31:
47:
158:
499:
237:(assembly programs) are commonly packaged into libraries for re-use, instead of relying on advanced compiler technologies to handle essential algorithms. Even with modern compiler optimizations hand-optimized assembly code is more efficient and many common algorithms involved in DSP calculations are hand-written in order to take full advantage of the architectural optimizations.
356:, in which operations that produce overflows will accumulate at the maximum (or minimum) values that the register can hold rather than wrapping around (maximum+1 doesn't overflow to minimum as in many general-purpose CPUs, instead it stays at maximum). Sometimes various sticky bits operation modes are available.
141:. Most general-purpose microprocessors can also execute digital signal processing algorithms successfully, but may not be able to keep up with such processing continuously in real-time. Also, dedicated DSPs usually have better power efficiency, thus they are more suitable in portable devices such as
681:
produces a multi-core multi-threaded line of processor well suited to DSP operations, They come in various speeds ranging from 400 to 1600 MIPS. The processors have a multi-threaded architecture that allows up to 8 real-time threads per core, meaning that a 4 core device would support up to 32 real
177:
Most general-purpose microprocessors and operating systems can execute DSP algorithms successfully, but are not suitable for use in portable devices such as mobile phones and PDAs because of power efficiency constraints. A specialized DSP, however, will tend to provide a lower-cost solution, with
631:
The main improvement in the third generation was the appearance of application-specific units and instructions in the data path, or sometimes as coprocessors. These units allowed direct hardware acceleration of very specific but complex mathematical problems, like the
Fourier-transform or matrix
787:
Most DSPs use fixed-point arithmetic, because in real world signal processing the additional range provided by floating point is not needed, and there is a large speed benefit and cost benefit due to reduced hardware complexity. Floating point DSPs may be invaluable in applications where a wide
616:
presented in 1983, proved to be an even bigger success. It was based on the
Harvard architecture, and so had separate instruction and data memory. It already had a special instruction set, with instructions like load-and-accumulate or multiply-and-accumulate. It could work on 16-bit numbers and
218:
By the standards of general-purpose processors, DSP instruction sets are often highly irregular; while traditional instruction sets are made up of more general instructions that allow them to perform a wider variety of operations, instruction sets optimized for digital signal processing contain
204:
The architecture of a DSP is optimized specifically for digital signal processing. Most also support some of the features of an applications processor or microcontroller, since signal processing is rarely the only task of a system. Some useful features for optimizing DSP algorithms are outlined
169:
typically require a large number of mathematical operations to be performed quickly and repeatedly on a series of data samples. Signals (perhaps from audio or video sensors) are constantly converted from analog to digital, manipulated digitally, and then converted back to analog form. Many DSP
668:), perform eight operations per clock-cycle and are compatible with a broad range of external peripherals and various buses (PCI/serial/etc). TMS320C6474 chips each have three such DSPs, and the newest generation C6000 chips support floating point as well as fixed point processing.
635:
The fourth generation is best characterized by the changes in the instruction set and the instruction encoding/decoding. SIMD extensions were added, and VLIW and the superscalar architecture appeared. As always, the clock-speeds have increased; a 3 ns MAC now became possible.
788:
dynamic range is required. Product developers might also use floating point DSPs to reduce the cost and complexity of software development in exchange for more expensive hardware, since it is generally easier to implement algorithms in floating point.
193:, and can be replaced with specialised DSPs with significant benefits to the satellites' weight, power consumption, complexity/cost of construction, reliability and flexibility of operation. For example, the SES-12 and SES-14 satellites from operator
674:
produces a multi-core DSP family, the MSC81xx. The MSC81xx is based on StarCore
Architecture processors and the latest MSC8144 DSP combines four programmable SC3400 StarCore DSP cores. Each SC3400 StarCore DSP core has a clock speed of 1 GHz.
820:
passed acceptance. With a processing speed of 0.4 TFLOPS, the chip can achieve better performance than current mainstream DSP chips. The design team has begun to create Huarui-3, which has a processing speed in TFLOPS level and a support for
624:; they also had an addressing unit capable of loop-addressing. Some of them operated on 24-bit variables and a typical model only required about 21 ns for a MAC. Members of this generation were for example the AT&T DSP16A or the
682:
time threads. Threads communicate between each other with buffered channels that are capable of up to 80 Mbit/s. The devices are easily programmable in C and aim at bridging the gap between conventional micro-controllers and FPGAs
219:
instructions for common mathematical operations that occur frequently in DSP calculations. Both traditional and DSP-optimized instruction sets are able to compute any arbitrary operation but an operation that might require multiple
688:
produces and licenses three distinct families of DSPs. Perhaps the best known and most widely deployed is the CEVA-TeakLite DSP family, a classic memory-based architecture, with 16-bit or 32-bit word-widths and single or dual
659:
series DSPs, which have clock speeds of 1.2 GHz and implement separate instruction and data caches. They also have an 8 MiB 2nd level cache and 64 EDMA channels. The top models are capable of as many as 8000 MIPS
648:
circuitry, and a wider bus system. Not all DSPs provide the same speed and many kinds of signal processors exist, each one of them being better suited for a specific task, ranging in price from about US$ 1.50 to US$ 300.
336:
to support several accesses per instruction cycle – typically supporting reading 2 data values from 2 separate data buses and the next instruction (from the instruction cache, or a 3rd program memory) simultaneously.
