So this will big big Navi, not consumer Navi?

What's the rest of that in the photo translate to English, does anyone know?

Only $10K, if some enterprise is insane enough to pay it.

Xavier Zepherious
Alleged AMD Next-Gen Flagship Navi ‘Radeon RX’ GPU Specifications Leaked – 5120 Cores, 24 GB HBM2e Memory, 2 TB/s Bandwidth
 

 
Surprisingly, the document lists the complete GPU and memory configurations. For the GPU itself, we are looking at 5120 stream processors packed in 80 Compute Units. The GPU consists of a total of 320 TMUs and 96 ROPs. The L2 cache size on the chip also sits at a mammoth 12 MB.
 
In terms of memory, we are looking at 24 GB of high-performance HBM2e. The GPU would feature a 4096-bit bus interface and reportedly up to 2 Terabytes per second worth of bandwidth
 
https://wccftech.com/amd-next-gen-flagship-radeon-rx-navi-gpu-specs-leak-5120-cores-24-gb-memory/
 
 
---------------------------------------------------
 
thats only 2 stacks of HDM2e - 4 would have been 48GB
 
Ampere(titan) is allegds to have 8 stacks for 96GB 
 

atfrico
Told ya about memory size😼.
Please show yourselves nonbelievers and say i was wrong😹😹

You are both wrong: https://www.tomshardware.com/news/amd-big-navi-gpus-are-fake-sk-hynik-refutes-2tbps-hbm2e-claims
Once again showing how wccftech is the garbage news of the internet that I have stated time and time again yet you guys hold onto their information as gold.  But hey, NVIDIA socks information is fake news and should be removed from the Computer Industry News Section, but yet wccftech is acceptable when ive stated countless times it should not be in this section.  Here we are once again with the obvious.

boylerya

Xavier Zepherious
Alleged AMD Next-Gen Flagship Navi ‘Radeon RX’ GPU Specifications Leaked – 5120 Cores, 24 GB HBM2e Memory, 2 TB/s Bandwidth
 

 
Surprisingly, the document lists the complete GPU and memory configurations. For the GPU itself, we are looking at 5120 stream processors packed in 80 Compute Units. The GPU consists of a total of 320 TMUs and 96 ROPs. The L2 cache size on the chip also sits at a mammoth 12 MB.
 
In terms of memory, we are looking at 24 GB of high-performance HBM2e. The GPU would feature a 4096-bit bus interface and reportedly up to 2 Terabytes per second worth of bandwidth
 
https://wccftech.com/amd-next-gen-flagship-radeon-rx-navi-gpu-specs-leak-5120-cores-24-gb-memory/
 
 
---------------------------------------------------
 
thats only 2 stacks of HDM2e - 4 would have been 48GB
 
Ampere(titan) is allegds to have 8 stacks for 96GB 
 

atfrico
Told ya about memory size😼.
Please show yourselves nonbelievers and say i was wrong😹😹

You are both wrong: https://www.tomshardware.com/news/amd-big-navi-gpus-are-fake-sk-hynik-refutes-2tbps-hbm2e-claims
Once again showing how wccftech is the garbage news of the internet that I have stated time and time again yet you guys hold onto their information as gold.  But hey, NVIDIA socks information is fake news and should be removed from the Computer Industry News Section, but yet wccftech is acceptable when ive stated countless times it should not be in this section.  Here we are once again with the obvious.

I am still sticking on my saying that the new flagship GPUs will have above 11gb of memory...😉😼
Fake or not it depends on how you view it hippie! Im optimistic even when other members are wrong.😸😼

Ban the socks!

Just to point out, SK Hynix does indeed have 4 Gb/s HBM2e stacks which they showcased last week at ISSCC-2020 as stated in a reply to the leaker by Twitter user Hans De Vries.
 
so HK Hynix saying they don't have the HBM2e that can go that fast = they just showcased it at isscc-2020 
 

 
so HK Hynix is Misrepresenting the truth....maybe
 
 
That's 512 GB/s for the die stack(16GB per chip here- 128/8bits=16GB) on a 1024 Bus
on a 4096 bus that would be 512GB's x4(for 4x1024) = 2 TB/s
 
but 12GB Vid card means 4GB sticks and 4 of them - one used for ECC so instead of 16GB you have 12 GB.... maybe
i mean you are NOT going to have 6GB HBM2E or 3GB HBM2E stacks
 
 
otherwise you use 1 chip - that means 1024 bus or 2 chips on 2048 bus - that means 8GB,16GB,24GB,32GB or 48GB
the larger the bus the better for bandwidth and more chips means more ram - and you might as well go for denser ram
 
they also said they were doing 12-high stack - which would be 768GB/s min per stack in a previous story - which if pushed like samsung could maybe do 1TB/s
 
 
 posted Dec 
 
For HBMe, manufacturers such as Micron, Samsung or SK Hynix simply use 2 GB dies instead of those with 1 GB, which increases the storage density per stack to 16 GB. The new high bandwidth memory with 24 GB, on the other hand, is a 12-Hi design with an I / O die and ergo twelve DRAM chips with 2 GB each. The data rate is 2.4 GBit / s per stack and thus the data transfer rate of a 3D stack on a 1,024-bit channel is around 307 GByte / s.
 
https://www.golem.de/news/high-bandwidth-memory-jedec-spezifiziert-stapelspeicher-mit-24-gbyte-1812-138299.html
 
if they can achieve 512GB/s per stack then with 4 chips at you would have 2 TB/s and 96GB ram or 192GB with 8 stacks - like ampere titan
 
HK might be sly and say it's fake because it not in full production yet and just test production
 
 

SK Hynix Announces 3.6 Gbps HBM2E Memory For 2020: 1.8 TB/sec For Next-Gen Accelerators

SK Hynix this morning has thrown their hat into the ring as the second company to announce memory based on the HBM2E standard. While the company isn’t using any kind of flash name for the memory (ala Samsung’s Flashbolt), the idea is the same: releasing faster and higher density HBM2 memory for the next generation of high-end processors. Hynix’s HBM2E memory will reach up to 3.6 Gbps, which as things currently stand, will make it the fastest HBM2E memory on the market when it ships in 2020.
 
