Watch on YouTube: https://www.youtube.com/watch?v=aoF98_XLL3U
Investigating A Very Rare Atari Sparrow FX-1 Prototype Motherboard - Should I Restore It?
Introduction
I’m now the proud owner of an Atari Sparrow / FX-1, the prototype for the Falcon, which was released in 1992. Only 21 of these are known to still exist! But should I restore it, or frame it as a museum piece? I’m open to suggestions!
Script
Hi everyone, Rees here, and welcome to a very special episode of ctrl-alt-rees.
In this video we’re going to be taking a close look at this incredibly rare Atari Sparrow prototype motherboard from 1992, which is one of only 21 that is still known to exist.
I also want to ask for your help as a community in tracking down as much information as we can on the Sparrow project to preserve it before it’s lost to history, and also to help me to decide what the best course of action is with this board - should I attempt to get it up and running again, or should I just frame it and stick it on the wall?
So what is the story behind the Sparrow project?
Well, after the success of 1985’s Atari ST, which was based around the Motorola 68000 CPU, and the later enhanced Atari STE, Atari started looking at more powerful CPUs to compete against the likes of the NeXTStation and the Amiga 1200, and that came in the form of the Motorola 68030.
After the ST had gained a very strong foothold in the world of music production thanks to the clever inclusion of MIDI ports, Atari decided to take the same route with its successor, which would eventually be known as the Falcon, and be released in 1992.
While the new machine was under development, Atari identified it internally using two different code names - FX-1 and Sparrow.
When Atari collapsed in 1996 the company was stripped for its assets and everything was sold off to the highest bidder - companies like Best Electronics in California and WizzTronics in New York snapped up anything and everything they could from Atari’s warehouses and offices, and that happened to include some prototypes that they’s been hoarding, and that’s how I came to be in possession of this Revision A Sparrow prototype motherboard, which was being sold by WizzTronics on eBay, who happens to be the source of most of the known Sparrow motherboards in existence.
I wasn’t the first to buy one of these - in fact there are 20 other people who’ve gone before me - and I’m not really sure how many more motherboards WizzTronics are sitting on, but for now there are only 21 known Sparrow motherboards in existence, and there are only actually five Revision A boards like this one.
I believe that the Revision A boards are actually much rarer, and the reason for that being that Atari was strapped for cash at the time that the Falcon development was taking place and so these boards would have been cannibalized for parts to build the Revision B boards.
There are working Revision A and Revision B machines out there including bare motherboards that are populated with all of the chips and everything required to make the computer boot - of course it’s worth bearing in mind that there is no commercial software that actually runs on these boards - in fact the only software written for them probably would have been internal testing tools used by Atari themselves which are probably long gone.
It would appear based on the information that I’ve seen on Atari-Forum that the Sparrow actually started life as an expansion card for the Mega STE, which was used to develop the custom ASIC chips which are found on these boards, and of course eventually made their way into the Falcon.
One of those Atari forum users - known as “Mikro” - owns at least one working Sparrow motherboard and has taken the initiative to set up a website cataloging the known serial numbers of these boards.
I’ve been in touch with him and he’s been incredibly helpful in my search for information, and he informs me that he’s going to be adding a lot more information to the site in future, so I’ll link to that down in the description along with any other information that I’ve managed to dig up so far.
On that note, when I first bought this motherboard all I had to go on were some very low resolution photos from eBay, and I could see that a lot of bits were missing, but I couldn’t really tell what the condition was like.
So I decided to number all of the missing parts that I could see and then go on a hunt for high resolution photos on the internet and try to put together a spreadsheet identifying what they might be.
After a lot of hunting, I found some decent photos on the various Atari forums and on an incredibly helpful website called atarimuseum.de, which had some very high resolution photos indeed, and armed with those photos I put together a spreadsheet listing all of the missing chips, and it’s those photos and that spreadsheet that I’m going to use as a reference while we explore this motherboard and see what it’s all about.
So I think without further ado, let’s have a closer look.
So this is the Atari Sparrow Revision A prototype motherboard. We’re going to have a look around this motherboard now, just show you some of the interesting things that I’ve discovered about this and the chips that I’ve managed to identify so far.
I’m also going to be using a USB microscope just to get a closer look at some of these components on here.
First things first, we have the copyright silkscreened onto the board here dating this board to 1992.