430:
DSPs can sometimes rely on supporting code to know about cache hierarchies and the associated delays. This is a tradeoff that allows for better performance. In addition, extensive use of
644:
Modern signal processors yield greater performance; this is due in part to both technological and architectural advancements like lower design rules, fast-access two-level cache, (E)
693:. The CEVA-X DSP family offers a combination of VLIW and SIMD architectures, with different members of the family offering dual or quad 16-bit MACs. The CEVA-XC DSP family targets
809:
In
Communications a new breed of DSPs offering the fusion of both DSP functions and H/W acceleration function is making its way into the mainstream. Such Modem processors include
632:
operations. Some chips, like the
Motorola MC68356, even included more than one processor core to work in parallel. Other DSPs from 1995 are the TI TMS320C541 or the TMS 320C80.
339:
Special loop controls, such as architectural support for executing a few instruction words in a very tight loop without overhead for instruction fetches or exit testing—such as
784:, such as motor control and in power supplies. The dsPIC runs at up to 40MIPS, and has support for 16 bit fixed point MAC, bit reverse and modulo addressing, as well as DMA.
571:
as an "analog signal processor". It had an on-chip ADC/DAC with an internal signal processor, but it didn't have a hardware multiplier and was not successful in the market.
1268:
1242:
620:
About five years later, the second generation of DSPs began to spread. They had 3 memories for storing two operands simultaneously and included hardware to accelerate
1189:
419:
DSPs are usually optimized for streaming data and use special memory architectures that are able to fetch multiple data or instructions at the same time, such as the
774:
produces the Quatro family of SoCs that contain one or more custom
Imaging DSPs optimized for processing document image data for scanner and copier applications.
3524:
2553:
727:
family of embedded digital signal processors combine the features of a DSP with those of a general use processor. As a result, these processors can run simple
1525:
1386:
1407:
3468:
704:
174:; that is, for the system to work, the DSP operation must be completed within some fixed time, and deferred (or batch) processing is not viable.
2664:
1847:
523:, the industry's first digital signal processor. It also set other milestones, being the first chip to use linear predictive coding to perform
544:(AMI) released the S2811. The AMI S2811 "signal processing peripheral", like many later DSPs, has a hardware multiplier that enables it to do
2366:
780:
produces the PIC24 based dsPIC line of DSPs. Introduced in 2004, the dsPIC is designed for applications needing a true DSP as well as a true
591:
519:' Dallas research facility. Two years later in 1978, they produced the first Speak & Spell, with the technological centerpiece being the
1644:
568:
1459:
2523:
2089:
1906:
1037:
1341:
2877:
1372:
1869:
185:
where hundreds or even thousands of analog filters, switches, frequency converters and so on are required to receive and process the
17:
2518:
556:, V-groove MOS), a technology that had previously not been mass-produced. It was designed as a microprocessor peripheral, for the
3623:
2590:
1276:
1105:
1002:
1177:
1118:
2872:
2343:
1153:
756:
technology, optimized for audio and video processing. In some products the DSP core is hidden as a fixed-function block into a
1226:
3287:
2411:
1674:
1518:
3461:
3297:
2438:
661:
3779:
3774:
1565:
973:
914:
817:
609:
The
Altamira DX-1 was another early DSP, utilizing quad integer pipelines with delayed branches and branch prediction.
3628:
2605:
2433:
2406:
1785:
1328:
1312:
760:, but NXP also provides a range of flexible single core media processors. The TriMedia media processors support both
720:
690:
595:
545:
302:
246:
3663:
3546:
3420:
2983:
1876:
1842:
1837:
1756:
1721:
534:
83:
3395:
3292:
2693:
2600:
2401:
1622:
1511:
35:
3799:
3454:
2421:
2140:
1575:
798:
Embedded general-purpose RISC processors are becoming increasingly DSP like in functionality. For example, the
3784:
3618:
2595:
2443:
2416:
2277:
1891:
1852:
1709:
792:
3789:
3032:
2794:
2270:
2231:
1886:
1881:
1815:
1627:
791:
Generally, DSPs are dedicated integrated circuits; however DSP functionality can also be produced by using
1431:
427:, which use separate program and data memories (sometimes even concurrent access on multiple data buses).
2659:
2356:
2054:
1751:
711:(million floating-point operations per second) to 400 MHz/2400 MFLOPS. Some models support multiple
575:
3529:
3309:
2956:
2373:
1864:
1832:
1602:
1590:
1570:
697:
modem designs and leverages a unique combination of VLIW and Vector architectures with 32 16-bit MACs.
665:
395:
131:
617:
needed 390 ns for a multiply–add operation. TI is now the market leader in general-purpose DSPs.
181:
Such performance improvements have led to the introduction of digital signal processing in commercial
3724:
3556:
3400:
3363:
3353:
1741:
1203:
834:
765:
750:
79:
1202:
Uh, Gang-Ryung; Wang, Yuhong; Whalley, David; Jinturkar, Sanjay; Burns, Chris; Cao, Vincent (2000).
3693:
3519:
3493:
3415:
2822:
2735:
2585:
2547:
2383:
2333:
2328:
1805:
1699:
1607:
839:
613:
424:
198:
182:
178:
better performance, lower latency, and no requirements for specialised cooling or large batteries.
103:
3678:
3368:
3151:
3045:
3009:
2926:
2910:
2752:
2541:
2500:
2488:
2351:
2265:
2186:
1951:
1612:
1555:
1069:"Architectures and Design techniques for energy efficient embedded DSP and multimedia processing"
1010:
822:
487:
284:
220:
95:
1243:"Speak & Spell, the First Use of a Digital Signal Processing IC for Speech Generation, 1978"
1066:
329:
DSPs sometimes use time-stationary encoding to simplify hardware and increase coding efficiency.
3573:
3568:
3174:
3146:
3056:
3021:
2770:
2764:
2746:
2480:
2474:
2378:
2282:
2173:
2112:
1974:
1617:
1469:
854:
799:
761:
694:
512:
482:
359:
323:
289:
250:
227:
instructions to compute might require only one instruction in a DSP optimized instruction set.
