We have had very extensive coverage on PCIe solutions from VIA and SiS, and today we complete the circle with some in depth information from NVIDIA's CK8-04; a chipset many have already decided to call nForce4. NVIDIA takes a firm stance in its roadmaps that this next revision of core logic will not use the nForce4 misnomer, so we will continue to call it CK8-04 until NVIDIA calls it something else. We first saw CK8-04 at Computex in June - and even then it supported dual PEG adaptors.

As expected, CrushedK8-04 will come in three flavors, CK8-04 SLI, CK8-04 Ultra and CK8-04. The Ultra and SLI chipsets are simply improvements upon one another, with the SLI chipset being the highest end solution. Vanilla flavored CK8-04 is very much the same as nForce3 250Gb, with the addition of 7.1 high definition audio and PCI Express. We also get four SATA 150 ports. RAID, 10 USB ports. Gigabit Ethernet and a hardware firewall.

The Ultra revision makes things a little more interesting; offering SATA 3Gb/s and an obscure device called the Secure Networking Processor. NVIDIA claims the "processor" enhances networking security, reduces CPU overhead and contains specialized features that defend against hacker attacks. Although we will have to see it to believe it, this journalist suspects it is probably nothing more than a tweaked ruleset for QoS and *maybe* some denial of service protection (hopefully outbound as well as inbound).

Finally, the SLI version of CK8-04 ties everything together with an additional switching PEG solution. Even though the CK8-04 only supports 20 PCIe lanes, NVIDIA's elegant graphic solution runs 16 lanes into what appears to be a separate switching bridge chip. This bridge can be electrically configured to either run all 16 lanes to one PEG interface, or 8 lanes to two PEG interfaces. Remember, PCIe supports 250MBps per lane, so as long as the video card can electrically support itself on 8 lanes, the theoretical 2GBps (full duplex) per video card of a dual x8 configuration is more than enough for upcoming video card solutions for many revisions to come. Current 8X AGP solutions run at 2.1GBps (half duplex) video bandwidth without coming in reach of taxing out the bus.

NVIDIA makes note in the roadmap that the Ultra and Non-Ultra revisions will only support single CPUs. We can only assume the SLI version will not widely be marketed for multiple Opterons, but it sure would be nice to give AMD and VIA some competition in that field. Samples of the new cores are shipping now and should launch early September. In reality, we probably won't see working cores for a few weeks still, but definitely expect to see boards on the shelves before Q3 is out.



View All Comments

  • Falco. - Saturday, August 21, 2004 - link

    does " with the addition of 7.1 high definition audio " = soundstorm or just regular onboard audio ?? Reply
  • Anemone - Friday, August 20, 2004 - link

    I think Nvidia will be floored by how many of these boards fly off the shelves. Tons of folks at least that I've been reading, have been holding for pci-e on A64's, and to offer them that plus SLI, will be quite amazing to watch the sales.

    Even just upgrading the mobo, lots of folks will be doing. These things are going to sell like crazy.
  • danidentity - Friday, August 20, 2004 - link

    Boards on shelves by the end of Q3? If that's true, you just made my day. I've been waiting for this for so long, and it looks like it's right around the corner. WOOHOO! :) Reply
  • johnsonx - Friday, August 20, 2004 - link

    OK, now that we're being picky about all of this, I don't think Half-Duplex is quite the right way to describe AGP.

    Half-Duplex usually describes a link that can be used full speed in either direction, but only one way at a time, such as Ethernet when used with a hub rather than a switch.

    AGP doesn't work this way. The 2.1GBps is downstream (from the host bus to the 8xAGP interface on the graphics card). Upstream, which isn't used for much at all, is standard 66Mhz PCI signalling, or only 266MBps. So it's assymetrical. I guess it is still half-duplex from the point of view that the link can still only be used one way at a time.

    Of course, as in many things (everything according to my wife), I could be wrong here...
  • KristopherKubicki - Friday, August 20, 2004 - link

    Ive been here 20 some years and im still having trouble with the language :'(

  • johnsonx - Friday, August 20, 2004 - link

    Kristopher: Fair enough ;)

    Mino: Your grammar is pretty good. You're far easier to read than half of the supposedly educated Americans that post online.



    p.s. No, that isn't America(n) bashing. I am one, and proud of it. I'm bashing the general lack of grammar used in online posting, which becomes particularly evident when a foreign speaker shows better English skills than most of the 'natives'.
  • KristopherKubicki - Friday, August 20, 2004 - link

    Fixed the numbers, i already wrote 3 other articles about PCIe so i dont know how i managed to mess that up other than lack of coffee.

  • Anemone - Friday, August 20, 2004 - link

    Put a CK8 ultra on the shelf from a major board maker, and I'll be buying it. I've waited a long time for these to arrive and was losing faith that they were ever going to arrive.

    The use of a chip to interpret 1 slot 2 slot SLI capable systems is extremely helpful, since I won't initially commit to SLI, but a second 6800 ultra would make a nice self present down the road a bit.

    If these chips were sitting in boards right now, I'd own one.
  • mino - Friday, August 20, 2004 - link

    #7 You're 99% right.

    I have no idea how could chipset support defined amount off A64's since the hypertansport interface DOES NOT make difference betwen CPU, NB, SB or any other HT device. therefore it should not depend on desing of NB wheter it supports multi-CPU systems. As I see it only way it may be incompatible with multi-CPU systems may be driver issue even which is highly unlikely.

    I suspect they just are not targeting CK4-08 to these markets.
  • themelon - Friday, August 20, 2004 - link

    Sorry, I mean the SLI version... Reply

Log in

Don't have an account? Sign up now