[Videocardz rumour] Vega pushed forward to October

[tential]

This.

Polaris (as a 'mainstream' option) is great in theory but late. If they wanted to tap the VR demand, this should have been released in Q1 when everyone was building, ordering or prepping for VR. Instead, its launching after VR and after NV's 1080/1070. I do think a lot of people will just grab the 1070 as a 'minimum' VR card, or a used 390x/980/970 rather than wait for Polaris. On the other hand, if AMD can get this launched a few months before the 1060 is out, they could get some nice sales in the ~$200-250 range.

I almost expect NV to just hold onto their 1060/1060Ti and release it as a surprise at the same time to spoil the Polaris launch. Would NOT be surprised at all.

Yup, my idea of Polaris being successful hinged on it being first. With Nvidia marketing already in full force and Pascal being released before AMD NDA is off on Polaris, it's not going to matter. People aren't aware of both companies, they're mostly aware of Nvidia. AMD needs to be the only thing people are talking about to gain market share.

Funny enough, someone posted a slide on the CPU side, which showed AMD wasn't expecting a large revenue increase from the GPU division this year, so even AMD doesn't believe Polaris will gain massive traction.

 

[airfathaaaaa]

Yup, my idea of Polaris being successful hinged on it being first. With Nvidia marketing already in full force and Pascal being released before AMD NDA is off on Polaris, it's not going to matter. People aren't aware of both companies, they're mostly aware of Nvidia. AMD needs to be the only thing people are talking about to gain market share.

Funny enough, someone posted a slide on the CPU side, which showed AMD wasn't expecting a large revenue increase from the GPU division this year, so even AMD doesn't believe Polaris will gain massive traction.

no they said that cpu department in general has more revenue than the gpu in general

 

[PPB]

If polaris only goes uonto 850mhz, then p10 should be damn fast to run hitman 1440p min 60fps at those speeds

Sent from my XT1040 using Tapatalk

 

[tential]

no they said that cpu department in general has more revenue than the gpu in general

Single digit growth from GPU. So not expecting that much extra from GPU, unless AMD has been selling tons of Fury X, and now they're replacing those sales with P10....

But in the real world, they aren't expecting to gain that much traction with P10....

 

[railven]

Single digit growth from GPU. So not expecting that much extra from GPU, unless AMD has been selling tons of Fury X, and now they're replacing those sales with P10....

But in the real world, they aren't expecting to gain that much traction with P10....

Judging by the prographics estimates, that's where they are funneling Fiji chips to.

Would Radeon Pro Duo count as Pro or GPU?

EDIT:

From a reddit poster an hour ago

https://www.reddit.com/r/Amd/commen...ris_validation_failed_might_launch_in/d32uwx7

seems it was an investors only stream

Summer can also include September. Which is a stone's throw away from October. The plot thickens.

 

[RussianSensation]

Single digit growth from GPU. So not expecting that much extra from GPU, unless AMD has been selling tons of Fury X, and now they're replacing those sales with P10....

But in the real world, they aren't expecting to gain that much traction with P10....

The irony in your post of using an outdated slide that doesn't align with the real world recent data and then trying to use that same slide to try and project Polaris 10/11 sales growth.

"Advanced Micro Devices, Inc. (NASDAQ:AMD) saw a double digit appreciation in total volume of discrete graphics processing unit (GPU) shipments during the quarter ended March 2016. The company also guided for 15% sequential in the next quarter as the demand for its semi-custom and GPU products reached new heights. AMD’s ROG initiative has become increasingly popular among gamers worldwide and is one of the catalysts behind the rising demand for company’s products, specially, in the gaming business."
http://www.thecountrycaller.com/456...a-corporations-nvda-market-share-wells-fargo/

Not trying to divert this into a financial/investor discussion but the point is those slides don't need to align with current reality as they are just projections made in the past.

It also should be obvious that even if Polaris 10 is a great product, it may hardly gain traction in the first place. After all, $299 280X failed out outsell $380-450 770 2-4GB, $250-280 R9 290 failed out outsell 960 and the entire 750/750Ti/950/960 lines easily outsold far superior from a performance and price/performance perspective R9 270/270X/280X/380/380X. Trying to extrapolate how successful Polaris 10 may or may not be from market share growth may not be an accurate representation of its engineering successes.

I feel bad for all the hardworking engineers who worked on Polaris 10 when online it seems the entire Internet is desperately trying to position Polaris 10 against 1070/1080 when AMD had repeatedly stated that they are aiming at high-end laptop and mainstream desktop market segments with Polaris 10.

Based on 1070's price of $379, if AMD is truthful, Polaris 10 should be nowhere close to even 90% as fast. It makes no sense why it would be with its rumoured 232mm2 die.

Even if 1070 = 970 SLI = 309%, AMD can easily price Polaris 10 = 390X for $249 and it would be great value until 1060/1060Ti show up. (Assuming 1070 = $379 = 309% / 223% = Polaris 10 ~ 390X => $272. That means AMD can price Polaris 10 at $269 and still be a linear value proposition).

