NVidia announces x86 chip?

That is more an indication that Via cannot engineer its way through process issues, but must skip a process node for a moderate-performance x86 design. An uninspiring result that makes the small team size loom large.

I think the most desired outcome when hitting process problems would be to have the resources to engineer a solution, instead of punting the whole design rollout down a process node and a different fab and negating all tentative design plans by customers.
 
That must be a misunderstanding - they did fix the issue, and it's either in mass production right now or ready for mass production. I suspect INKster simply meant that this negative experience will push them to go for another foundry on the next generation...

Although I have no idea where INKster got the idea that VIA didn't need an Intel license - they definitely got one, and had more than a few issues (to say the least) back in the early 00s when they didn't have an official one. And it can't be transfered in a purchase, so my favorite strategy is still for NVIDIA to "license" some GPU IP to VIA for a SoC; i.e. they really do all the design work and can work as much together as they want, but legally speaking the money goes to VIA first which then transfers large royalty payments (presumably proportional to the GPU's die size rather than the IP's apparent value like in the handheld world) back to NV.
 
That must be a misunderstanding - they did fix the issue, and it's either in mass production right now or ready for mass production. I suspect INKster simply meant that this negative experience will push them to go for another foundry on the next generation...

Although I have no idea where INKster got the idea that VIA didn't need an Intel license - they definitely got one, and had more than a few issues (to say the least) back in the early 00s when they didn't have an official one. And it can't be transfered in a purchase, so my favorite strategy is still for NVIDIA to "license" some GPU IP to VIA for a SoC; i.e. they really do all the design work and can work as much together as they want, but legally speaking the money goes to VIA first which then transfers large royalty payments (presumably proportional to the GPU's die size rather than the IP's apparent value like in the handheld world) back to NV.

That sounds like a really good deal. I really hope an improved IPC Dual Core Nano come through. Intel have been forcing us with crap chipset and IGFX for far too long.
 
What do you mean? I never had a single problem with any Intel chipset, as opposed to pretty much every chipset for nV, VIA or AMD.
 
What do you mean? I never had a single problem with any Intel chipset, as opposed to pretty much every chipset for nV, VIA or AMD.

And despite such glowing reviews I have had trouble with intel chipsets on and off since my abit BH6 which was undoubtedly perferct for me. I am certainly relieved to have finally got rid of my Asus p5b based on an intel chipset as it was ridiculously unstable.
 
Back
Top