Alleged Ampere GPU performance numbers appear on the 3DMARK database

Everyone’s waiting for the next generation of graphics cards with eager anticipation, be it “Big Navi” from AMD or Ampere from Nvidia. Now, results from unknown graphics cards are appearing on the 3DMARK database, revealing performance numbers that exceed that of Nvidia’s RTX 2080 Ti. 

Through the website HardwarLeaks, three GA102 GPU designs have been detailed; one of which features 24GB of VRAM, another with 12GB of VRAM while another has 10GB of VRAM, closely matching the specification of the already leaked (but unverified) RTX 30 series. 

An unknown Nvidia Ampere graphics card has been uncovered which features1935MHz clock speeds and a 3DMARK Time Spy graphics score of 18257. This score is said to be around 31% higher than an RTX 2080 Ti Founders Edition and 21.07% faster than an MSI RTX 2080 Ti Lightning Z. At this time, it is unknown if this result is of Nvidia’s rumoured RTX 3080 or 3090/3080 Ti graphics card.  

What this points towards is a notable generational leap for Nvidia’s next generation of graphics cards in rasterisation workloads, despite what appears to be modest core clock speeds of 1935MHz. Remember that this test result is of an early GPU model with unfinalised drivers and clock speeds, so performance for Nvidia’s next-generation graphics cards could be greater. 

Below are some recently leaked specifications for Nvidia’s RTX 30 series. These specifications are unverified. 

 

Â	RTX Titan (Ampere)	RTX 3090 /3080 Ti	RTX 3080	Titan RTX (Turing)	RTX 2080 Ti	RTX 2080 Super	RTX 2080
GPU	GA102-400	GA102-300	GA102-200	TU102-400	TU102-300	TU104-450	TU104-400
CUDA Cores	5376	5248	4352	4652	4352	3072	2944
Memory	24GB GDDR6X	12GB GDDR6X	10GB GDDR6X	24GB GDDR6	11GB GDDR6	8GB GDDR6	8GB GDDR6
Memory Speed	17Gbps	21Gbps	19Gbps	14Gbps	16Gbps	15.5 Gbps	14 Gbps
Memory Bus	384-bit	384-bit	320-bit	384-bit	352-bit	256-bit	256-bit
Memory Bandwidth	816GB/s	1008GB/s	760 GB/s	672 GB/s	616 GB/s	496 GB/s	448GB/s