Vega/Navi Rumors (Updated)

[french toast]

I was referring to level of "secrecy". It was ridiculous in case of RV770 and even more strict in case Evergreen family and EyeFinity.

But they never released a card of same uarch a month before that sucked at gaming though?

 

[Krteq]

Yes, you are right, but they are NOT advertising Vega FE as a gaming card.

 

[CatMerc]

A return to RV770 would match with what I've heard about the current RTG staff being ATi fanboys, but I have doubts about it actually happening

 

[tviceman]

Yes, you are right, but they are NOT advertising Vega FE as a gaming card.

That is because it's $1,000 and doesn't even come close to Nvidia's $1,000 offering. If it matched or trumped Nvidia's best, then you can be sure AMD would be making every comparison possible.

 

[EXCellR8]

That is true... you could probably manage a titan xp for around that price; that card eats big Vega for breakfast and brunch

 

[Despoiler]

That is because it's $1,000 and doesn't even come close to Nvidia's $1,000 offering. If it matched or trumped Nvidia's best, then you can be sure AMD would be making every comparison possible.

Yah it's only going toe to toe with Nvidia's $6000 card in the workloads it was really designed for.

http://www.tomshardware.de/vega-ben...ngsaufnahme-gaming,testberichte-242375-6.html

 

[french toast]

Yes, you are right, but they are NOT advertising Vega FE as a gaming card.

But they are advertising Rx as such which uses the same uarch, besides titan started -and is the market leader in prosumer GPUs, and the titan is judged on a mix of gaming and semi pro, it excels in gaming- Vega Fe should be judged accordingly.

An anology would be Vauxhall bringing out a new hatchback car, and saying to everyone it is targeted for carrying the shopping/kids to school and shouldn't be judged on its driving experience-when the market leader ford fiesta is judged on such things and excels- just like every hatchback is judged on certain parameters that's the standard in its class. (Jack of all trades)
In the GPU world the prosumer cards are the hatchbacks, except they are expensive premium products unlike small cars.

Consumer = gaming.
Prosumer= uber gaming/gpgpu/semi professional mix (uber price)
Professional= uber gpgpu compute, stable certified drivers (crazy price)
What you and other people are suggesting is that AMD purposely turned off great gaming performance for that card ie artificially segmenting against the market expectations, which is daft.
You better believe if Vega Fe had uber titan like gaming capabilities they would enable it and market it as such, they are not because frankly it cant-hence this rubbish about just for game development.

The alternative to that Is what I mentioned in an earlier post, that AMD is pulling some masterful sandbagging for an early Christmas present, not impossible but extremely unlikely imo.

Im dismissing entirely the notion that AMD is doing neither of those and is working on an unready driver that is going to increase performance by 35℅ in one month of work- they have had 8 months to work on drivers- most of the gains would be in code already.
Edt; Just to restate- I would love some crazy driver action as I would probably go buy one, also I still think the bandwidth limitation rumours are fishy as hell, I expect if true just fixing that would bring 20℅ or so I would think, but would also hamper compute, so they would be shooting themselves in the foot with Vega Fe either way.

 

[EXCellR8]

RX Vega non-reference might get us closer to 1080 performance but my guess is that power consumption is going to be brutal.

Lot of sources stating Vega is "stillborn" and I'm starting to agree

 

[Zstream]

Which team created this GPU? I heard they have two teams, or at least at some point did, and they would take turns developing GPU's.

 

[vissarix]

Thats what you get when some folks who kept telling us Gtx1080ti + 30% = Vega
Partly Amd helped with their brilliant marketing #PoorVolta

Well you guys got pretty close with your conservative predictions ...just a few % short, but hey, drivers might fix that...

 

[Kenmitch]

RX Vega non-reference might get us closer to 1080 performance but my guess is that power consumption is going to be brutal.

Lot of sources stating Vega is "stillborn" and I'm starting to agree

Journalist using the term stillborn to describe a functional product should be cyber-slapped silly!

Only thing for sure is Vega RX is what it is. Whatever that is we won't know until the official launch and reviews!

