• 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

meldarthx

meldarthx

meldarthx

Associate
Joined
18 Oct 2013
Posts
1,475
Location
far side of the moon
N19h7m4r3 said:
Issue there is it's been a long time since a single card could push the numbers and settings I want :(

Maybe big Polaris will be that card, but they haven't shown anything indicating that. Just that demo of Polaris using less power to push 60fps than the equivalent NV card.

It's cool and all, but showing it doing much better at about the same power usage would have been better.
Click to expand...

No they wouldn't shown that as that would be showing off too much of the chip and tip their hand towards Nvidia.

showing this; shows -AMD serious improved power functions massively - but doesn't give their hand away. Only reason Nvidia got as much power functions this time as they ripped everything out of Maxwell to do it; guess what they are doing with Pascal; putting it back in......so you're going to see a jump in usage of power with Pascal as its only going to a modification of Maxwell.

What else AMD has in their bag of tricks for Polaris; we'll find out soon. As for a launch in april with big Pascal; no way not going to happen. Nvidia's never ever been on time with node shrinks and specially adding in new memory. Earliest is 2017 for big. I have a feeling AMD's going to pull a surprise chip out of air......as they are masters of misdirection when it comes to their graphic chips....
 

Kaapstad

Kaapstad

Kaapstad

Man of Honour
Joined
21 May 2012
Posts
31,922
Location
Dalek flagship
looncraz said:
Wow, you are so full of it!

nVidia's advantage vanishes when crossfire/SLI comes into play. AMD's CF scales notably better. The 4GB of RAM on Fury has proven to not be an issue - even at 4k.

http://wccftech.com/amd-radeon-r9-f...ia-geforce-gtx-titan-quad-sli-uhd-benchmarks/

Notice that your statement would be "more true" at 1080p, but NOT at 4k (or even 1440p).

And that was before the updated drivers and their improved memory management on Fury.
Click to expand...

I use both TXs and Fury Xs but unlike wccftech I don't need clickbait.

The 4gb on the Fury X is a very big issue @2160p in multi GPU setups. For a single card 4gb is not too bad but for more GPUs than that it can be a pain.
 

Kaapstad

Kaapstad

Kaapstad

Man of Honour
Joined
21 May 2012
Posts
31,922
Location
Dalek flagship
flopper said:
Yep, FuryX owns anything Nvidia has.

Polaris?
Its heaven on earth realized brightly.:D
Click to expand...

The TitanX in single usage beats the Fury X in nearly every synthetic and game benchmark when both are overclocked and pushed to the max.

In MGPU setups when there is proper driver support for both cards again the TXs come out on top.

If you use something like Tomb Raider (a gaming evolved title) where the drivers work for both camps it is an easy win for the TX.

Having said that it is unfair to compare the TXs and Fury Xs. The TXs with their 12gb frame buffer are built for 2160p MGPU use in multi GPU setups, the Fury Xs are not.

It is also unfair as the Fury Xs are a lot cheaper and offer far better value for money and for single card usage and I would recommend them every day of the week over the TitanXs for that reason.
 

D.P.

D.P.

D.P.

Caporegime
Joined
18 Oct 2002
Posts
32,623
meldarthx said:
No they wouldn't shown that as that would be showing off too much of the chip and tip their hand towards Nvidia.

showing this; shows -AMD serious improved power functions massively - but doesn't give their hand away. Only reason Nvidia got as much power functions this time as they ripped everything out of Maxwell to do it; guess what they are doing with Pascal; putting it back in......so you're going to see a jump in usage of power with Pascal as its only going to a modification of Maxwell.

What else AMD has in their bag of tricks for Polaris; we'll find out soon. As for a launch in april with big Pascal; no way not going to happen. Nvidia's never ever been on time with node shrinks and specially adding in new memory. Earliest is 2017 for big. I have a feeling AMD's going to pull a surprise chip out of air......as they are masters of misdirection when it comes to their graphic chips....
Click to expand...


AMD didn't really show that much. The 14/16nm process has about half the power requirements as 28nm, it looks like the Polaris chip has caught up with Maxwell but their demo is far to vague to give any indication since it was capped at 60FPS.

And no, Maxwell's efficiency comes from a huge amount of improvements in efficiency gains, power cycling, frequency control and architectural improvements. The removal of double precision performance doesn't affect the efficiency at all, it just allows for a smaller chip size.
 

Boomstick777

Boomstick777

Boomstick777

Soldato
OP
Joined
2 Jan 2012
Posts
11,978
Location
UK.
Picking up a Radeon Nano, something to play with until the Nano X (Polaris) comes out :p

I really liked the Fury X but the Eeeeee sound was a deal breaker. Gone for the Sapphire Nano, hopefully will be Eeeee free.
 

AlamoX

AlamoX

AlamoX

Associate
Joined
8 May 2014
Posts
2,288
Location
france
Boomstick777 said:
Picking up a Radeon Nano, something to play with until the Nano X (Polaris) comes out :p

I really liked the Fury X but the Eeeeee sound was a deal breaker. Gone for the Sapphire Nano, hopefully will be Eeeee free.
Click to expand...

are you sure ? nano cooler doesn't seem to be any better, it's even louder at load if you try to overclock it, not that they overclock well to start with...
the reason i wouldnt get Fury personally, is the absence of partner coolers for FuryX and nano, and Fury for having 30% extra useless size.
i really like the Fury line up but i can't help being disapointed at the poor decisions AMD made for them, they are rectifying one by adjusting the price, but they still need to let partners play with them, we would probably get better overclokers and quieter options.
 

Boomstick777

Boomstick777

Boomstick777

Soldato
OP
Joined
2 Jan 2012
Posts
11,978
Location
UK.
Haha guys, I am a GPU whore it's no secret I like to try them and then discard them for the next thing :p. Always said I would pick up a Nano when the priced dropped. Good time to get one now mess with for next few months until the big boys launch. I would like to see a Nano 2 in Polaris, maybe 8GB HBM 2.0. Same tiny form factor low power etc.
 

AlamoX

AlamoX

AlamoX

Associate
Joined
8 May 2014
Posts
2,288
Location
france
btw i wanted to make a poll about a subject, but let's just do it here instead.

Title : Would you buy a Fury rebrand with the next Polaris line up ?

choices are as follow :

1- i would buy a Fury rebrand, only if it comes with HBM2 8Go+
2- i would buy a Fury rebrand , only if i can choose different coolers from partners
3- i would buy a Fury rebrand, only if they have both HBM2 and partners coolers
4- i wouldn't buy AMD with the current state of their Drivers
5- i would buy AMD cards when chickens grow teeth
6- Other : specify what
even if you wanna reply to another post just throw in a choice Number,
thx!
 

JediFragger

JediFragger

JediFragger

Caporegime
Joined
18 Oct 2002
Posts
30,185
Boomstick777 said:
Haha guys, I am a GPU whore it's no secret I like to try them and then discard them for the next thing :p. Always said I would pick up a Nano when the priced dropped. Good time to get one now mess with for next few months until the big boys launch. I would like to see a Nano 2 in Polaris, maybe 8GB HBM 2.0. Same tiny form factor low power etc.
Click to expand...

My money is on another Ti within 6-8 weeks ;) :p
 
You must log in or register to reply here.
Back
Top Bottom