Sometimes a mix of old and new is better than either the old or new alone. That’s what [Brad Carter] learned when he was given an old 1990s sound board with a noisy SCSI drive in it. In case you don’t know what a sound board is, think of a bunch of buttons laid out in front of you, each of which plays a different sound effect. It’s one way that radio DJ’s and podcasters intersperse their patter with doorbells and car crash sounds.
Before getting the sound board, [Brad] used a modern touchscreen table but it wasn’t responsive enough to get a machine gun like repetition of the sound effect when pressing an icon in rapid succession. On the other hand, his 1990s sound board had very responsive physical buttons but the SCSI hard drive was too noisy. He needed the responsiveness of the 1990s physical buttons but the silence of modern solid state storage.
And so he replaced the sound board’s SCSI drive with an SD card using a SCSI2SD adaptor. Of course, there was configuration and formatting involved along with a little trial and error to get the virtual drive sizes right. To save anyone else the same difficulties, he details all his efforts on his webpage. And in the video below you can see and hear that the end result is an amazing difference. Pressing the physical buttons gives instant sound and in machine gun fashion when pressed in rapid succession, all with the silence of an SD card.
A SCSI2SD card is a nice off-the-shelf solution but if you want something a little more custom then there’s a Raspberry Pi SCSI emulator and one which uses a Teensy with a NCR5380 SCSI interface chip.
Our thanks to [radiodork] for sending in a tip about this.
That drive does not sound:
A) bad
B) like it’s gonna fail
C) like it needs a defrag
It’s a late 90’s SCSI drive they last FOREVER!!! I use to work at a place called WeirdStuff Warehouse (they just went out of business) while working there the older drives we came across the MFM, and IDE and even early SATA were all on their last legs. SCSI drives they take a lick’n and keep tick’n! What kills a SCSI drive is the cost to operate to storage capacity ratio. Eventually a 9.1GB SCSI drive is useless… I would hazzard a bet that the original SCSI hard drive in that soundboard will outlast the SD card he put in there, maybe even the SD card adaptor too.
The ticking noises it makes are the heads seeking sectors, and the initilization routine.
I say all this of course, but I did put an SD card in my Apple IIgs so… you know…
You’re at least wrong about A) it not sounding bad. There was an audible high pitched whining sound produced by the hard drive the entire time the machine was on, making the 360 impossible to use during recordings. Having it click on every button press is not a good thing if you’re producing audio. And I’m 100% sure that SCSI drives do not last forever. You can buy an unlimited supply of broken ones from eBay.
I will concede that the sound of the drive is not conducive to a recording studio. And I will agree that SCSI drive do not in fact last forever.
However…
SCSI drives especially from the time-frame that one was produced were designed for 24×7 server-based operations in fact that drive he has pictured there was par for the course in HP, Compaq, and DEC servers of the time. With a MTBF far in excess of any IDE drives of the time (which had a horrible habit of disintegrating heads). The berings on the drive itself might be getting a bit stuck as it’s probably sat for ages. I remember attempting to spin up a drive that was year neglected and it would just tick, the platters wouldn’t spin. I gave it a solid whack to the base of my palm while rotating along the axis of the platters. Plugged it in, and it fired right up.
I think I’m beginning to sound like an old timer…
Anyways you are correct in his scenario the drive noise was unacceptable. In a sound studio the bird would have been placed in the operators room, and the recording room would have been sound-proofed.
He did in fact improve the device to make it suits his needs much better.
A hint to the OP about the mystery of his drive sizes: I bet the drive controller is configured for fixed partition tables that would match common drive sizes of the day. 2.1GB and 3.2GB sizes are common sizes that correlate the platter capacities 1.05, 2.1, 3.2, 4.3, 6.4, 7.5, 8.6, then there was a technology change and platters got a little denser, 9.2GB was the common 4-platter configuration in 3.5 1/4 height drives….
