· Content
· News
· Articles
· Mailinglists
· Knowledgebase
· Trouble Tickets
· Files
· Glossary
· Links
· Compatibility Lists
· Forums
Welcome to our website
To take full advantage of all features you need to login or register. Registration is completely free and takes only a few seconds.
Tyan S2895 supports AMD dual chips, two Nvidia chipsets
Posted by: duke on: 09/15/2004 07:58 PM [ Print | 7 comment(s) ]
The Inquirer also has a blip about a new board inbound from Tyan, the S2895.
A THREAD SUGGESTS Tyan dual motherboards will have two Nvidia CK8-04 chips working in tandem. What does this mean? It implies that dual core Opterons might well be able to use two PCI Express X16 lanes - which would be a boon for workstation users. The S2895 Tyan motherboard implementing this sounds like an interesting bit of innovation. And this motherboard will also be able to support four way systems in 2005.You can check out this thread for yourself right here.
« Seagate 3GBps S-ATA and RAID 3 unveiled · Tyan S2895 supports AMD dual chips, two Nvidia chipsets
· Opteron Continues to Outsell Itanium »
Comment
sAvAgE69 Unregistered |
![]() mmmmm yummy ![]() |
Comment
Hydra Performance Junkie Posts: 174 Joined: 2001-03-09 |
![]() Awesome! :cool: Overclocking is a blast! Do I smell something burning? ![]() |
Comment
[312]Seppuku SMP Aficionado Posts: 152 Joined: 2001-01-20 |
![]() Yeah, I'm calling dibs, good thing it's a while before those come out, cause i'm totally broke, lol. Intel Q6700, Asus p5e-vm hdmi with vdroop mod, 4x 2048MB Gskill pc8000 5-5-5-15, EVGA 8800gts 512 pci-e, Seagate 7200.11 500GB Sata2, Seagate 7200.10 500GB Sata2, Asus 18xDVD+-RW, all housed in a Silverstone TJ08 with a Corsair HX620w. (3.2GHZ @ 8x400). |
Comment
sAvAgE69 Unregistered |
![]() This will be an excellent board me thinks. I know My Wife will be needing an upgrade. I searched the web and no pics that I can see. ![]() |
Comment
opus13 misanthrope. Posts: 1574 Joined: 2002-04-05 |
![]() if there is a CK8-04 slung off of each proc, does that mean that if you are troubleshooting your board and have only one chips in, that one set of pcie slots becomes disabled, and potentially half of the board? sounds like it would be a nightmare to work with. |
Comment
jonykaos SMP savy Posts: 167 Joined: 2002-01-29 |
![]() This is the opteron board I've been waiting for. Paired with dual-core processors = 4-way goodness ![]() Can't wait "Where a calculator on the ENIAC is equipped with 18,000 vacuum tubes and weighs 30 tons, computers in the future by the year 2000, may have only 1,000 vacuum tubes and weigh only 1.5 tons" - Popular Mechanics, March 1949 heatware ebay |
Comment
i4004 Registered User Posts: 1 Joined: 2004-09-20 |
![]() I sent the original blurb to the Inq. This conversation (http://www.clustermatic.org/pipermail/linuxbios/2004-August/008796.html) that I happened upon lists Tyan as having the upcoming Tyan S2895 Dual Opteron motherboard coming up utilizing the CK8-04. That isn't strange on itself - but the fact that drivers/LinuxBIOS are being made to manage *two* Nvidia ck8-04 chips working in tandem certainly is news. This shines a bit of light on the fact that the a64 arch still has a few surprises up it's sleeve, as support for this has to be built into and handled by CPU logic - i can only imagine what the interrupt structure should be like between processors, but with HT links it shouldn't be a problem, infact it should shine as each proc has a dedicated upstream/downstream links, so contention access methods are not required (unlike netburst..ahem) . Overall, I have never seen anything like this - I can only imagine how transactions are ordered between both on board Nforce4 chipsets - are there HT link(s) between them? If not, then what? (heck VIA has V-Link for n/sbridge comms, but native HT would obviously be better in this case) One obvious benefit, this would feasibly give you the potential of two full blown x16 PCI-E lanes instead of the split 2x-x8 lanes mechanically compatible with x16 slots going around for consumer SLI. Think of the workstation possibilities here (dual PCI-E Wildcats/rendering farms.. ahh what visuals one will be able to produce!). This fully leaves another x8 lanes for other PCI-E devices. Heck in reality any PCI-E combination you can imagine is possible. To further cover your arse expansion-wise, the added PCI-X HT bridge dishes out a more traditional and readily supported 133Mhz/64bit I/O which is still very much viable; offering 1066Mb/sec at low latency. I found no info on the Tyan web or ftp sites about the S2895. The fact that this motherboard will be upgradeable to a 4-way system in 2005 only sweetens the pot further. Dell should really be feeling the pain soon - they usually jump on new tech. That's what put them where they are today. I really do hope those Intel Inside(R) 'bottom-line' millions they get now are really worth more to them than a long term relationship with AMD, but somehow it doesn't figure. The only one apparently really loosing out seems to be their (primarily IT) client base, really. gerson ricardo network.admin |