890:
552:
chip specifically designed as a DSP, and fabricated using vertical metal oxide semiconductor (
3794:
3714:
3658:
3650:
3503:
3485:
3477:
3348:
3257:
3003:
2715:
2533:
2292:
2260:
2218:
2130:
1931:
1746:
1736:
1726:
1716:
1686:
1669:
1534:
1348:
1298:
1165:
963:
849:
716:
541:
353:
340:
171:
3683:
3606:
3590:
3378:
3314:
2900:
2622:
2512:
2459:
1991:
1704:
1560:
1542:
1436:
777:
645:
431:
420:
388:
366:
257:
816:
In May 2018, Huarui-2 designed by
Nanjing Research Institute of Electronics Technology of
8:
3731:
3425:
3410:
3230:
3081:
3063:
3027:
3015:
2669:
2616:
2393:
2309:
2191:
2046:
1941:
1800:
1068:
968:. Instrument Engineers' Handbook. Vol. 2 (4th ed.). CRC Press. pp. 11–12.
583:
451:
333:
146:
119:
560:, and it had to be initialized by the host. The S2811 was not successful in the market.
3741:
3709:
3498:
3282:
3274:
3126:
3101:
2905:
2780:
2304:
2245:
2125:
1857:
1585:
1464:
746:
549:
315:
115:
99:
91:
906:
446:
or memory protection. Operating systems that use virtual memory require more time for
3638:
3585:
3235:
3202:
3118:
3050:
2951:
2941:
2931:
2862:
2857:
2852:
2775:
2704:
2610:
2570:
2203:
2153:
2103:
2079:
1961:
1901:
1896:
1778:
1694:
1366:
1308:
1222:
1122:
1067:
Ingrid
Verbauwhede; Patrick Schaumont; Christian Piguet; Bart Kienhuis (2005-12-24).
969:
936:
928:
920:
910:
712:
652:
599:
516:
447:
231:
3719:
3633:
3561:
3405:
3338:
3324:
3179:
3086:
3040:
2847:
2842:
2837:
2832:
2827:
2817:
2687:
2654:
2565:
2560:
2469:
2321:
2316:
2299:
2287:
2226:
1790:
1768:
1654:
1632:
1550:
1214:
1092:
902:
844:
757:
728:
524:
39:
768:, and have specific instructions to deal with complex filters and entropy coding.
528:
137:
The goal of a DSP is usually to measure, filter or compress continuous real-world
3753:
3748:
3319:
3304:
3252:
3156:
3131:
2968:
2961:
2812:
2807:
2802:
2741:
2649:
2639:
2361:
2196:
2148:
1911:
1795:
1763:
1664:
1659:
1580:
944:
781:
471:
462:
402:
319:
30:
3736:
3688:
3551:
3430:
3264:
3247:
3240:
3136:
2993:
2730:
2644:
2575:
2158:
2120:
2069:
2064:
2059:
1773:
1597:
1042:
898:
700:
625:
443:
379:
138:
75:
59:
55:
579:
3768:
3225:
3141:
2181:
2163:
1956:
1649:
1218:
940:
924:
803:
557:
442:
DSPs frequently use multi-tasking operating systems, but have no support for
281:
Fundamental DSP algorithms depend heavily on multiply–accumulate performance
274:
194:
123:
2084:
948:
739:
while operating on real-time data. The SHARC-based ADSP-210xx provides both
3435:
3373:
3189:
3166:
2978:
2699:
1637:
979:
736:
723:
instructions and audio processing-specific components and peripherals. The
587:
142:
3220:
3184:
2895:
2867:
2725:
2580:
1503:
656:
269:
263:
127:
3446:
1498:
1387:"NEC Electronics Inc. μPD77C20A, 7720A, 77P20 Digital Signal Processors"
149:
that are able to fetch multiple data or instructions at the same time.
3536:
3106:
3096:
3091:
3073:
2973:
2946:
2208:
2041:
2011:
1731:
1213:. Lecture Notes in Computer Science. Vol. 1781. pp. 157–172.
870:
740:
685:
621:
531:
234:
46:
732:
606:
applications, was one of the most commercially successful early DSPs.
3668:
3197:
3194:
2936:
2006:
1984:
1304:
1300:
Embedded and
Networking Systems: Design, Software, and Implementation
1190:"Understanding Advanced Processor Features Promotes Efficient Coding"
1142:
671:
603:
405:, to reduce or eliminate the overhead required for looping operations
166:
1499:
Pocket Guide to Processors for DSP - Berkeley Design Technology, INC
157:
3673:
3611:
3212:
2031:
1204:"Techniques for Effectively Exploiting a Zero Overhead Loop Buffer"
724:
383:
190:
78:
chip, with its architecture optimized for the operational needs of
51:
27:
Specialized microprocessor optimized for digital signal processing
2021:
1979:
771:
520:
515:
concept to Paul Breedlove, Larry Brantingham, and Gene Frantz at
230:
One implication for software architecture is that hand-optimized
145:
because of power consumption constraints. DSPs often use special
932:
62:
with 25 MHz which was directly accessible via an interface.
3578:
2036:
2001:
1966:
865:
708:
186:
87:
498:
2494:
2026:
1996:
1408:"Introduction of ADSP-21000 Family digital signal processors"
810:
564:
111:
107:
3358:
2506:
2426:
2016:
859:
753:
678:
553:
307:
1493:
1946:
1936:
707:-based DSP and range in performance from 66 MHz/198
224:
1003:"1979: Single Chip Digital Signal Processor Introduced"
1201:
1038:"30 years of DSP: From a child's toy to 4G and beyond"
594:'80. Both processors were inspired by the research in
997:
995:
993:
991:
989:
465:
to be implemented without having to test for wrapping
889:
Dyer, Stephen A.; Harms, Brian K. (13 August 1993).
