AMD’s Big Navi is “going to disrupt 4K gaming" once it’s got the mainstream market dealt with

[cucker tarlson]

Like Ryzen, AMD’s Big Navi is “going to similarly disrupt 4K” gaming

"All of us need a thriving Radeon GPU ecosystem. So, are we going after 4K, and going to similarly disrupt 4K? Absolutely."

www.pcgamesn.com

Radeon product management lead, Mithun Chandrasekhar, hinted at the red team’s plan for 4K domination once it’s got the mainstream market dealt with

"all of us need a thriving Radeon GPU ecosystem. So, are we going after 4K, and going to similarly disrupt 4K? Absolutely, you can count on that. But that’s all I can say right now".

 

[lexluthermiester]

Let's hope! That would be great for the industry as a whole!

 

[cucker tarlson]

well it's 2020,shouldn't they have some sort of timeframe even for a TU104 competitor ?

 

[Space Lynx]

good times are a coming, I just hope they can prove better drivers and proper Freesync.

 

[Frick]

Let's hope! That would be great for the industry as a whole!

sure, but we should also not take their word for it until it's actually done.

 

[Deleted member 67555]

well it's 2020,shouldn't they have some sort of timeframe even for a TU104 competitor ?

^This
I'm willing to wait until I have the money (in March when I get my bonus) to make a decision but if they don't at least have something on paper that's been reviewed then it's one more customer that will not buy that generation.

 

[silentbogo]

"once it’s got the mainstream market dealt with" is the key phrase here, cause all of these recent rumors of "Big Navi coming soon" were debunked.
In that same interview Mithun Chandrasekhar never says "Big Navi", he just says that 4K-capable GPUs will eventually happen (not necessarily Navi, and not necessarily in 2020).

 

[Countryside]

Sounds good but it would be more interesting after it goes through a test bench and we see the results.

 

[Rahnak]

Uh huh.. I'll believe it when I see it. And I hope their mid-range refresh for the 5700 series is more than what they did with the 480 > 580.

 

[cucker tarlson]

Uh huh.. I'll believe it when I see it. And I hope their mid-range refresh for the 5700 series is more than what they did with the 480 > 580.

->590

 

[bug]

Last time AMD swore their cards were made for 4k, not for FHD, we ended up with Vega. Here's hoping it will be different this time around.

 

[cucker tarlson]

Last time AMD swore their cards were made for 4k, not for FHD, we ended up with Vega. Here's hoping it will be different this time around.

v64/v56 were both 1440p cards

good cards nevertheless by amd's standards
I love how the red fanbase turned on Raja but since Vega they can't provide a competitor for the 104 die despite going a node down or stable drivers.

 

[Chomiq]

Seeing how team red GPU division so far managed to compromise their own launches with broken drivers or bad decisions I don't doubt that big Navi will disrupt something on launch.

 

[64K]

I love how the red fanbase turned on Raja but since Vega they can't provide a competitor for the 104 die despite going a node down or stable drivers.

The AMD fans turned on Raja unfairly imo. He wanted Lisa Su to put more R&D money into RTG but she wanted to put more into Ryzen. As it turns out she was right but blaming Koduri for going to Intel wasn't a bad thing to do to AMD. Intel just had more to offer Koduri and Lisa Su had different priorities.

 

[ppn]

Has the mining died, then forget about disruptions. I mean $399 384bit card would be nice, and 5nm shrinks to be prohibited for 5 years. So we can have nice and stable love for the card, but no, what we have is money grab after money grab forever. 5nm already enters volume production in 2020, video cards in 2021. So no you can't disrupt, only wallets, whoever wanted already got 2080Ti and was disrupting himself with it for more than 18 months to a year.

 

[MrGRiMv25]

I wonder if it is the same GPU that appeared in GPU-L on my Android app. It lists an unreleased 5900xt on RDNA 2.0/NAVI 21 with 5120 CU, 320TMU's, 96 ROP's, 12GB GDDR5 on a 384bit bus @ 14,000mhz for 672GB/s bandwidth. [505mm die size] - Launch is vague at 2020.

Half Precision FP16 (Boost) is: 31437 GFlops
Single Precision FP32 (Boost) is : 15718 GFlops
Double Precision FP64 (Boost) is: 982 GFlops

Hope it's real as that would be quite a powerhouse.

 

[cucker tarlson]

I wonder if it is the same GPU that appeared in GPU-L on my Android app. It lists an unreleased 5900xt on RDNA 2.0/NAVI 21 with 5120 CU, 320TMU's, 96 ROP's, 12GB GDDR5 on a 384bit bus @ 14,000mhz for 672GB/s bandwidth. [505mm die size] - Launch is vague at 2020.

Half Precision FP16 (Boost) is: 31437 GFlops
Single Precision FP32 (Boost) is : 15718 GFlops
Double Precision FP64 (Boost) is: 982 GFlops

Hope it's real as that would be quite a powerhouse.

Fake

 

[MrGRiMv25]

Fake

More than likely as it says unreleased, was just a small shred of my psyche that wanted it to be true haha... though they have predicted quite a few of the recent RX launches so there might be an inkiling of a chance.

But, cosidering how AMD have eben treating their board partners the shred of hope in me is dwindling. Haha....

 

[cucker tarlson]

More than likely as it says unreleased, was just a small shred of my psyche that wanted it to be true haha... though they have predicted quite a few of the recent RX launches so there might be an inkiling of a chance.

But, cosidering how AMD have eben treating their board partners the shred of hope in me is dwindling. Haha....

it's impossible to do a 16tflop cartd on rdna1 unless you produced a 400w one.

 

[Darmok N Jalad]

Anandtech said they mentioned a Navi refresh, in their earnings call, of all places. More details are rumored in March. That would certainly help explain an RDNA 2.0 test mule, though it’s all with a grain of salt these days.

 

[Chomiq]

I wonder if it is the same GPU that appeared in GPU-L on my Android app. It lists an unreleased 5900xt on RDNA 2.0/NAVI 21 with 5120 CU, 320TMU's, 96 ROP's, 12GB GDDR5 on a 384bit bus @ 14,000mhz for 672GB/s bandwidth. [505mm die size] - Launch is vague at 2020.

Half Precision FP16 (Boost) is: 31437 GFlops
Single Precision FP32 (Boost) is : 15718 GFlops
Double Precision FP64 (Boost) is: 982 GFlops

Hope it's real as that would be quite a powerhouse.

5ghz Zen2 called, it wants its outrageous rumors back.

 

[MrGRiMv25]

it's impossible to do a 16tflop cartd on rdna1 unless you produced a 400w one.

The specs mentioned in GPU-L DB was that it was RDNA ,2 so if for some miracle GPU-L is right then it could be in that ballpark. All hearsay at this point though, but one thing I can defo agree on is AMD need to get big Navi/RDNA2 out asap.

 

[kings]

Talk is cheap. AMD always has something fantastic under its sleeve. We've been hearing this for years.

We want release dates and performance figures, no more empty promises like "wait for us, we will have something fantastic" that they say almost every year.

And honestly, after AMD said that the R9 390 series were 4K GPUs, I think they’re a little lost on what that really means.

 

[moproblems99]

We want release dates and performance figures

Bar graphs with no scales for reference incoming!

 

[windwhirl]

Anandtech said they mentioned a Navi refresh, in their earnings call, of all places. More details are rumored in March. That would certainly help explain an RDNA 2.0 test mule, though it’s all with a grain of salt these days.

Tom's Hardware asked for clarification and AMD answered that the plan is to launch RDNA2 GPUs in 2020, not a RX5000 refresh...