Knowledge (XXG)

Project Denver

Source 📝

397:
SoC offering a total of 8 cores (or 4 dual-core pairs). The Carmel CPU core supports full Advanced SIMD (ARM NEON), VFP (Vector Floating Point), and ARMv8.2-FP16. First published testings of Carmel cores integrated in the Jetson AGX development kit by third party experts took place in September 2018
398:
and indicated a noticeably increased performance as should expected for this real world physical manifestation compared to predecessors systems, despite all doubts the used quickness of such a test setup in general an in particular implies. The Carmel design can be found in the Tegra model T194 ("
431:
ARM code is translated either by a hardware translator or through software emulation to an instruction set that is internal to Project Denver. ARM instructions can be reordered, removed if they do not contribute to the end result, or otherwise optimized if software emulation is
315:) where "Denver's binary translation layer runs in software, at a lower level than the operating system, and stores commonly accessed, already optimized code sequences in a 128 MB cache stored in main memory". Denver is a very wide in-order 558:
According to Charlie Demerjian, the Project Denver CPU may internally translate the ARM instructions to an internal instruction set, using firmware in the CPU. Also according to Demerjian, Project Denver was originally intended to support both
582:, was announced on October 15, 2014. The tablet was manufactured by HTC and features the dual-core Tegra K1 SoC. The Nexus 9 was the first 64-bit Android device available to consumers. 1604: 1946: 512:
backward compatibility. Project Denver was started in Stexar Company (Colorado) as an x86-compatible processor using binary translation, similar to projects by
955: 743: 393:", based on ARMv8 (64-bit; variant: ARM-v8.2 with 10-way superscalar, functional safety, dual execution, parity & ECC) got integrated into the Tegra 880: 1041: 1020: 765: 827: 936: 475:
is pairing an Nvidia Volta-GPU and several special purpose accelerators with 8 CPU cores with the Carmel design. In this design 4 Carmel
468:(ARMv8 64bit) cores using a coherent HMP (Heterogeneous Multi-Processor Architecture) approach. This pairs the units with a Parker-GPU. 320: 571:
technology from Transmeta, but was changed to the ARMv8-A 64-bit instruction set because Nvidia could not obtain a license to Intel's
720: 895: 842: 780: 425:
Denver also sets aside 128 MiB of main memory as an interpretation cache, which is inaccessible to the main operating system.
1534: 993: 457:
tablet, released November 3, 2014. Note, however, that the quad-core Tegra K1, while using the same name, isn't based on Denver.
627: 1863: 1464: 655: 896:"64-bit Nvidia Tegra 6 "Parker" Chip May Arrive in 2014. Devices with a 64-bit Tegra 6 could launch before the end of 2014" 1769: 1539: 802:
NVIDIA Unveils Tegra Parker SOC at Hot Chips – Built on 16nm TSMC Process, Features Pascal and Denver 2 Duo Architecture
705: 250: 112: 1504: 1479: 1469: 591: 552: 1529: 1524: 1519: 1509: 1499: 1474: 548: 446: 1544: 1514: 1494: 1489: 1484: 1417: 1329: 1199: 488: 479:
macro blocks (with each having 2 cores) are matched to each other with one more crossbar and 4 MiB of L3 memory.
