In 2026, power isn’t just a buzzword it is quantified, compared, and recalculated in ways that can swing the final results dramatically. When you shift from raw totals to Calculating Power statistics, the gaps between teams and players often change from “close” to clearly separated. This post breaks down how those power numbers are computed and what the key statistics mean when you line them up side by side.
Key Takeaways
- AMD Ryzen Threadripper PRO 5995WX scores 100 GFLOPS peak single CPU FP64
- Frontier supercomputer holds the current TOP500 #1 at 1.194 exaFLOPS Rmax as of June 2023
- Frontier supercomputer efficiency is 52.72 gigaFLOPS/W Green500 #1 June 2023
- Moore's Law predicts doubling of transistors every 2 years, implying ~1.86x computing power
- The ENIAC computer, completed in 1945, had a peak performance of approximately 0.0000001 gigaFLOPS (100 kiloFLOPS)
Discover how calculating power impacts performance and helps you make smarter decisions faster.
Related reading
01 · Category
CPU and GPU Performance24 stats
01
AMD Ryzen Threadripper PRO 5995WX scores 100 GFLOPS peak single CPU FP64
02
Intel Core i9-13900K achieves 1.7 TFLOPS FP32 peak with AVX-512
03
NVIDIA H100 SXM GPU delivers 67 TFLOPS FP64 Tensor Core performance
04
AMD Instinct MI300X GPU reaches 163.4 TFLOPS FP16 peak
05
Apple M2 Ultra chip peaks at 31.6 TFLOPS FP32 GPU performance
06
Intel Xeon Platinum 8592+ offers 2.9 TFLOPS FP64 per socket peak
07
NVIDIA A100 80GB GPU achieves 19.5 TFLOPS FP64 with Tensor Cores
08
AMD EPYC 9754 (Genoa) peaks at 3.2 TFLOPS FP64 dual socket
09
Qualcomm Snapdragon 8 Gen 2 GPU at 3.2 TFLOPS FP32 peak for mobile
10
IBM Power10 processor delivers 5 TFLOPS FP64 per chip
11
NVIDIA RTX 4090 GPU reaches 82.6 TFLOPS FP16 peak shader
12
AMD Radeon RX 7900 XTX at 61 TFLOPS FP32 peak performance
13
Intel Arc A770 GPU delivers 17.2 TFLOPS FP16 peak
14
ARM Neoverse V2 core in AWS Graviton3 peaks at 0.4 TFLOPS FP32 per core
15
Google Tensor Processing Unit v4 (Trillium) achieves 2.7 petaFLOPS FP16 per pod
16
Cerebras Wafer-Scale Engine 2 (WSE-2) delivers 20 petaFLOPS FP16 AI performance
17
Graphcore IPU Colossus MK2 GC200 at 350 TOPS INT8 per chip
18
SambaNova SN40L chip reaches 2 petaFLOPS FP16 per card
19
Tenstorrent Grayskull at 114 TOPS INT8 peak performance
20
SiPearl Rhea CPU for HPC peaks at 1.7 TFLOPS FP64 per socket
21
Frontier's HPE Slingshot-11 interconnect enables 200 Gb/s per node for calculating power
22
NVIDIA DGX H100 system with 8 H100 GPUs reaches 32 petaFLOPS FP8 AI
23
AMD MI250X dual-GPU die at 47.9 TFLOPS FP64 peak
24
Intel Ponte Vecchio (Max 1550) GPU at 56 TFLOPS FP64 Tensor
Interpretation
CPU and GPU Performance Interpretation
This dizzying array of silicon bragging rights reveals a computational arms race where the only universal truth is that your laptop is now officially a glorified abacus compared to these number-crunching behemoths.
02 · Category
Current Supercomputers24 stats
01
Frontier supercomputer holds the current TOP500 #1 at 1.194 exaFLOPS Rmax as of June 2023
02
Aurora supercomputer ranks #2 at 1.012 exaFLOPS Rmax in June 2023 TOP500 list
03
Eagle supercomputer at 561.2 petaFLOPS Rmax, #3 on June 2023 TOP500
04
Fugaku at 442.0 petaFLOPS Rmax, #4 in June 2023
05
LUMI at 381.0 petaFLOPS Rmax, #5 June 2023 TOP500
06
Frontier's Rp peak is 1.707 exaFLOPS in June 2023
07
El Capitan projected at over 2 exaFLOPS, but current Leonardo at 233.3 petaFLOPS #6 June 2023
08
Alps supercomputer at 204.8 petaFLOPS #7 June 2023 TOP500
09
MareNostrum 5 at 175.5 petaFLOPS #9 June 2023
10
Frontier uses 37,888 AMD Instinct MI250X GPUs contributing to its exaFLOPS performance
11
Aurora employs Intel Xeon Max CPUs and Data Center GPU Max for 1.012 exaFLOPS
12
Summit at Oak Ridge has 27,648 NVIDIA V100 GPUs for 148.6 petaFLOPS Rmax current rank #13
13
Perlmutter at NERSC delivers 64.6 petaFLOPS Rmax with AMD GPUs, rank #20 June 2023
14
Frontier consumes 20.99 MW power for 1.194 exaFLOPS, efficiency 56.9 gigaFLOPS/W
15
Japan's ABCI-Q at 95.2 petaFLOPS Rmax for quantum simulation, rank #26 June 2023
16
China's OceanLite at 1.3 exaFLOPS AI performance but 125.4 petaFLOPS HPL #27
17
Microsoft Azure Eagle at 561 petaFLOPS but HPL 561.2 petaFLOPS #3
18
Nvidia-powered Isambard-AI at 132.0 petaFLOPS #18 June 2023 TOP500
19
HPC6 at 110.4 petaFLOPS Rmax #24 June 2023
20
AMD EPYC 7763 CPU in Frontier nodes contributes to overall calculating power
21
HPE Cray EX architecture in Frontier enables 9.2 million cores
22
Japan's Fugaku with A64FX processors at 442 petaFLOPS sustained
23
European LUMI uses AMD MI250X GPUs for 381 petaFLOPS
24
Selene at 63.5 petaFLOPS Rmax with NVIDIA A100 GPUs #33 June 2023
Interpretation
Current Supercomputers Interpretation
The global supercomputing race has now reached the exascale frontier, where power has become a public competition of precision engineering, national pride, and enormous electricity bills, all to make really, really difficult math look easy.
