Ferranti ULA doc may be of interest

[Moggy]

I posted this some time ago in the old ts1000 forum, may be of interest even if only academic.


Regards
Moggy

[Andy Rea]

Having done a search this thread seems like a good match... i stumbled upon this today interesting stuff

https://ia801203.us.archive.org/21/item ... asheet.pdf

now then if any is clever enough to decap a ULA and photograph the final layer, it would be possible to reconstruct the exact logic hiding inside the little vlack chip

regards Andy

[Moggy]

I'm almost sure Mark posted something re this PDF some time ago and gave an explanation of the ULA's heat output being due to,amongst other things, a collection of voltage regs scattered about the chips periphery.

As for the decapping I have seen a collection of stills for the 81(I'm aware of the many spectrum ones) but have lost the link I thought I'd bookmarked.

I wouldn't bother though because some dodgy geezer in the Midlands is shipping them out by the bucketful.

[Andy Rea]

As for the decapping I have seen a collection of stills for the 81(I'm aware of the many spectrum ones) but have lost the link I thought I'd bookmarked.

I carry on digging

[RetroTechie]

Pretty useless in practical sense. But nice reading material anyway.

now then if any is clever enough to decap a ULA and photograph the final layer, it would be possible to reconstruct the exact logic hiding inside the little vlack chip

Getting at the bare IC die isn't the hard part... Probably would have sacrificed a few IC's for this purpose myself, IF there was some easy way to tell what you're looking at.

Seeing some metal strips on the IC die or following connections from a bond pad around is one thing. But is that a P-channel or N-channel FET? Or just an integrated resistor / tiny capacitor? Or for a ROM area, yeah you could perhaps spot the 1's and 0's in there. But what's the rows & what's the columns? How are those bits used inside the IC? Never mind that some structures may only become visible after 1 or more layers of material are etched away.

Wish there was some easy guide for this. Several examples to be found these days (with pics ). But a how-to for someone who's never seen an IC die up close.. haven't come across that (yet).

[Andy Rea]

Hi Mark,

I believe in this instance you only need the interconnect layer to decide the whole thing, each cell is identical and the doc tells you what each cell is. Yes it's hard work but not impossible.

With regards to an unknown device then yes I agree a high level of expertise would be needed to decipher the different layers.

Regards Andy

[Moggy]

Found it!!

A video of a 2c184 ULA being de-capped and showing its various interconnections, it's French language but the picture should suffice.


The relevant bit starts about 8 mins in and if you pause the video at 10:07 you can compare the ULA cell matrix to the one in the pdf posted by Andy, they look similar but not quite the same.

https://www.youtube.com/watch?time_cont ... xLtgs-2mwA

[Andy Rea]

hmm very interesting... bummer that the cell layout isn't identical...

@Mark 1 helping of humble pie i must eat

[gammaray]

Trying to contact user but form does not work for me...

[Andy Rea]

i also tried... seems no way to answer the " are you human? " question... wonder if it requires flash ?