1941: 692: 668: 610: 340: 332: 324: 1936: 1742: 1648: 1435: 286: 174: 1823: 1447: 1242: 921: 544: 524: 505:
development on which hundreds of engineers had already worked for three and half years and which also has
1888: 312: 857: 1150: 986: 442: 265: 127: 1371: 1217: 1627: 1592: 1222: 1212: 1207: 1188: 1183: 1178: 1173: 1168: 355: 328: 308: 1813: 1693: 1261: 1256: 1251: 1232: 1227: 1131: 351: 965:"Nvidia's First CPU Is a Winner. Denver Uses Dynamic Translation to Outperform Mobile Rivals" 1873: 1868: 1708: 1393: 1359: 1340: 1312: 1163: 1158: 1141: 1136: 1126: 1121: 1099: 1094: 560: 532: 344: 297: 379: 1683: 1665: 1411: 979: 632: 528: 336: 304: 240: 102: 801: 680: 1641: 1636: 1405: 492: 465: 964: 1930: 1893: 1727: 1688: 1653: 1399: 1109: 568: 359: 1898: 1818: 1808: 1722: 1612: 1423: 1306: 1300: 1113: 418:
128 KiB instruction + 64 KiB data L1 cache per core (both 4-way), 2 
1698: 1578: 1089: 540: 412: 316: 744:"Tegra K1 64-bit Denver core analysis: Are Nvidia's x86 efforts hidden within?" 1828: 1617: 1441: 1365: 177: 828:"Q&A: Nvidia chief explains his strategy for winning in mobile computing" 17: 1074: 813: 766:"Nexus 9 storms through Geekbench, Tegra K1 outperforms Apple iPhone 6's A8" 536: 513: 193: 67: 937:"Google announces Nexus 6, Nexus 9, Nexus Player, and Android 5.0 Lollipop" 695:
by Dustin Franklin (Nvidia development team for Jetson), December 12, 2018
693:
NVIDIA Jetson AGX Xavier Delivers 32 TeraOps for New Era of AI in Robotics
671:
by Dustin Franklin (Nvidia development team for Jetson), December 12, 2018
669:
NVIDIA Jetson AGX Xavier Delivers 32 TeraOps for New Era of AI in Robotics
613:
by Dustin Franklin (Nvidia development team for Jetson), December 12, 2018
611:
NVIDIA Jetson AGX Xavier Delivers 32 TeraOps for New Era of AI in Robotics
1069: 1064: 450: 419: 227: 215: 199: 89: 73: 956:"NVIDIA Project Boulder Revealed: Tegra's Competitor Hides in GPU Group" 1878: 1761: 1717: 1015: 579: 499: 454: 453:; the dual-core 2.3 GHz Denver-based K1 was first used in the HTC 294: 1587: 1429: 1002: 572: 506: 496: 301: 290: 166: 54: 366:
SoC series from Nvidia. Initially Denver cores was designed for the
578:
The first consumer device shipping with Denver CPU cores, Google's
1883: 1858: 1732: 1660: 1376: 1351: 1345: 520: 509: 472: 461: 399: 394: 383: 378:
was an improved design that built for the smaller, more efficient
371: 367: 363: 319:
pipeline. Its design makes it suitable for integration with other
307:
using a combination of simple hardware decoder and software-based
881:"New Tegra Roadmap Reveals Logan, Parker and Kayla CUDA Strategy" 539:
on the Denver team, and they have extensive experience designing
1853: 1335: 476: 206:(128 KiB I-cache with parity, 64 KiB D-cache with ECC) 80:(128 KiB I-cache with parity, 64 KiB D-cache with ECC) 1790: 1565: 1282: 1013: 975: 1055: 1050: 843:"NVIDIA Project Denver "Lost in Rockies", to Debut in 2014-15" 781:"NVIDIA Announces Tegra K1 SoC with Optional Denver CPU Cores" 564: 502: 656:""PROJECT DENVER" PROCESSOR TO USHER IN NEW ERA OF COMPUTING" 628:"Nvidia claims Haswell-class performance for Denver CPU core" 464:
