Talk:DEC Alpha

Archives (Index)

2011 2014 2017 2020 2021 2023

This page is archived by ClueBot III.

Memory semantics

DEC Alpha is notorious for being particularly "weak", in the sense of requiring particular inter-thread synchronisation that is implied by other architectures, in a trade-off between implied synchronisation (easier for programmers) and potential performance (when comparing ideal implementations), with PA-RISC at the other end of the spectrum. There seems to be an opportunity for somebody with specific knowledge to describe this issue where it seems to belong, to help make sense of C++ memory_order_consume. If you are that person, you need no further explanation. RFST (talk) 06:51, 16 September 2024 (UTC)

DEC Alpha listed as 'Open' despite no prevalent evidence

The summary card on this article states that Alpha is open, however from a quick google search I can find more claims that it isn't, and the few stating that it is are forum posts. The cited source itself (https://web.archive.org/web/20131212011841/http://alasir.com/articles/alpha_history/dec_collapse.shtml) only passingly mentions that Alpha is open, and doesn't have any source for that.. I think we should edit it out, or find a more credible citation for that claim. Kierteiyu (talk) 05:44, 18 November 2025 (UTC)

The source said

Although Alpha was declared an open architecture right from the start, there was no consortium to develop it. All R&D actions were handled by DEC itself in cooperation with Mitsubishi sometimes. In fact, though the architecture was free de jure, most important hardware designs of it were pretty much closed de facto, so had to be paid licenced if could be at all. Of course, it wasn't that thing helping to promote the architecture. For instance, if they kept EV4 and EV5 designs proprietary but let LCA4 go free without any licence fees or other obligations implied, it would be a strong move shaking the desktop market well. Soon after the introduction of EV4, DEC offered to licence manufacturing rights to Intel, Motorola, NEC and Texas Instruments. However, all these companies were involved in different projects of their own and didn't want to play the second fiddle at this party, so they refused. Perhaps the conditions could be also unacceptable or something else.

which sounds like "DEC declared it to be open but didn't do much of anything to open it up", i.e. "open in theory but not in practice". Not sure where DEC declared it to be open.

Maybe the person who marked it open meant "you could buy Alpha chips on the merchant market and build systems from them", which, from reading the documentation for {{Infobox CPU architecture}}, doesn't sound to me as if it's what "open" is intended to mean.

As for what DEC said, this page claims that DEC announced that

First, the Alpha architecture, which is full 64-bit, open to any operating system or language (including Open VMS, OSF/1, and in the future NT and real-time operating systems), and scalable to 1,000 times its initial iteraton (with the help of multi-processing). Second, the initial product, the 21064 running at 150MHz, sampling immediately and shipping by mid-year. Third, a newly inaugurated open business practice, described as fundamental departures from DEC’s conventional ways of doing business – namely the licensing of Alpha to other semiconductor partners; the licensing of the manufacturing and design of the chip, allowing copies of Alpha built in accordance to DEC’s design rules, the licensing of DEC software to any Alpha partner; sales at chip, board and system-level; and customer and OEM support from DEC to Alpha users.

none of which are "the architecture is open to independent implementations", just open to other semiconductor partners dabbing DEC's designs.

So I'd say "no, not open in the sense of the template". Guy Harris (talk) 07:58, 17 December 2025 (UTC)