Over this side we also have the serial number, this is a known serial number, it’s already on Mikro’s list that I mentioned before.
Interesting that this one doesn’t follow the same pattern as all the others but I think it’s just a mistake that whoever wrote the label made when they were writing that.
Also down here on the bottom of the board here we can see that the part number - the model number - obviously that’s designed to be handwritten and hasn’t been filled in.
Now, when I was identifying the parts of this board I put together a diagram and like I mentioned I went and tracked down as many photos of these things as I could on the internet and tried to sort of match up the numbers to the components in those photos.
So the one that I haven’t managed to identify as of yet is in this top corner, which is a small surface mount chip, so I’m not quite sure what’s supposed to go in here. I assume it’s probably something to do with the MIDI ports as it’s in that area of the board, so it’ll be interesting if there are any other Sparrow owners out there who could provide that information.
Like I say, it’s an area of the board that I don’t have any close-up photos of.
There are actually quite a few chips - obviously the most obvious thing on here will be the missing components - but actually there are quite a few smaller chips that are still present so we’ll have a look at those as well as we go around.
So, just coming around to this area of the board here, we have the cartridge port. Obviously that’s not present - the fact that they’ve populated all of these links kind of indicates to me that it was, and this port’s been removed - probably to be used on a Revision B board just for testing.
In this socket here - this is labeled u14 on my diagram - I labeled that as number two - that definitely looks like there was something in this position at some point.
So this one was actually relatively easy to identify it’s in the same position on the Falcon, on the final version of this board, it’s a Motorola MC68901FN, which is a a multi-purpose I/O controller chip. It’s more than likely for the cartridge port. There’s quite a lot of traces going in that direction on the prototypes. On the Sparrow boards that was actually populated with an ST TS68HC901CFN4B chip - which is quite a mouthful - but that’s actually just a clone of the Motorola chip so I assume that’s just what they could get their hands on at the time when they were building these, and in this next position here there was obviously a timing crystal - based on the photos that I’ve seen that was a 2.4576MHz crystal, which was probably used to clock this chip - just as a communications - just to clock the cartridge port there. So that’s a very very common component.
This motorola ST chip here is also very common off the shelf component so it should be an easy one to track down.
So, just moving a bit further down the left-hand side here, we can see we have two sockets here. These two spare sockets - just have a look in there - that one’s labeled “Spare 3”.
Now it’s interesting that they’ve fitted a socket on here because I found a few pictures of these boards - the Revision As and the Revision Bs - and Atari never fitted a chip in this position. So maybe this was for to make their lives easier when they were attaching probes and things to test communications between various different components.
I’m not really sure.
So moving just a bit further down this left-hand side, we see the joystick ports, and I’ll flip over to the other side of the board as well in a second just to show you how that’s wired up, because that’s quite interesting.
It looks like there were probably two joystick ports fitted to this board, we can see quite an impressive bodge here where we’ve got these two filters - they’ve obviously managed to wire those up the wrong way around so the engineers have desoldered those and resoldered them crossed over which is quite interesting to see.
Quite a few missing capacitors here as well on this joystick port - and the reason for those if I just flip the board over…
…that’s because they’ve managed to mix up some of the pins on the joystick port, so they’ve just put some bodge wires in underneath here just to connect them up to the right places.
Now, that heavily implies to me that this board was fully up and running and working and they were testing the joystick functionality on this, so perhaps that’s a strong argument for restoring this and getting it working again.
I think just while we are on the other side of the board - and I’m probably going to flip this over a few times over the course of this investigation - but just just as this is the first we’ve kind of seen of it, as you can see there are plenty of bodges on here: we’ve got some capacitors on top of capacitors, and presumably values that they’re just trying to tweak and trying to sort of finalize the values of those.
There’ll be things like decoupling capacitors and just smoothing out the power supply to the chips.
We also have some bodge wires here. These will have been turned into traces - actual motherboard traces on the PCB - in the final version of this board. But obviously the whole purpose of these prototypes is to just get these connections finalized and work out what connects to what.
Most of them are still connected.
I have come across a few that have just come loose, I wouldn’t want to guess where these connect to, I mean some of them I think it’s quite obvious.
That one there, you can see you can see where the solder’s kind of burnt there as well, so I think that’s probably safe to assume that that went to there.
But yet again, it’s kind of another good argument to get hold of some high resolution photos of working versions of these boards to try and work out where these are supposed to go.
