AMD Fury Series disappointment

[3DVagabond]

Oh yeah, the "GameBarelyWorks" controversy

Except most of the Gameworks games perform poorly on both brands and are buggy as hell. Batman AK even got pulled and refunds offered it was so bad.

They would have been better off simply doing a straight port and forgetting about Gameworks. The console version doesn't suffer the same issues as the PC does.

Hopefully they didn't get their mitts into the code on Rise of the Tomb Raider and it works as well as it did when they were GE titles.

 

[AtenRa]

Fury Nano is a little faster than GTX 970 at 1440p in GW titles like Dying Light and Project Cars. Even in Witcher 3 with GameWorks enabled Nano is faster than GTX 970 at 1440p.
In other titles the performance difference is higher, especially in BF Battlefront.

#

 

[Sable]

I'd would hope so. Because I could nearly get two 970's for the price of a nano.

 

[sontin]

Fury Nano is a little faster than GTX 970 at 1440p in GW titles like Dying Light and Project Cars. Even in Witcher 3 with GameWorks enabled Nano is faster than GTX 970 at 1440p.
In other titles the performance difference is higher, especially in BF Battlefront.

Seriously? You are comparing a 600mm^2 monster with HBM to a GTX970? :\

 

[3DVagabond]

I'd would hope so. Because I could nearly get two 970's for the price of a nano.

That post was in response to some BS post that 970 beat Nano.

We have a recent review from [H] that shows even with both cards manually O/C'd the 970 isn't overall faster than a 390. Never mind Nano. And [H] uses a lot of GW titles.

http://www.hardocp.com/article/2015/12/22/asus_r9_390_strix_directcu_iii_video_card_review/1

Some people just spread FUD and hope nobody contradicts them.

 

[AtenRa]

Those two GE games are the worst performance scaling for the NANO and the best for the GTX 970.

As i have said, if you look other games the performance difference is bigger.

edit:

examples,

Witcher 3 without GameWorks.

 

[Jacky60]

The reference is pretty tame, ~1.2ghz boost.

When you look on the market and buy cards, custom 980Ti is plain faster even at 4K.

Fury X doesn't get to beat it until we get multi-GPU scenarios. There, it is more than competitive.

.

Both my Zotac reference 980tis run happily at 1380 boost all day long although I did manually configure that.

 

[Stuka87]

With the Nano now being $499, that makes it FAR more competitive on a priceerformance ratio.

 

[MrTeal]

With the Nano now being $499, that makes it FAR more competitive on a priceerformance ratio.

Sorry, where can you get a Nano for $499? It might eventually get down there, but as far as what's available now the cheapest Nano is still within $50 of a 980Ti.

 

[Stuka87]

It may take a little while for the price to be reflected at retailers, but here is the article announcing it: http://anandtech.com/show/9938/amd-cuts-price-of-radeon-r9-nano-to-499

 

[AtenRa]

Yea Nano at $499 is the right price and better overall card than GTX 980 at comparable price.

 

[railven]

Yea Nano at $499 is the right price and better overall card than GTX 980 at comparable price.

Holy cow I just saw nvidia scored tomb raider 2 bundle. Course that doesn't mean gameworks but id be surprised if it doesn't.

Also surprised game got Jan 2016 launch. Rumors are it was due to bombing so bad on xbone. Didn't even crack top 10 for sales on launch. Who ever decided to bundle with xbone and worst launch same day as fallout 4 should be fired.

Interesting to see benchmarks now. Fury nano price is amazing. What it should have been day 1. Guess that SFF premium didn't help.

 

[Headfoot]

Fury was disappointing because there is only one release that is the "Titan Killer" and 980 Ti got there first, along with how good of an overclocker it is, at an attractive price point made to seem even more attractive due to the terrible pricing of the Titan. If 980 Ti was $850 instead of $650, the story would be totally different. But nVidia came in and preemptively priced it aggressively, anticipating the competition. First to market advantage can't be understated and nvidia seems to fully understand this. I'm sure AMD does too but for whatever reason they haven't been able to execute in this regard faster than nVidia since HD 5xxx.

I must also agree the Tomb Raider 2 situation was pretty insane. Who would launch an exclusive like that on the same day as Fallout 4, given the truly massive levels of hype?

 

[AtenRa]

I'm sure AMD does too but for whatever reason they haven't been able to execute in this regard faster than nVidia since HD 5xxx.

You mean after HD79xx

 

[ultron]

We need more game profiling results to see. (how many pixels drawn, how many instructions executed, what amount of data read/write) Otherwise we are not be able to go beyond predictions/assumptions/theroies etc. Yeah, i know that this kind of data is not enough to understand all the facts, but knowing that stuffs will be big development for me at least.