Not to be left behind, SK Hynix is now also readying their own HBM2E memory. In terms of performance, SK Hynix says that their memory will be able to clock at up to 3.6 Gbps/pin, which would give a full 1024-pin stack a total of 460GB/sec of memory bandwidth, and in the process the lead for HBM2E memory speeds. And for more advanced devices which employ multiple stacks (e.g. server GPUs), this means a 4-stack processor could be paired up with as much as 1.84TB/sec of memory bandwidth, a massive amount by any measure. Meanwhile their capacity is doubling, from 8 Gb/layer to 16 Gb/layer, allowing a full 8-Hi stack to reach a total of 16GB. It’s worth noting that the revised HBM2 standard actually allows for 12-Hi stacks, for a total of 24GB/stack, however we’ve yet to see anyone announce memory quite that dense.
 
 
so 1.8T/s is acheiveable with 4 chips with this ram
and the one above is over 2TB/s
460GB/s vs 512 GB/s above
 
samsung can do 640GB/s on one chip thats 2.5TB/s with 4 chips on a 4096 bus
 
https://www.anandtech.com/show/14733/sk-hynix-announces-36-gbps-hbm2e-memory-for-2020
 
 

SK Hynix Licenses DBI Ultra Interconnect for Next-Gen 3DS and HBM DRAM

 
https://www.anandtech.com/show/15498/sk-hynix-licenses-dbi-ultra-interconnect-for-nextgen-3ds-hbm-dram
 
 

atfrico
You hear that? 😳Its the crickets telling the hippies i told you so😹😹😹😼

Actually its me not caring enough to bother.  Ill just wait for the actual launch before coming back and pointing out who is lacking intelligence.

AMD’s Next-Gen RDNA2 Radeon RX “Navi 2x” GPUs To Feature 50% Better Performance Per Watt, HW-Accelerated Ray Tracing, VRS & More
 
 
https://wccftech.com/amd-radeon-rx-navi-2x-rdna2-gpus-2x-performance-ray-tracing-2020-launch/
 
 
Watch The AMD Financial Analyst Day 2020 Webcast Livestream Here – Zen 3, Zen 4, RDNA2, RDNA3, Next-Gen EPYC & More
 
https://wccftech.com/watch-the-amd-financial-analyst-day-2020-webcast-livestream-here-zen-3-rdna2-next-gen-epyc-more/
 
 

 
Sneak peak of card
 

Finally not a garbage blower card for the reference model?  Hopefully the thing GTXJack posted will no longer hold true.
 
Until now, the reference AMD cards have been terrible blower models with bad thermals, and AMD GPU buyers always had to wait months for a decent partner card to come out.  Maybe we will get some decent cooling at launch.  As always, wait for reviews.

I sure hope the sites claiming these cards are languishing until a launch at end of year are wrong.

Sure would be a shame of they drop a "Nvidia killer" that Nvidia already surpassed releasing ampere a few weeks earlier.

I feel another poor Volta moment coming.

GTXJackBauer
The only way they can be a true 'Nvidia Killer' is, if they beat the next gen within that % increase.  I would welcome that but until than, I'm going to keep throwing banana peels at the AMD fanboys parading in high heels and speedos.  That's how silly they sound.

Not as bad as the Intel/Nvidia boys defending these shady companies🙄😒

GTXJackBauer

How any fan, in any spot in my case sounds to me ;-)

AMD Radeon ‘Big Navi’ RX Gamma Flagship GPU Specs And Benchmarks Leaked – An Absolute Beast 
https://wccftech.com/amd-radeon-rx-gamma-gpu-big-navi-benchmarks-specs/
 

AMD Radeon RX Gamma XT GPU: A 26 TFLOPs Monster With 1.5 TB/s of Bandwidth And 21.2 Gigarays Ray Tracing Support

Overall, the card contains 80 compute units. Since the slide mentions 5120 stream processors, we can now lay to rest rumors about an increase CU to SP ratio. It is still 64. The clock speeds are divided into 3 categories just like small Navi and the highest clock speed goes all the way up to 2633 MHz. Based on this clock speed and assuming the same instructions as the last architecture, this card can deliver a resounding 26.6 TFLOPs of single-precision compute. This is a phenomenal jump over the last generation and driven not only by an increase in core count but by an increase in clock speeds as well.
 
The card will be based on a DisplayPort 2.0 and PCIe 5.0 and a board power of just 350W. A 512-bit bus will complement 32 GB of GDDR7 memory clocked at 24 Gbps. HDMI 8k is also supported for those that have compatible displays. The card is built using AMD's RDNA2 architecture and a 7nm process (likely TSMC's 7nm EUV).
 
---------------------------------
 
watch AMD fanbois think that this is the real deal
 
 
the telling point is GDDR7 and pcie 5.0 - let alone anything else - its fake -it's April 1st
 
 
good laugh for APRIL 1ST