Another interesting thing about this mod by the way - just incidentally - is that it is slightly warped, and some of these components on the bottom you can see have kind of scrape marks where the printing has been scraped off.
I think that’s probably just where it’s been in storage and it’s had something stacked on top of it. Obviously once they moved on from the Revision A to the Revision B boards these things probably wouldn’t have been a priority anymore and were just used as a source of parts for salvage - for scavenging - for building later prototypes. But these chips these did make their way through to the final Falcon design.
I’m not really sure what they do.
Also: this is potentially quite interesting.
These are actually just jumpers rather than chip sockets, but I assume this was just the easiest way for them to make them reconfigurable. Obviously Atari were at the stage where they were kind of debugging some of these connections down here so based on the numbering of my diagram we’ll go back up to the top again and we’ll just sort of investigate this area where these ports are.
So we have two ports missing here - this is the network port, these are the 9-pin serial and parallel 25-pin parallel printer port.
Looking at it, I think it’s pretty obvious to see that these were fitted at one point and they’ve been scavenged - like I said before - probably for a Revision B board.
There’s quite an interesting chip in this area actually which I’ll just get a close-up shot of. Now as you can see, it’s partially lifted out of its socket, but none of the traces or pads have been damaged and this is a theme that we’re going to come across quite a few times as we have a look at this board, and I think what’s happened is when the Atari engineers needed to salvage the chips and the ports and various other parts from this board is they’ve just chucked it into a reflow oven and um just plucked off the parts that they needed, and obviously this chip wasn’t needed, but must have been dislodged in the process and it’s ended up kind of half hanging out of its socket but still firmly soldered into place, which is quite interesting.
I think another piece of evidence for this theory would be these capacitors around here - these have some heat damage to them - you can see the outer packaging is all melted on these. Perhaps someone being a bit overzealous with a heat gun, or like I say, maybe it’s been overheated in something like a reflow oven.
I’m not really sure.
I’d be interested to hear your theories on that down in the comments.
Also in this area there’s a chip labeled U20 - I’ve labeled that as number 10 on my diagram. I think this is probably just a bog-standard TTL 7400 chip, but I can’t find any decent close-up pictures of this chip, so if you are familiar with the Sparrow motherboard it would be very useful to know what this is.
So moving down a bit further, we get to U87 and U29 just here. U29 is quite an easy one - that’s a Dallas real-time clock chip.
That is something that did make it through to the final Falcon design. It was moved to sort of this area of the board down here.
These are very problematic - they have a battery built into them and the battery eventually dies, and there are modern solutions to fit an external 2032 battery, or you can even - I think - cut them open and hardwire an external battery into them.
For authenticity’s sake, if I did go down the route of rebuilding this motherboard I’d probably want to fit an actual proper Dallas module.
While we’re in this area, we have these two sockets here labeled U32 and U87. These - well I say sockets, these are obviously pads for SMD ICs - on my homemade diagram they’re labeled number 8 and number 12, so number 8 in position U87 here, this is a Zilog serial communications controller. The part number is Z85C30VSC - that’s a very common off-the-shelf part, so that would be easy enough to get on eBay for a couple of pounds.
This chip labeled U32, which is position 12 on my diagram, is a Yamaha Y3439 chip, and that is actually a sound chip, and it’s the surface mount version of the YM2149 which was used in the Atari ST and that is a chip that did make it through to the final Falcon design as per the Zilog serial communications controller and the Dallas real-time clock.
So one theme that is obviously starting to emerge with this motherboard is that the vast majority of the chips that we’re coming across were finalized and did make it through to the final design, so perhaps that’s an argument in favour of rebuilding it because they should be relatively easy to track down.
In fact I have actually sourced some of these chips, and the pretty much all of them are available from Best Electronics in California and should be not all that expensive to get hold of and to rebuild. Or if there are any “spares or repair” boards out there that we could get these from perhaps that’s an option that could be investigated as well.
So moving along - we have this chip in position 59 here, this is just a 7400 series TTL logic chip. I’m not sure what the purpose of that is.
It’s interesting that the socket is obviously far too big for the chip and that they’ve put a resistor across here. Yet again I think this is just a prototype thing, just a way of routing signals to different places.
There are also a few jumpers on this motherboard - some of these made it through to the final Falcon design and I haven’t yet investigated the purpose of these, so I’d be interested to learn about those.
