. It offers the possibilities of analogue sampling using the 18-bit converters or digital using the SP/DIF interface. Additionally the synthesizer processes MIDI-sounds and plays MOD-files without loading the CPU. All the synthesizer outputs can be processed using the onboard effects.

Contribution by Purple Snail

The sync button is the button on the Control Panel that tells you what the sample rate is (e.g. 'Int 48kHz') The 'Int' part means the card is generating the signal that is used to control  'time'.
If you plug a digital device into one of the digital INs or OUTs, some more buttons will pop up next to this sync button, one letting you turn auto sync on or off, the other telling you whether you are sync'd internally or externally. In general you should sync to whatever device you are recording from. If you are recording from a digital device to the EWS you should sync externally, conversely if you are recording the EWS output on a digital device you should sync internally. If you just leave auto sync on, the EWS should handle these settings for you.

Contribution by Purple Snail

The Clip on/off button acts as a compressor. If Clip is On, and output produced by the EWS exceeds 0dB (which causes clicking and heavy distortion), the EWS will compress the waveform, i.e. flatten it out at the top, so that it doesn't exceed 0dB and no clicking or distortion will occur. The EWS will also amplify low volume sounds closer to the maximum output when Clip is On.
When Clip is Off, the distortion and clicking will be free to occur.
So why would you want Clip Off then? Well, the algorithm that Clip uses can't predict exactly how to flatten the waveform to prevent it exceeding 0dB, so it plays it safe, and compresses sounds that are close to exceeding 0dB, even if they end up not exceeding that volume. Thus, with Clip On any high volume sounds will be changed in shape slightly, this was the cause of people originally reporting that the EWS couldn't make 1:1 digital copies.
As for the button next to the Clip: if Clip is off, this button tells you whether the output or input signal has exceeded the 0dB range.

Additional information by us:

In some situations the Clip On/Off button is absent:

• If you don't have the EWS64 Hardware rev 1.1 or later.
• If you choose the option 'Small mixer' from the EWS mixer icon in the taskbar and after that enlarge the mixer again. Minimize the mixer to the taskbar once more and let it pop up again: buttons are back.
• If you did not upgrade to drivers version beta 2.0x

Contribution by Purple Snail)

To record the FX, you have to have the button that says 'Eq/V-space Send' turned on. This is the button below the chorus 'rate' and 'depth' settings in the FX panel, not the buttons under each audio device (midi, wave, mod..etc). This doesn't make any sense, I know, but I think it is just a mislabeling, and it should read 'Effects Send'. If you have it turned off, you will still hear effects, but they won't be recorded by Ed!son.

Contribution by Kristofer Maad

The EWS64 has two sound mixing parts:

• One part, the Mixer, is analog.
• The other, the Synthesizer, is digital. Digital mixing, as opposed to analog, gives no noise except for the input noise.

If you like, one of these two mixers can function as a main mixer

• the output from this mixer will go to the OUT-1 output jack
• the other one will function as a sub-mixer (mixing several signals and then giving the resulting signal an input of the main mixer).

You can also use the two mixers independently of each other. The choice you make depends on if and how you want to record things, more about that later.

• If you would like to use the analog mixer as the main mixer, set the OUT-1 routing switch (the lower right A-B-C one) to the 'A' position.
• If you prefer to use the digital mixer as the main mixer, set the OUT-1 routing switch to the 'B' position.
• If you want to use the mixers independently, you should put the switch in the 'A' position, to route the analog mixer to OUT-1. OUT-2 can then be used by the digital mixer, but without reverb/chorus effects unfortunately.

Of course it doesn't suffice to determine how the mixers should be connected, you need some signals to mix, too! Each mixer has a number of 'hardwired' inputs, each with separate volume controls.

The analog mixer has the following inputs:

• Wave - from the Codec, for DOS-games and the Codec Play driver.
• Syn - from the Synthesizer and the FM Synthesizer. Use this if you want to utilize the Synthesizer as a sub-mixer.
• CD/Syn2 - from the CD-player or wavetable daughterboard (XL only).
• In-1 - from stereo-in 1.
• Beep - from the mainboard Speaker output.

The digital mixer has the following inputs:

• MIDI - from the sounds of the MIDI data sent to the Synthesizer.
• Wave - from the Wave Play driver(s).
• MOD - from the sounds of module data.
• Synth-In - either analog stereo-in 1, digital stereo-in (XL only), or the
• Analog mixer, depending on the setting (A-D-M) in the control panel. Use M here to put the analog mixer as a sub-mixer. Unfortunately, you can only record the analog mixer output from the Wave Record in this mode.

Note from Kristofer: Well, that's about what I've figured out - and it turned out to be quite a lot, actually. I guess there might be some errors in this description, so I take no responsibility for correctness here! :)

Note by Computer ConText

As you all probably know we have  the program AudioRoute by Morten Egelund Rasmussen now. This excellent program makes the routing more visual and at the same time enables you to control the EWS. The program is still in 'alpha' phase, but you can download this program from Morten's website, http://silverjam.home.ml.org
In the near future Morten will make more contributions to The Stress Control Panel about the routing topic.