For example hardware.fr profiled 3D Mark11(Stage 3, 4X MSAA, Extreme Settings) on Radeon HD 6970 couple of years ago.

Render time: 125.9 ms (= 8 fps) Vertices before tessellation: 5.36 million Vertices after tessellation: 11970000 Primitives: 12.61 million Primitives ejected from the rendering: 6,190,000 Pixel: 179 290 000 items exported by the PS 151 180 000 Texels: 2390000000 Instructions executed: 14.40 billion Quantity of data read: 4.73 GB Amount of data written: 1.08 GB

http://www.hardware.fr/articles/845-1/comprendre-rendu-3d-etape-par-etape-avec-3dmark11.html

I overloved this article.

So, are there any websites/articles that have this kind of stats? For example why FAR CRY 4 loves AMD cards than NVIDIA's even it bundled for GeForces. Why Crysis 3's particles effects relatively more damage on Furys? Why weaker AMD cards perform better with some SweetFX on some games?(sorry for the link, i forgot where i read that)

 

[Madpacket]

I keep hearing how the 980Ti's hit 1500Mhz no problem yet when you look at the feedback of the aftermarket cards on places like NewEgg it paints a different picture. Many complaining about getting crappy ASIC ratings with terrible overclocking etc.

I'm guessing the truth lies somewhere in the middle. If the average overclock is around 1350Mhz how much of an improvement would that be over stock cards? 10%?

 

[Silverforce11]

Holy cow I just saw nvidia scored tomb raider 2 bundle. Course that doesn't mean gameworks but id be surprised if it doesn't.

Crystal Dynamics changed TressFX3.0 to "PureHair" labeling and added NV's new HBAO+, they call it VXAO...

The last time they shoveled HBAO+ in Dying Light, AMD GPUs took a 40% performance loss to enable it. lol

 

[2is]

I keep hearing how the 980Ti's hit 1500Mhz no problem yet when you look at the feedback of the aftermarket cards on places like NewEgg it paints a different picture. Many complaining about getting crappy ASIC ratings with terrible overclocking etc.

I'm guessing the truth lies somewhere in the middle. If the average overclock is around 1350Mhz how much of an improvement would that be over stock cards? 10%?

Mine hits low 1400's but is factory overclocked to around 1360 as it is.

 

[kami]

I keep hearing how the 980Ti's hit 1500Mhz no problem yet when you look at the feedback of the aftermarket cards on places like NewEgg it paints a different picture. Many complaining about getting crappy ASIC ratings with terrible overclocking etc.

I'm guessing the truth lies somewhere in the middle. If the average overclock is around 1350Mhz how much of an improvement would that be over stock cards? 10%?

If you look here you can see the various overclocks they got with the different cards. Obviously it varies from sample to sample, but in general 1400MHz is a given with aftermarket cards. Heck, a card like the MSI Lightning boosts to 1400MHz out of the box I believe.

Performance gains over reference clocks are much more than 10%. More like 25%+ for a good overclock. Probably closer to 15-20% for a modest overclock.

 

[3DVagabond]

These are boost clocks, of course. Not stable base clocks.

 

[2is]

These are boost clocks, of course. Not stable base clocks.

As far as my particular card is concerned, they are one in the same. I can game all day and it will maintain the 1360mhz boost clock. It does not come close to hitting it's power envelope nor does it get warm enough to ever fall back to base clocks. Perhaps if I run furmark it would, but it hasn't happened with any of the games I play regardless of how long I'm playing them.

 

[kami]

These are boost clocks, of course. Not stable base clocks.

It's the clock speed they run at while playing a game. It stays there as long as it doesn't thermal throttle or hit a power limit. The whole base/boost clock thing is a misnomer really.

 

[3DVagabond]

It's the clock speed they run at while playing a game. It stays there as long as it doesn't thermal throttle or hit a power limit. The whole base/boost clock thing is a misnomer really.

I know that. Just want to be clear the tech that allows this.

 

[moonbogg]

I found 1440mhz on both cards in SLI is the spot where no games crash. I can go 1500mhz but there is an occasional crash, so perhaps with a custom bios and more voltage that could fix it. I'm good though. 1440mhz is still Jedi class performance.

 

[raghu78]

Sorry, where can you get a Nano for $499? It might eventually get down there, but as far as what's available now the cheapest Nano is still within $50 of a 980Ti.

Prices have started falling.

http://www.newegg.com/Product/Produ...ption=R9 NANO&bop=And&Order=PRICE&PageSize=30