AMD Confirms Delay of R9 Fury X2/Gemini to 2016 for Better Alignment with VR Market

[HKZeroFive]

At E3 AMD announced that its dual Fiji board would launch sometime in the Autumn of 2015 ahead of the holiday season, however this has not happened and the company shed some light on the matter via earlier today via an email to our friend Damien from Hardware.fr.

Q. On the E3 Livecast, Lisa committed to shipping Fiji Gemini by Xmas. What happened? Is Fiji Gemini delayed?

A. The product schedule for Fiji Gemini had initially been aligned with consumer HMD availability, which had been scheduled for Q415 back in June. Due to some delays in overall VR ecosystem readiness, HMDs are now expected to be available to consumers by early Q216. To ensure the optimal VR experience, we’re adjusting the Fiji Gemini launch schedule to better align with the market.

Working samples of Fiji Gemini have shipped to a variety of B2B customers in Q415, and initial customer reaction has been very positive.

In my opinion, AMD delaying the Fury X2 to 2016 minimises the chance of selling more cards, especially since Pascal is virtually around the corner. It also bugs me that they promised to bring this card around the holiday season at E3, and while it's understandable they want it to coincide with the growth of the VR market, they should have had a bit more transparency and not having to be asked the question of 'where is it' in the first place.

Overall, it just seems to me by delaying the Fury X2 to 2016, AMD will be selling a significantly fewer cards as Pascal/Arctic Islands will be releasing soon after. The VR market alignment justification honestly sounds like an excuse.

Sources:

http://wccftech.com/amd-fury-x2-delayed/

http://www.hardware.fr/news/14451/amd-repousse-fiji-gemini-bi-radeon-fury.html

[Kobathor]

Pascal is not around the corner. More like Q4 2016.

The Fury X2 will still be a beast after Pascal. After all, Pascal isn't some kind of magic GPU architecture that will be 200% faster than Maxwell.

[PlayStation 2]

Pascal is not around the corner. More like Q4 2016.

The Fury X2 will still be a beast after Pascal. After all, Pascal isn't some kind of magic GPU architecture that will be 200% faster than Maxwell.

This. Workstation Pascal is around the corner but this is a desktop GPU.

[Enderman]

Pascal is not around the corner. More like Q4 2016.

The Fury X2 will still be a beast after Pascal. After all, Pascal isn't some kind of magic GPU architecture that will be 200% faster than Maxwell.

and the fury X2 also isnt some kind of magical architecture that will be any faster than the fury X...

its just two fury Xs

pascal will actually be a performance improvement over the 980ti

 

so if two 980tis perform the same as two fury Xs right now (or even better when OCd) what do you think will happen when you put two 1080tis against two fury Xs (the same as a fury X2)

[PlayStation 2]

and the fury X2 also isnt some kind of magical architecture that will be any faster than the fury X...

its just two fury Xs

pascal will actually be a performance improvement over the 980ti

 

so if two 980tis perform the same as two fury Xs right now (or even better when OCd) what do you think will happen when you put two 1080tis against two fury Xs (the same as a fury X2)

that's not really a relevant argument

maybe Arctic Islands x2 vs. Pascal x2.

[SurvivorNVL]

At the rate things are going.  May as well snag up a pair of Fury Nitro's and rock that until Volta and whatever AMD brings to the table.

[Enderman]

that's not really a relevant argument

maybe Arctic Islands x2 vs. Pascal x2.

exactly

so the fury X2 will already be outdated within half a year of release

[PlayStation 2]

exactly

so the fury X2 will already be outdated within half a year of release

Q1 2016 isn't bad. Chances are, Pascal isn't hitting shelves until November.

[Enderman]

Q1 2016 isn't bad. Chances are, Pascal isn't hitting shelves until November.

he said Q2...

and pascal are coming in Q4

4-2=2

the fury X2 will be relevant for half a year before new architectures are released with better performance per card

[PlayStation 2]

he said Q2...

and pascal are coming in Q4

4-2=2

the fury X2 will be relevant for half a year before new architectures are released with better performance per card

Or the Fury X turns out like the 7970 or the 290X and is still relevant for another 5 years.

[Atmos]

and the fury X2 also isnt some kind of magical architecture that will be any faster than the fury X...

its just two fury Xs

pascal will actually be a performance improvement over the 980ti

 

so if two 980tis perform the same as two fury Xs right now (or even better when OCd) what do you think will happen when you put two 1080tis against two fury Xs (the same as a fury X2)

 

But would you be able to fit both of those 1080tis in an m-atx case?

Because the fury x2 is only slightly larger than a 980ti as it is right now...

 

For the m-atx market, and even some itx cases this will undoubtedly be the fastest gpu you can get, price irregardless.

[Enderman]

Or the Fury X turns out like the 7970 or the 290X and is still relevant for another 5 years.

what??

the 7970 and 295x2 were top of the line for a couple years before new architectures were released

the 295x2 was released almost 2 years ago and it not been top of the line for quite a while, and nobody buys it now because its crap compared to newer GPUs

[PlayStation 2]

what??

the 7970 and 295x2 were top of the line for a couple years before new architectures were released

the 295x2 was released almost 2 years ago and it not been top of the line for quite a while, and nobody buys it now because its crap compared to newer GPUs

Nobody buys the 295x2 because there's little to no availability of it.

And the fact that you need a PSU with actual requirements to use it.

[Kobathor]

and the fury X2 also isnt some kind of magical architecture that will be any faster than the fury X...

its just two fury Xs

pascal will actually be a performance improvement over the 980ti

 

so if two 980tis perform the same as two fury Xs right now (or even better when OCd) what do you think will happen when you put two 1080tis against two fury Xs (the same as a fury X2)

 

that's not really a relevant argument

maybe Arctic Islands x2 vs. Pascal x2.

