Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Computer performance by orders of magnitude
open-in-new
<h2 id="milliscale-computing-103">Milliscale computing (10−3)</h2> <ul><li>2×10−3: average human multiplication of two 10-digit numbers using pen and paper without aids<a class="footnote-ref" id="fnref:1" href="#fn:1"><sup>1</sup></a></li></ul> <h2 id="deciscale-computing-101">Deciscale computing (10−1)</h2> <ul><li>1×10−1: multiplication of two 10-digit numbers by a 1940s electromechanical desk calculator<a class="footnote-ref" id="fnref:2" href="#fn:2"><sup>2</sup></a></li> <li>3×10−1: multiplication on <a href="/facts/Z3_(computer)/JNPXGqDp">Zuse Z3</a> and <a href="/facts/Z4_(computer)/RNjK5G7f">Z4</a>, first programmable <a href="/facts/Digital_computer/SQe1aK05">digital computers</a>, 1941 and 1945 respectively</li> <li>5×10−1: computing power of the average human mental calculation for multiplication using pen and paper</li></ul> <h2 id="scale-computing-100">Scale computing (100)</h2> <ul><li>1.2 OP/S: addition on Z3, 1941, and multiplication on Bell <a href="/facts/Model_V/IfWE4EaQ">Model V</a>, 1946</li> <li>2.4 OP/S: addition on Z4, 1945</li></ul> <h2 id="decascale-computing-101">Decascale computing (101)</h2> <ul><li>1.8×101: <a href="/facts/ENIAC/EcfmM1Y0">ENIAC</a>, first programmable electronic digital computer, 1945<a class="footnote-ref" id="fnref:3" href="#fn:3"><sup>3</sup></a></li> <li>5×101: upper end of serialized human perception computation (light bulbs do not flicker to the human observer)</li> <li>7×101: <a href="/facts/Whirlwind_I/IbG8pxYt">Whirlwind I</a> 1951 vacuum tube computer and <a href="/facts/IBM_1620/yRqdvqt0">IBM 1620</a> 1959 transistorized scientific <a href="/facts/Minicomputer/g8hvpQxe">minicomputer</a><a class="footnote-ref" id="fnref:4" href="#fn:4"><sup>4</sup></a></li></ul> <h2 id="hectoscale-computing-102">Hectoscale computing (102)</h2> <ul><li>1.3×102: <a href="/facts/PDP-4/oP3eBKVA">PDP-4</a> commercial minicomputer, 1962<a class="footnote-ref" id="fnref:5" href="#fn:5"><sup>5</sup></a></li> <li>2×102: <a href="/facts/IBM_602/eDBwoqPb">IBM 602</a> electromechanical calculator (then called computer), 1946</li> <li>6×102: <a href="/facts/Manchester_Mark_1/S8wpt6U5">Manchester Mark 1</a> electronic general-purpose stored-program digital computer, 1949<a class="footnote-ref" id="fnref:6" href="#fn:6"><sup>6</sup></a></li></ul> <h2 id="kiloscale-computing-103">Kiloscale computing (103)</h2> <ul><li>2×103: <a href="/facts/UNIVAC_I/LQzozqY2">UNIVAC I</a>, first American commercially available electronic general-purpose stored program digital computer, 1951<a class="footnote-ref" id="fnref:7" href="#fn:7"><sup>7</sup></a></li> <li>3×103: <a href="/facts/PDP-1/PSeIBDJU">PDP-1</a> commercial minicomputer, 1959<a class="footnote-ref" id="fnref:8" href="#fn:8"><sup>8</sup></a></li> <li>15×103: <a href="/facts/IBM_Naval_Ordnance_Research_Calculator/XspPRv68">IBM Naval Ordnance Research Calculator</a>, 1954</li> <li>24×103: <a href="/facts/AN%2fFSQ-7_Combat_Direction_Central/5kS3IRb0">AN/FSQ-7 Combat Direction Central</a>, 1957<a class="footnote-ref" id="fnref:9" href="#fn:9"><sup>9</sup></a></li> <li>30×103: <a href="/facts/IBM_1130/zQ9Acd05">IBM 