The Ferranti ULA6C001E is a custom circuit used in ZX Spectrum computers and some of its clones, notably Didaktik Gama.
ULA is basically an "Uncommitted Logic Array", a kind of predecessor of today's CPLDs and FPGAs. This was a technology developed by Ferranti of which Sinclair was one of the early adopters in the ZX81 (incorporating all the ZX80 discrete logic plus NMI generation circuitry into the original ZX81 2C158E ULA).
A similar implementation of ZX-Spectrum compatible logic is the ULA1 in many eastern clones, but it is not a replacement, it just implements similar functionality.
Mr. Drexler's article in ZXM 2/91 states:
About 2 years ago, Tesla Rožnov managed to construct (a more accurate term would probably be "copy") the ULA circuit, which is the basic, so far difficult to replace, heart of the classic ZX Spectra, but also ZXS+ and Didaktik Gama. The circuit is marked MH 6001 and we had the opportunity to test it for a short time. In general it worked O.K., only sometimes when turning on the computer froze and the screen remained dark, which did not happen even once with the original ULA. Since no more of these IOs were available, it could only be a defect in the given sample. The temperature of the case after longer operation is comparable to the classic ULA, smaller differences were found in the sensitivity when recording programs, the image frame (BORDER) was moved a little more to the left. However, it must be said that similar small differences can also be found between individual types of original ULA. When reading the MGF input empty from address 254, we get 191, which corresponds to the newer ULA types. Because of this, some programs, e.g. SLOWLOADER, do not work, fortunately, this feature or, if you want to additionally remove the error, can either be modified by SW or HW.
Also on the discussion forum OLDComp.cz:
A good thing happened.. For a very long time I tried to contact the person at Onsemi, if he had any information about the ULA for the ZXS that was supposed to be produced in Tesla Rožnov... And here is the result:
At the time, I was working on the design group of digital customer circuits. (ZIO, approx. 3 people) .
We tried to copy the circuit for Didaktik on the basis of the gate field, which at the time was also developed (not copied) for ZVT (computer technology plants)
First, we reverse engineered the original ULA, we drew the wiring logic from a micro photo layout (2x2m). We never simulated the overall ULA scheme, it was really a copy of the gate connection.
This method of IO construction was exceptional for us, we have never used it before or since. We always developed customer logic circuits according to the scheme (keyboard controllers for Zbrojovka Brno, CRC controllers, and multipliers for ZVT Zilina screen computers.....)
The attempt to develop a copy of ULA was rather an unofficial event of enthusiasts, without official support and deadlines. As a development after working hours... The second or third pass started to work a little in didactics. At that time, ZIO did not even have comprehensive testing support, after encapsulation, didactics were tested by inserting them into the board. This approach to ZIO development has never been practiced before or since. The didactic then backed away from further cooperation. - development never finished. I think they found a channel to get the original ULA..
The development of the 200 gate field has been completed, the development of the 1000 gate field remains in progress...
Hello Joseph
PS I still have the trial Didaktik from that time hidden away..