Latest Rumors: "Polaris 10" performs "close" to a 980ti

I hope they perform as well as the rumors say.

Would make for an awesome price drop on the current cards.

I'll take a second 390 for $125, thank you very much.
 
"Polaris 10 performance close to a 980ti"
Based on who's claims though i can take 1 guess at who makes that claim and probably certain benchmarks that was used with settings that are questionable to start with.
 
I'm going to take a stab at trying to understand what you just said.

I don't actually know of any 'AMD benchmarks', unless you mean 'benchmarks performed by AMD'.

And everyone knows the Fury line of cards scale better with resolution than Nvidia's cards: so that's not exactly controversial.

That may not apply anymore, now that we'll see cards with the exact same bus width in similar performance ranges.

Kepler cards had between 1/3 less and 1/4 less bandwidth than GCN 1.0

Maxwell had about half the bandwidth of GCN 1.1 and 1.2.

If AMD launches their new parts at 256-bit, it will be the first time they've had memory bus parity in fucking forever.
 
Last edited:
This is what everyone is talking about.

The Polaris 10 GPU is said to have 3DMark Firestrike Ultra performance around 4000 points which is about what a Radeon R9 Fury X and GeForce GTX 980 Ti score. By 4000 points, we don’t mean exactly 4000 but it’s actually a bit less than that.

Which is a meaningless statement to me.
In the real world, most 980 tis sold were AIB custom which out of the box were significantly faster than Fury X which had little OC headroom.
It's amazing how much vaporware gets some people riled up. Actual products will be out soon enough.
Well the more obvious question is what's the score for 1440p and 1080p given fury X was touted to beat the 980ti in AMD benchmarks which seemed to be very specific why proper benchmarks and reviews can mean so much.
980ti outsold Fury by about 20 to one according to Steam.
 
From the article:

If AMD can manage to launch a card at around $329 US with better efficiency and performance close to 980 Ti with twice the VRAM of Fury X, they can have a great product in their hands which many users will be willing to upgrade to. AMD Polaris looks to be an incremental step for team red in the efficiency game, it will be great to see competitive products from AMD in the coming months which will be built for DirectX 12 and VR Gaming. You can see the leaked specifications in the table below:

Read more: http://wccftech.com/amd-polaris-10-desktop-polaris-11-notebook-gpu/#ixzz46rIFHozd


It appears this is where they are going now, with the $1500 Radeon Pro Duo being their current flagship "Halo" until Vega next year, if the article is to be believed.
This is not meant as a smart ass comment. Why would I buy a card that is ALMOST as fast as a 980ti? Is it a lot cheaper?
 
Vega is the replacement of Fury X which will come out in Q1 of '17. You guys should check out the New UEFI bios for Fury X and Nano. Better overclocking stability. :)
GPU Firmware Download
 
This is not meant as a smart ass comment. Why would I buy a card that is ALMOST as fast as a 980ti? Is it a lot cheaper?

The expectation is for it to some in from $300-$450, with a far lower power usage, and I would expect a smaller size.
 
Whatever you say, Neil Degrasse Tyson.
1350-0-1447882000.jpg
 
Is this good or bad?

I mean, the GTX1080 is supposed to be significantly faster than the GTX980Ti and the 1070 is supposed to be on par.
 
Is this good or bad?

I mean, the GTX1080 is supposed to be significantly faster than the GTX980Ti and the 1070 is supposed to be on par.

define signifcantly

index.php


fury X 16% faster than 390x here
980ti 5% faster than 390x here
 
Last edited:
Significantly is more than close, right?

I mean, it's a fine line we're treading there :p By most accounts the 980ti is significantly faster than a 390x, yet it's pretty "close" in hitman.

Similarly we know 390x is significantly slower tha fury x, but it's only 16% lead here. 16% isn't quite as significant as i would expect frankly
 
Yep, that's why reputable sites always do multiple benchmarks, because many games are extremely well optimized for either red or green.
 
If it's even remotely close the Fury cards will all be discontinued. Based on die size and the fact that it doesn't use HBM Polaris 10 cards will be much cheaper to make.
 
Back
Top