Just coming down a bit, this is the 68030 expansion connector for the CPU.
Next to that is a standard Atari Falcon power connector, so the power supply should be relatively easy to sort out, no problems there.
Down in this area of the board we have this interesting collection of components.
I’m not really sure what’s going on here - obviously lots of bodging - so it seems whoever designed this motherboard got these transistors the wrong way around - pretty much all of them - because they’ve all been soldered in backwards here and they all have an extra insulator on one of the legs, obviously because they’re all kind of twisted out of position just to stop them shorting out.
So that’s quite interesting.
They’re between the 68030 expansion connector up here and the actual 68030 socket, so maybe they’re related to that. Obviously they’re also next to the COMBEL chip socket so perhaps there’s something to do with the interface between the two.
This is probably something that Atari were in the process of debugging at the time when they were designing and building these prototype motherboards, so interesting area there.
Obviously there’s a small surface mount chip.
It’s good to see that this is all still intact.
I assume these jumpers are original as you can see quite a few bent bits here as well where the board had stuff stacked on top of it over the years.
Just to the left of this area there are four sockets, these are for the GAL chips.
GAL chips are quite interesting in that they’re programmable logic chips, so they’re a bit like EPROMs but rather than holding data they they’re actually logic. So AND gates, NAND gates, that kind of thing.
I’m working on the assumption that these were finalized by this point and that the Falcon chips should just drop straight into these sockets and just work.
I have a list on my spreadsheet - I did manage to find a reasonably high quality photo from Atari-Forum which actually showed which of these chips were which, and next to those we have the 68030 CPU socket so that’s a relatively easy and common chip to get hold of. It should just drop straight in.
There are two different versions - there’s one that’s - I’m not sure of the pin count on those - but there’s a smaller version and a larger version. This is the smaller version.
It’d be good to get hold of a nice ceramic chip to go into this socket, just because it would look the part, and that’s what they used originally.
There is a 68000 socket just here as well. I’ve never seen this socket populated on any of the pictures that I can see. I’m not sure if it’s something that Atari were going to fit to provide backwards compatibility with the ST machines, and in that vein there’s also a socket for the 68882 which is the FPU - I think - for the 68000 series CPUs.
Again, this was an optional extra in the Mega STE machines and I’ve never seen it populated in any of these Sparrow motherboards that I’ve come across, So I’m working on the assumption that I wouldn’t need to track down chips for either of these sockets if I did go down the route of rebuilding this.
Also in this area of the board we have the two TOS roms.
These machines ran TOS 2.07, which was a slightly enhanced and improved version of the Mega STE’s TOS which was 2.06.
The final Falcon machines ran TOS 4.04 - I think - or 4.4.0.
I think maybe there are earlier machines that run 4.02, but 4.04 was the final Falcon TOS.
Someone has imaged these, so if - I do have an EPROM burner - if I find the right size EPROMs which I think are relatively common I could burn the image to those and we could have it booting TOS 2.07, obviously provided we have all of the other chips required to make that possible.
Just while we’re down here we have these jumpers - these are to do with wait states I think - I do have good photos of these in other machines so I can set the jumper settings as they’re supposed to be, although I have seen some reports from people that they’ve tried changing these on their working prototype Sparrows and they haven’t made any difference at all, so perhaps for a feature that hadn’t been implemented yet.
Sparrow prototype motherboards use the same RAM expansion connector that is used on the Falcon on the final design, which is a daughter board that plugs into these two pin headers here, and just sits over this area.
Also in this area we have two sockets labeled “Spare 4” and “Spare 2”.
I’ve never seen these populated on any of the other Sparrow motherboards - whether working or not - so I think we can quite safely disregard those.
So we got ahead of ourselves a bit there…
…just moving back up to the top - or the the rear - of the motherboard, this is obviously where the ports are so this is the rear of the case in the machine, obviously.
There’s an RF modulator just here, that’s quite a nice small compact unit, and I think that’s the same as in the final Falcon design.
Next to that is the video connector, which is a DB-19 connector and that was used on the final Falcon design.
They aren’t actually manufactured anymore.
They’re also used for the hard disk connector on the ST, I believe, so probably salvageable from another machine.
So, moving down from there we see another couple of transistors which were designed the wrong way around - or have been fitted the other way around - I’m not sure if Atari changed their mind on the type of transistors that they wanted to use or if they went from PNP to NPN or - I’m not really sure with that. It’s a bit strange that pretty much all of the transistors on this board have been fitted that way around these resistors are on here.