 

Also, it will most likely be like the R9 295X2 is. It's still the most powerful single PCB GPU on the market, and the Fury X2 will take its place. The Fury X2 will (looking at the evolution of nVidia GPU's) be faster than the best Pascal card, plus more. It'll be relevant for a long time.

[Enderman]

But would you be able to fit both of those 1080tis in an m-atx case?

Because the fury x2 is only slightly larger than a 980ti as it is right now...

 

For the m-atx market, and even some itx cases this will undoubtedly be the fastest gpu you can get, price irregardless.

yes you can fit SLI in an Matx case

http://www.gamespot.com/articles/building-the-ultimate-matx-sli-pc-with-intels-5960/1100-6423349/

 

if you mean mini itx, then yeah a x2 would be more powerful, but still the amount of people building mitx cases is very small, and the people who can afford a X2 is even smaller

[Enderman]

Nobody buys the 295x2 because there's little to no availability of it.

nobody buys it because there are gpus that perform better, use less power, and create less heat

 

Also, it will most likely be like the R9 295X2 is. It's still the most powerful single PCB GPU on the market, and the Fury X2 will take its place. The Fury X2 will (looking at the evolution of nVidia GPU's) be faster than the best Pascal card, plus more. It'll be relevant for a long time.

if you consider 1 year to be a "long time" then sure...

[PlayStation 2]

nobody buys it because there are gpus that perform better, use less power, and create less heat

 

if you consider 1 year to be a "long time" then sure...

Name a single GPU that performs faster than the 295x2.

Not two separate cards, single PCB.

[Kobathor]

nobody buys it because there are gpus that perform better, use less power, and create less heat

 

if you consider 1 year to be a "long time" then sure...

Uh

 

The R9 295X2 is two years old, and again, it's still the most powerful single-PCB graphics card on the market. 

 

There are no nVidia competing dual-GPU cards, if there was a GTX 990 I'm sure it'd be more powerful. But, nVidia doesn't make those anymore. The last dual-GPU card they made was the Titan Z, and that's not even the powerhouse that it was 2 years ago anymore.

 

Name a single GPU that performs faster than the 295x2.

Not two separate cards, single PCB.

^^^^this

 

The trick is that there isn't.

[Enderman]

Name a single GPU that performs faster than the 295x2.

Not two separate cards, single PCB.

the titan z perform better in several cases

 

but that's not my point

 

my point is that people who are not restricted by SFF will never buy the 295x2 because you can have two cards that perform better in SLI/CFX for less money

and almost everybody is not restricted by SFF

 

just look at how many people have a 295x2 in their PC on this forum...lol

[PlayStation 2]

the titan z perform better in several cases

-snip-

 

but that's not my point

 

my point is that people who are not restricted by SFF will never buy the 295x2 because you can have two cards that perform better in SLI/CFX for less money

and almost everybody is not restricted by SFF

 

just look at how many people have a 295x2 in their PC on this forum...lol

Well for one, @Prysin for damn sure is happy with his 295x2.

It may not be the most practical card, or even the smartest card to buy, but look at where the Titan X beats the 295x2. It takes two of them and OC'd and it BARELY edges out the 295x2. By 0.2 fps and that's the margin of error.

Just look. People who would rather have two separate GPUs are gonna go that route. But you have people like me who'd rather prefer one chunk of PCB.

[Enderman]

Uh

 

The R9 295X2 is two years old, and again, it's still the most powerful single-PCB graphics card on the market. 

 

There are no nVidia competing dual-GPU cards, if there was a GTX 990 I'm sure it'd be more powerful. But, nVidia doesn't make those anymore. The last dual-GPU card they made was the Titan Z, and that's not even the powerhouse that it was 2 years ago anymore.

 

 

^^^^this

 

The trick is that there isn't.

it DOESNT MATTER that it is the most powerful "single card" on the market

99% of people have space for two GPUs or more

and 99% of people would rather spend less on something that performs better because they cant afford a 295x2

 

btw the titan Z can compete with the 295x2 when OCd

 

 

Well for one, @Prysin for damn sure is happy with his 295x2.

 

ok cool, one out of 500 000

i bet if you gave him $1000 right now he would buy something other than a 295x2

[PlayStation 2]

ok cool, one out of 500 000

i bet if you gave him $1000 right now he would buy something other than a 295x2

I'm sure he would.

But you don't have a point. It still edges out a $1500 more expensive GPU (at both of their launch MSRPs). At $649 it fucking slaughters a 980 Ti.

[Kobathor]

it DOESNT MATTER that it is the most powerful "single card" on the market

99% of people have space for two GPUs or more

and 99% of people would rather spend less on something that performs better because they cant afford a 295x2

 

btw the titan Z can compete with the 295x2 when OCd

 

Just because you have space doesn't mean it's necessary.

The "spend less on something that performs better" argument doesn't make any sense. Spend $1.2K on a 980ti SLI? Spend $700 on a 970 SLI? You can't spend less and get more raw power than a 295X2.

 

The Titan Z is also a Kepler GPU, Kepler is on it's way out. A 970 already beats a 780ti.

[Swndlr]

that's not really a relevant argument

maybe Arctic Islands x2 vs. Pascal x2.

But then again it actually is a valid argument if AMD are holding off the release long enough

[CommandMan7]

Pascal is not around the corner. More like Q4 2016.

The Fury X2 will still be a beast after Pascal. After all, Pascal isn't some kind of magic GPU architecture that will be 200% faster than Maxwell.

 

Well, Pascal is on 16nm and Maxwell is on 28nm. Just due to the fact there will be way more transistors plus the HBM2, yes, there is a good chance that Pascal will be nearly 200% faster (closer to 150%).