(Updated) AMD Navi GPU to Offer GTX 1080 Class Performance at ~$250 Report Claims

[leadeater]

7 minutes ago, Taf the Ghost said:

1070, 1070 Ti & 1080 are all the same die. It's just that the top-end of the GPU was relatively much stronger, so Nvidia charged more. And keeps charging more. The die itself is only slightly bigger than the 680/770 die. Pascal & the node it's on just happened to turn out really, really good.

I know the die is the same but the configuration of them is different for each of those products. Sharing the same base die hardly makes them in the same class of GPU.

 

Either way boring topic and not about AMD so meh, we're all enthusiasts so call it mid if we want. Just that product stack wise no it isn't, just the way you view it I guess.

[minervx]

I'm keeping expectations low.  The RAM and GPU shortage, along with increasing mining and gaming demand are hard limits on what's possible.  This is what hurt their entire line and made Vega hard to produce at a reasonable price for many.

 

As long as it's a card that's worth buying, it's a step up.  I'm not expecting a $250 GTX 1080.  Maybe a $400 1080.

[GoldenLag]

4 minutes ago, huilun02 said:

Speculative news regarding Radeon graphics? Nah lets bash AMD instead.

We always need to compare their products to Nvidia's flagship, because everyone already has or is able to afford a 1080Ti.

Forget about AMD's intention to bring better price/perf because they can't put a card in my hand at launch MSRP.

Its not the miners fault because Nvidia's mid range to to high end cards are in greater supply easily found at MSRP.

Nvidia is better because GPP is wonderful and Gameworks titles have always looked and performed better than anything else.

How are we bashing AMD here? What else are we going to compare them to? Or are you talking about the article?

[Arika]

 

[Dylanc1500]

6 minutes ago, leadeater said:

Dreams.... .

 

Edit:

Err edited that joke, too far lol

Lol well that's nicer than what was in the back of my mind.

 

I wouldn't say no though if someone "donated" one to me.

[Ryujin2003]

1 hour ago, Princess Cadence said:

Hey it only took AMD 3 years to make Vega have the same performance of the TITAN X Maxwell... seems like a pattern, AMD's often 3 years behind on the highest end.

There is also a chance that they weren't aiming, but only comparing a future product against something current with known performance metrics.

 

[TheRandomness]

And I shall just sit and wait for the performance benchmarks to come out. I need my hype suppression pills.

[GoldenLag]

Just now, TheRandomness said:

And I shall just sit and wait for the performance benchmarks to come out. I need my hype suppression pills.

Im putting down railtracks for the incoming hypetrain.

[leadeater]

5 minutes ago, GoldenLag said:

Im putting down railtracks for the incoming hypetrain.

Hypetrain need no tracks, it goes where it wants. 

 

[Okjoek]

1 minute ago, leadeater said:

Aaaaaaand.... not clicking on that .

 

It's not overly complicated though, you just need to look at the product stack (in full) and the performance they give then it's a simple case of group cards in to performance tiers. Really the only thing that actually makes a 1080 look not high end is just how much bigger and faster the GP102 die is and the 3 products cut from that, if you consider just how close those 3 top products actually are to each other (1080Ti, Titan X, Titan Xp) it's hardly worth treating them as separate cards but rather one sane purchase and two status symbols .

 

I could support the idea of calling a 1080 mid range if we're going to class those top 3 as high end but that creates a giant problem for every card below a 1080 and what to call those, is a 1030 low end but then is a 1060 6GB also low end? How can those possibly be in the same class name they are nothing alike. Only way 1080 being mid range works is if we ignore the existance of anything less than 1060 6GB.

The minimalist in me makes me want to rank things from the bottom up based on power consumption. I've come to trust Adored knows his shit though. IMO He's a great journalist and market analyst.

 

 

[ravenshrike]

1 hour ago, leadeater said:

Myself I doubt the MCM theory, I just don't think the technology, design and fab processing are all collectively at the point to be able to pull it off for a GPU with the bandwidth requirements those have. If you look at the Infinity Fabric as in example it's only 38GB/s and I don't see it jumping from that to the order of 300GB/s+ in the space of time without requiring far too much die space and transistors.

 

We might see a more CCX style but I'm not counting on MCM at all.

How does that shake out if you assume that RTG is licensing EMIB tech from Intel for use only on the graphics side of things? I mean, you'd think there had to be some sort of side deal going on to just let Koduri go over to Intel.