I assume these are to to do with video output and just getting that to the right levels. As you can see, some of them have kind of popped out and are in at funny angles - I think they’re also victims of the reflow oven.
So just moving down from there, we have these two SMD chips which are labeled U43 and U40, and these are numbers 21 and 15 on my diagram and on my spreadsheet.
U43 here, based on the picture that I’ve seen online, is a triple 8-bit D/A converter - a digital to analog converter - so that’s perhaps something to do with the video output, or of course we have the DSP connector just here so maybe something to do with audio output.
But that’s a very common off-the-shelf part so it shouldn’t be too difficult to track down.
As you can see, there’s quite a lot of flux in this area - I’m not sure if this has been reworked at some point when it was being built and when it was being debugged.
All of the pads, as per most of this board, do seem to be in good condition and intact so it wouldn’t be a big job to attach another chip there.
U40 here is the first of the Atari custom chips on this board that we’re going to talk about and it’s quite an interesting one, because I think this one is actually the final surface mount version of the chip rather than the prototype versions as per the other two chips on the board, which I’ll talk about in a minute. This is the Atari SDMA chip, which of course is the DMA controller chip - Direct Memory Access controller chip - it seems to be exactly the same version that was used in the Falcon in the final version.
But they are available from best electronics, and they don’t cost a fortune, so that’s something that could be replaced relatively easily.
So just moving down…
Just below the SDMA chip we have this socket here - or I say socket, it’s actually two sockets next to each other of different sizes - it’s one of my favorite things on the board because it’s obviously where they didn’t have the correct size socket in stock when they assembled this thing, so they’ve just cobbled two together!
This is the fifth of the five GAL chips.
I think this one is to do with memory control but I will put the information up online as I mentioned before.
This was actually moved slightly on the Falcon motherboard design - on the final design - over to this side, but it seems to be exactly the same chip.
So there are two custom chips here which I think may present a problem if I was to go down the route of rebuilding this board, and at the point that these Revision A Sparrows were manufactured and I think the Revision Bs as well actually, these were socketed chips and these are the COMBEL chip here - which is kind of a general I/O controller chip that talks to the I/O and the CPU and the RAM and all that kind of stuff - kind of ties all that together - kind of like a northbridge in a modern PC I guess, and there’s also the VIDEL chip here which is to do with video output, and both of these were custom designs by Atari, and at the point that these boards were made these were socketed prototypes.
Now what I’m not sure about is whether these are actually the final chips, and it’s just that the the final surface mount components weren’t available yet so they built these boards around the socketed versions, in which case it would be possible to build an adapter to convert the Falcon surface mount chips to work in these sockets.
This is a 208-pin socket here and this is a 120-pin socket.
I have had a quick look for adapters and haven’t managed to find anything of the appropriate size for these chips if I were to go down that route, so something would need to be designed, but I think that’d be relatively easy.
So that one’s maybe up in the air of course. If there are any Sparrow owners out there or any other Atari collectors who have managed to get their hands on these prototype chips I think that would be a really really useful thing for me to have - maybe even just to borrow - just to try and get this thing up and running.
So if you’re watching this video and you do know a source of these I would be very interested.
Another thing that’s kind of interesting while we’re in the region of this COMBEL chip here is this IDE header just here.
This is something that did make it through to the final Falcon design. These had a 2.5” internal laptop drive which plugged into this IDE header, so that’s obviously been populated. I assume the IDE controller is part of the COMBEL chip so that’s something that must have been actively being tested while this motherboard was working.
One header that does seem to be missing is the floppy drive connector over here - I’m not sure if this is - it looks like there probably was something there and that’s been removed.
A bit strange that they would have removed that, but maybe it was a cable that was soldered directly to the motherboard and they didn’t have any spares.
Another interesting thing in this general area is the SCSI connector here.
Now as you can see, that’s melted, and obviously the Atari engineers in their mad rush to kind of cannibalize as many parts as they could from this board and start building up those Revision B boards have managed to melt this socket and it’s kind of half hanging off.
But much like the other stuff that’s been removed - and been partially removed - the traces and the pads underneath are actually perfectly fine and this could be soldered back into position, and would probably work.
Maybe in the name of preservation and preserving the history and the story of this board, maybe it’s more interesting to leave it hanging off.