1130</a> commercial minicomputer, 1965<a class="footnote-ref" id="fnref:10" href="#fn:10"><sup>10</sup></a></li> <li>40×103: multiplication on <a href="/facts/Hewlett-Packard_9100A/Tvm11Jy1">Hewlett-Packard 9100A</a> early desktop electronic calculator, 1968</li> <li>53×103: <a href="/facts/Lincoln_TX-2/Yvk3A10o">Lincoln TX-2</a> transistor-based computer, 1958<a class="footnote-ref" id="fnref:11" href="#fn:11"><sup>11</sup></a></li> <li>92×103: <a href="/facts/Intel_4004/kAoetjSU">Intel 4004</a>, first commercially available full function <a href="/facts/Central_processing_unit/Jxp1bqMV">CPU</a> on a chip, released in 1971</li> <li>500×103: <a href="/facts/Colossus_computer/pbkHLPwp">Colossus computer</a> <a href="/facts/Vacuum_tube/VHd1BNLq">vacuum tube</a> cryptanalytic supercomputer, 1943</li></ul> <h2 id="megascale-computing-106">Megascale computing (106)</h2> <ul><li>1×106: computing power of the <a href="/facts/Motorola_68000/3p2wt2D6">Motorola 68000</a> commercial computer introduced in 1979.</li> <li>1.2×106: <a href="/facts/IBM_7030_Stretch/V8SdUp00">IBM 7030 "Stretch"</a> transistorized supercomputer, 1961</li> <li>5×106: <a href="/facts/CDC_6600/QpnT1Ltl">CDC 6600</a>, first commercially successful supercomputer, 1964<a class="footnote-ref" id="fnref:12" href="#fn:12"><sup>12</sup></a></li> <li>11×106: <a href="/facts/I386/ag5eSgWv">Intel i386</a> microprocessor at 33 MHz, 1985</li> <li>14×106: <a href="/facts/CDC_7600/Rgu6spkG">CDC 7600</a> supercomputer, 1967<a class="footnote-ref" id="fnref:13" href="#fn:13"><sup>13</sup></a></li> <li>40×106: <a href="/facts/I486/2zfEAXAw">i486</a> microprocessor at 50 MHz, 1989</li> <li>86×106: <a href="/facts/Cray-1/4afuzp0D">Cray 1</a> supercomputer, 1978<a class="footnote-ref" id="fnref:14" href="#fn:14"><sup>14</sup></a></li> <li>100×106: <a href="/facts/Pentium_(original)/dqChAmxz">Pentium (i586)</a> microprocessor, 1993</li> <li>400×106: <a href="/facts/Cray_X-MP/BmPqwbta">Cray X-MP</a>, 1982<a class="footnote-ref" id="fnref:15" href="#fn:15"><sup>15</sup></a></li></ul> <h2 id="gigascale-computing-109">Gigascale computing (109)</h2> <ul><li>1×109: <a href="/facts/ILLIAC_IV/CX8GQQgI">ILLIAC IV</a> 1972 supercomputer does first <a href="/facts/Computational_fluid_dynamics/zuWdtNJp">computational fluid dynamics</a> problems</li> <li>1.4×109: <a href="/facts/Intel_Pentium_III/sV1v2Frq">Intel Pentium III</a> microprocessor, 1999</li> <li>1.6×109: <a href="/facts/PowerVR/lTtwow8j">PowerVR</a> <a href="/facts/PowerVR/lTtwow8j">MBX</a> Lite 3D <a href="/facts/Graphics_processing_unit/PTK1RQVp">GPU</a> on <a href="/facts/IPhone_(1st_generation)/xwVatAHK">iPhone 1</a>, 2007</li> <li>8×109: PowerVR SGX535 GPU on <a href="/facts/IPad_(1st_generation)/FCH0AXW9">iPad 1</a>, 2010</li> <li>136×109: PowerVR GXA6450 GPU on <a href="/facts/IPhone_6/vB2CIFx9">iPhone 6</a> and <a href="/facts/IPhone_SE_(1st_generation)/wIN7r90A">iPhone SE</a>, 2014</li> <li>148×109: <a href="/facts/Gulftown/qq85u06f">Intel Core i7-980X Extreme Edition</a> commercial computing 2010<a class="footnote-ref" id="fnref:16" href="#fn:16"><sup>16</sup></a></li></ul> <h2 id="terascale-computing-1012">Terascale computing (1012)</h2> <ul><li>1.34×1012: <a href="/facts/Intel_ASCI_Red/4vP9Hmko">Intel