03 · Category
Energy Efficiency and Power Consumption22 stats
01
Frontier supercomputer efficiency is 52.72 gigaFLOPS/W Green500 #1 June 2023
02
Aurora at 49.03 gigaFLOPS/W #2 on Green500 June 2023
03
Eagle achieves 46.18 gigaFLOPS/W efficiency #3 Green500 June 2023
04
Alps at 40.60 gigaFLOPS/W #4 Green500, consumes less power per FLOP
05
LUMI supercomputer at 38.99 gigaFLOPS/W #5 Green500 June 2023
06
NVIDIA H100 GPU efficiency up to 1.98 TFLOPS/W FP64 Tensor
07
AMD MI300X at 81.7 TFLOPS/W FP16 efficiency claimed
08
Google TPU v5e at 2.5x better efficiency than v4, ~400 TFLOPS/W INT8
09
Cerebras CS-3 wafer-scale at 1200 TFLOPS/W sparsity FP16
10
Graphcore Bow IPU efficiency 500+ TOPS/W INT8
11
Frontier total power draw 21 MW for 1.194 exaFLOPS
12
Fugaku consumes 29.9 MW at 442 petaFLOPS, 14.8 gigaFLOPS/W
13
Summit power 10.1 MW for 148.6 petaFLOPS, 14.7 gigaFLOPS/W
14
Sunway TaihuLight used 15.37 MW for 93 petaFLOPS, 6.05 gigaFLOPS/W historical
15
NVIDIA A100 SXM4 400W TDP for 19.5 TFLOPS FP64, ~48.75 GFLOPS/W
16
AMD EPYC 9754 400W TDP dual socket ~8 GFLOPS/W FP64
17
Intel Xeon 8592+ 350W TDP ~8.3 GFLOPS/W FP64 per socket
18
Apple M1 Max 60W for 10.4 TFLOPS FP32, 173 GFLOPS/W GPU
19
Qualcomm Snapdragon 8 Gen 2 5nm process 4nm effective, ~640 GFLOPS/W mobile GPU
20
IBM Power10 at 20.6 gigaFLOPS/W in TOP500 systems
21
SiPearl Rhea 2.0 nm process target 50+ GFLOPS/W FP64
22
El Capitan projected 2 exaFLOPS at under 30 MW, ~66 gigaFLOPS/W target
Interpretation
Energy Efficiency and Power Consumption Interpretation
In the relentless pursuit of computational might, the supercomputing arena reveals a stark hierarchy of efficiency, where the Frontier system's crown for doing the most with each watt is being challenged by specialized accelerators claiming efficiency numbers so high they seem to belong to a different league entirely.