I’m not sure, I mean that’s kind of one of the things I’m looking for some feedback on.
Of course next to this port we have component U77 here - this is number 22 on my diagram and on my spreadsheet and that’s a Logic l53C80JC which is a SCSI controller chip.
Obviously it makes perfect sense that there would be a SCSI controller chip here next to the SCSI port, the Falcon and the Sparrow motherboards only support external SCSI - they don’t support internal SCSI drives.
Quite interesting that they went with IDE for the internal hard drive and SCSI for external.
That’s quite an interesting mix there, the IDE controller is infamously not a a brilliant implementation in the Falcon, the SCSI is much, much faster and people using the Falcons for music production and that kind of thing would usually have gone with an external SCSI drive over an internal IDE drive just because of the much better performance.
Also just moving down from there, we have the one custom Atari chip that they did decide to leave in place and not salvage, and that is this C302096 which is the AJAX high density disk controller chip.
So this is the floppy drive controller chip.
As you can see, there’s some bodge work been going on here as well - quite a few bodge wires in this general area.
There’s a couple of those I just skipped over that looked like they connected to that SCSI controller chip as well, so would need to work out where they actually need to go to do their job properly.
Just next to that floppy controller chip we have the video chip - of course prototype chip - I was just talking about, and moving down from there we have this labeled U500.
Now I’m not sure what this is - I don’t really want to remove this sticker because it is part of the history of the machine and all of the Sparrow prototype pictures that I have seen do have this sticker on top of this chip, so I’m not really sure what it is.
I assume maybe it’s something to do with the SRAM - so we have socketed SRAM chips here, three of these these exact chips did make it through to the final Falcon design.
They’re fairly common off the shelf Alliance SRAM chips. So the part number for those is AS7C256-25PC 25 being 250ns timing of course.
The 25s do seem a bit harder to get hold of than the other speed ratings, but I think if I was going to rebuild this board I would use the 250ns parts just because they’re the ones that I’ve seen in all of the Sparrow prototypes that I’ve come across so far, and obviously that would be the most authentic way to do it.
From this chip here, we can also see that there’s another bodge wire, which comes up to U18.
U18 is number 24 on my diagram and on my spreadsheet, and that’s the Motorola 5600FE33, which is a DSP chip - a Digital Signal Processor - that did make it through to the final Falcon design as well.
So I think the general theme that we’re coming across here is that the vast majority - in fact pretty much all of the chips - that we’ve come across so far are the final chips for the final production machine.
These Sparrow prototype boards were fairly late on in the development cycle for the Falcon - I think it was probably six months before the the Falcon was actually finalized and went into production when these boards were made - so very much just a case of just using these to finalize some of the traces and some of the routing and that kind of thing, I think.
So just back up to the rear of the board here again - this is the DSP connector, which is obviously the final Falcon design for the DSP connector.
The fact that this is all wired up kind of implies that they were testing this on this actual board.
We have the audio in and audio out connectors here - obviously one’s present and one’s missing - I’m not quite quite sure what the part number is for this and how I would go about tracking another one down but I assume that’s a fairly common off-the-shelf connector that I could I could probably find as per the rest of the board.
There’s some quite interesting bodge work around this area - a resistor there that’s kind of been patched in and has some insulation on there just to stop it shorting out.
Another thing that I wanted to leave until last because I think it’s one of the most interesting things on here, and I think it’s also something that’s going to be impossible to track down now: this socket is for the audio codec chip, and this socket is in the same position as it is on the Falcon, but it’s a different pin count and the final audio codec chip on the Falcon was a Cirrus Logic CS4216-KL.
But the pictures that I’ve seen of the Sparrow motherboards - and I’ve only managed to find this chip on one picture and it’s kind of in the background, it’s quite hard to see - but it’s in one of the pictures from atarimuseum.de, and it’s a Toshiba TC24-SC160ES I think, the writing’s not entirely clear.
But it’s a chip that I’ve not really managed to track down and I think - looking at that actual part number - I did manage to find some information to do with a chip architecture and that kind of thing, and I think it refers to the architecture of that chip rather than an actual part number of a final chip.
Which implies to me that maybe it’s a prototype chip that came from Toshiba - maybe Atari were working with toshiba to finalize that audio codec chip and for maybe time or cost reasons they changed their mind very late on in the development process and decided to go with the Cirrus Logic chip instead.
