Path: csiph.com!usenet.pasdenom.info!aioe.org!.POSTED!not-for-mail From: Marc Schlensog Newsgroups: comp.os.linux.alpha Subject: Re: WTD: UP1500 or UP2000+ motherboard Date: Tue, 13 Nov 2012 10:02:14 +0100 Organization: Aioe.org NNTP Server Lines: 55 Message-ID: <20121113100214.714906a8@Crunch> References: <56214f6d-8b62-4e21-a4fa-49e62f41481d@googlegroups.com> <20121111093737.6a64132d@Crunch> <20121112071113.0380e66a@Crunch> <344e47af-bc69-434f-86ee-e006b50161a9@googlegroups.com> NNTP-Posting-Host: fWz6a8WIGNRStvDWHOW2VA.user.speranza.aioe.org Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@aioe.org X-Notice: Filtered by postfilter v. 0.8.2 X-Newsreader: Claws Mail 3.8.1 (GTK+ 2.24.13; x86_64-redhat-linux-gnu) Xref: csiph.com comp.os.linux.alpha:14 On Mon, 12 Nov 2012 15:54:48 -0800 (PST) tony@tonyjones.com wrote: > On Sunday, November 11, 2012 10:11:17 PM UTC-8, Marc Schlensog wrote: > > Steven Hirsch replied to me personally, I hope he doesn't mind that > > I share his valuable information here too: > > > > "A significant number of the UP2000+ boards have chipsets that have > > succumbed to electro-migration in varying degrees. I gave away my > > system last year to another collector. It was progressively losing > > its ability to work with multiple memory modules installed and was > > down to > > > > 512MB when we parted ways. > > "Even if you can find a working UP2000, don't count on it for > > long-term reliability." > > > > This is consistent with my experience with a lot of CS20 (which > > basically are UP2000+ systems in a 1U enclosure) and various other > > reports. While the API systems are to varying degrees* pretty nice > > on paper, they aren't built to last. The CS20 had another flaw in > > form of an unreliable PSU. > ... > > *) This most certainly doesn't include UP1000 and UP1100 :) > > Is this saying that the UP1000/UP1100 isn't nice on paper or don't > have the same long-term reliability issues as the UP2000(+) ? The UP1000/UP1100 aren't even nice on paper. What makes them borderline interesting, is the use of AMD chipsets, but the benefits of that never came to fruition. The UP1000/UP1100 are very memory limited, in size as well as in bandwidth (max. 768MB, single channel SD-RAM). The UP1500 on the other hand is much more useful in that regard (up to 4GB DDR266). > I'd heard several other reports of the UP2000(+) not being long term > reliable. Thanks for confirming. Is the unobtanium UP1500 similarly > afflicted? I don't think so. > If someone wanted an Alpha workstation which supports Linux, what > would be the intersection of "fast" (for 2001 era), quiet > (workstation), reliable and reasonably obtainable be? ATX form > factor isn't essential, the UP1500 just sounded like a nice system > due to EV68/AGP/DDR. The UP1500's AGP port is a bit... special. Don't expect wonders from it. If I had to chose a decent system, it would probably be a DS10 or, if you can get a hold of one cheaply (yeah right, haha!), a DS15. The largest I'd go for a hobbyist system, is a DS20E, DS25, if you can live with them consuming ~500W doing nothing. b.r., Marc