Johnathan Newman: makes me feel better about the CAD workstation from 200x at work that my boss refuses to throw a solid state drive in, thanks buddy
Johnathan Newman: makes me feel better about the CAD workstation from 200x at work that my boss refuses to throw a solid state drive in, thanks buddy
That drive sounds perfectly normal for a SCSI drive from the era! The head init sounds and the motor whine are all normal, and don’t indicate any problems at all! The drives are bulletproof, and quantum was a pretty reputable brand. They are, however, quite loud compared to modern drives, and if you have never dealt with them, all the extra noises are not expected. I can understand his desire to have it be quieter, but holy crap dude! At least do a nice job of retrofitting! Clearly a lot of DGAF going on here, and in such a cool piece of gear! Old pro gear from late 90’s to early 2000’s always has cool stuff like blinky lights and buttons, vfd displays etc. Everything is ‘touch screen’ or similar nowadays, and it seems to lose a lot of its charm. I guess this just dates me though!
As someone who has used these specific devices in a broadcast setting for 20+ years, I can assure you that the SCSI drives can, and certainly do, fail. As a whole the IR360 is an incredibly well-built machine. But the early units (like the one featured here) would either fail due to power supply failure or SCSI drive failure. Granted, this is likely due to heavy use. But they certainly do NOT last forever.
It should be noted that later models of the 360 Systems IR used IDE drives instead of SCSI.
And I also agree that his mounting solution is pretty bad. Looks like a hobo did it. And we all know, hobos are well known to do substandard repair work. Just look at those patches on their dungarees…
In all seriousness, anything that keeps these machines going is a positive. I applaud him for taking the time to document the process, sloppy as it may appear. The real value here is working out the configuration of the new drive, allowing others to complete the retrofit without headache.
I was really hoping for more details and pictures of the main board. I am also not really clear as to why he left the original hard drive in the system…
From his commentary seems to be an acute case of DGAF.
The adaptor board has 3.5″ mounting holes so it would have fit with stand-offs…. it would have been cleaner….
Taking the original drive out means that you’d have to put it somewhere and it would get in the way or get lost, and your classic instrument would no longer have all the “original” parts, for whatever that’s worth.
I left the drive in there so the new board would have something to sit on. Jonathan Newman says the mounting holes were in standard 3.5″ hard drive positions, but they were nowhere near standard positions. If they were I definitely would have mounted it that way. You need to buy a kit to mount it in place of a hard drive. And I did put some extra circuit board pictures up – they’re linked on the very bottom of that page.
It looks like the screws match 90-degrees from original orientation, at least from the pictures… not much in the way of technical board data on their websites. I do see the adaptors, look like external style headers, for mounting where you’d put a floppy drive. So you could swap SD cards as you want.
I’ve seen these 360 consoles and other strange 90’s-00’s era tech in my time at small TV stations, many of which were handed tech from the larger affiliates when they upgrade. I imagine the disc noise wasn’t an issue in radio studios as there was allot of effort put into the design of the furniture to isolate the mics from bumps and knocks. In a TV station, the audio was normally a separate booth from the microphone source (in the studio proper or a voice over booth).
As for the lack of files on the SD card: it’s quite possible the audio data is being recorded as raw bits on the disc, accessed by track number. That would also explain the constant noise from the hard disc and the fast access time to samples (apart from the SIMM in the pictures).
One other cart system I used was based on 3.5″ floppies. It was pulled out of storage when the MiniDisc unit normally used as a cart needed a repair (that had memory as well to pre-load 10 tracks from the MD). It was like using a true tape cart, so I imagine it was commonly used with a small number of samples per disc. Then you don’t really need a file system to manage them, just the cart style ‘Cue’ and ‘Play’ buttons. In our case they were used for looping foley sounds (crowd noise etc).
In live theater organ with 100 years of practice, the Toy Counter is where all of the hoof beats and squealing tires came from. Then came Foley in the sound movies. The toy counter lived on in live radio even though it has changed so much.
I wonder if an i-pad could do this without lag. I looked into the most awesome touch screen instrument I have ever seen on YouTube. It is called Geoshred, but it is only available for i-OS not Android. Lag is the reason. Apple did it right, Android is a poor emulation of it when it comes to sound.
Keeping audio latency down in iOS was definitely a win for Apple.
Maybe Geoshred should look into https://superpowered.com/ who claim to have solved audio latency on Android.
Not entirely off-topic: has anyone implemented the RaSCSI hack (RPi as SCSI device) linked in the article?
a) I need a WiFi-connected SCSI drive
b) Would make a cool article
The write up was hilarious and just my style :) When in doubt, copy all of the files lol
Great work and I wish I had saved all of the SCSI drives I came across early on in my career. Those would be retirement gold.
cactus cactus