has two Denver2 (ARMv8 64bit) cores inside and another four
721:"Nvidia reveals PC-like performance for 'Denver' Tegra K1" 971: 487:
The existence of Project Denver was revealed at the 2011
389:
In 2018, Nvidia released an improved design (codename: "
519:
According to Tom's Hardware, there are engineers from
1846: 1801: 1759: 1741: 1707: 1676: 1626: 1603: 1576: 1457: 1386: 1322: 1293: 1241: 1197: 1149: 1108: 1040: 1033: 706:"A Quick Test of NVIDIA's "Carmel" CPU Performance" 264: 259: 249: 239: 234: 222: 210: 191: 186: 172: 162: 154: 149: 126: 121: 111: 101: 96: 84: 65: 60: 50: 40: 35: 402:") that is designed with a 12 nm structure size. 362:. Respective cores have found integration in the 350:Project Denver is targeted at mobile computers, 987: 683:by Hassan Mujtaba on Jan 8, 2018 via WccfTech 8: 915: 913: 495:revealed that Project Denver is a five-year 144: 30: 1798: 1787: 1573: 1562: 1290: 1279: 1037: 1030: 1010: 994: 980: 972: 814:http://www.nvidia.com/object/ces2011.html 516:. Stexar was acquired by Nvidia in 2006. 491:. In a March 4, 2011 Q&A article CEO 621: 619: 603: 742:Anthony, Sebastian (January 6, 2014). 143: 29: 920:Demerjian, Charlie (August 5, 2011). 7: 1947:Very long instruction word computing 370:process node (Tegra model T132 aka " 107:28 nm (Denver 1) to 16 nm (Denver 2) 954:Valich, Theo (September 20, 2012). 894:Parrish, Kevin (October 14, 2013). 963:Linley Gwennap (August 18, 2014). 922:"What is Project Denver based on?" 898:. Tom's Hardware & ExtremeTech 858:"NVIDIA has x86 CPU in the works?" 841:Valich, Theo (December 12, 2011). 25: 856:Miller, Paul (October 19, 2006). 826:Takahashi, Dean (March 4, 2011). 816:Nvidia's press conference webcast 779:Shimpi, Anand (January 5, 2014). 719:Hachman, Mark (August 11, 2014). 626:Wasson, Scott (August 11, 2014). 935:Amadeo, Ron (October 15, 2014). 681:NVIDIA Drive Xavier SOC Detailed 449:-based GPU solution to form the 879:Valich, Theo (March 20, 2013). 654:Dally, Bill (January 5, 2011). 411:Pipelined processor with 7-way 235:Architecture and classification 97:Architecture and classification 1444:(framebuffer in system memory) 1: 445:Denver CPU was paired with a 428:Running at up to 2.5 GHz 382:node. (Tegra model T186 aka " 592:Comparison of ARMv8-A cores 553:simultaneous multithreading 549:very long instruction words 510:ARM instruction set (ARMv7) 27:Microarchitecture by Nvidia 1963: 1797: 1786: 1649:RSX 'Reality Synthesizer' 1572: 1561: 1289: 1284:Software and technologies 1278: 1029: 1009: 489:Consumer Electronics Show 422:L2 cache (16-way shared) 1436:Scalable Link Interface 1402:(variable refresh rate) 1294:Multimedia acceleration 658:. Official Nvidia blog. 287:central processing unit 260:Physical specifications 122:Physical specifications 1458:GPU microarchitectures 1450:(live video upscaling) 1448:Video Super Resolution 1368:(ray tracing platform) 1198:Unified shaders & 545:out-of-order execution 1889:Mellanox Technologies 313:dynamic recompilation 285:is the codename of a 182:to 2.3 GHz  1034:Fixed pixel pipeline 967:. MPR, Linley Group. 