NVIDIA's CK8-04; The Best K8 MCP (On Paper)
by Kristopher Kubicki on August 20, 2004 3:50 AM EST- Posted in
- CPUs
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.
Facebook
Twitter
RSS
47 Comments
View All Comments
langles - Tuesday, October 19, 2004 - link
#46Nope. It's not TCPA.
Press release day is here:
http://www.nvidia.com/object/feature_activearmor.h...
http://www.microsoft.com/whdc/device/network/TCP_C...
It seems to be a few steps in the direction I was wondering about. But no details yet on whether or not it supports IPSEC offloading.
quanta - Friday, October 8, 2004 - link
#45, I believe the 'Secure Networking Processor' is just another fancy name for the TCPA, which has nothing to do with protecting owner of the machine. See http://www.notcpa.org/faq.html for details.langles - Saturday, September 18, 2004 - link
Regarding this part of the original article:"... 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)."
IF this version of NForce incorporates the EtherMax technology from IReady which they acquired in April, then it could do all that and more. (That's what I would wish for!)
See:
http://www.nvidia.com/object/IO_12881.html
http://www.national.com/appinfo/networks/files/Mar...
http://www.national.com/appinfo/networks/files/whi...
But I would speculate that it's too soon for this version of NForce to have incorporated technoogy from IReady, unless NVidia had already begun this process before the acquisition occurred.
Myrandex - Monday, September 13, 2004 - link
also gtech, whats up with 'raid using morons'? Is having faster performance that bad of a thing? Or is having increased drive security bad? I personally use raid, and I have enjoyed every moment of it.Myrandex - Monday, September 13, 2004 - link
Reflex, it would be a pretty dumb idea to remove PCI altogether. No one would want to have to buy all new stuff just because PCI wouldn't be supported anymore. I wouldn't want to go out and buy a new TV tuner card, and a new Creative Audigy 2 sound card. Yea I think there needs to be less PCI slots, but not elimination altogether.KristopherKubicki - Tuesday, August 31, 2004 - link
thebluesgnr:1394 can still be implemented by the manufacturer. All these documents say here is that it isnt integrated onto the core logic chip.
Kristopher
ceefka - Tuesday, August 31, 2004 - link
#39 Why? Please elaborate. An additional FireWire PCI-card would of course be possible.Does an add-on FireWire chip also use PCI-bandwith? If not, well great, if so then I still would not want a 1394-PCI card sit in a slot that I'd like to use otherwise. If in fact USB 2.0 would do the trick for me, I'd be OK, but it doesn't. I've been told however that USB is heavier on the CPU than 1394a/b. So if this is true, 1394a/b is still a good feature to add-(on). With or without it, the CK8-04 looks like a smash.
In the end it all depends on your priorities. I wouldn't choose PCI-E over 1394b though I would gladly accept PCI-E if it came with 1394b on a new 754/939 mobo. Let's just bring it on nVidia!
TimTheEnchanter25 - Monday, August 30, 2004 - link
I wanted to upgrade my system last October, but decided to wait when I read early reports of socket 775 and pci-e. Unfortunately when these boards finally arrived their performance didn't compare to amd 64 boards. So, I decided to wait for socket 939 boards, only to be disappointed in the lack of pci-e. I can't justify buying an AGP card today, knowing that there won't be a slot for it if I update my mobo later. The exciting thoughts of SLI make pci-e all the more attractive to me.Nforce 4 is finally what I've been waiting almost a year for and more (I didn't even dream of SLI), but here I am a year later, still waiting.... My computer isn't going to be able to run Doom3 or HL2. I don't know how much longer I can wait to upgrade after HL2 is released (assuming it really will happen)!!
If the Nforce4 boards are really out by the end of Q3, I would be happy. But, a recent post in the ABIT forums by their marketing team isn't as optimistic:
"NF4 Fatal1ty board will be available late this year or early next year depends on when the chipset is available. And yes it will have PCI-E, SATA 300 and other ABIT innovative features that I can not talk about right now."
Kris, is the Q3 info reliable enough to convince me to wait another month?? Or, should I just give up and go with the msi neo2 now and be stuck with an AGP card?
Thanks
thebluesgnr - Saturday, August 28, 2004 - link
With PCIe the lack of 1394b support is no big deal.ceefka - Friday, August 27, 2004 - link
#36 Marlin1975: So no point in adding something most will not use.So most DV-camera's, TC FireWorks, RME Fireface and outboard FireWire HDD will be obsolete? Are you unaware of 1394b? IEEE1394b = 800Mb vs USB 2.0 = 480Mb. Then why the hell would they even invent that? Maybe average househould PC's do not rely on FireWire. For audio/video people it's a different story.
Oh well, this is just the chipset, right? 1394b will probably come on a Texas Instruments add-on chip then.
37 I'd be dissapointed if it weren't.