ASCI Red</a> 1997 supercomputer</li> <li>1.344×1012 <a href="/facts/GeForce_400_series/Bf82VpYV">GeForce GTX 480</a> in 2010 from Nvidia at its peak performance</li> <li>2.15×1012: <a href="/facts/IPhone_15_Pro/q3JtyMPX">iPhone 15 Pro</a> September 2023 <a href="/facts/A17_Pro/tf5mmo8D">A17 Pro</a> processor</li> <li>4.64×1012: <a href="/facts/Radeon_HD_5000_series/RvyDSDdz">Radeon HD 5970</a> in 2009 from <a href="/facts/Advanced_Micro_Devices/CB1G7QTb">AMD</a> (under ATI branding) at its peak performance</li> <li>5.152×1012: <a href="/facts/Nvidia_Tesla/SsHo1w98">S2050/S2070 1U GPU Computing System</a> from Nvidia</li> <li>11.3×1012: <a href="/facts/GeForce_10_series/rEQwmDqC">GeForce GTX 1080 Ti</a> in 2017</li> <li>13.7×1012: <a href="/facts/AMD_RX_Vega_series/j13wIHR5">Radeon RX Vega 64</a> in 2017</li> <li>15.0×1012: <a href="/facts/Volta_(microarchitecture)/bxD4hpFL">Nvidia Titan V</a> in 2017</li> <li>80×1012: <a href="/facts/Watson_(computer)/xub4UvVx">IBM Watson</a><a class="footnote-ref" id="fnref:17" href="#fn:17"><sup>17</sup></a></li> <li>170×1012: <a href="/facts/Nvidia_DGX-1/IernkzmQ">Nvidia DGX-1</a> The initial Pascal based DGX-1 delivered 170 teraflops of half precision processing.<a class="footnote-ref" id="fnref:18" href="#fn:18"><sup>18</sup></a></li> <li>478.2×1012 IBM <a href="/facts/BlueGene%2fL/y4EvbyS4">BlueGene/L</a> 2007 Supercomputer</li> <li>960×1012 <a href="/facts/Nvidia_DGX-1/IernkzmQ">Nvidia DGX-1</a> The Volta-based upgrade increased calculation power of <a href="/facts/Nvidia_DGX-1/IernkzmQ">Nvidia DGX-1</a> to 960 <a href="/facts/Teraflop/6zYlH68F">teraflops</a>.<a class="footnote-ref" id="fnref:19" href="#fn:19"><sup>19</sup></a></li></ul> <h2 id="petascale-computing-1015">Petascale computing (1015)</h2> <p class="note">Main article: <a href="/facts/Petascale_computing/HbIWaQ0L">Petascale computing</a></p> <ul><li>1.026×1015: <a href="/facts/IBM_Roadrunner/tXDaFkQM">IBM Roadrunner</a> 2009 Supercomputer</li> <li>1.32×1015: <a href="/facts/GeForce_40_series/ye41auEe">Nvidia GeForce 40 series</a>' RTX 4090 consumer graphics card achieves 1.32 petaflops in AI applications, October 2022<a class="footnote-ref" id="fnref:20" href="#fn:20"><sup>20</sup></a></li> <li>2×1015: <a href="/facts/Nvidia_DGX-2/IernkzmQ">Nvidia DGX-2</a> a 2 Petaflop Machine Learning system (the newer <a href="/facts/Nvidia_DGX/IernkzmQ">DGX A100</a> has 5 Petaflop performance)</li> <li>11.5×1015: <a href="/facts/Google/GT9Sugza">Google</a> <a href="/facts/Tensor_processing_unit/aVlg0LFF">TPU</a> pod containing 64 second-generation TPUs, May 2017<a class="footnote-ref" id="fnref:21" href="#fn:21"><sup>21</sup></a></li> <li>17.17×1015: <a href="/facts/IBM_Sequoia/ToKXt78K">IBM Sequoia</a>'s LINPACK performance, June 2013<a class="footnote-ref" id="fnref:22" href="#fn:22"><sup>22</sup></a></li> <li>20×1015: roughly the hardware-equivalent of the human brain according to <a href="/facts/Ray_Kurzweil/1DYRFcVe">Ray Kurzweil</a>. Published in his 1999 book: The Age of Spiritual Machines: When Computers Exceed Human Intelligence<a class="footnote-ref" id="fnref:23" href="#fn:23"><sup>23</sup></a></li> <li>33.86×1015: <a href="/facts/Tianhe-2/SwTCG6Hp">Tianhe-2</a>'s LINPACK performance, June 2013<a class="footnote-ref" id="fnref:24" href="#fn:24"><sup>24</sup></a></li> <li>36.8×1015: 2001 estimate of computational power required to <i>simulate</i> a human brain in real time.