500:ARMv8-A architecture 293:that implements the 251:Instruction set 241:Technology node 113:Instruction set 103:Technology node 1372:Nvidia System Tools 150:General information 146: 36:General information 32: 1379:(video decode API) 1332:(shading language) 1245:& tensor cores 768:. 16 October 2014. 415:execution pipeline 352:personal computers 329:display controller 309:binary translation 1924: 1923: 1920: 1919: 1916: 1915: 1814:Chris Malachowsky 1782: 1781: 1778: 1777: 1694:Shield Android TV 1557: 1556: 1553: 1552: 1348:(ray tracing API) 1274: 1273: 1270: 1269: 1084: 1083: 1024: 804:, August 22, 2016 339:, etc.) into one 280: 279: 142: 141: 31:Nvidia Denver 1/2 16:(Redirected from 1954: 1874:Cumulus Networks 1869:Bright Computing 1854:3dfx Interactive 1799: 1788: 1669: 1657: 1645: 1574: 1563: 1394:Nvidia 3D Vision 1360:Nvidia Omniverse 1341:Nvidia GameWorks 1315:(video decoding) 1309:(video decoding) 1303:(video encoding) 1291: 1280: 1038: 1031: 1018: 1011: 996: 989: 982: 973: 968: 959: 941: 940: 932: 926: 925: 917: 908: 907: 905: 903: 891: 885: 884: 876: 870: 869: 867: 865: 853: 847: 846: 838: 832: 831: 823: 817: 811: 805: 799: 793: 792: 790: 788: 776: 770: 769: 762: 756: 755: 753: 751: 739: 733: 732: 730: 728: 716: 710: 709: 702: 696: 690: 684: 678: 672: 666: 660: 659: 651: 645: 644: 642: 640: 623: 614: 608: 345:system on a chip 305:instruction sets 230:@ 8 cores, T194) 147: 33: 21: 1962: 1961: 1957: 1956: 1955: 1953: 1952: 1951: 1942:Nvidia products 1927: 1926: 1925: 1912: 1842: 1836:Ranga Jayaraman 1833:Debora Shoquist 1793: 1774: 1755: 1737: 1703: 1684:Shield Portable 1672: 1666:Nintendo Switch 1663: 1651: 1639: 1622: 1599: 1568: 1549: 1453: 1426:(module/socket) 1420:(module/socket) 1414:(multi-monitor) 1412:Nvidia Surround 1408:(GPU switching) 1382: 1318: 1285: 1266: 1237: 1193: 1151:Unified shaders 1145: 1104: 1080: 1079: 1060: 1025: 1005: 1000: 962: 953: 950: 945: 944: 934: 933: 929: 924:. Semiaccurate. 919: 918: 911: 901: 899: 893: 892: 888: 878: 877: 873: 863: 861: 855: 854: 850: 840: 839: 835: 825: 824: 820: 812: 808: 800: 796: 786: 784: 778: 777: 773: 764: 763: 759: 749: 747: 741: 740: 736: 726: 724: 718: 717: 713: 704: 703: 699: 691: 687: 679: 675: 667: 663: 653: 652: 648: 638: 636: 633:The Tech Report 625: 624: 617: 609: 605: 600: 588: 485: 462:Nvidia Tegra X2 439: 408: 343:constituting a 337:image processor 276: 205: 138: 79: 46:2016 (Denver 2) 45: 28: 23: 22: 15: 12: 11: 5: 1960: 1958: 1950: 1949: 1944: 1939: 1937:ARM processors 1929: 1928: 1922: 1921: 1918: 1917: 1914: 1913: 1911: 1910: 1907: 1904: 1901: 1896: 1891: 1886: 1881: 1876: 1871: 1866: 1861: 1856: 1850: 1848: 1844: 1843: 