For the mainstream market, a reference 1070 would be $449 at launch. If Polaris 10 ~ 390X for $249, it would mean the 1070 costs 80% more for just a 39% increase in performance. Bad deal for mainstream gamers to buy a 1070 for 1080p 60Hz if the competitor's card is fast enough for that peasant resolution for $150-200 less.

24" 1080P IPS monitors are now $100. Who the hell spends $380-450 USD to game on a $100 monitor?!

What's more impressive a $100-200 gaming monitor and a $700 1080 OR a $650 4K A-Sync monitor and a 390? In 3 years that $700 videocard is worth $200 (November 2013 $700 780Ti -> Now we are already seeing sales of 970 for $225) and that monitor is still top notch.

If review sites don't start encouraging gamers to upgrade to 1440p or 1080p 120-144Hz with next gen $380+ cards, they are not being truthful to their consumers.

Where AMD is going wrong is marketing Polaris 10 for VR. What they should be doing is targeting 1080p 60Hz gaming for the mainstream $199-249 market.

 

[airfathaaaaa]

Judging by the prographics estimates, that's where they are funneling Fiji chips to.

Would Radeon Pro Duo count as Pro or GPU?

EDIT:



Summer can also include September. Which is a stone's throw away from October. The plot thickens.

pretty stupid to make a launch some months ago then do a launch again on computex and then do another on september

 

[xthetenth]

I feel bad for all the hardworking engineers who worked on Polaris 10 when online it seems the entire Internet is desperately trying to position Polaris 10 against 1070/1080 when AMD had repeatedly stated that they are aiming at high-end laptop and mainstream desktop market segments with Polaris 10.

It's going to be positioned against the 1070 just because this generation is a crazy disjoint mess of a launch. However it's going to be competing on price/performance unless something goes horribly right for AMD.

 

[railven]

pretty stupid to make a launch some months ago then do a launch again on computex and then do another on september

I dunno, with AMD involved, anything is possible.

 

[RussianSensation]

It's going to be positioned against the 1070 just because this generation is a crazy disjoint mess of a launch. However it's going to be competing on price/performance unless something goes horribly right for AMD.

That makes no sense unless Polaris 10 is a 1070 performance competitor. We know that AMD has Vega 11 (GP100/102 HBM2 competitor) and Vega 10 GDDR5X most likely (GP104 competitor).

Expecting Polaris 10 to compete with GP104 is like expecting HD7850/7870 to compete with GTX670/680.

TPU has relased GPU-Z and it shows GP104 as a 314mm 16nm chip, with 1733mhz paper boost. That means in the real world the card will boost to 1860-1900mhz. How the hell is a Polaris 10 card with a sub-250mm2 die size (rumoured) supposed to be a competitor to that? It's not, never was designed to be either.

 

[tential]

The irony in your post of using an outdated slide that doesn't align with the real world recent data and then trying to use that same slide to try and project Polaris 10/11 sales growth.

"Advanced Micro Devices, Inc. (NASDAQ:AMD) saw a double digit appreciation in total volume of discrete graphics processing unit (GPU) shipments during the quarter ended March 2016. The company also guided for 15% sequential in the next quarter as the demand for its semi-custom and GPU products reached new heights. AMD’s ROG initiative has become increasingly popular among gamers worldwide and is one of the catalysts behind the rising demand for company’s products, specially, in the gaming business."
http://www.thecountrycaller.com/456...a-corporations-nvda-market-share-wells-fargo/

Not trying to divert this into a financial/investor discussion but the point is those slides don't need to align with current reality as they are just projections made in the past.

15% GPU volume growth isn't 15% revenue growth. Two different things....

 

[xthetenth]

That makes no sense unless Polaris 10 is a 1070 performance competitor. We know that AMD has Vega 11 (GP100/102 HBM2 competitor) and Vega 10 GDDR5X most likely (GP104 competitor).

Expecting Polaris 10 to compete with GP104 is like expecting HD7850/7870 to compete with GTX670/680.

TPU has relased GPU-Z and it shows GP104 as a 314mm 16nm chip, with 1733mhz paper boost. That means in the real world the card will boost to 1860-1900mhz. How the hell is a Polaris 10 card with a sub-250mm2 die size (rumoured) supposed to be a competitor to that? It's not, never was designed to be either.

Sorry. I was unclear. I meant positioned relative to, but said positioned against.

 

[Krteq]

Reviving this thread from a dust.

AMD was supposed to launch their high-end class VEGA (codename) product in early 2017, the latest rumor was that the launch was being moved forward towards October. However today on Facebook Chris Hook, Sr. Director, Global Marketing and Public Relations, Radeon Technologies Group at AMD posted a photo with the comment "This is the Vega launch venue. Shh, don't tell the press". Considering half the worlds tech press is on his facebook, that obviously was a rather subtle joke slash hint...

...If this isn't a complete pun (and I do not expect it to be) then AMD is preparing a launch event for VEGA. Let's say you start planning such an even two months up-front, then indeed you can expect launch time to be October ...

Guru3D - AMD VEGA Launch Imminent?

 

[happy medium]

Reviving this thread from a dust.
Guru3D - AMD VEGA Launch Imminent?