<a class="footnote-ref" id="fnref:25" href="#fn:25"><sup>25</sup></a></li> <li>93.01×1015: <a href="/facts/Sunway_TaihuLight/7TLEatpK">Sunway TaihuLight</a>'s LINPACK performance, June 2016<a class="footnote-ref" id="fnref:26" href="#fn:26"><sup>26</sup></a></li> <li>143.5×1015: <a href="/facts/Summit_(supercomputer)/BmKWTpEk">Summit's</a> LINPACK performance, November 2018<a class="footnote-ref" id="fnref:27" href="#fn:27"><sup>27</sup></a></li></ul> <h2 id="exascale-computing-1018">Exascale computing (1018)</h2> <p class="note">Main article: <a href="/facts/Exascale_computing/HEb6YV2v">Exascale computing</a></p> <ul><li>1×1018: <a href="/facts/Fugaku_(supercomputer)/jmuyc2sW">Fugaku</a> 2020 Japanese supercomputer in single precision mode<a class="footnote-ref" id="fnref:28" href="#fn:28"><sup>28</sup></a></li> <li>1.1x1018: <a href="/facts/Frontier_(supercomputer)/NTUlSjry">Frontier</a> 2022 U.S. supercomputer</li> <li>1.72×1018: operations per second of <a href="/facts/El_Capitan_(supercomputer)/hQTdofIV">El Capitan</a>, the fastest <a href="/facts/Distributed_computing/o5nyhqTF">non-distributed</a> supercomputer in the world as of November 2024<a class="footnote-ref" id="fnref:29" href="#fn:29"><sup>29</sup></a></li> <li>1.88×1018: U.S. Summit achieves a peak throughput of this many operations per second, whilst analysing genomic data using a mixture of numerical precisions.<a class="footnote-ref" id="fnref:30" href="#fn:30"><sup>30</sup></a></li> <li>2.43×1018: <a href="/facts/Folding%40home/Kuuo4ISC">Folding@home</a> distributed computing system during <a href="/facts/COVID-19_pandemic/0xTsh7Ut">COVID-19 pandemic</a> response<a class="footnote-ref" id="fnref:31" href="#fn:31"><sup>31</sup></a></li></ul> <h2 id="zettascale-computing-1021">Zettascale computing (1021)</h2> <p class="note">Main article: <a href="/facts/Zettascale_computing/qo2IBfvB">Zettascale computing</a></p> <ul><li>1×1021: Accurate global weather estimation on the scale of approximately 2 weeks.<a class="footnote-ref" id="fnref:32" href="#fn:32"><sup>32</sup></a> Assuming <a href="/facts/Moore%2527s_law/90CSekhl">Moore's law</a> remains applicable, such systems may be feasible around 2035.<a class="footnote-ref" id="fnref:33" href="#fn:33"><sup>33</sup></a></li></ul> <p>A zettascale computer system could generate more single floating point data in one second than was stored by any digital means on Earth in the first quarter of 2011. </p> <h2 id="beyond-zettascale-computing-1021">Beyond zettascale computing (>1021)</h2> <ul><li>1.12×1036: Estimated computational power of a <a href="/facts/Matrioshka_brain/z7JTxY3m">Matrioshka brain</a>, assuming 1.87×1026 <a href="/facts/Watt/VK8oqRrm">watt</a> power produced by solar panels and 6 <a href="/facts/Performance_per_watt/rnXFPlME">GFLOPS/watt</a> efficiency.