Hippo Sound Digitizer
An easy-to-use sound digitizer for your ST
Reviewed by Patrick Bass
I want this product. I can't use this product. No, the above is not a textbook example of dichotomy, it's how I feel about the Hippo Sound Digitizer. First, however, let's explore it and then get back to my petty wants and I desires.
The Hippo Sound Digitizer (herein after known as HSD) is a eight-bit resolution hardware/software combination which allows you to digitize sound with your Atari ST computer. The HSD hardware consists of a dove-grey box about half the size of an ST disk drive, which plugs into the cartridge slot of the computer. The box has two external control knobs for adjusting input and output level. On the back you'll find 1/4 inch standard phone jacks for sound input and output.
Notice that word output. You'll need an external amplifier to hear the sounds you've digitized, since there seems to be no way to play the sounds back through the TV speaker. I spent about 15 frustrating minutes before I found a reference to "adjusting the level of your amplifier" buried in the HSD manual. They should have made this apparent in the "Setting Up The Unit" section. Incidentally, cords are not provided.
The HSD software comes on a 3.5 inch disk, and contains both the digitizer software and sample sounds. HSD works on either color or monochrome Atari ST systems. To actually operate the digitizer, from the desktop, double-click on SOUND.PRG, and in a few seconds you'll see the HSD title screen pop up. There are four choices presented along the bottom of the screen.
On the left is "Command", which will present the main control panel for interacting with HSD. Next comes "O-scope", which presents a digital oscilloscope onscreen so you may watch the waveform of the incoming sound. To the right of that is "Rack," which presents a screen from which you may add reverb or echo to incoming or existing sounds, and finally, on the right is "Exit", which drops you back to the GEM desktop.
The command screen is where you'll spend most of your time. Let's examine it closely. Most of the upper half of the screen contains an editing window, where the sound waveform is displayed and edited from. The upper right of the screen contains a "gas gauge" display of memory used for the sound and memory left in the cut/paste buffer. The lower half of the screen contains buttons for 24 different functions.
First, you may LOAD or SAVE a sound file. The STATS button updates the display concerning playing time, number of samples, sample rate, and percentage of modulation concerning the average amplitude of the signal. The MIDI button sends sampling out through the MIDI port in the back. You may ADJUST RATE, either for the entire sound or just for a selected section of the sound, and vay the VOLUME, again either for the entire sound or just for a selected section.
HSD allows you to define your own wave shape and ADSR envelope, for completely unique sounds created from your own imagination. Also, at any time during editing of a sound, you may UNDO your last action.
You may MIX, or overlay, (sound with sound) one voice or instrument on top of another one. For example, select one sentence of a person talking, cut out a copy and save it in the cut buffer. Then, MIX the sound in the buffer back in with the original sound many times, each one starting a fraction of a second sooner or later than the last. The result? A "crowd" of people saying the original sentence (more or less) in unison.
You may select a section of sound and SQUEEZE it to raise the pitch up one note, or you may STRETCH it out to lower the pitch. For multiple SQUEEZES and STRETCHES, you may double-click on the buttons, then type the number of times to repeat into the resolution dialog box.
Ever play Beatles' records backwards? With the REVERSE button, you could record and select the voice saying "Number Nine... Number Nine. . ." from the White Album and play it in reverse, confirming for yourself what the Beatles were really saying. (I have, and no, I'm not saying.)
The next four buttons all work together. CUT will remove the selected section of sound and place it into the cut buffer. COPY just makes a copy of the sound and places it into the cut buffer, without removing it from your original sound. INSERT copies the sound in the cut buffer to wherever the cursor is in the current sound being edited. REPLACE will remove the selected section of sound on the edit screen and replace it with the sound stored in the cut buffer.
When you click on SHOW ALL, the entire sound is displayed in the edit window along the top. The left side of the edit display shows the time the selected sound starts, and the right side of the display shows what time the sound ends. SHOW ALL is needed, because using the ZOOM IN and ZOOM OUT buttons, you may zoom in on a selected section of sound until the individual peaks and valleys of the sound waveform itself are visible.
This can have a very important function. For example, record a selection of music off an old record. Now play the section of music back, and if the record was old enough, you'll hear clicks and pops in the sound where the record has deterioated. On the edit screen, you can see the clicks and pops, as they are louder than the music around them. Now ZOOM IN on a single click or pop, and cut it out of the music. The two ends of the song will seal together automatically, leaving just music, no clicks or pops.
Finally, you may PLAY the entire recorded sound, or just a selected piece of it. To select a section of sound, point the cursor at the left edge of the sound you wish to select, then click-drag the mouse to the right. The selected section of sound will be highlighted.
Above the PLAY button is DIGITIZE, which starts the HSD sampling the input sound and stores it in memory. Last is the SET TIME button, which allows you to determine how long the digitization lasts.
Click on O-SCOPE, and the screen becomes a digital oscilloscope displaying the incoming sound as a waveform on your TV screen. This screen is used for setting the level of the incoming sound for use with the digitizer.
RACK is a screen containing control for inserting echo or reverb information into a previously digitized and stored sound, or for adding reverb or echo to the sound currently digitized.
Just how much sound can be stored at one time? Basically, it depends on how fast you sample the sound when you digitize it. There are rules in digital sound recording which state that in order to properly digitize the incoming sound, you must take twice as many samples (per second) as the highest expected incoming frequency. The HSD will sample sound up to 200 kilohertz (Khz) and signals up to five volts.
This, for example, means that to record a high-fidelity sound whose highest note is around 20 Khz, we must take 40 thousand samples per second. For voice, where the highest note is more like 5 Khz, a sampling rate of 10 Khz should work fine.
Each "sample" of sound is stored in one byte of memory. This means 20 thousand samples would take up 20 thousand bytes of memory. In an Atari ST with one megabyte of memory, there are roughly 800,000 bytes free for sound storage.
So, with a little math we can see that high-fidelity music sampled at 40Khz should give us (800,000 /40,000) 20 seconds of recording capability. A slower sampling rate, for instance, gives a correspondingly longer recording time. For example, voice sampled at 10 Khz should result in (800,000 /10,000) 80 seconds, or nearly a full minute and a half, of digitized, editable sound.
Now this may be a case of not seeing the Emperor's new clothes, but I can tell very little difference (if at all) between music sampled at 40 Khz and music sampled at 20 Khz. If you can't tell the difference either, then you've just found a way to double your recording time.
Overall, the product is solid and usable. . .to an extent. For me, I was disappointed to find there was no apparent easy way to include my digitized sounds into my own programs. Like in Jez Sans' StarGlider, I'd like my panel to say "Systems alert!" instead of beeping at me. (With that English accent, of course. Who is she?) On the other hand, Hippo does reveal the file structure of the sound files created and they do include programming examples of how to capture digitized sound and how to drive the digitizer with your own sound data.
But . . . If I were running a radio or TV station, or worked in a recording studio, this one package would allow me to compose and edit complete commercials without ever putting sound on tape. I could cut, paste, and distort any sound I could hear, and even some I couldn't. I would want this package greatly, even to the point of buying an ST just to drive it.
Think of this . . . a complete package of HSD and the Atari 1040ST costs less than one-tenth the cost of a studio digital tape recorder. And the digital tape recorder won't play Time Bandits or Flight Simulator II...
HIPPO SOUND DIGITIZER
985 University Avenue Suite 12
Los Gates, California