 

[EXCellR8]

uh, [certainly].

except the cards that matter aren't going to be released right away. AMD hasn't had the greatest record with reference design in recent years. that's all I used to buy too but it seems like there's something holding back each ref pcb

Profanity isn't allowed in the technical forums.
-- stahlhart

 

[davide445]

I'm starting thinking ASD has simply focus all his effort into workstation and server performance for both CPU and GPU design, leaving gaming as secondary market.
Considering both performance for Ryzen and Vega are really competitive everywhere in professional application this might be a good strategy, Professional market is much more lucrative than gaming, so let's Nvidia have the edge on this and focus the much smaller resources on the other.

 

[iwulff]

I'm starting thinking ASD has simply focus all his effort into workstation and server performance for both CPU and GPU design, leaving gaming as secondary market.
Considering both performance for Ryzen and Vega are really competitive everywhere in professional application this might be a good strategy, Professional market is much more lucrative than gaming, so let's Nvidia have the edge on this and focus the much smaller resources on the other.

The Pro market always had more focus, also for Nvidia. Vega's drivers are probably not mature yet, or just becoming before the RX launch. Raja did comment that they were facing quite some difficulties with developing the Vega drivers. Expect to see large increases between the Vega FE now and the RX for the RX launch or the months afterwards.

 

[swilli89]

Which team created this GPU? I heard they have two teams, or at least at some point did, and they would take turns developing GPU's.

AMD has not discerned but I believe it is their team located in China that has done the bulk of the work. At least it seemed this way when the Vega team picture was posted and it appeared to be mainly Chinese engineers.

 

[lobz]

But they never released a card of same uarch a month before that sucked at gaming though?

no, they provided the press with samples that had like 40-50% lower performance than the final product - read anandtech's review

 

[Zstream]

AMD has not discerned but I believe it is their team located in China that has done the bulk of the work. At least it seemed this way when the Vega team picture was posted and it appeared to be mainly Chinese engineers.

Isn't that the team that's notorious for high power GPU's that suck?

 

[swilli89]

Isn't that the team that's notorious for high power GPU's that suck?

Well if they aren't then they might soon be

 

[Elixer]

This remains interesting.
I do not know what to think.
I do get some worry that the draw stream binner rasterizer may need application support after all to function.
But it may be my not flawless translation and understanding efforts.

http://www.freshpatents.com/-dt20161222ptan20160371873.php&usg=ALkJrhjunwhetA17bfzQv6lvdRfNrx0n8g

That would be doubtful, since all the major current API's wouldn't make use of it.
That is why it most likely is just a generic "black box", takes geometry as input, sees if it need culling, and discards what isn't seen.

While new functions can be exposed to things like Vulkan, it won't make it into DX anytime soon.

 

[Elixer]

AMD has not discerned but I believe it is their team located in China that has done the bulk of the work. At least it seemed this way when the Vega team picture was posted and it appeared to be mainly Chinese engineers.

Yes they have.

By this time for both the 10xx & Fury launch we already saw box art, so... with only 9 days away, this seems odd.
Really, really hope they don't do a paper launch.

 

[tajoh111]

A very interesting comparison here: http://www.3dmark.com/compare/fs/13008898/fs/13151493/fs/12065201

Click the results. The person cheated with the vega card my modifying the tesselation level.

They went out of their way to hide this fact by burying it in a comparison. Sneaky.

 

[Malogeek]

no, they provided the press with samples that had like 40-50% lower performance than the final product - read anandtech's review

Bodes well for RX Vega then lol

 

[crashtech]

Click the results. The person cheated with the vega card my modifying the tesselation level.

They went out of their way to hide this fact by burying it in a comparison. Sneaky.

I don't see anything out of order, just the 1080ti taking top honors. You might want to provide further substantiation for your claim.

 

[tajoh111]

http://www.3dmark.com/fs/13151493

These are the results if you look in detail.

Look at the right corner, it says results have a modified tessellation load.

From the the original comparison, it things things are apples to apples, but if you look closer at the results, the radeons have a cheat enabled to boost the score.

 

[swilli89]

Yes they have.

By this time for both the 10xx & Fury launch we already saw box art, so... with only 9 days away, this seems odd.
Really, really hope they don't do a paper launch.

Ah so according to Raja this is the same team that brought Polaris.. So I wonder if Navi will be from the stateside team that built Hawaii/Fiji?