Great to see you here Hans!

Thank you. However many people preferred to run Unix on their Vaxen. IMHO it was no good that DEC artificially suspended the development of the PDP-11 in favor of the VAX.

I have no data on the number of VAX's machines running Unix versus VMS. Many did run Unix (universities etc), most did use VMS.

Since the VAX/VMS systems opened the door to many businesses and the amount of support requested by customers in DEC at the time for VMS versus Unix, I can safely assume VMS was more widely used. VAX/VMS made DEC the second computer company in the 80ties, highly profitable and a great place to work.

Now all that changed in the 90ties and the server world now runs on Intel, Microsoft Windows server and Unix/Linux.

Of course Unix also benefitted from the instruction set of the VAX-11.

As I recall, at university the Electrical Engineering faculty ran Unix on VAX, the Computer Science people ran VMS on VAX. But when I went to work in industry, it was always VMS.

For me the VAX is surrounded by two mysteries:
1) Why was it so slow? The first RISC systems outperformed it much. Even the x86 had almost the same CPU performance.
2) Why did DEC miss the Internet? I remember that in the first half of the 90s many people used Sun Workstations for it.

Slow? It was the heart of a minicomputer, often clustered, so comparing CPU speed is difficult, peripherals also count.
With a VT serial terminal it was a nice system to work with. We as engineers were all CLI users. Business apps used a Forms Management system still on serial ASCII VT terminals.
One could get an impressive amount of work with many users out of with good response time.

The VT220 serial terminal was such a pleasure to work with. Good keyboard, sharp display.
When DECWindows came along things changed, the central CPU was replaced for UI with a VAX workstation and ethernet networking to the mini, still fast enough with local processing and central processing.

You re right that the VAX CPU lost to Risc architectures. The Alpha came (too) late due to lack of strategy in DEC. It performed well, but lost out even earlier than the SUN Sparc and others to Intel.

And that answers you second question: lack of strategy made DEC a very slow mover. So Internet was ready for DEC, but DEC was not ready for such a paradigm shift, too conservative, too much directed to developing inhouse (everything was made in DEC factories!) and so used to huge profit margins. And once on the slope down only cost reductions and selling out remained. A great service organization and a very large customer base were bought by Compaq and that was the end of DEC engineering and R&D.

This is a little morbid, but on the plus side ARM was given two big assists by DEC: first by developing StrongARM, which showed the way to faster implementations, and second by failing to take over the world with StrongARM. (Strong ARM went to Intel as part of an out-of-court settlement, and they didn't take it forward because they had Titanium on their minds.)

My computer room needs cleaning ... send in the VAX.
So why did DEC promote the VAX and drop the PDP 10?
Ben?

Intel is known for its crush projects. Their most known project of such kind was about Motorola. IMHO Itanium project was a crush project for Russian Elbrus 3 architecture because they lured almost all its people to work with the Itanium. IMHO when Bill Mensch says that he can make 20 GHz 6502 he implies that he doesn't want to be crushed too.
IMHO Intel still actively uses the StrongARM descendants - so it is still their main RISC processor. It is interesting why didn't Commodore or DEC realize potential of so many good hardware they had?

The answer for both Commodore and DEC is management.

I'm afraid to offer the opinion that Bill Mensch is talking nonsense when he suggests such high clock rates for the 6502. He might well very much want to sell his company, so he always has to make it sound promising.

May be but we all know that IT is forced to mobile device direction and the development of desktops seems to be slowed down much. So maybe if Bill Mensch had several young initiative engineers in his company they would make the next 6502 revolution under his guide. IMHO large companies very slow accepts changes...
But maybe the main problem is mass of people who are satisfied with the current desktop designs.

Would you like to help me to understand software availability for VAX/VMS in the 80s? I know there were good compilers and text editors but I know little about DBMs, spreadsheets, ... It is odd for me that they made Oracle for the PDP-11 but not for the VAX/VMS. They didn't even port it there. I know that there was DEC Rdb/VMS but were there other options? Thank you

At the CSU Sacramento campus in the mid- to late 1980s the converse was true with our two VAXen. TITAN was typically in the realm of the EE crowd and ran VMS. Athena was typically for the CS crowd, and ran Ultrix-32. I assume that both are long gone by now.

An interesting reversal! Perhaps it's like siblings, there's a need to make some distance between yourself and your closest competitor.

I have just added part 14, AIM PowerPC.