So I’m working on the assumption that that will be impossible to track down unless anyone has any. I’m not sure, and to be honest I don’t think that the machine will actually need that chip to be able to boot.
I think it will probably quite happily boot with that socket unpopulated, and that’s probably the case with quite a few of the chips on this board, other than the the main things like the CPU and the custom Atari ASICs on here.
We could probably get away with actually missing quite a bit of this - and I think some of the working boards out there are indeed missing some of those parts - so yeah, an interesting thing to end on and an interesting thing to think about, I think.
So that’s the tour, and a look at some of the components which are present on the board and obviously some of the components that are missing, just as a higher level overview.
Just another couple of interesting things to note about this Sparrow motherboard - the keyboard connector here looks the same as the Atari ST keyboard connector.
I know that Atari actually started producing cases and keyboards for these machines - the Sparrow keyboard I think has a different font and slightly different colour print on the keycaps, and that Best Electronics actually came to be in possession of some of those and are now sold out of them.
So there are keyboards out in the wild that will plug into this and weill work with it, but to be honest I think the ST keyboard will probably work with it as well.
This area here is where a cooling fan went in the Falcon, this is an off the shelf component. There’s just like a metal shroud and a standard fan that go in here.
This doesn’t look like it was ever populated and to be honest it makes sense that they would never have bothered to populate that for testing purposes because it probably wasn’t in a case.
All of these Revision A boards are loose, there are only Revision B boards in cases if that makes sense.
So that probably is some indication of the timing of kind of the development process and and where they were with that at the time.
So yet another aspect of this prototype Sparrow motherboard, and an interesting part of its story and its history, and definitely something that I’d be interested in learning a bit more about.
So now we’re all intimately familiar with the condition of this board and the parts that would be needed to rebuild it I’d love to know what you think.
I said at the beginning of this video that I want this to be a community effort - do you think it would be a massive waste of time and money to try to rebuild this thing and get it back to working order? Or maybe do you think there’s a chance that it could be rebuilt?
Also the historical perspective: is this an important piece of history with a story to tell in its current condition, or would that story be better told with it fully rebuilt?
I’m on the fence.
To be honest, as a tinkerer, I think I would probably rather get it up and running again but that’s purely for selfish reasons of wanting to go through with the the soldering and the debugging and everything else.
But if there’s a strong feeling against that I’ll quite happily just stick it in a box frame and have it on the background as part of my set - that’s perfectly fine as well.
I’ve also tracked down a lot of information about this already - like I’ve already said - so I will link to that down in the description just so it’s out there and documented.
I’d also like to put together some more information on my website, I think, and also make that available to Mikro on his website just so it’s preserved and there’s as much information out there as possible about this project, because I would hate for it to get lost to history.
If you enjoyed this video and found it educational and/or entertaining, please do give it a thumbs up - it helps me to increase the visibility of my videos and helps me to grow my channel in the future.
Of course, if you’re not subscribed already and you’d like to keep up to date on this project and my other projects please hit that subscribe button, and hopefully YouTube will keep you informed.
Finally I’d just like to thank you very much as always for joining me, and I hope to see you again soon.
Original Video Links
Relevant Links:
Mikro’s Sparrow serial list: https://mikrosk.github.io/sparrow
Atarimuseum.de FX-1 / Sparrow photos: http://www.atarimuseum.de/falcon030.htm#fx-1
Sparrow Schematic: https://www.chzsoft.de/asic-web
Atari ICs from Best Electronics: http://www.best-electronics-ca.com/custom-i.htm
Falcon motherboard pics: https://www.flickr.com/photos/116852484@N05/with/47986521323
Sparrow Mega STE plugin prototype: https://www.atari-forum.com/viewtopic.php?t=38517
Mikro Sparrow thread with links to more information: https://www.atari-forum.com/viewtopic.php?f=27&t=31148
Another big Sparrow thread with pics of running boards: https://www.atari-forum.com/viewtopic.php?f=33&t=32073
My attempt at a Sparrow BOM spreadsheet: https://ctrl-alt-rees.com/downloads/atarisparrow/Atari%20Sparrow%20Rev%20A%20BOM.csv
My labelled diagram referenced in the video: https://ctrl-alt-rees.com/downloads/atarisparrow/Atari%20Sparrow%20Labelled%20Diagram.jpg
If you liked this video please consider subscribing to ctrl.alt.rees on YouTube!