<a class="footnote-ref" id="fnref:34" href="#fn:34"><sup>34</sup></a></li> <li>4×1048: Estimated computational power of a Matrioshka brain whose power source is the <a href="/facts/Sun/CadPCVHh">Sun</a>, the outermost layer operates at 10 <a href="/facts/Kelvin/E3trzC4C">kelvins</a>, and the constituent parts operate at or near the <a href="/facts/Landauer_limit/G6Y0m3FX">Landauer limit</a> and draws power at the efficiency of a <a href="/facts/Carnot_engine/UgsqwvQI">Carnot engine</a></li> <li>5×1058: Estimated power of a <a href="/facts/Galaxy/zCyUvEf7">galaxy</a> equivalent in luminosity to the <a href="/facts/Milky_Way/ulF6lccN">Milky Way</a> converted into Matrioshka brains.</li></ul> <h2 id="see-also">See also</h2> <ul><li><a href="/facts/Futures_studies/h59ennpW">Futures studies</a> – study of possible, probable, and preferable futures, including making projections of future technological advances</li> <li><a href="/facts/History_of_computing_hardware_(1960s%25E2%2580%2593present)/Eq0nC7aD">History of computing hardware (1960s–present)</a></li> <li><a href="/facts/List_of_emerging_technologies/00drktJ6">List of emerging technologies</a> – new fields of technology, typically on the cutting edge. Examples include genetics, robotics, and nanotechnology (GNR) <ul><li><a href="/facts/Artificial_intelligence/lJGauQwX">Artificial intelligence</a> – computer mental abilities, especially those that previously belonged only to humans, such as <a href="/facts/Speech_recognition/z7S7Pgk6">speech recognition</a>, <a href="/facts/Natural_language_generation/qudtLU6Y">natural language generation</a>, etc. <ul><li><a href="/facts/History_of_artificial_intelligence/1wNLDHbB">History of artificial intelligence</a> (AI)</li> <li><a href="/facts/Artificial_general_intelligence/vbxwaq8t">Strong AI</a> – hypothetical AI as smart as a human</li></ul></li> <li><a href="/facts/Quantum_computing/nbbcgmvq">Quantum computing</a> <ul><li><a href="/facts/Timeline_of_quantum_computing_and_communication/s1pp1CP2">Timeline of quantum computing and communication</a></li></ul></li></ul></li> <li><a href="/facts/Moore%2527s_law/90CSekhl">Moore's law</a> – observation (not actually a <a href="/facts/Scientific_law/aVtc4wpk">law</a>) that, over the <a href="/facts/History_of_computing_hardware/83xXSmuW">history of computing hardware</a>, the number of <a href="/facts/Transistor/Bmr8zwL0">transistors</a> on <a href="/facts/Integrated_circuit/fjn6vg7C">integrated circuits</a> doubles approximately every two years. The law is named after Intel co-founder <a href="/facts/Gordon_Moore/AMW52edX">Gordon Moore</a>, who described the trend in his 1965 paper.<a class="footnote-ref" id="fnref:35" href="#fn:35"><sup>35</sup></a></li> <li><a href="/facts/Supercomputer/HT9NwGAN">Supercomputer</a> <ul><li><a href="/facts/History_of_supercomputing/6Sgp4vCJ">History of supercomputing</a></li></ul></li> <li><a href="/facts/Superintelligence/VL1NaEPS">Superintelligence</a></li> <li><a href="/facts/Timeline_of_computing/bEcQv72r">Timeline of computing</a></li> <li><a href="/facts/Technological_singularity/xcg4M6AF">Technological singularity</a> – hypothetical point in the future when computer capacity rivals that of a human brain, enabling the development of <a href="/facts/Artificial_general_intelligence/vbxwaq8t">strong AI</a> — artificial intelligence at least as smart as a human <ul><li><i><a href="/facts/The_Singularity_Is_Near/50wmkNHS">The Singularity Is Near</a></i> – book by <a