612:
Another DSP produced by Texas Instruments (TI), the
365:
Single-cycle operations to increase the benefits of
1178:"ADSP-BF533 Blackfin Processor Hardware Reference"
986:
548:in a single instruction. The S2281 was the first
3766:
574:In 1980, the first stand-alone, complete DSPs –
437:
362:is often used to speed up arithmetic processing.
1266:
582:based on the modified Harvard architecture and
1143:"Architecture of the Digital Signal Processor"
3462:
1519:
1062:
1060:
592:International Solid-State Circuits Conference
58:(25 MHz) and a digital signal processor
1296:
1096:Broadgate Publications (September 2016) pp22
2524:Computer performance by orders of magnitude
3469:
3455:
1533:
1526:
1512:
1297:Khan, Gul N.; Iniewski, Krzysztof (2017).
1057:
1035:
3476:
888:
1119:""DSP processors: memory architectures""
497:
409:
213:
156:
45:
34:An L7A1045 DSP chip, as used in several
29:
14:
3767:
1451:
1371:: CS1 maint: archived copy as title (
961:
511:In 1976, Richard Wiggins proposed the
414:
332:Multiple arithmetic units may require
3450:
1507:
1036:Taranovich, Steve (August 27, 2012).
1031:
1029:
1027:
955:
322:and bit-reversed addressing mode for
197:launched in 2018, were both built by
2495:Floating-point operations per second
1432:"国产新型雷达芯片华睿2号与组网中心同时亮相-科技新闻-中国科技网首页"
527:. The chip was made possible with a
347:
662:millions of instructions per second
470:Bit-reversed addressing, a special
240:
161:A typical digital processing system
24:
1024:
818:China Electronics Technology Group
502:TRW TDC1010 multiplier-accumulator
382:unit integrated directly into the
25:
3811:
3547:Hardware random number generation
1487:
1329:"Digital Graphic Audio Equalizer"
596:public switched telephone network
256:used extensively in all kinds of
170:applications have constraints on
3421:Semiconductor device fabrication
965:Process Control and Optimization
893:. In Yovits, Marshall C. (ed.).
201:with 25% of capacity using DSP.
165:Digital signal processing (DSP)
3396:History of general-purpose CPUs
1623:Nondeterministic Turing machine
1457:
1424:
1400:
1379:
1334:
1321:
1290:
1260:
1235:
1195:
1183:
1171:
1159:
1147:
373:
208:
94:chips. They are widely used in
1576:Deterministic finite automaton
1267:Bogdanowicz, A. (2009-10-06).
1136:
1111:
1099:
1085:
882:
639:
602:. The μPD7720, introduced for
506:
13:
1:
2367:Simultaneous and heterogenous
1269:"IEEE Milestones Honor Three"
907:10.1016/S0065-2458(08)60403-9
876:
793:field-programmable gate array
546:multiply–accumulate operation
438:Addressing and virtual memory
314:Specialized instructions for
3051:Integrated memory controller
3033:Translation lookaside buffer
2232:Memory dependence prediction
1675:Random-access stored program
1628:Probabilistic Turing machine
695:Software-defined Radio (SDR)
7:
2507:Synaptic updates per second
1106:"Memory and DSP Processors"
891:"Digital Signal Processing"
828:
576:Nippon Electric Corporation
476:useful for calculating FFTs
458:Hardware modulo addressing
454:, which increases latency.
189:signals and ready them for
152:
10:
3816:
2911:Heterogeneous architecture
1833:Orthogonal instruction set
1603:Alternating Turing machine
1591:Quantum cellular automaton
1154:"ARC XY Memory DSP Option"
813:ModemX and CEVA's XC4000.
743:and non-delayed branches.
666:very long instruction word
493:
343:and hardware loop buffers.
132:high-definition television
3780:Digital signal processors
3775:Digital signal processing
3702:
3649:
3599:
3542:Digital signal processing
3512:
3484:
3401:Microprocessor chronology
3388:
3364:Dynamic frequency scaling
3337:
3273:
3211:
3165:
3117:
3072:
2992:
2919:
2888:
2793:
2714:
2678:
2632:
2532:
2519:Cache performance metrics
2458:
2392:
2342:
2253:
2244:
2217:
2172:
2139:
2111:
2102:
1922:
1825:
1814:
1685:
1541:
835:Digital signal controller
766:floating-point arithmetic
183:communications satellites
88:metal–oxide–semiconductor
80:digital signal processing
18:Digital signal processors
3494:Universal Turing machine
3416:Hardware security module
2759:Digital signal processor
2736:Graphics processing unit
2548:Graphics processing unit
1460:"全国产芯片华睿2号通过"核高基"验收-新华网"
1219:10.1007/3-540-46423-9_11
840:Graphics processing unit
590:– were presented at the
425:von Neumann architecture
199:Airbus Defence and Space
104:digital image processing
68:digital signal processor
3552:Artificial intelligence
3369:Dynamic voltage scaling
3152:Memory address register
3046:Branch target predictor
3010:Address generation unit
2753:Physics processing unit
2542:Central processing unit
2501:Transactions per second
2489:Instructions per second
2412:Array processing (SIMT)
1556:Stored-program computer
1011:Computer History Museum
823:artificial intelligence
488:Address generation unit
396:multiplier–accumulators
118:systems, and in common
96:audio signal processing
3574:Custom hardware attack
3175:Hardwired control unit
3057:Memory management unit
3022:Memory management unit
2771:Secure cryptoprocessor
2765:Tensor Processing Unit
2747:Vision processing unit
2481:Cycles per instruction
2475:Instructions per cycle
2422:Associative processing
2113:Instruction pipelining
1535:Processor technologies
1275:. IEEE. Archived from
962:Liptak, B. G. (2006).