[GoldenLag]

2 minutes ago, leadeater said:

Hypetrain need no tracks, it goes where it wants. 

 

The US F-35 TRAINer plane. Best plane in the world, but allways dissapoints

[Dylanc1500]

1 minute ago, leadeater said:

Hypetrain need no tracks, it goes where it wants. 

 

Great, so now it'll never derail. I guess there is still a possibility of ending up in a big ball of flames.

[Trixanity]

4 minutes ago, ravenshrike said:

How does that shake out if you assume that RTG is licensing EMIB tech from Intel for use only on the graphics side of things? I mean, you'd think there had to be some sort of side deal going on to just let Koduri go over to Intel.

AMD isn't licensing it. They've said they'll continue to develop their interposer tech in collaboration with partners. I doubt it'll ever be as cost effective though which is a problem.

[GoldenLag]

Just now, Dylanc1500 said:

Great, so now it'll never derail. I guess there is still a possibility of ending up in a big ball of flames.

Not another GTX 480 hopefully

[leadeater]

8 minutes ago, ravenshrike said:

How does that shake out if you assume that RTG is licensing EMIB tech from Intel for use only on the graphics side of things? I mean, you'd think there had to be some sort of side deal going on to just let Koduri go over to Intel.

Hmmm.

 

Quote

We do know that a single chip can support up to 20,000 EMIB connections with up to 2Gbps of throughput each.

http://www.tomshardware.co.uk/intel-emib-interconnect-fpga-chiplet,news-56570.html

 

20,000 x 2 = 40,000 Gbps = 5,000 GB/s.

 

Um yea sure EMIB sounds perfect bandwidth wise.

[JediFragger]

8 minutes ago, leadeater said:

Hypetrain need no tracks, it goes where it wants. 

 

Dr. Emmet Brown: Roads? Where we’re going we don’t need…roads.

[leadeater]

10 minutes ago, GoldenLag said:

The US F-35 TRAINer plane. Best plane in the world, but allways dissapoints

 

5 minutes ago, Dylanc1500 said:

Great, so now it'll never derail. I guess there is still a possibility of ending up in a big ball of flames.

 

3 minutes ago, JediFragger said:

Dr. Emmet Brown: Roads? Where we’re going we don’t need…roads.

 

Achievement Unlocked: Thread derail

 

[Dylanc1500]

1 minute ago, leadeater said:

 

 

 

Achievement Unlocked: Thread derail

 

PULL UP!!!

 

Can't derail what was never on tracks!

[Rune]

Yawn.

[Drak3]

3 minutes ago, leadeater said:

 

 

 

Achievement Unlocked: Thread derail

 

Why is it fine for you mods to derail topics, but when Drak3 does it, it's "a violation of community standards"

 

Spoiler

Other than the fact that we can go from Power supplies to what's between Drak3's legs in the span of 30 minutes.

 

[ravenshrike]

9 minutes ago, Trixanity said:

AMD isn't licensing it. They've said they'll continue to develop their interposer tech in collaboration with partners. I doubt it'll ever be as cost effective though which is a problem.

I realize AMD isn't licensing it for CPUs, mainly because there's no way Intel would agree to it. Which means they would still need to develop their own tech for anything involving the CPU side. However, that doesn't mean that RTG hasn't licensed it solely for a MCM GPU. Again, AMD would have wanted something significant to just release Koduri to go over to Intel like they did. Using EMIB solely on their discrete graphics cards fits.

[asus killer]

44 minutes ago, Sauron said:

That's a bit like saying that there's no point in making a gtx 1060 when the gtx 780 exists.

Nvidia didn't just made a 1060, they made a 1070, a 1080, a 1080ti, the titans all on that architecture. According to the article they don't intent to do anything faster than vega already out there.

[bcredeur97]

2019 huh?

After Nvidia releases Turing it'll probably only be on par with the 60/70 series cards

 

I think we're basically looking at another rx480/580 (once it's released)

[Doobeedoo]

Hah, I literally said yday at other post that Navi may offer Vega 64 performance for half od it's expected msrp price. Nothing unexpected giving that Navi is still year off too and giving that by then next successor should be a larger increase in performance. Seeing that Navi uses MCM design. 

 

Either way I just hope graphics cards are back to actual normal pricings by then. Could use an upgrade next year.