href="/facts/Raymond_Kurzweil/1DYRFcVe">Raymond Kurzweil</a> dealing with the progression and projections of development of computer capabilities, including beyond human levels of performance</li></ul></li> <li><a href="/facts/TOP500/rz8zbyss">TOP500</a> – list of the 500 most powerful (non-distributed) computer systems in the world</li></ul> <h2 id="external-links">External links</h2> <ul><li><a href="http://www.futuretimeline.net/subject/computers-internet.htm#supercomputer-future-trends">Historical and projected growth in supercomputer capacity</a></li></ul> <h2 id="references">References</h2> <ol> <li id="fn:1"><p>Neumann, John Von; Brody, F.; Vamos, Tibor (1995). The Neumann Compendium. World Scientific. ISBN 978-981-02-2201-7. <a href="978-981-02-2201-7" target="_blank">978-981-02-2201-7</a> <a href="#fnref:1" class="footnote-back-ref">↩</a></p></li> <li id="fn:2"><p>Neumann, John Von; Brody, F.; Vamos, Tibor (1995). The Neumann Compendium. World Scientific. ISBN 978-981-02-2201-7. <a href="978-981-02-2201-7" target="_blank">978-981-02-2201-7</a> <a href="#fnref:2" class="footnote-back-ref">↩</a></p></li> <li id="fn:3"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:3" class="footnote-back-ref">↩</a></p></li> <li id="fn:4"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:4" class="footnote-back-ref">↩</a></p></li> <li id="fn:5"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:5" class="footnote-back-ref">↩</a></p></li> <li id="fn:6"><p>Copeland, B. Jack (2012-05-24). Alan Turing's Electronic Brain: The Struggle to Build the ACE, the World's Fastest Computer. OUP Oxford. ISBN 978-0-19-960915-4. <a href="978-0-19-960915-4" target="_blank">978-0-19-960915-4</a> <a href="#fnref:6" class="footnote-back-ref">↩</a></p></li> <li id="fn:7"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:7" class="footnote-back-ref">↩</a></p></li> <li id="fn:8"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:8" class="footnote-back-ref">↩</a></p></li> <li id="fn:9"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:9" class="footnote-back-ref">↩</a></p></li> <li id="fn:10"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:10" class="footnote-back-ref">↩</a></p></li> <li id="fn:11"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:11" class="footnote-back-ref">↩</a></p></li> <li id="fn:12"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:12" class="footnote-back-ref">↩</a></p></li> <li id="fn:13"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:13" class="footnote-back-ref">↩</a></p></li> <li id="fn:14"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:14" class="footnote-back-ref">↩</a></p></li> <li id="fn:15"><p>"Cost of CPU Performance Through Time 1944-2003". www.jcmit.net. Retrieved 2024-01-15. <a href="https://www.jcmit.net/cpu-performance.htm" target="_blank">https://www.jcmit.net/cpu-performance.htm</a> <a href="#fnref:15" class="footnote-back-ref">↩</a></p></li> <li id="fn:16"><p>"Intel 980x Gulftown | Synthetic Benchmarks | CPU & Mainboard | OC3D Review". www.overclock3d.net. March 12, 2010. <a href="https://www.overclock3d.net/reviews/cpu_mainboard/intel_980x_gulftown/4" target="_blank">https://www.overclock3d.net/reviews/cpu_mainboard/intel_980x_gulftown/4</a> <a