More related reading
04 · Category
Future Projections and Theoretical Limits23 stats
01
Moore's Law predicts doubling of transistors every 2 years, implying ~1.86x computing power
02
Exascale computing achieved 2022, zettascale targeted by 2030 at 10^21 FLOPS
03
Quantum supremacy demonstrated by Google Sycamore at 53 qubits, 200s vs classical 10k years
04
IBM roadmap to 100k+ logical qubits by 2033 for fault-tolerant quantum computing
05
Landauer limit theoretical minimum energy 2.8 kT ln2 per bit erasure ~3 zJ/op at room temp
06
Dennard scaling ended 2006, but 3D stacking to continue power efficiency gains
07
Optical computing could reach 10^15 FLOPS/W vs electronic 10^12
08
Neuromorphic computing like Intel Loihi 2 at 10^12 ops/W synaptic
09
Frontier to El Capitan 2x performance at same power by 2025
10
AMD roadmap MI400 series 5x AI performance over MI300 by 2026
11
NVIDIA Rubin platform R100 GPU 30x inference perf over Hopper by 2026
12
Intel 18A process 1.8nm for Xeon 6 by 2025, 20% perf/W gain
13
TSMC A16 1.6nm node 10% speed 15-20% power reduction 2026
14
Quantum annealers like D-Wave Advantage 5000+ qubits solve optimization 10^6x faster
15
Photonic chips Lightmatter Passage 36 petaFLOPS FP16 at 10 kW
16
Global supercomputing capacity to hit 10 exaFLOPS aggregate by 2025
17
AI training FLOPS doubling every 6 months, 10x every 2 years per OpenAI
18
Bekenstein bound limits info density 10^69 bits/m^3 black hole, theoretical compute limit
19
Reversible computing could approach Landauer limit, 10^42 ops/J theoretical
20
Planetary computing limit Bremermann's 10^50 FLOPS/kg matter
21
Margolus-Levitin theorem 6×10^33 ops/J energy-time limit per op
22
ExaEnergy project targets 60 gigaFLOPS/W sustainable by 2030
23
Post-Moore photonics-neuromorphic hybrid 1000x efficiency by 2040
Interpretation
Future Projections and Theoretical Limits Interpretation
We are in the breathtakingly clever phase where our computing ambitions have outpaced even our best metaphors, simultaneously chasing the ghostly potential of quantum supremacy and the sobering physical limits of thermodynamics, all while patching the fading legacy of Moore's Law with a dazzling quilt of quantum, photonic, and neuromorphic architectures.
05 · Category
Historical Milestones30 stats
01
The ENIAC computer, completed in 1945, had a peak performance of approximately 0.0000001 gigaFLOPS (100 kiloFLOPS)
02
The Manchester Mark 1, operational in 1949, performed about 1.2 kiloFLOPS in floating-point operations
03
The UNIVAC I, delivered in 1951, achieved around 0.000001 gigaFLOPS (1 kiloFLOPS) peak performance
04
The IBM 701, introduced in 1953, delivered approximately 0.000016 gigaFLOPS (16 kiloFLOPS)
05
The CDC 6600, launched in 1964, reached 3 megaFLOPS peak performance
06
The Cray-1 supercomputer, released in 1976, had a peak speed of 160 megaFLOPS
07
The Cray X-MP, introduced in 1982, achieved up to 940 megaFLOPS in multi-processor configuration
08
The Connection Machine CM-5, deployed in 1991, scaled to 1.056 teraFLOPS with 1024 processors
09
ASCI Red, completed in 1997, became the first teraFLOPS supercomputer at 1.338 teraFLOPS
10
ASCI White, operational in 2000, peaked at 7.226 teraFLOPS
11
Earth Simulator, launched in 2002, achieved 35.86 teraFLOPS on TOP500 list
12
Blue Gene/L, reached 280.6 teraFLOPS in 2006
13
Roadrunner supercomputer hit 1.026 petaFLOPS in 2008
14
Tianhe-1A achieved 2.566 petaFLOPS in 2010
15
Fujitsu K computer reached 10.51 petaFLOPS in 2011
16
Titan supercomputer delivered 17.59 petaFLOPS in 2013
17
Tianhe-2 peaked at 33.86 petaFLOPS in 2014
18
Sunway TaihuLight achieved 93.01 petaFLOPS in 2016
19
Summit supercomputer reached 122.3 petaFLOPS in 2018
20
IBM Power9-based Sierra hit 94.64 petaFLOPS in 2018
21
Frontier became the first exaFLOPS machine at 1.102 exaFLOPS in 2022
22
The first TOP500 list in June 1993 was topped by TMC CM-5/1024 at 59.7 gigaFLOPS
23
Intel Paragon XP/S 140 at 143.4 gigaFLOPS topped November 1993 list
24
Numerical Wind Tunnel at 170.0 gigaFLOPS in June 1994
25
Intel Paragon at 281.0 gigaFLOPS in November 1996
26
ASCI Red at 1.340 teraFLOPS in June 1997
27
ASCI Red sustained 1.064 teraFLOPS in November 1997
28
ASCI Red at 2.382 teraFLOPS in June 1999
29
ASCI Q reached 4.944 teraFLOPS projected in November 2001
30
Earth Simulator at 35.860 teraFLOPS in June 2002
Interpretation
Historical Milestones Interpretation
The breathtaking speed at which we've rocketed from needing an entire room to calculate a single artillery trajectory to casually simulating supernovas on a desktop proves that humanity's appetite for computational power is the ultimate exponential curve—our ambitions outrace our machines almost as soon as we build them.
Reference
Cite This Report
This report is designed to be cited. We maintain stable URLs and versioned verification dates. Copy the format appropriate for your publication below.
APA
Diana Reeves. (2026, February 13). Calculating Power Statistics. Gitnux. https://gitnux.org/calculating-power-statistics
MLA
Diana Reeves. "Calculating Power Statistics." Gitnux, 13 Feb 2026, https://gitnux.org/calculating-power-statistics.
Chicago
Diana Reeves. 2026. "Calculating Power Statistics." Gitnux. https://gitnux.org/calculating-power-statistics.
Sources & references
32 datasets cited across this report · attribution is report-level
aws.amazon.com+1 additional datasets cited (not shown individually)