1841: 1840: 1839:Jonah M. Alben 1837: 1834: 1831: 1826: 1821: 1816: 1811: 1809:Jen-Hsun Huang 1805: 1803: 1795: 1794: 1791: 1784: 1783: 1780: 1779: 1776: 1775: 1773: 1772: 1766: 1764: 1757: 1756: 1754: 1753: 1751:Project Denver 1747: 1745: 1739: 1738: 1736: 1735: 1730: 1725: 1720: 1714: 1712: 1705: 1704: 1702: 1701: 1696: 1691: 1686: 1680: 1678: 1674: 1673: 1671: 1670: 1658: 1646: 1633: 1631: 1624: 1623: 1621: 1620: 1615: 1609: 1607: 1601: 1600: 1598: 1597: 1596: 1595: 1584: 1582: 1570: 1569: 1567:Other products 1566: 1559: 1558: 1555: 1554: 1551: 1550: 1548: 1547: 1542: 1537: 1532: 1527: 1522: 1517: 1512: 1507: 1502: 1497: 1492: 1487: 1482: 1477: 1472: 1467: 1461: 1459: 1455: 1454: 1452: 1451: 1445: 1439: 1433: 1427: 1421: 1415: 1409: 1406:Nvidia Optimus 1403: 1397: 1390: 1388: 1384: 1383: 1381: 1380: 1374: 1369: 1363: 1357: 1356: 1355: 1349: 1338: 1333: 1326: 1324: 1320: 1319: 1317: 1316: 1310: 1304: 1297: 1295: 1287: 1286: 1283: 1276: 1275: 1272: 1271: 1268: 1267: 1265: 1264: 1259: 1254: 1248: 1246: 1239: 1238: 1236: 1235: 1230: 1225: 1220: 1215: 1210: 1204: 1202: 1195: 1194: 1192: 1191: 1186: 1181: 1176: 1171: 1166: 1161: 1155: 1153: 1147: 1146: 1144: 1139: 1134: 1129: 1124: 1119: 1117: 1106: 1105: 1103: 1102: 1097: 1092: 1085: 1082: 1081: 1078: 1077: 1072: 1067: 1061: 1059: 1058: 1053: 1047: 1046: 1044: 1035: 1027: 1026: 1014: 1007: 1006: 1001: 999: 998: 991: 984: 976: 970: 969: 960: 949: 948:External links 946: 943: 942: 927: 909: 886: 871: 848: 833: 818: 806: 794: 771: 757: 734: 711: 697: 685: 673: 661: 646: 615: 602: 601: 599: 596: 595: 594: 587: 584: 493:Jen-Hsun Huang 484: 481: 438: 435: 434: 433: 429: 426: 423: 416: 407: 404: 360:supercomputers 283:Project Denver 278: 277: 275: 274: 270: 268: 262: 261: 257: 256: 253: 247: 246: 243: 237: 236: 232: 231: 224: 220: 219: 212: 208: 207: 196: 189: 188: 184: 183: 180: 170: 169: 164: 160: 159: 156: 152: 151: 140: 139: 137: 136: 132: 130: 124: 123: 119: 118: 115: 109: 108: 105: 99: 98: 94: 93: 86: 82: 81: 70: 63: 62: 58: 57: 52: 48: 47: 42: 38: 37: 26: 24: 14: 13: 10: 9: 6: 4: 3: 2: 1959: 1948: 1945: 1943: 1940: 1938: 1935: 1934: 1932: 1908: 1905: 1902: 1900: 1897: 1895: 1894:Mental Images 1892: 1890: 1887: 1885: 1882: 1880: 1877: 1875: 1872: 1870: 1867: 1865: 1862: 1860: 1857: 1855: 1852: 1851: 1849: 1845: 1838: 1835: 1832: 1830: 1827: 1825: 1822: 1820: 1817: 1815: 1812: 1810: 1807: 1806: 1804: 1800: 1796: 1789: 1785: 1771: 1768: 1767: 1765: 1763: 1758: 1752: 1749: 1748: 1746: 1744: 1740: 1734: 1731: 1729: 1726: 1724: 1721: 1719: 1716: 1715: 1713: 1710: 1706: 1700: 1697: 1695: 1692: 1690: 1689:Shield Tablet 1687: 1685: 1682: 1681: 1679: 1677:Nvidia Shield 1675: 1667: 1662: 1659: 1655: 1654:PlayStation 3 1650: 1647: 1643: 1638: 1635: 1634: 1632: 1629: 1625: 1619: 1616: 1614: 1611: 1610: 1608: 1606: 1602: 1594: 1591: 1590: 1589: 1588:Nvidia Quadro 1586: 1585: 1583: 1580: 1575: 1571: 1564: 1560: 1546: 1543: 1541: 1538: 1536: 1533: 1531: 1528: 