href="#fnref:16" class="footnote-back-ref">↩</a></p></li> <li id="fn:17"><p>Tony Pearson, IBM Watson - How to build your own "Watson Jr." in your basement, Inside System Storage <a href="https://www.ibm.com/developerworks/mydeveloperworks/blogs/InsideSystemStorage/entry/ibm_watson_how_to_build_your_own_watson_jr_in_your_basement7?lang=en" target="_blank">https://www.ibm.com/developerworks/mydeveloperworks/blogs/InsideSystemStorage/entry/ibm_watson_how_to_build_your_own_watson_jr_in_your_basement7?lang=en</a> <a href="#fnref:17" class="footnote-back-ref">↩</a></p></li> <li id="fn:18"><p>"DGX-1 deep learning system" (PDF). NVIDIA DGX-1 Delivers 75X Faster Training...Note: Caffe benchmark with AlexNet, training 1.28M images with 90 epochs <a href="http://images.nvidia.com/content/technologies/deep-learning/pdf/Datasheet-DGX1.pdf" target="_blank">http://images.nvidia.com/content/technologies/deep-learning/pdf/Datasheet-DGX1.pdf</a> <a href="#fnref:18" class="footnote-back-ref">↩</a></p></li> <li id="fn:19"><p>"DGX Server". DGX Server. Nvidia. Retrieved 7 September 2017. <a href="https://www.nvidia.com/en-us/data-center/dgx-server/" target="_blank">https://www.nvidia.com/en-us/data-center/dgx-server/</a> <a href="#fnref:19" class="footnote-back-ref">↩</a></p></li> <li id="fn:20"><p>"NVIDIA GeForce-News". 12 October 2022. <a href="https://www.nvidia.com/de-de/geforce/news/rtx-40-series-graphics-cards-announcements" target="_blank">https://www.nvidia.com/de-de/geforce/news/rtx-40-series-graphics-cards-announcements</a> <a href="#fnref:20" class="footnote-back-ref">↩</a></p></li> <li id="fn:21"><p>"Build and train machine learning models on our new Google Cloud TPUs". 17 May 2017. <a href="https://blog.google/topics/google-cloud/google-cloud-offer-tpus-machine-learning/" target="_blank">https://blog.google/topics/google-cloud/google-cloud-offer-tpus-machine-learning/</a> <a href="#fnref:21" class="footnote-back-ref">↩</a></p></li> <li id="fn:22"><p>"Top500 List - June 2013 | TOP500 Supercomputer Sites". top500.org. Archived from the original on 2013-06-22. <a href="https://web.archive.org/web/20130622060404/http://www.top500.org/list/2013/06/" target="_blank">https://web.archive.org/web/20130622060404/http://www.top500.org/list/2013/06/</a> <a href="#fnref:22" class="footnote-back-ref">↩</a></p></li> <li id="fn:23"><p>Kurzweil, Ray (1999). The Age of Spiritual Machines: When Computers Exceed Human Intelligence. New York, NY: Penguin. ISBN 9780140282023. <a href="9780140282023" target="_blank">9780140282023</a> <a href="#fnref:23" class="footnote-back-ref">↩</a></p></li> <li id="fn:24"><p>"Top500 List - June 2013 | TOP500 Supercomputer Sites". top500.org. Archived from the original on 2013-06-22. <a href="https://web.archive.org/web/20130622060404/http://www.top500.org/list/2013/06/" target="_blank">https://web.archive.org/web/20130622060404/http://www.top500.org/list/2013/06/</a> <a href="#fnref:24" class="footnote-back-ref">↩</a></p></li> <li id="fn:25"><p>"Brain on a Chip". 