862:– a multiprocessor DSP
855:Vision processing unit
802:processors include an
762:fixed-point arithmetic
749:produce DSPs based on
503:
483:memory management unit
360:Fixed-point arithmetic
299:related instructions:
290:Fast Fourier transform
162:
63:
43:
3800:Hardware acceleration
3504:Distributed computing
3478:Hardware acceleration
3258:Sum-addressed decoder
3004:Arithmetic logic unit
2131:Classic RISC pipeline
2085:Epiphany architecture
1932:Motorola 68000 series
1327:Alberto Luis Andres.
1211:Compiler Construction
1166:"Zero Overhead Loops"
1074:. rijndael.ece.vt.edu
895:Advances in Computers
850:Hardware acceleration
542:American Microsystems
501:
410:Hardware architecture
354:Saturation arithmetic
341:zero-overhead looping
275:polynomial evaluation
214:Software architecture
160:
49:
33:
3785:Computer engineering
3607:High-level synthesis
3379:Performance per watt
2957:replacement policies
2623:Package on a package
2513:Performance per watt
2417:Pipelined processing
2187:Tomasulo's algorithm
1992:Clipper architecture
1848:Application-specific
1561:Finite-state machine
1468:. 南京. Archived from
778:Microchip Technology
421:Harvard architecture
401:Hardware-controlled
334:memory architectures
247:multiply–accumulates
147:memory architectures
3790:Integrated circuits
3732:Digital electronics
3684:In-memory computing
3664:Transport triggered
3411:Digital electronics
3064:Instruction decoder
3016:Floating-point unit
2670:Soft microprocessor
2617:System in a package
2192:Reservation station
1722:Transport-triggered
901:. pp. 59–118.
535:fabrication process
415:Memory architecture
120:consumer electronic
74:) is a specialized
3742:Hardware emulation
3710:Programmable logic
3499:Parallel computing
3283:Integrated circuit
3127:Processor register
2781:Baseband processor
2126:Operand forwarding
1586:Cellular automaton
1465:Xinhua News Agency
1007:The Silicon Engine
747:NXP Semiconductors
600:telecommunications
550:integrated circuit
504:
253:, FMA) operations
251:fused multiply–add
163:
116:speech recognition
100:telecommunications
92:integrated circuit
64:
44:
3762:
3761:
3639:Network on a chip
3444:
3443:
3333:
3332:
2952:Instruction cache
2942:Scratchpad memory
2789:
2788:
2776:Network processor
2705:Network on a chip
2660:Ultra-low-voltage
2611:Multi-chip module
2454:
2453:
2240:
2239:
2227:Branch prediction
2204:Register renaming
2098:
2097:
2080:VISC architecture
1902:Quantum computing
1897:VISC architecture
1779:Secondary storage
1695:Microarchitecture
1655:Register machines
1228:978-3-540-67263-0
729:operating systems
653:Texas Instruments
517:Texas Instruments
513:Speak & Spell
448:context switching
348:Data instructions
326:cross-referencing
249:(MACs, including
134:(HDTV) products.
16:(Redirected from
3807:
3754:Embedded systems
3634:System on a chip
3471:
3464:
3457:
3448:
3447:
3406:Processor design
3298:Power management
3180:Instruction unit
3041:Branch predictor
2990:
2989:
2688:System on a chip
2630:
2629:
2470:Transistor count
2394:Flynn's taxonomy
2251:
2250:
2109:
2108:
1912:Addressing modes
1823:
1822:
1769:Memory hierarchy
1633:Hypercomputation
1551:Abstract machine
1528:
1521:
1514:
1505:
1504:
1482:
1481:
1479:
1477:
1455:
1449:
1448:
1446:
1444:
1428:
1422:
1421:
1419:
1418:
1412:
1404:
1398:
1397:
1395:
1394:
1383:
1377:
1376:
1370:
1362:
1360:
1359:
1353:
1347:. Archived from
1346:
1338:
1332:
1325:
1319:
1318:
1294:
1288:
1287:
1285:
1284:
1264:
1258:
1257:
1255:
1254:
1239:
1233:
1232:
1208:
1199:
1193:
1187:
1181:
1175:
1169:
1163:
1157:
1151:
1145:
1140:
1134:
1133:
1131:
1130:
1121:. Archived from
1115:
1109:
1103:
1097:
1093:Beyond Frontiers
1089:
1083:
1082:
1080:
1079:
1073:
1064:
1055:
1054:
1052:
1050:
1033:
1022:
1021:
1019:
1017:
999:
984:
983:
959:
953:
952:
897:. Vol. 37.