1526: 1523: 1521: 1518: 1516: 1513: 1511: 1508: 1506: 1503: 1501: 1498: 1496: 1493: 1491: 1488: 1486: 1483: 1481: 1478: 1476: 1473: 1471: 1468: 1466: 1463: 1462: 1460: 1456: 1449: 1446: 1443: 1440: 1437: 1434: 1431: 1428: 1425: 1422: 1419: 1416: 1413: 1410: 1407: 1404: 1401: 1400:Nvidia G-Sync 1398: 1395: 1392: 1391: 1389: 1385: 1378: 1375: 1373: 1370: 1367: 1364: 1362:(3D graphics) 1361: 1358: 1354:(physics SDK) 1353: 1350: 1347: 1344: 1343: 1342: 1339: 1337: 1334: 1331: 1328: 1327: 1325: 1321: 1314: 1311: 1308: 1305: 1302: 1299: 1298: 1296: 1292: 1288: 1281: 1277: 1263: 1260: 1258: 1255: 1253: 1250: 1249: 1247: 1244: 1240: 1234: 1231: 1229: 1226: 1224: 1221: 1219: 1216: 1214: 1211: 1209: 1206: 1205: 1203: 1201: 1196: 1190: 1187: 1185: 1182: 1180: 1177: 1175: 1172: 1170: 1167: 1165: 1162: 1160: 1157: 1156: 1154: 1152: 1148: 1143: 1140: 1138: 1135: 1133: 1130: 1128: 1125: 1123: 1120: 1118: 1115: 1111: 1107: 1101: 1098: 1096: 1093: 1091: 1087: 1086: 1076: 1073: 1071: 1068: 1066: 1063: 1062: 1057: 1054: 1052: 1049: 1048: 1045: 1043: 1039: 1036: 1032: 1028: 1022: 1017: 1012: 1008: 1004: 997: 992: 990: 985: 983: 978: 977: 974: 966: 961: 957: 952: 951: 947: 938: 931: 928: 923: 916: 914: 910: 897: 890: 887: 882: 875: 872: 859: 852: 849: 844: 837: 834: 829: 822: 819: 815: 810: 807: 803: 798: 795: 782: 775: 772: 767: 761: 758: 746:. ExtremeTech 745: 738: 735: 727:September 19, 722: 715: 712: 707: 701: 698: 694: 689: 686: 682: 677: 674: 670: 665: 662: 657: 650: 647: 635: 634: 629: 622: 620: 616: 612: 607: 604: 597: 593: 590: 589: 585: 583: 581: 576: 574: 570: 569:code morphing 566: 562: 556: 554: 550: 546: 542: 538: 534: 530: 526: 522: 517: 515: 511: 508: 504: 501: 498: 494: 490: 482: 480: 478: 474: 469: 467: 463: 458: 456: 452: 448: 444: 436: 430: 427: 424: 421: 417: 414: 410: 409: 405: 403: 401: 396: 392: 387: 385: 381: 377: 373: 369: 365: 361: 358:, as well as 357: 353: 348: 346: 342: 338: 334: 330: 326: 322: 318: 314: 310: 306: 303: 299: 296: 292: 288: 284: 272: 271: 269: 267: 263: 258: 254: 252: 248: 244: 242: 238: 233: 229: 225: 221: 217: 213: 209: 204: 201: 197: 195: 190: 185: 181: 179: 176: 171: 168: 165: 161: 157: 153: 148: 145:Nvidia Carmel 134: 133: 131: 129: 125: 120: 116: 114: 110: 106: 104: 100: 95: 91: 87: 83: 78: 75: 71: 69: 64: 59: 56: 53: 49: 44:2014 (Denver) 43: 39: 34: 19: 18:Nvidia Carmel 1899:PortalPlayer 1847:Acquisitions 1819:Curtis Priem 1750: 1711:and embedded 1661:Tegra NX-SoC 1613:Nvidia Tesla 1535:Ada Lovelace 1387:Technologies 1021:List of GPUs 930: 900:. Retrieved 889: 874: 862:. Retrieved 851: 836: 821: 809: 797: 785:. Retrieved 774: 760: 748:. Retrieved 737: 725:. Retrieved 714: 700: 688: 676: 664: 649: 637:. Retrieved 631: 606: 577: 557: 518: 486: 473:Tegra Xavier 470: 459: 440: 400:Tegra Xavier 390: 388: 375: 349: 323:cores (e.g. 289:designed by 282: 281: 202: 76: 1699:GeForce Now 1593:Quadro Plex 1579:Workstation 1438:(multi-GPU) 1396:(stereo 3D) 1243:Ray tracing 