I think we agree, but I would say computers are used _more_ to crunch numbers. Inside the system, everything is binary; you only need decimal to present data to humans.
With old time COBOL workloads, a lot of the stuff you do was presenting stuff to humans. The typical banking application, for instance, reads numbers, does a few additions, and shows them on a terminal or prints a report. Using fixed-point arithmetic was normal.
Nowadays, we do way more floating point arithmetic (for graphics, for machine learning)
Also, I think the reason those complex instructions aren’t created anymore is pipelining. I can’t find timing information for the ED instruction, but I think it is a safe bet it isn’t constant cycle count, let alone single-cycle. It not only writes a character buffer, it also has to read the bytes containing the pattern defining how to format the number.
What you are saying is only partly true. Yes, the packed decimal instructions are horrible today from performance perspective, because they require direct access to memory.
However, recently, IBM has added decimal floating-point arithmetic to z/Architecture CPUs (and also made it part of updated IEEE 754 standard). These modern instructions do not suffer from the problems of old instructions, and in fact should be preferred to packed decimal for applications that need decimal point arithmetic.
And more broadly, IBM still adds a lot of specialized instructions to mainframe CPUs. For example, calls to cryptographic accelerators.
With old time COBOL workloads, a lot of the stuff you do was presenting stuff to humans. The typical banking application, for instance, reads numbers, does a few additions, and shows them on a terminal or prints a report. Using fixed-point arithmetic was normal.
Nowadays, we do way more floating point arithmetic (for graphics, for machine learning)
Also, I think the reason those complex instructions aren’t created anymore is pipelining. I can’t find timing information for the ED instruction, but I think it is a safe bet it isn’t constant cycle count, let alone single-cycle. It not only writes a character buffer, it also has to read the bytes containing the pattern defining how to format the number.