• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

More Polaris10 and Polaris11 Specifications Revealed

D9FV2w6.png



...only trolls and AMD fanboys try to make "Pascal 10X Maxwell" applicable...

Ah, oh...
I thought it was harmless to copy&paste an nVidia slide...
I mean, clearly, anyone who'd spot that slide anywhere on the internet, would immediately be clear about the context.

Anyhow, what about "1070 is faster than Titan X"?
Just curious.

Or this one:

PascalCraft_575px.png


You mean a year later, they have now come up with something to compete against a 980ti?

You might need to check Fury X benchmarks on reasonable (for cards of such power) resolutions.

And 480 (that's the fastest card that is expected in June) is definitelly NOT supposed to compete against 980Ti. It's a competitor to 1050/1060 (yet to be announced).
 
Last edited:
At least I can spell patience correctly. Now buzz off trollio. I don't have time for your shit.

Sure you do, you are on a forum looking for some news, you have oceans of time
 
I said LIKE NVLINK.
It helps with Sli

Tom Peterson from Nvidia already addressed it.
  • Old bridges were running at below 600mhz
  • Newer bridges (LED & HB) bridges can run at +600mhz
Old SLI used a combo of SLI bridge along with PCI-E talk to achieve SLI in resolutions exceeding 1440p@60 (ie 1440@120 was SLI+PCI-E)

HB Bridges will allow up to 5k over SLI Bridge without the need for PCI-E talk in 2-way SLI.

sli-configuration-100661339-large.png


He also points out that Nvidia prefers LDA Explicit for multi GPUs instead of MDA. Not sure if that will be part of their GameWorks initiative but not good for consumers that want to mix different GPUs.
 
Last edited:
NVLink is a data-bus that drives 4 GPU cards, interconnected with a theoretical max of 40GB per device.

Pcie Express has 82 pins, an SLI connector is 26 pins, an actualy NVLink Interface consists of 2x 400-pins connectors.
 
Last edited:
Spoken like a true fanboy.

And that's the problem I'm quickly seeing with these forums. If you make a comment one side doesn't like - even if it's backed up by facts, people throw out the mindless "you're just a fanboy" attack. And that definitely mutes much intelligent and fair discussion based on factual information. BTW, I'm an AMD owner - just to show how wrong that comment was.
 
And that's the problem I'm quickly seeing with these forums. If you make a comment one side doesn't like - even if it's backed up by facts, people throw out the mindless "you're just a fanboy" attack. And that definitely mutes much intelligent and fair discussion based on factual information. BTW, I'm an AMD owner - just to show how wrong that comment was.
When a person generalises and states Nvidia will always be better, my comment cannot be wrong.
Stick to facts and I don't have an issue with what you say.
 
Back
Top