1208:GeForce 600 1090:GeForce 256 1042:Pre-GeForce 902:October 19, 864:October 19, 783:. Anandtech 567:code using 551:(VLIW) and 541:superscalar 413:superscalar 317:superscalar 163:Designed by 51:Designed by 1931:Categories 1829:Bill Dally 1824:David Kirk 1802:Key people 1630:components 1465:Fahrenheit 1442:TurboCache 1432:(protocol) 1366:Nvidia RTX 1252:GeForce 20 860:. Engadget 787:January 6, 750:January 7, 723:. PC World 639:August 14, 598:References 543:CPUs with 178:clock rate 1760:Computer 1577:Graphics 1540:Blackwell 1313:PureVideo 1159:GeForce 8 1122:GeForce 3 537:Transmeta 514:Transmeta 443:dual-core 255:ARMv8.2-A 218:@ 2 cores 198:192  92:@ 2 cores 72:192  1762:chipsets 1323:Software 1070:RIVA TNT 1065:RIVA 128 586:See also 451:Tegra K1 406:Overview 384:Tegra X2 376:Denver 2 372:Tegra K1 226:(4  223:L3 cache 211:L2 cache 203:per core 155:Launched 85:L2 cache 77:per core 41:Launched 1879:DeepMap 1792:Company 1718:GoForce 1628:Console 1505:Maxwell 1480:Rankine 1470:Celsius 1116:shaders 1016:GeForce 580:Nexus 9 573:patents 555:(SMT). 483:History 455:Nexus 9 356:servers 347:(SoC). 295:ARMv8-A 214:2  117:ARMv8-A 88:2  1909:Stexar 1906:MediaQ 1903:Exluna 1770:nForce 1728:Jetson 1530:Hopper 1525:Ampere 1520:Turing 1510:Pascal 1500:Kepler 1475:Kelvin 1430:NVLink 1110:Vertex 1088:  1003:Nvidia 507:32-bit 497:64-bit 447:Kepler 395:Xavier 391:Carmel 302:32-bit 291:Nvidia 167:Nvidia 55:Nvidia 1884:Icera 1859:Ageia 1733:Tegra 1723:Drive 1605:GPGPU 1581:cards 1545:Rubin 1515:Volta 1495:Fermi 1490:Tesla 1485:Curie 1377:VDPAU 1352:PhysX 1346:OptiX 1307:NVDEC 1301:NVENC 1114:pixel 521:Intel 437:Chips 432:used. 380:16 nm 368:28 nm 364:Tegra 266:Cores 245:12 nm 194:cache 187:Cache 173:Max. 128:Cores 68:cache 61:Cache 1743:CPUs 1709:SoCs 1642:Xbox 1637:NV2A 1336:CUDA 1218:800M 1200:NUMA 1127:4 Ti 1112:and 1100:4 MX 1075:TNT2 904:2013 866:2013 789:2014 752:2014 729:2014 641:2014 563:and 535:and 477:ASIC 471:The 460:The 386:"). 374:"). 321:SIPs 158:2018 1864:ULi 1618:DGX 1424:SXM 1418:MXM 1223:900 1213:700 1189:500 1184:400 1179:300 1174:200 1169:100 1056:NV2 1051:NV1 565:x86 561:ARM 533:Sun 525:AMD 503:CPU 466:A57 420:MiB 341:die 333:DSP 325:GPU 228:MiB 216:MiB 200:KiB 192:L1 175:CPU 90:MiB 74:KiB 66:L1 1933:: 1330:Cg 1262:40 1257:30 1233:16 1228:10 1132:FX 912:^ 630:. 618:^ 575:. 547:, 531:, 529:HP 527:, 523:, 441:A 354:, 335:, 331:, 327:, 298:64 1668:) 1664:( 1656:) 1652:( 1644:) 1640:( 1164:9 1142:7 1137:6 1095:2 1023:) 1019:( 995:e 988:t 981:v 958:. 939:. 906:. 883:. 868:. 845:. 830:. 791:. 754:. 731:. 708:. 643:. 311:( 300:/ 273:2 135:2 20:)

Index

Nvidia Carmel
Nvidia
cache
KiB
MiB
Technology node
Instruction set
Cores
Nvidia
CPU
clock rate
cache
KiB
MiB
MiB
Technology node
Instruction set
Cores
central processing unit
Nvidia
ARMv8-A
64
32-bit
instruction sets
binary translation
dynamic recompilation
superscalar
SIPs
GPU
display controller

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.