• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Polaris architecture – GCN 4.0

arc@css said:
Will be nice to see the back end of 28nm.
Click to expand...

I think we can all agree on that one. :)

My lad may just get an early upgrade of my Fury Tri-X if these cards produce the power we are all hoping for.

Exciting times ahead for Gfx Cards this year..combined with VR...Hopefully 2016 will be golden for PC gaming.

:D
 
Am looking forward to this.

Will I at last be able to ditch my two 290x for a single card? and still play on 4k?

Thats the dream
 
It's hard to believe how long we've been stuck with 28nm, and the incremental upgrades that came with it. It will be good to get a new process, and with HBM, this next generation will hopefully be a game changer.
 
fs123 said:
A benchmark slide comparing power consumption of a Polaris gpu with a GTX 950 @ 1080P

mlqbh0.jpg


the test is dated 2nd December 2015 so they have working samples of a next gen gpu which seems to be a low power notebook gpu. Hopefully more leaks coming in the next few months. Both cards are on medium preset for Battlefront but for comparison we know that ultra preset on a gtx 950 gets about 48 fps.

http://www.techspot.com/review/1096-star-wars-battlefront-benchmarks/
Click to expand...

So much for the 2x performance per watt increase that was mentioned last year.
 
Whichever Polaris vs GTX 950 by the looks of the fine details. but is still near half the power usage at 86W compared to 140W.

Freddie1980 said:
So much for the 2x performance per watt increase that was mentioned last year.
Click to expand...

That 2x Perf/Watt is compared to thier older uArch.

Also, GCN does a lot more managment in hardware compared to Maxwell, Kepler etc which still rely more heavily on drivers.
 
Last edited:
Its 16 Watts over half the power consumption, it actually equates to 60% of the 950's power consumption.

The quote is vs their own Radeon GPU's anyway, not the competition.

At that a GTX 980TI level card would be about 150 Watts
 
Last edited:
Well its also early silicon and drivers too - there is at least another six months before its released.

So,if we equate the performance/watt improvements at this early stage,it would mean the equivalent to the GTX960 Nvidia I have now would be around R9 290 level performance at 1080P, using a single 6 pin PCI-E power connector.

Edit!!

I screwed up the calculations.

That would be GTX980TI level performance at under 150W.
 
Last edited:
CAT-THE-FIFTH said:
Well its also early silicon and drivers too - there is at least another six months before its released.

So,if we equate the performance/watt improvements at this early stage,it would mean the equivalent to the GTX960 Nvidia I have now would be around R9 290 level performance at 1080P, using a single 6 pin PCI-E power connector.

Edit!!

I screwed up the calculations.

That would be GTX980TI level performance at under 150W.
Click to expand...

die shrink alone should be a midrange gpu allowing above furyx performance at a lower watt than current leader Nano.
 
Interesting.... Encouraging to see a mid-range GTX 950 in the test being compared to the upcoming Polaris GPU. Hoping that means we'll see a new range from top-to-bottom instead of older cards being used in a confusing rebranding exercise to fill the different price points.
 
So the Demo was a measurement of full sytem power, which includes an i7.

So the GPU by its self was probably only pulling about 40 watts, maybe even less, that puts a whole new perspective on it because we don't know how much the rest of the system is using and there in how much the rest of the system makes up the power consumption of both, the 950 system is pulling <60 Watts more.

Think about that for a minute. ;)

In the live press demonstration we saw the Polaris system average 88.1W while the GTX 950 system averaged 150W. Meanwhile in RTG’s own official lab tests (and used in the slide above) they measured 86W and 140W respectively. Keeping in mind that this is wall power – PSU efficiency and the power consumption of other components is in play as well – the message RTG is trying to send is clear: that Polaris should be a very power efficient GPU family thanks to the combination of architecture and FinFET manufacturing. That RTG is measuring a 54W difference at the wall is definitely a bit surprising as GTX 950 averages under 100W to begin with, so even after accounting for PSU efficiency this implies that power consumption of the Polaris video card is about half that of the GTX 950. But as this is clearly a carefully arranged demo with a framerate cap and a chip still in early development, I wouldn’t read too much into it at this time.
Click to expand...
Also, apparently the whole architecture is brand new. its not just a die shrink.
 
Last edited:
TonyTurbo78 said:
Really couldn't care about power saving. Just give us a brut of a card, if it needs the power of a nuclear station then so be it.
Click to expand...

There are always limits. I'm sure you wouldn't want a card that pulls 5 KW, even if some people would be fine with that. Once you agree that there are limits, it's just a question of where you set them, and AMD are going to set them to satisfy the largest market.

Plus if you can cut your power usage in half, then that means you can have twice as many transistors, so you trade the power efficiency for more graphical power. Power efficiency isn't just about saving power, it's about being able to do more with the power you have because you're more efficient.
 
The industry wants to head into the realm of multi-gpu/neg. overhead/~100% scaling regardless of gpu count, this allows them to sell more units obviously.

Big unwieldy power hungry behemoths are a dead end.
 
I have a bad feeling about this.

"optimized" finfet design
touting unimpressive bench figures and focusing on ppw

Of course I predicted all this. Pascal will be the same way. 980 all over again. But with no Titan this time.
 
Last edited:
ooch, that is some nice performance, that Polaris sample only running at 850mhz at 0.8V against the stock 950 for same performance.
 
You must log in or register to reply here.
Back
Top Bottom