Seems more like a ("hey we might have a Vega card in a few months,don't everyone buy the gtx1070)" type of marketing leak. I don't expect to see a Vega card till at least December.

 

[3DVagabond]

Reviving this thread from a dust.
Guru3D - AMD VEGA Launch Imminent?

If that's the launch venue it'll be years before we get it. lol

 

[Cookie Monster]

I wonder if Vega will have any models with GDDR5X. If their entire VEGA lineup is HBM based.. i can't think of the margins being so good unless these things perform well.

 

[3DVagabond]

I wonder if Vega will have any models with GDDR5X. If their entire VEGA lineup is HBM based.. i can't think of the margins being so good unless these things perform well.

I hope they give us HBM2 for $400.

 

[Keysplayr]

If that's the launch venue it'll be years before we get it. lol

Might need a Tetanus shot after the launch. Venue doesnt look very "up to code" if ya know what I mean.

 

[Keysplayr]

Is Vega supposed to be AMDs flagship? I mean at best it should compete with 1070/1080. Do they have anything lined up to eventually take on Titan X (p) or a 1080Ti?

 

[linkgoron]

Is Vega supposed to be AMDs flagship? I mean at best it should compete with 1070/1080. Do they have anything lined up to eventually take on Titan X (p) or a 1080Ti?

Supposedly Vega 11

 

[raghu78]

Is Vega supposed to be AMDs flagship? I mean at best it should compete with 1070/1080. Do they have anything lined up to eventually take on Titan X (p) or a 1080Ti?

http://www.anandtech.com/show/10145/amd-unveils-gpu-architecture-roadmap-after-polaris-comes-vega

"With Polaris confirmed to use GDDR5, Vega is notable for being the first AMD architecture to use HBM2, and the first parts in general to use HBM tech since Fiji. I’m presuming these are higher-end GPUs to complement the Polaris GPUs (the smaller of which we know to be a low-power laptop design), which is where HBM would be more cost-effective, at least at current prices.

Meanwhile AMD has also confirmed the number of GPUs in the Vega stack and their names. We’ll be seeing a Vega 10 and a Vega 11. This follows Polaris GPU naming – which has finally been confirmed – with Polaris 10 and Polaris 11. I have also been told that Polaris 11 is the smaller of the Polaris GPUs, so at this point it’s reasonable to assume the same for Vega."

there are 2 Vega chips. I would guess the smaller Vega chip will launch in early Q1 2017 (CES) and the bigger Vega in mid-2017 (Computex launch). Nvidia will probably have no competition for GP102 for another 9 months. But as long as we can see competition for GTX 1070 / GTX 1080 that would be nice. Right now Nvidia is laughing its way to the bank.

 

[Glo.]

I don't know who have written this, I think it might have been Fottemberg, but he was saying that Vega 10 might launch this year, as part of AMD own Gaming Platform - Project Quantum with Vega 10 and Zen CPU. It would also be launched around the time of Battlefield 1 release. Which is on 21st October.

Coincidence?

Lets be clear. If Vega 10 is competitor for GP102, then we have to expect high prices. 899-999$ at least. Thats why for launching this GPU this year, AMD does not need high quantity. Even if this GPU would go to Apple, for Example to updated Mac Pro all would Apple need is at most 50 thousand GPUs, for this year(25 thousand computers sold in highest end spec on GPU front, when the computer would launch in October. That is realistic expectation, and I would even say too optimistic, especially if tier 3 GPU update would cost there around 1500 - 2000$.).

 

[Arachnotronic]

I don't know who have written this, I think it might have been Fottemberg, but he was saying that Vega 10 might launch this year, as part of AMD own Gaming Platform - Project Quantum with Vega 10 and Zen CPU. It would also be launched around the time of Battlefield 1 release. Which is on 21st October.

Coincidence?

Fottemberg is not a reliable source. Lisa Su said Vega launching within the next two quarters, which is Q1 2017. If they were planning a limited EOY launch, they would say so in the investor calls just as they do with Summit Ridge.

 

[Glo.]

Fottemberg is not a reliable source. Lisa Su said Vega launching within the next two quarters, which is Q1 2017. If they were planning a limited EOY launch, they would say so in the investor calls just as they do with Summit Ridge.

Within means, exactly that. Within. Not at the end of two upcoming quarters. It can be either Q4 2016 or Q1 2017.

Mostly that would mean paper launch of this GPU, with small availability in 2016 and wide in 2017.
Especially if you pair this with possible production of this GPU on TSMC process, and knowing that it may lack capacity, because of production of Apple A10 and A10X CPUs.

 

[raghu78]

Right now AMD relies on marketing and hype to sell product as the inherent product competitiveness is pretty weak as we saw with polaris. We are seeing more of the same with Vega. Nvidia meanwhile is executing like a juggernaut raking in the moolah with minimum noise or fuss. AMD talked for 6 months about Polaris and delivered a disappointing product. Nvidia spoke about Pascal 3 weeks before GTX 1080 launch and knocked it out of the park. This is basically the difference between a company which is executing flawlessly with strong R&D (Nvidia) and one which is struggling to compete due to weak R&D and financial state.