• 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 VEGA confirmed for 2017 H1

Status
Not open for further replies.
Its stupid but the longer I hold out for Vega, the more and more tempted I am to pick up an X80 Polaris card instead.

The closer it gets, the more tempted I become.

Sick of waiting now :mad:
 
If Intel one day (or some other 3rd party) came to market with their own GPU, would you buy it?

What we need to keep everyone honest is another player at the high end.

Surprised no one's tried jumping in. (I know of Knights Landing)

Yes but we were all left high and dry waiting for Larrabee. A duopoly is better than a monolpoy, but not much better at the moment!
 
Its stupid but the longer I hold out for Vega, the more and more tempted I am to pick up an X80 Polaris card instead.

The closer it gets, the more tempted I become.

Sick of waiting now :mad:

I bought a 480 as a temporary card over a month ago now. Gave up waiting and upgraded to a used 1080.

The 580 release and now this new pro duo card just tells me Vega is far off or not as good as people hope it will be.
 
AMD update on VEGA launch.

14a35a5e64.png




Source https://www.reddit.com/r/Amd/comments/4zr8co/amd_said_vega_is_coming_in_h1_of_2017_in_a/

https://forums.anandtech.com/threads/amd-vega-coming-in-1h-2017.2484851/

still no news? First wccftech said it was April, then wccftech said May, now wccftech said it's end of June?

Fixed that for you. Above however is the first post in this thread, posted in August 2016, referencing what AMD said: 1H 2017

So this about sums it up:

First it was H1 2017 then it was H1 2017 and now it's H1 2017 if we go by the official AMD statements.
 
Last edited:
First it was H1 2017 then it was H1 2017 and now it's H1 2017 if we go by the official AMD statements.
Well said. I was thinking the exact same thing. Where does he get his news from I wonder? Wccftech? Lol :p
 
I hope Vega slots in between the 1080 and the 1080ti. Pascal is over a year old and I want to build a Ryzen+Vega rig :D

Personally I will be very impressed if Vega gets anywhere close to 1080ti with DX11. Let's not forget it is still using the sophisticated hardware scheduler approach, meaning its strong point will still be DX12/Vulkan.

If it closes the gap in DX11 it'll be the same as with the RX480/RX580 vs the 1060: you'd be getting slightly worse DX11 (not significant) but much better DX12/Vulkan. Forward looking people will gladly go for it (though obviously t will never sell as much as it should based on price/perf).
 
AMD's roadmap they published last year said a 2017 release. They later clarified by saying first half of 2017, they've never said anything different. It's just rumour sites and what not that have been promoting that.

it's difficult to get anything real in all this fake noise made by all the sites on the planet ... and that's ignoring wccftech!
 
pretty sure, it was Q1 in the beginning not first half, unless I am imagining things! oh well, so end of June it is!
It was never Q1, if was Q1 hen AMD would have said that. AMD explicitly said H1 because it sounds better than saying end of Q2 but everyone know that when AMD said H1 it meant May to June.
 
It was never Q1, if was Q1 hen AMD would have said that. AMD explicitly said H1 because it sounds better than saying end of Q2 but everyone know that when AMD said H1 it meant May to June.

This. They said H1 in the very same events where they said Q1 for Ryzen.

If they meant Q1, they would've said Q1... H1 basically means Q2 in business-speak.
 
This. They said H1 in the very same events where they said Q1 for Ryzen.

If they meant Q1, they would've said Q1... H1 basically means Q2 in business-speak.
It's such a lousy PR move, too. I guess marketing data has shown that H1 sounds better than Q2, but I think all it does is create disappointment for consumers who are waiting patiently and basically always get it at the very end of Q2, where that patience is really at its wits end.

Maybe it's been shown to help investor confidence or something, I dunno.
 
It's such a lousy PR move, too. I guess marketing data has shown that H1 sounds better than Q2, but I think all it does is create disappointment for consumers who are waiting patiently and basically always get it at the very end of Q2, where that patience is really at its wits end.

Maybe it's been shown to help investor confidence or something, I dunno.
The information is not really meant for consumers but for investors and business people in the industry.

It is one of these silly things where investors/industry knows that H1 means end of Q2 but companies always have to keep up the pretense that its not a code word.
 
yes, but they do not have a choice, it's either perpetual underdog, or try and get out of this cycle, and after these last couple years i think AMD got a clear message of the global mindshare, they will never sell enough product no matter how amazing they are, even if they bring Vega beating the 1080Ti and price it the same or even 20-30$ less, Nvidia will still sell twice as many Ti than vega, you can be 100% sure of it, but if Navi manages to do what vega did, the mindshare will shift and navi will start selling well as good or even better than nvidia, but if Navi comes out short, it would be like hiting a reset botton on mindshare, ppl will still see AMD product as incappable.
AMD cannot drop HBM, because it's a huge part of all their future products, from APU's to Navi and beyond, and it's all because of the HB memory controller and cache, in many instances AMD explained how they want to make scalable GPUs, CPUs and APUs, part of that is using HBM memory stacks over the graphic unit it self not along side it, and just scale those units on interposer, so HBM isn't going anywhere, and the price isn't that high if you manage the size, especialy if you come up with a way to cancel the down side of it, let's say by having 512TB or virtual adresses.
beside HBM is more expensive but how much expensive is it ? is 4GB HBM cheaper than 8 or 12GB of GDDR5/X ? intel is starting to use it, nvidia too, so AMD wont be the only company paying for it's bad yields.
googled it but couldn't find the price of HBM2 and DDR5/X per GB, if anyone has the info...


AMD is not a charity, they have to make healthy profit to reinvest in R&D, start paying off their $2bn loans, and start making their investors happy. Now trading some profits for market share may help in the future its no guarantee of future revenue, e.g. if nvidia make a more compelling product then AMD will still loose sales. Besides which, market share will come naturally if there is a good top-to-bottom lineup of competitive products on time, with good release drivers and closer developer relations. Not having an answer to pascal for a year is costing market share and mind share. Having similar products to Nvidia at similar prices would have netted them plenty of sales and profits. Trying to undercut Nvidia pricing is very risky, nothing stops nvidia lowering prices if NVidia feel that they are loosing sales. the fact that Nvidia frequently don't hit AMD's price levels is a sure fact that Nvidia know they wont gain more sales so ultimately it is just costing AMD profits. Which is exactly why AMD have said they need to escape form being known as this budget brand, it doesn't ultimately gain them much.


You also seem somewhat confused about Navi. The availability of Navi with multiple dies on a single interpose is for HPC/compute applications, it doesn't work for games without massive driver support. Look how well crossfire works currently.

As for prices, you have to compare like for like. how much does 12GB of HBM2 cost compared to 12GB GDDR5X? It is a complete fallacy to believe faster memory allows to to use less.
 
It seems pretty obvious to me,that Amd are doing exactly what they did with polaris,they will announce Rx Vega release date and price at computex show on the 30th may.
 
Yes but we are talking reference coolers here, and you have to admit that AMD haven't got a great track record in this regard!! Hopefully it will change with Vega (although it didn't seem to with the 480) :)
Yea but you was talking in reference to power draw. U was making out they was hot because of thier power draw. And why it needed a watercooler etc. Which now you have just admitted its because of the reference cooler which is what my point was. Lol. 290x was not hot because of power draw. It was the reference cooler that made them seem hot. Aib cards was not hot at all. My 980ti produced far more heat than my 290x ever did.
 
Status
Not open for further replies.
Back
Top Bottom