NVidia's Dispute over Nehalem Licensing - SLI not involved

X58 with an extra chip from nvidia .. soudns expensive. Also I have read that Intel will not be supporting motherboard manufacturers who do that.

Best for nvidia to put the SLi on the card and have dual gpu cards so they can forget about Sli on intel chipsets.
 
our customers are experiencing exciting new ways to interact with their photos
Interacting with still images? Awwrite, I need that GT280 NOW :LOL:









Yes, I do realise they are referring to the upcoming GPU accelerated Photoshop, its just a poor use of words.
 
How can you disagree with something you have no experience with or documented data on? Are you aware of Broadcast which was built onto the Nforce 200 chipset? The entire purpose of this technology was too get rid of what you are talking about. Allowing the system memory and CPU to write to each GPU with 1 command. Its built into all Nforce 200 chipsets.

((Which the 780A/780I support so I wont be surprised if future intel chipsets have an Nforce 200 or a chipset like it)) Its an extremely efficient way of running all the GPUS at the same time without a need for alot of latency/CPU overhead.



Okay, digging back a little way I know, but this clearly illustrates that people calling nForce 200 a PCIe switch are dead wrong.

It's more like a PCIe hub :runaway:
 
So, the conclusion appears to be that NVidia does not have a QPI licence and seemingly won't have one any time soon.

http://www.xbitlabs.com/news/chipse...rm_Fails_to_Get_License_to_Make_Chipsets.html


Jawed

Not quite.
Digitimes says that mainstream LGA 1160 is a go for Nvidia-designed chipsets.
The trouble of developing one for the high-end might not be worth it to them (and it would be difficult to achieve sufficiently differentiating performance in an QPI-enabled integrated memory controller architecture), or perhaps it would collide with their own strategy of downplaying the role of the CPU in modern high-end gaming.
 
I thought KGA 1160 chipset were supposed to have a more cost friendly DMI interface for the chipset (and integrated PCI 16X to make without a northbridge) ?
 
I thought KGA 1160 chipset were supposed to have a more cost friendly DMI interface for the chipset (and integrated PCI 16X to make without a northbridge) ?

Yes.

That doesn't stop Nvidia from doing their own ***-tastic southbridge for it, or perhaps just renaming the PCH, adding the so-called 'SLI processors' to that one as well and market it as the 'less expensive CPU, more expensive GPU' 'balanced PC' solution they've been touting.
 
HybridPower's transfer scheme is based on the bizarre assumption that anyone sane would play a game with either vsync off or triple buffering on. I'm sorry, but you're supposed to play your games with vsync on and triple buffering off, and you should optimize your settings so that you remain below a given multiple of the minimum frametime as often as possible. Anything else is heresy, and I am deeply unhappy that technologies made with other approaches in mind are being encouraged *shrugs*.

Mind you, this is still substantially less heretic than the complete joke that is Hybrid SLI/CrossFire (the performance aspects, not the power ones). Long live the 3D industry, where naive reviewers and overly excited users who misunderstand the very basics of 3D Graphics are the only judges! And no, I'm not bitter. No, really. I swear. Maybe. With all due respect.

What's your problem with triple buffering? The increased latency? Isn't that somewhat counteracted by being able to run at say 50fps instead of 30fps?
 
vsync on double buffering is ideal but you can only reasonably use it on older games where the framerate is locked. Else the framerate always suddenly gets cut in half, then doubles again etc. ; I hate it, it's very jerky. I don't get how using something better is an heresy!
for me it's vsync off with a refresh rate of 85Hz or higher. the 60Hz limitation of LCD is the heresy.
 
It's not needed, but NVIDIA needs to figure out a way to make money while still adding some value so that it's semi-justifiable. A BR04 does add value in the Socket 1160 market (instead of 2x8 PCIe, you get 2x16 PCIe; this is equivalent to the difference between the nForce 750a and the 780a, and the price gap can be pretty big amusingly enough!) - although it pretty much doesn't in the Socket 1366 market (except 2xBR04 for Tri-SLI, I guess) and you'd need something like my speculative BR05 to do the trick.

I don't see how turning 1 x 16x into 2 x 16x helps in any way, you aren't magicing bandwidth out of thin air, bar some data which is sent to both cards and I'll admit I have no idea how much that is.

I believe testing with X58 Tri-SLI has shown a native X58 solution to be faster than one using BR04.

Edit: Holy crap I quoted something from about 8 months ago :runaway:
 
Last edited by a moderator:
I don't understand why Intel is the one doing the filing. The press release:

http://www.intel.com/pressroom/archive/releases/20041119corp.htm

says "Additionally, the companies signed a multi-year chipset agreement for NVIDIA to license Intel's front-side bus technology."

Nehalem doesn't have an FSB, therefore NVidia doesn't have a licence. So, why is Intel filing?

Jawed

Just because there's no formal "front-side bus", it doesn't necessarily mean that it did away with all remnants of the "front-side bus technology".
Even Intel itself officially names the DDR3 memory controller, Quickpath Interconnects and L3 cache as the "uncore" part of the CPU design...
 
Just because there's no formal "front-side bus", it doesn't necessarily mean that it did away with all remnants of the "front-side bus technology".
Even Intel itself officially names the DDR3 memory controller, Quickpath Interconnects and L3 cache as the "uncore" part of the CPU design...
:LOL: I'm not going to argue with you - leave that to the lawyers, that's why it's ended up in court it seems.

Jawed
 
Is this related (yes i'm being lazy not reading the entire thread)

Intel filed a lawsuit against NVIDIA last night, reports bit-tech.net, saying the CPU giant is suing the GPU giant over NVIDIA's license to create motherboard chipsets, alleging their agreement does not cover current and future Intel CPUs, including the Nehalem family. The report states NVIDIA claims this filing does not impact any of their currently shipping products.
http://www.bit-tech.net/news/hardware/2009/02/18/intel-files-lawsuit-against-nvidia/1
 
:LOL: I'm not going to argue with you - leave that to the lawyers, that's why it's ended up in court it seems.

Jawed

Think about it.
Why would Intel sue over something that doesn't exist yet (Nehalem-ready chipsets from Nvidia) ?
And how could Nvidia have started development unless they had been given direct access to Intel's brand new tech, including sensitive trade secrets ? Who guards those secrets ? Intel, of course.
Notice how Intel didn't start suing VIA over their P4-bus chipsets until they were pretty well into their market life cicle.

It's pretty obvious, isn't it ? It's a preemptive strike, a scare tactic if you will, but i doubt JHH will back off this one.
 
Back
Top