Contribution by Purple Snail

The dB scale is just a logarithmic scale that is used to measure audio volume. Logarithmic means that a signal at 0dB is 10 times as loud as one at -10dB, which is 10 times as loud as one at -20dB...and so on, so that a signal at 0dB will be 100 times as loud as one at -20dB, but 1000 times as loud as one at -30dB, and 10000 times as loud as one at -40dB. There really is no way to 'use' a dB :) It is just a scale. All I can say is that you should try to produce music that is as close as possible to 0dB in volume, without actually exceeding 0dB (where clicking and distortion will occur).

Contribution by Purple Snail

The wave slider in the control panel only affects the Codec device.
Thus to notice any difference you must have the Out-1 'abc' switch set to a. Then, any programs that are set to use the 'EWS64XL Codec Play' (not EWS64XL Wave Play #) will have their volumes controllable by the wave slider.

Contribution by Purple Snail

I assume you mean without a sequencer. The answer is no. This is just common sense. If you wanna play the ews synthesizer just bootup your favorite sequencer and play away (making sure of course that the sequencers midi in devices include the Ews64 Midi In, and its output devices include the Ews64 Midi Out). The old bankloader used to have an option that let you route midi In to Midi Out, so effectively it was acting as the sequencer, but the new bankloader doesn't have this option.
If you wanna just jam away with your keyboard without having to boot up a sequencer every time, I suggest you get Hubi's LoopBack device that will allow you to make a 'virtual' connection between the keyboard and the ews synth.

(Clarified by Philip Koester)

Which synthesizer do you mean? The EWS' or your synth's? OK I'll explain both directions ...

1. Listen to the EWS' sound while playing on the keyboard.
   • Connect the keyboard's MIDI out to the EWS' MIDI in.
   • Use a program that routes incoming MIDI signals to a MIDI output device (e. g. Cakewalk or
     Cubase).
   • Use MIDI output device 'EWS MIDI Play #1'
2. Listen to the keyboard's sound while having a MIDI sequencer play.
   • Connect the EWS' MIDI out to your keyboard's MIDI in.
   • Make sure the keyboard is set to 'Multi mode' (or whatever the name is on your keyboard).
   • Use a MIDI output device that sends MIDI signals to the EWS' MIDI out.
   • Start your sequencer (such as Windows Media Player, Cakewalk, Cubase).

Contribution by Leon der Profi i.e. Leon the Professional

SP/DIF means: Sony/Philips Digital Interface (optical & coaxial)

Official answer by Terratec as stated in their FAQ of 11/11/97:
(Translated from German by Computer ConText)

Output 1 and 2 (OUT 1 & OUT 2) can be set separately.
This way reproduction of several WAVE-files through the Synthesizer can be supplied with different effect-units. Or, when using the 4-channel mode of the EWS in WAVE-reproduction, you can use the outputs separately, e.g. 4 mono-outputs. The MIDI-volume and the panning can be set separately from WAVE-audio of course.

When you're interested in further information about this subject, you can download a file about the use of Virtual Channels (German version).

Note by us: This file has been translated in English by Andrew Morgan. Go to our Download Page to get it. We also have another English translation available there by Tat Tsitsiklis.

(Clarified by Dave O'Flynn)

MME (MultiMedia Extensions) is the standard specification for Windows soundcard drivers. They originally appeared in the Multimedia add-on for Windows 3.0. They were subsequently included in Windows 3.1 and Windows '95. All soundcards have MME drivers. They are 16bit and have a very high latency. Latency is the time difference between a program telling the soundcard to play something and the time it actually starts playing. They're fine for 'business audio', but too slow for anything else. Although many people do use them for music, there is a noticeable lag between the time a sound is supposed to start and it actually starting.

DirectSound is a subset of DirectX. DirectSound drivers are Microsoft's replacement for the MME drivers. They're targeted at the games market, and are 32bit with a lower latency (~170mS) than MME drivers. However this is still too slow for serious audio work, which means that there is no standardized, capable audio driver spec. for the Windows platform.

ASIO drivers are a Steinberg standard. They only work in Steinberg products (mainly Cubase). They are technically much better than the MME drivers, as they're designed for professional musician's usage.
However, very little semi-pro equipment comes with  ASIO drivers, at least as of yet. On July 2, 1999 Terratec released the first beta of ASIO drivers for the EWS64. See our Download Page.
ASIO is a proprietary Steinberg standard, so you won't find a non-Steinberg program that uses ASIO drivers.

As we said earlier: a stupid question does not exist. So feel free to email us your questions. But please try to keep the question a 'one liner', that's easier to process.

© 1998-1999, Computer ConText, Amsterdam, The Netherlands, except for contributions made by others.

Latest update on page: 19-04-99

© 1998-1999, Computer ConText, Amsterdam, The Netherlands, except for contributions made by others.

Latest update on page: 19-04-99