ALESIS NANOTRACKER

[macg]

All Electronics, a salvage company in california, has a stock of 80 units, as is, no manual, warranty etc.

 

a few of us on a yahoo group (rolandbossdr5supportgroup) bought them with full awareness of their shortcomings, just to see what was up. they are, indeed, possessed of very buggy software, which is capable of rendering them almost useless, except to an embedded engineer type, such as myself.

 

i have been working on hacking and repairing the units for a few weeks, and have made a lot of progress and plan to patch the code to make it possible to use the working features of the product and prevent it from self-destructing.

 

the software was penned by amateurs. we are using just the download/playback features and i am planning to design a little product that performs only that portion of the function set of the NanoTracker.

 

alesis had a good idea, and this is not the first example of an outside service provider torpedoing a worthwhile design. most of it is actually pretty good, hardware wise.

 

alas, the software sucks mightily.

 

if you wanna read a bunch of posts on the topic, head over to the yahoo group and read away! i am posting the hack progress periodically, and expect to have some patched software in a few weeks or so.

 

regards,

 

MACG

[Rod S]

Originally posted by macg:

a few of us on a yahoo group (rolandbossdr5supportgroup) bought them with full awareness of their shortcomings, just to see what was up. they are, indeed, possessed of very buggy software, which is capable of rendering them almost useless, except to an embedded engineer type, such as myself.

Why? Why bother? Curious minds want to know....

 

The functionality exists in other instruments, this thing wasn't exactly ground breaking. Why take the time to try to fix this?

 

Out of curiousity, how did you get into the o/s? Most of these units have some kind of o/s or firmware that's downloaded into an eprom or something, and it's hard (if not impossible) to reverse the process. How did you get into the code?

 

I'm an engineer too... and have the annoying habit of trying to fix things But... I'd guess I'd rather find an old b-3, prophet 5 or minimoog and try to get that working... Just curious why you guys are going thru the trouble. I don't think this thing will have any resale value if it's 'fixed'.

[macg]

hey rod,

 

all valid observations! my interest is two-fold. first, i am working on because i can and i find it a challenge. the other is that i am interested in developing a product with a subset of the nanotracker functions and this is giving me some exposure to the right and wrong ways to do this, and exposure to some potential users.

 

i have about 100 8051 designs under my belt, so when i saw this one, much of it was familiar territory. in attacking a lock-up problem brought to my attention by one of the other buyers of this klunker, i discovered that the firmware is contained in the flash memory also used for sone storage. i recognized certain recurring characteristics of assembly code (such as a long jump at the reset vector at 0000, timer setups, interrupt service routines, and literals.) i saw inside the memory initially because i wanted to see what was stored in the flash after an initialization sequence, and a unit in which i stored a very simple midi sequence. once i realized that the firmware was in there, too, i took the eprom binary files and ran the through an 8051 disassembler to get a rudimentary listing of the firwmare, which is accurate, but not very clear. it has no symbol names or variable names, of course, but it does have to manipulate resources like the IO, the stack pointer, external memory, etc. and i can determine what parts of the 16K firmware do what. so i am working on that.

 

i have a bunch of assets that i use for microcontroller development that work just as well in reverse as forward, and i supplement the technology with 25 years of design experience, a lot of which is 8051 related.

 

will i make any money on it? maybe, but who knows. i could make 4 or 5 thousand bux just repairing and reselling the remaining inventory, but i am not sure i want that headache for that amount of money. i would rather piece my own design together and make a few bux doing that. this project is helpful on the way to that end.

 

where do you work as an engineer and what do you do? are you involved in electronic music in your work?

 

regards,

 

forrest macregor

[Peake]

As someone who was part of the NT team, I find it suprising how much a batch of 80 unfinished units have caused so much interest on the net. Even when it is well-known that it doesn't work, was never finished, has no manual, and will never be supported by the parent company... There was a long time where "NT" (the product name, such as QS) was a forbidden word at the company... It's just suprising that it has developed such an interest among DIY'ers. I suppose that I have a comparable interest in unreleased analog synth gear but that doesn't quite gel with the NT in my mind ;-)

 

Now that I know that there is a market for unfinished products, I'm starting my own company right away ;-)

[macg]

thanks for the input, mr. peake. i have heard you r name once or twice in regard to this product.

 

orphaned/abandoned/obsolete and/or troubled products are a staple of the ham radio world. heck, you can still find people using osborne, kaypro and commodore computers for some stuff. i wonder, too what the fascination is. however, i also have a working 1938 capehart radio in my foyer that i brought back from the dead, and it seems right at home in my 1891 house!

 

i enjoy 'mind reading' other people's designs. that has little to do with the nanotracker, but having been involved in design and test for a few decades, and with several hundred products worth of production experience, i am used to being critical and working with way too little information. so this is really not all that difficult. for less than 12 hours of work, i prety much know how this thing goes together, though i would dearly love to have some more info on it, such as a coffee-stained source code listing or the maps for the xilinx part, if you happen to know where they are buried!

 

none of the fellows i know who are interested in this expect alesis to support it. most are content to have a paperweight. i am only lifting pieces of the design, which i repeat, is really not all that bad, except the xilinks seems like overkill. a phillips 80c451 would have done the whole job for less cost, real estate, complexity and schedule. the software in the box is not impressive, but it is buggy and unfinished. but the hardware is ok.

 

and this function of a tiny midi playback unit is useful. that is why the interest is there. smaller than a laptop, and cheaper, not floppy based, simple to use.

 

do you know anyone who might have kept and software listings? there are german literals in the code. was there a german subcontractor working on the unit? any leads at all on former co-workers with dusty file cabinets full of career memories who might be willing to share??

 

thanks for the post! i'll let you know how i do on this.

 

regards,

 

forrest macgregor