30 November 2001. <a href="http://hplusmagazine.com/2009/04/07/brain-chip/" target="_blank">http://hplusmagazine.com/2009/04/07/brain-chip/</a> <a href="#fnref:25" class="footnote-back-ref">↩</a></p></li> <li id="fn:26"><p>http://top500.org/list/2016/06/ Top500 list, June 2016 <a href="http://top500.org/list/2016/06/" target="_blank">http://top500.org/list/2016/06/</a> <a href="#fnref:26" class="footnote-back-ref">↩</a></p></li> <li id="fn:27"><p>"November 2018 | TOP500 Supercomputer Sites". www.top500.org. Retrieved 2018-11-30. <a href="https://www.top500.org/lists/2018/11/" target="_blank">https://www.top500.org/lists/2018/11/</a> <a href="#fnref:27" class="footnote-back-ref">↩</a></p></li> <li id="fn:28"><p>"June 2020 | TOP500". <a href="https://www.top500.org/lists/top500/2020/06/" target="_blank">https://www.top500.org/lists/top500/2020/06/</a> <a href="#fnref:28" class="footnote-back-ref">↩</a></p></li> <li id="fn:29"><p>Eadline, Doug (2024-11-19). "An Inside Look at El Capitan: Facts Beyond the Numbers". HPCwire. Retrieved 2024-12-01. <a href="https://www.hpcwire.com/2024/11/19/an-inside-look-at-el-capitan-facts-beyond-the-numbers/" target="_blank">https://www.hpcwire.com/2024/11/19/an-inside-look-at-el-capitan-facts-beyond-the-numbers/</a> <a href="#fnref:29" class="footnote-back-ref">↩</a></p></li> <li id="fn:30"><p>"Genomics Code Exceeds Exaops on Summit Supercomputer". Oak Ridge Leadership Computing Facility. Retrieved 2018-11-30. <a href="https://www.olcf.ornl.gov/2018/06/08/genomics-code-exceeds-exaops-on-summit-supercomputer/" target="_blank">https://www.olcf.ornl.gov/2018/06/08/genomics-code-exceeds-exaops-on-summit-supercomputer/</a> <a href="#fnref:30" class="footnote-back-ref">↩</a></p></li> <li id="fn:31"><p>Pande lab. "Client Statistics by OS". Archive.is. Archived from the original on 2020-04-12. Retrieved 2020-04-12. <a href="https://archive.today/20200412111010/https://stats.foldingathome.org/os" target="_blank">https://archive.today/20200412111010/https://stats.foldingathome.org/os</a> <a href="#fnref:31" class="footnote-back-ref">↩</a></p></li> <li id="fn:32"><p>DeBenedictis, Erik P. (2005). "Reversible logic for supercomputing". Proceedings of the 2nd conference on Computing frontiers. ACM Press. pp. 391–402. ISBN 1-59593-019-1. <a href="1-59593-019-1" target="_blank">1-59593-019-1</a> <a href="#fnref:32" class="footnote-back-ref">↩</a></p></li> <li id="fn:33"><p>"Zettascale by 2035? China Thinks So". 6 December 2018. <a href="https://www.hpcwire.com/2018/12/06/zettascale-by-2035/" target="_blank">https://www.hpcwire.com/2018/12/06/zettascale-by-2035/</a> <a href="#fnref:33" class="footnote-back-ref">↩</a></p></li> <li id="fn:34"><p>Jacob Eddison; Joe Marsden; Guy Levin; Darshan Vigneswara (2017-12-12), "Matrioshka Brain", Journal of Physics Special Topics, 16 (1), Department of Physics and Astronomy, University of Leicester <a href="https://journals.le.ac.uk/ojs1/index.php/pst/article/view/2559" target="_blank">https://journals.le.ac.uk/ojs1/index.php/pst/article/view/2559</a> <a href="#fnref:34" class="footnote-back-ref">↩</a></p></li> <li id="fn:35"><p>Moore, Gordon E. (1965). "Cramming more components onto integrated circuits" (PDF). Electronics Magazine. p. 4. Retrieved 2006-11-11. <a href="http://www.cs.utexas.edu/~fussell/courses/cs352h/papers/moore.pdf" target="_blank">http://www.cs.utexas.edu/~fussell/courses/cs352h/papers/moore.pdf</a> <a href="#fnref:35" class="footnote-back-ref">↩</a></p></li> </ol>