|
|
|
|
— |
ai_timelines:costs_of_human-level_hardware [2022/09/21 07:37] (current)
|
| | ====== Costs of human-level hardware ====== |
| | |
| | // Published 26 July, 2015; last updated 28 September, 2017 // |
| | |
| | <HTML> |
| | <p>Computing hardware which is equivalent to the brain –</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <ul> |
| | <li><div class="li">in terms of FLOPS probably costs between $1 x 10<sup>5</sup> and $3 x 10<sup>16</sup>, or $2/hour-$700bn/hour.</div></li> |
| | <li><div class="li">in terms of TEPS probably costs $200M – $7B, or or $4,700 – $170,000/hour (including energy costs in the hourly rate).</div></li> |
| | <li><div class="li">in terms of secondary memory probably costs $300-3,000, or $0.007-$0.07/hour.</div></li> |
| | </ul> |
| | </HTML> |
| | |
| | |
| | |
| | ===== Details ===== |
| | |
| | |
| | ==== Partial costs ==== |
| | |
| | |
| | === Computation === |
| | |
| | |
| | <HTML> |
| | <p><em>Main articles: <a href="/doku.php?id=ai_timelines:brain_performance_in_flops">Brain performance in FLOPS</a>, <a href="/doku.php?id=ai_timelines:current_flops_prices">Current FLOPS prices</a>, <a href="/doku.php?id=ai_timelines:trends_in_the_cost_of_computing">Trends in the costs of computing</a></em></p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p><a href="https://en.wikipedia.org/wiki/FLOPS">FLoating-point Operations Per Second</a> (FLOPS) is a measure of computer performance that emphasizes computing capacity. The human brain is estimated to perform between 10<sup>13.5</sup> and 10<sup>25 </sup>FLOPS. Hardware <a href="/doku.php?id=ai_timelines:current_flops_prices">currently costs</a> around $3 x 10<sup>-9</sup>/FLOPS, or $7 x 10<sup>-14</sup>/FLOPShour. This makes the current price of hardware which has equivalent computing capacity to the human brain between $1 x 10<sup>5</sup> and $3 x 10<sup>16</sup>, or $2/hour-$700bn/hour if hardware is used for five years.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>The price of FLOPS <a href="/doku.php?id=ai_timelines:trends_in_the_cost_of_computing">has probably</a> decreased by a factor of ten roughly every four years in the last quarter of a century.</p> |
| | </HTML> |
| | |
| | |
| | === Communication === |
| | |
| | |
| | <HTML> |
| | <p><em>Main articles: <a href="/doku.php?id=ai_timelines:brain_performance_in_teps">Brain performance in TEPS</a>, <a href="/doku.php?id=ai_timelines:the_cost_of_teps">The cost of TEPS</a> </em></p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p><a href="https://en.wikipedia.org/wiki/Traversed_edges_per_second">Traversed Edges Per Second</a> (TEPS) is a measure of computer performance that emphasizes communication capacity. The human brain <a href="/doku.php?id=ai_timelines:brain_performance_in_teps">is estimated</a> to perform at 0.18 – 6.4 x 10<sup>5</sup> GTEPS. Communication capacity <a href="/doku.php?id=ai_timelines:the_cost_of_teps">costs</a> around $11,000/GTEP or $0.26/GTEPShour in 2015, when amortized over five years and combined with energy costs. This makes the current price of hardware which has equivalent communication capacity to the human brain around $200M – $7B in total, or $4,700 – $170,000/hour including energy costs.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p><a href="/doku.php?id=ai_timelines:the_cost_of_teps">We estimate</a> that the price of TEPS falls by a factor of ten every four years, based the relationship between TEPS and FLOPS.</p> |
| | </HTML> |
| | |
| | |
| | === Information storage === |
| | |
| | |
| | <HTML> |
| | <p><em>Main articles: <a href="/doku.php?id=ai_timelines:information_storage_in_the_brain">Information storage in the brain</a>, <a href="/doku.php?id=ai_timelines:costs_of_information_storage">Costs of information storage</a>, <a href="/doku.php?id=ai_timelines:cost_of_human-level_information_storage">Costs of human-level information storage</a></em></p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p><a href="https://en.wikipedia.org/wiki/Computer_memory">Computer memory</a> comes in primary and secondary forms. Primary memory (e.g. RAM) is intended to be accessed frequently, while secondary memory is slower to access but has higher capacity. Here we estimate the secondary memory requirements ofthe brain. The human brain <a href="/doku.php?id=ai_timelines:information_storage_in_the_brain">is estimated</a> to store around 10-100TB of data. Secondary storage <a href="/doku.php?id=ai_timelines:costs_of_information_storage">costs around $30/TB</a> in 2015. <a href="/doku.php?id=ai_timelines:cost_of_human-level_information_storage">This means</a> it costs $300-3,000 for enough storage to store the contents of a human brain, or $0.007-$0.07/hour if hardware is used for five years.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>In the long run <a href="/doku.php?id=ai_timelines:costs_of_information_storage">the price of secondary memory has declined</a> by an order of magnitude roughly every 4.6 years. However the rate has declined so much that prices haven’t substantially dropped since 2011 (in 2015).</p> |
| | </HTML> |
| | |
| | |
| | ==== Interpreting partial costs ==== |
| | |
| | |
| | <HTML> |
| | <p>Calculating the total cost of hardware that is relevantly equivalent to the brain is not as simple as adding the partial costs as listed. FLOPS and TEPS are measures of different capabilities of the same hardware, so if you pay for TEPS at the aforementioned prices, you will also receive FLOPS.</p> |
| | </HTML> |
| | |
| | |
| | <HTML> |
| | <p>The above list is also not exhaustive: there may be substantial hardware costs that we haven’t included.</p> |
| | </HTML> |
| | |
| | |
| |
