NVIDIA GeForce RTX 5090 PCB Leak Reveals Huge GB202 “Blackwell” GPU Die, Samsung GDDR7 Memory, and Extensive VRM Configuration

Recently, a detailed leak regarding NVIDIA’s GeForce RTX 5090 has emerged, showcasing its PCB fitted with the powerful Blackwell GB202 GPU and GDDR7 memory. This flagship device aims to redefine performance standards in the upcoming RTX 50 series.

Unveiling the NVIDIA GeForce RTX 5090: A Look at Its Impressive PCB Design

Fresh information surfaced yesterday, offering insights into the PCB design of the NVIDIA GeForce RTX 5090, set to become the cornerstone of the RTX 50 “Blackwell”gaming lineup. This newly leaked image reveals a congested yet sophisticated PCB, highlighted by a remarkable array of components.

Discovered on Baidu Forums, this latest leak presents the PCB brimming with crucial elements, specifically the GPU, VRAM, and VRMs. Notably, the GPU is identified as the GB202-300-A1, affirming details from earlier leaks. It’s essential to note that this chip represents a qualification sample nearing the final design rather than a retail product. Its substantial size is emphasized by an oversized integrated heat spreader (IHS), ensuring optimal thermal contact with the heatsink.

NVIDIA GeForce RTX 4090 PCB — PCB of the NVIDIA GeForce RTX 4090 GPU for reference. (Image Source: NVIDIA)

Around the GPU, the board hosts 16 Samsung GDDR7 memory modules, operating at estimated speeds of 28 Gbps. Notably, the RTX 5080 is slated to have the highest GDDR7 speeds in this lineup. The layout of these GDDR7 modules confirms prior leak predictions, set to function across a 512-bit bus interface, yielding an astounding total bandwidth of 1792 GB/s—significantly surpassing the RTX 4090’s 1008 GB/s.

Upon examining the VRM section, a total of 28 VRMs can be counted, accompanied by a substantial number of capacitors, indicating a robust power delivery system. Power will be supplied through a single 16-pin (12V-2×6) connector. The anticipated Total Board Power (TBP) for the RTX 5090 is up to 600W, catering to the enormous performance demands.

The NVIDIA GeForce RTX 5090 is projected for an official launch between late January and early February, while the RTX 5080 is expected to be available by mid-January. Anticipation builds as CES 2025 approaches, where further details about this next-generation lineup are likely to emerge.

NVIDIA GeForce RTX 50 Series Specifications (Preliminary)

Graphics Card Name	NVIDIA GeForce RTX 5090	NVIDIA GeForce RTX 5080	NVIDIA GeForce RTX 5070 Ti	NVIDIA GeForce RTX 5070	NVIDIA GeForce RTX 5060 Ti	NVIDIA GeForce RTX 5060
GPU Name	Blackwell GB202-300	Blackwell GB203-400	Blackwell GB203-300-A1	Blackwell GB205-300-A1	Blackwell GB206?	Blackwell GB206?
GPU SMs	170 (192 Full)	84 (84 Full)	70 (84 Full)	50 (50 SM Full)	TBD	TBD
GPU Cores	21,760 (+33%)	10,752 (+11%)	8,960 (+16%)	6,144	TBD	TBD
Clock Speeds	TBD	TBD	TBD	TBD	TBD	TBD
L2 Cache	TBD	TBD	TBD	TBD	TBD	TBD
Memory Capacity	32 GB GDDR7 (+33%)	16 GB GDDR7 (0%)	16 GB GDDR7	12 GB GDDR7	16 GB GDDR7	8 GB GDDR7
Memory Bus	512-bit (+33%)	256-bit (0%)	256-bit (0%)	192-bit (0%)	128-bit (0%)	128-bit (0%)
Memory Speed	28 Gbps	30 Gbps	28 Gbps	28 Gbps	28 Gbps?	28 Gbps?
Bandwidth	1792 GB/s	1024 GB/s	896 GB/s	672 GB/s	448 GB/s	448 GB/s
TBP	600W (+33%)	400W (+25%)	300W (+10.5%)	250W (+14%)	TBD	TBD
Power Interface	1 12V-2×6 (16-Pin)	1 12V-2×6 (16-Pin)	1 12V-2×6 (16-Pin)	1 12VHPWR (16-Pin)	1 12VHPWR (16-Pin)	1 12VHPWR (16-Pin)
Launch	Q1 2025	Q1 2025	Q1 2025	Q1 2025	TBD	TBD