AMD Next-Gen Flagship Navi ‘Radeon RX’ GPU Specifications Leaked
AMD’s next-gen Big Navi GPU teased with 24GB of HBM2e on a 4096-bit memory bus with an insane 2TB/sec bandwidth.
AMD is hard at work on its fresh Big Navi GPU, and now we have a fresh new rumor that is very exciting — as it teases some truly monster specs that could really shake up the camp at Team Green, at least until Ampere is here.
An unreleased GPU that’s allegedly based on AMD’s next-generation ‘Radeon RX’ Navi GPU architecture provides leaked out by Twitter user CyberPunkCat who promises have sources over at SK Hynix who offered him with a details sheet listing down the specifications for this new 2020 flagship design.
The new report is coming from ‘CyberPunkCat’ on Twitter that reportedly has sources at SK Hynix that have supplied him with a details sheet on the specs of Big Navi — including an insane 24GB of HBM2e memory that has an even-more-insane 2GB/sec+ of memory space bandwidth.
The brand new spec sheet teases Big Navi with a much larger 5120 shading units, 320 TMUs, 96 ROPs, 80 compute units and 12MB of L2 cache. But it’s the gigantic 24GB of new HBM2e storage on a 4096-bit memory bus spitting out a huge 2048GB/sec (2TB/sec) of memory space bandwidth.
AMD Radeon RX 5950 XT “Big Navi” Specs So Far
- Optimized 7nm+ node
- RDNA2 architecture
- 80 compute units
- 5120 stream processors
- 17.5 TFLOPs compute performance
- Hardware ray tracing support
- 24GB HBM2e
We know that AMD is preparing Big Navi behind the scenes and that they want their brand-new enthusiast graphics card to do to NVIDIA with 4K gaming what Ryzen did to Intel in the CPU market. In a recent press briefing with AMD’s Radeon Product Management Lead, Mithun Chandrasekhar, we found out some info on Big Navi.
We also know that NVIDIA has its new Ampere GPU architecture in the oven right now, and should be ready for its big debut at the GPU Technology Conference (GTC) being held in San Jose, between March 22-26. I will be there on the ground providing insurance coverage from GTC where we have to see Ampere-structured Quadro RTX graphics cards revealed.
NVIDIA will end up being launching its fresh GeForce RTX 3080 and GeForce RTX 3070 images cards later in the year, with my resources telling me the current coronavirus issues are going to halt the launch. Manufacturing, shipping, and everything in between is affected which means gamers will be waiting longer for new GPUs sadly.
Just to point out, SK Hynix does indeed possess 4 Gb/s HBM2e stacks which they showcased last week at ISSCC-2020 as stated in a reply to the leaker by Twitter consumer Hans De Vries.
— CyberPunkCat (@CyberCatPunk) February 24, 2020
Coming to the juicy bits, the specifications leaked in a document specifically sourced back to SK Hynix. We can’t say how much legitimate this document is but it looks pretty credible at first sight and the specs do seem plausible. The document mentions the specific GPU config with its codename ‘D32310/15’. There’s no way to tell whether this is Navi or something else but I will get to that in a bit. The first issue that arises is how do we know that is indeed an AMD GPU?
If you go by our recent record, an AMD GPU with a similar codename just passed RRA certification this month. Based on the evidence my colleague had collected, it sure does look like a new flagship Navi GPU and one that might end up being faster than all of AMD’s Radeon RX video gaming graphics cards released to time.
Surprisingly, the record lists the complete GPU and storage configurations. For the GPU itself, we are looking at 5120 stream processors packed in 80 Compute Devices. 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. That’s 2x the cache of NVIDIA’s flagship Tesla V100 GPU and Titan RTX. It’s also 3x the cache size over AMD’s very own Radeon RX 5700 XT which is based on the Navi 10 GPU. There are no clock speeds described for this part but the memory configuration is listed.
With regard to memory, we are considering 24 GB of high-performance HBM2e. The GPU would feature a 4096-bit bus interface and reportedly up to 2 Terabytes per second worthy of bandwidth. The new HBM2e standard was unveiled a while back with both SK Hynix and Samsung accelerating production this year since we are expecting the arrival of brand-new HPC parts from both NVIDIA and AMD. Interestingly, if this leak retains any truth to it, then it looks like AMD is going back again to HBM2 style on its flagship gaming graphics cards.
AMD Next-Gen Radeon RX ‘Big Navi’ GPU Alleged Specifications:
|Graphics Card||AMD ‘Big Navi’ Graphics Card||Radeon RX 5700 XT 50th Anniversary||Radeon RX 5700 XT||Radeon RX 5700||Radeon RX 5600 XT||Radeon RX 5500 XT|
|GPU Architecture||7nm+ Navi (RDNA 2nd Gen)||7nm Navi (RDNA 1st Gen)||7nm Navi (RDNA 1st Gen)||7nm Navi (RDNA 1st Gen)||7nm Navi (RDNA 1st Gen)||7nm Navi (RDNA 1st Gen)|
|Stream Processors||5120 SPs||2560 SPs||2560 SPs||2304 SPs||2304 SPs||1408 SPs|
|TMUs / ROPs||320 / 96||160 / 64||160 / 64||144 / 64||144 / 64||88 / 32|
|Base Clock||TBD||1680 MHz||1605 MHz||1465 MHz||1130 MHz||1670 MHz|
|Boost Clock||TBD||1980 MHz||1905 MHz||1725 MHz||1560 MHz||1845 MHz|
|Game Clock||TBD||1830 MHz||1755 MHz||1625 MHz||1375 MHz||1717 MHz|
|Compute Power||~18-20 TFLOPs||10.14 TFLOPs||9.75 TFLOPs||7.95 TFLOPs||7.19 TFLOPs||5.19 TFLOPs|
|VRAM||24 GB HBM2e||8 GB GDDR6||8 GB GDDR6||8 GB GDDR6||6 GB GDDR6||8 GB GDDR6|
|Bandwidth||2 TB/s||448 GB/s||448 GB/s||448 GB/s||288 GB/s||224 GB/s|
|Price||TBD||$449 US||$399 US||$349 US||$279 US||$169 US (4 GB)|
$199 US (8 GB)
|Launch||2020||7th July 2019||7th July 2019||7th July 2019||21st January, 2020||7th October 2019|
Recently, AMD made the switch to GDDR6 on its entire Navi portfolio but there have been no high-end images cards in AMD’s range since the Radeon VII which do feature HBM2 design. Bear in mind that all AMD flagships because the Radeon R9 Fury series have got featured HBM/2 memory. It won’t be a surprise either to discover flagship Navi do the same.
However, there are some things we have to consider here. While 24 GB 4096-bit is possible on 12-hi stacks, AMD revealed some design hindrances when going with 12-hi stacks back in August 2019 so it is possible that they have changed plans and are going for the limits with what HBM2e has to offer.