886:
845:System on a chip
741:delayed branches
525:speech synthesis
463:circular buffers
394:Highly parallel
241:Instruction sets
122:devices such as
54:from 1990 had a
40:Hyper Neo Geo 64
21:
3815:
3814:
3810:
3809:
3808:
3806:
3805:
3804:
3765:
3764:
3763:
3758:
3749:Logic synthesis
3698:
3645:
3600:Implementations
3595:
3508:
3480:
3475:
3445:
3440:
3426:Tick–tock model
3384:
3340:
3329:
3269:
3253:Address decoder
3207:
3161:
3157:Program counter
3132:Status register
3113:
3068:
3028:Load–store unit
2995:
2988:
2915:
2884:
2785:
2742:Image processor
2717:
2710:
2680:
2674:
2650:Microcontroller
2640:Embedded system
2628:
2528:
2461:
2450:
2388:
2338:
2236:
2213:
2197:Re-order buffer
2168:
2149:Data dependency
2135:
2094:
1924:
1918:
1817:
1816:Instruction set
1810:
1796:Multiprocessing
1764:Cache hierarchy
1757:Register/memory
1681:
1581:Queue automaton
1537:
1532:
1494:DSP Online Book
1490:
1485:
1475:
1473:
1472:on May 26, 2018
1456:
1452:
1442:
1440:
1430:
1429:
1425:
1416:
1414:
1410:
1406:
1405:
1401:
1392:
1390:
1385:
1384:
1380:
1364:
1363:
1357:
1355:
1351:
1344:
1342:"Archived copy"
1340:
1339:
1335:
1326:
1322:
1315:
1295:
1291:
1282:
1280:
1265:
1261:
1252:
1250:
1247:IEEE Milestones
1241:
1240:
1236:
1229:
1206:
1200:
1196:
1188:
1184:
1176:
1172:
1164:
1160:
1152:
1148:
1141:
1137:
1128:
1126:
1117:
1116:
1112:
1104:
1100:
1090:
1086:
1077:
1075:
1071:
1065:
1058:
1048:
1046:
1034:
1025:
1015:
1013:
1001:
1000:
987:
976:
960:
956:
917:
887:
883:
879:
831:
806:and C6000 DSP.
795:chips (FPGAs).
782:microcontroller
735:, velocity and
642:
509:
496:
481:Exclusion of a
472:addressing mode
440:
417:
412:
376:
350:
243:
216:
211:
155:
28:
23:
22:
15:
12:
11:
5:
3813:
3803:
3802:
3797:
3792:
3787:
3782:
3777:
3760:
3759:
3757:
3756:
3751:
3746:
3745:
3744:
3737:Virtualization
3734:
3729:
3728:
3727:
3722:
3712:
3706:
3704:
3700:
3699:
3697:
3696:
3691:
3689:Systolic array
3686:
3681:
3676:
3671:
3666:
3661:
3655:
3653:
3647:
3646:
3644:
3643:
3642:
3641:
3631:
3626:
3621:
3616:
3615:
3614:
3603:
3601:
3597:
3596:
3594:
3593:
3588:
3583:
3582:
3581:
3571:
3569:Machine vision
3566:
3565:
3564:
3554:
3549:
3544:
3539:
3534:
3533:
3532:
3527:
3516:
3514:
3510:
3509:
3507:
3506:
3501:
3496:
3490:
3488:
3482:
3481:
3474:
3473:
3466:
3459:
3451:
3442:
3441:
3439:
3438:
3433:
3431:Pin grid array
3428:
3423:
3418:
3413:
3408:
3403:
3398:
3392:
3390:
3386:
3385:
3383:
3382:
3376:
3371:
3366:
3361:
3356:
3351:
3345:
3343:
3335:
3334:
3331:
3330:
3328:
3327:
3322:
3317:
3312:
3307:
3302:
3301:
3300:
3295:
3290:
3279:
3277:
3271:
3270:
3268:
3267:
3265:Barrel shifter
3262:
3261:
3260:
3255:
3248:Binary decoder
3245:
3244:
3243:
3233:
3228:
3223:
3217:
3215:
3209:
3208:
3206:
3205:
3200:
3192:
3187:
3182:
3177:
3171:
3169:
3163:
3162:
3160:
3159:
3154:
3149:
3144:
3139:
3137:Stack register
3134:
3129:
3123:
3121:
3115:
3114:
3112:
3111:
3110:
3109:
3104:
3094:
3089:
3084:
3078:
3076:
3070:
3069:
3067:
3066:
3061:
3060:
3059:
3048:
3043:
3038:
3037:
3036:
3030:
3019:
3013:
3007:
3000:
2998:
2987:
2986:
2981:
2976:
2971:
2966:
2965:
2964:
2959:
2954:
2949:
2944:
2939:
2929:
2923:
2921:
2917:
2916:
2914:
2913:
2908:
2903:
2898:
2892:
2890:
2886:
2885:
2883:
2882:
2881:
2880:
2870:
2865:
2860:
2855:
2850:
2845:
2840:
2835:
2830:
2825:
2820:
2815:
2810:
2805:
2799:
2797:
2791:
2790:
2787:
2786:
2784:
2783:
2778:
2773:
2768:
2762:
2756:
2750:
2744:
2739:
2733:
2731:AI accelerator
2728:
2722:
2720:
2712:
2711:
2709:
2708:
2702:
2697:
2694:Multiprocessor
2691:
2684:
2682:
2676:
2675:
2673:
2672:
2667:
2662:
2657:
2652:
2647:
2645:Microprocessor
2642:
2636:
2634:
2633:By application
2627:
2626:
2620:
2614:
2608:
2603:
2598:
2593:
2588:
2583:
2578:
2576:Tile processor
2573:
2568:
2563:
2558:
2557:
2556:
2545:
2538:
2536:
2530:
2529:
2527:
2526:
2521:
2516:
2510:
2504:
2498:
2492:
2486:
2485:
2484:
2472:
2466:
2464:
2456:
2455:
2452:
2451:
2449:
2448:
2447:
2446:
2436:
2431:
2430:
2429:
2424:
2419:
2414:
2404:
2398:
2396:
2390:
2389:
2387:
2386:
2381:
2376:
2371:
2370:
2369:
2364:
2362:Hyperthreading
2354:
2348:
2346:
2344:Multithreading
2340:
2339:
2337:
2336:
2331:
2326:
2325:
2324:
2314:
2313:
2312:
2307:
2297:
2296:
2295:
2290:
2280:
2275:
2274:
2273:
2268:
2257:
2255:
2248:
2242:
2241:
2238:
2237:
2235:
2234:
2229:
2223:
2221:
2215:
2214:
2212:
2211:
2206:
2201:
2200:
2199:
2194:
2184:
2178:
2176:
2170:
2169:
2167:
2166:
2161:
2156:
2151:
2145:
2143:
2137:
2136:
2134:
2133:
2128:
2123:
2121:Pipeline stall
2117:
2115:
2106:
2100:
2099:
2096:
2095:
2093:
2092:
2087:
2082:
2077:
2074:
2073:
2072:
2070:z/Architecture
2067:
2062:
2057:
2049:
2044:
2039:
2034:
2029:
2024:
2019:
2014:
2009:
2004:
1999:
1994:
1989:
1988:
1987:
1982:
1977:
1969:
1964:
1959:
1954:
1949:
1944:
1939:
1934:
1928:
1926:
1920:
1919:
1917:
1916:
1915:
1914:
1904:
1899:
1894:
1889:
1884:
1879:
1874:
1873:
1872:
1862:
1861:
1860:
1850:
1845:
1840:
1835:
1829:
1827:
1820:
1812:
1811:
1809:
1808:
1803:
1798:
1793:
1788:
1783:
1782:
1781:
1776:
1774:Virtual memory
1766:
1761:
1760:
1759:
1754:
1749:
1744:
1734:
1729:
1724:
1719:
1714:
1713:
1712:
1702:
1697:
1691:
1689:
1683:
1682:
1680:
1679:
1678:
1677:
1672:
1667:
1662:
1652:
1647:
1642:
1641:
1640:
1635:
1630:
1625:
1620:
1615:
1610:
1605:
1598:Turing machine
1595:
1594:
1593:
1588:
1583:
1578:
1573:
1568:
1558:
1553:
1547:
1545:
1539:
1538:
1531:
1530:
1523:
1516:
1508:
1502:
1501:
1496:
1489:
1488:External links
1486:
1484:
1483:
1450:
1423:
1399:
1378:
1333:
1320:
1313:
1289:
1259:
1234:
1227:
1194:
1182:
1170:
1158:
1146:
1135:
1110:
1098:
1084:
1056:
1023:
985:
975:978-0849310812
974:
954:
916:978-0120121373
915:
899:Academic Press
880:
878:
875:
874:
873:
868:
863:
857:
852:
847:
842:
837:
830:
827:
701:Analog Devices
641:
638:
626:Motorola 56000
508:
505:
495:
492:
491:
490:
485:
479:
478:
477:
468:
467:
466:
444:virtual memory
439:
436:
416:
413:
411:
408:
407:
406:
399:
392:
386:
380:Floating-point
375:
372:
371:
370:
363:
357:
349:
346:
345:
344:
337:
330:
327:
318:addressing in
312:
311:
310:
305:
297:
296:
295:
294:
293:
287:
279:
278:
277:
272:
267:
242:
239:
215:
212:
210:
207:
154:
151:
139:analog signals
76:microprocessor
60:Motorola 56001
56:Motorola 68040
26:
9:
6:
4:
3:
2:
3812:
3801:
3798:
3796:
3793:
3791:
3788:
3786:
3783:
3781:
3778:
3776:
3773:
3772:
3770:
3755:
3752:
3750:
3747:
3743:
3740:
3739:
3738:
3735:
3733:
3730:
3726:
3723:
3721:
3718:
3717:
3716:
3713:
3711:
3708:
3707:
3705:
3701:
3695:
3692:
3690:
3687:
3685:
3682:
3680:
3679:Heterogeneous
3677:
3675:
3672:
3670:
3667:
3665:
3662:
3660:
3657:
3656:
3654:
3652:
3651:Architectures
3648:
3640:
3637:
3636:
3635:
3632:
3630:
3627:
3625:
3622:
3620:
3617:
3613:
3610:
3609:
3608:
3605:
3604:
3602:
3598:
3592:
3589:
3587:
3584:
3580:
3577:
3576:
3575:
3572:
3570:
3567:
3563:
3560:
3559:
3558:
3555:
3553:
3550:
3548:
3545:
3543:
3540:
3538:
3535:
3531:
3528:
3526:
3523:
3522:
3521:
3518:
3517:
3515:
3511:
3505:
3502:
3500:
3497:
3495:
3492:
3491:
3489:
3487:
3483:
3479:
3472:
3467:
3465:
3460:
3458:
3453:
3452:
3449:
3437:
3434:
3432:
3429:
3427:
3424:
3422:
3419:
3417:
3414:
3412:
3409:
3407:
3404:
3402:
3399:
3397:
3394:
3393:
3391:
3387:
3380:
3377:
3375:
3372:
3370:
3367:
3365:
3362:
3360:
3357:
3355:
3352:
3350:
3347:
3346:
3344:
3342:
3336:
3326:
3323:
3321:
3318:
3316:
3313:
3311:
3308:
3306:
3303:
3299:
3296:
3294:
3291:
3289:
3286:
3285:
3284:
3281:
3280:
3278:
3276:
3272:
3266:
3263:
3259:
3256:
3254:
3251:
3250:
3249:
3246:
3242:
3239:
3238:
3237:
3234:
3232:
3229:
3227:
3226:Demultiplexer
3224:
3222:
3219:
3218:
3216:
3214:
3210:
3204:
3201:
3199:
3196:
3193:
3191:
3188:
3186:
3183:
3181:
3178:
3176:
3173:
3172:
3170:
3168:
3164:
3158:
3155:
3153:
3150:
3148:
3147:Memory buffer
3145:
3143:
3142:Register file
3140:
3138:
3135:
3133:
3130:
3128:
3125:
3124:
3122:
3120:
3116:
3108:
3105:
3103:
3100:
3099:
3098:
3095:
3093:
3090:
3088:
3085:
3083:
3082:Combinational
3080:
3079:
3077:
3075:
3071:
3065:
3062:
3058:
3055:
3054:
3052:
3049:
3047:
3044:
3042:
3039:
3034:
3031:
3029:
3026:
3025:
3023:
3020:
3017:
3014:
3011:
3008:
3005:
3002:
3001:
2999:
2997:
2991:
2985:
2982:
2980:
2977:
2975:
2972:
2970:
2967:
2963:
2960:
2958:
2955:
2953:
2950:
2948:
2945:
2943:
2940:
2938:
2935:
2934:
2933:
2930:
2928:
2925:
2924:
2922:
2918:
2912:
2909:
2907:
2904:
2902:
2899:
2897:
2894:
2893:
2891:
2887:
2879:
2876:
2875:
2874:
2871:
2869:
2866:
2864:
2861:
2859:
2856:
2854:
2851:
2849:
2846:
2844:
2841:
2839:
2836:
2834:
2831:
2829:
2826:
2824:
2821:
2819:
2816:
2814:
2811:
2809:
2806:
2804:
2801:
2800:
2798:
2796:
2792:
2782:
2779:
2777:
2774:
2772:
2769:
2766:
2763:
2760:
2757:
2754:
2751:
2748:
2745:
2743:
2740:
2737:
2734:
2732:
2729:
2727:
2724:
2723:
2721:
2719:
2713:
2706:
2703:
2701:
2698:
2695:
2692:
2689:
2686:
2685:
2683:
2677:
2671:
2668:
2666:
2663:
2661:
2658:
2656:
2653:
2651:
2648:
2646:
2643:
2641:
2638:
2637:
2635:
2631:
2624:
2621:
2618:
2615:
2612:
2609:
2607:
2604:
2602:
2599:
2597:
2594:
2592:
2589:
2587:
2584:
2582:
2579:
2577:
2574:
2572:
2569:
2567:
2564:
2562:
2559:
2555:
2552:
2551:
2549:
2546:
2543:
2540:
2539:
2537:
2535:
2531:
2525:
2522:
2520:
2517:
2514:
2511:
2508:
2505:
2502:
2499:
2496:
2493:
2490:
2487:
2482:
2479:
2478:
2476:
2473:
2471:
2468:
2467:
2465:
2463:
2457:
2445:
2442:
2441:
2440:
2437:
2435:
2432:
2428:
2425:
2423:
2420:
2418:
2415:
2413:
2410:
2409:
2408:
2405:
2403:
2400:
2399:
2397:
2395:
2391:
2385:
2382:
2380:
2377:
2375:
2372:
2368:
2365:
2363:
2360:
2359:
2358:
2355:
2353:
2350:
2349:
2347:
2345:
2341:
2335:
2332:
2330:
2327:
2323:
2320:
2319:
2318:
2315:
2311:
2308:
2306:
2303:
2302:
2301:
2298:
2294:
2291:
2289:
2286:
2285:
2284:
2281:
2279:
2276:
2272:
2269:
2267:
2264:
2263:
2262:
2259:
2258:
2256:
2252:
2249:
2247:
2243:
2233:
2230:
2228:
2225:
2224:
2222:
2220:
2216:
2210:
2207:
2205:
2202:
2198:
2195:
2193:
2190:
2189:
2188:
2185:
2183:
2182:Scoreboarding
2180:
2179:
2177:
2175:
2171:
2165:
2164:False sharing
2162:
2160:
2157:
2155:
2152:
2150:
2147:
2146:
2144:
2142:
2138:
2132:
2129:
2127:
2124:
2122:
2119:
2118:
2116:
2114:
2110:
2107:
2105:
2101:
2091:
2088:
2086:
2083:
2081:
2078:
2075:
2071:
2068:
2066:
2063:
2061:
2058:
2056:
2053:
2052:
2050:
2048:
2045:
2043:
2040:
2038:
2035:
2033:
2030:
2028:
2025:
2023:
2020:
2018:
2015:
2013:
2010:
2008:
2005:
2003:
2000:
1998:
1995:
1993:
1990:
1986:
1983:
1981:
1978:
1976:
1973:
1972:
1970:
1968:
1965:
1963:
1960:
1958:
1957:Stanford MIPS
1955:
1953:
1950:
1948:
1945:
1943:
1940:
1938:
1935:
1933:
1930:
1929:
1927:
1921:
1913:
1910:
1909:
1908:
1905:
1903:
1900:
1898:
1895:
1893:
1890:
1888:
1885:
1883:
1880:
1878:
1875:
1871:
1868:
1867:
1866:
1863:
1859:
1856:
1855:
1854:
1851:
1849:
1846:
1844:
1841:
1839:
1836:
1834:
1831:
1830:
1828:
1824:
1821:
1819:
1818:architectures
1813:
1807:
1804:
1802:
1799:
1797:
1794:
1792:
1789:
1787:
1786:Heterogeneous
1784:
1780:
1777:
1775:
1772:
1771:
1770:
1767:
1765:
1762:
1758:
1755:
1753:
1750:
1748:
1745:
1743:
1740:
1739:
1738:
1737:Memory access
1735:
1733:
1730:
1728:
1725:
1723:
1720:
1718:
1715:
1711:
1708:
1707:
1706:
1703:
1701:
1698:
1696:
1693:
1692:
1690:
1688:
1684:
1676:
1673:
1671:
1670:Random-access
1668:
1666:
1663:
1661:
1658:
1657:
1656:
1653:
1651:
1650:Stack machine
1648:
1646:
1643:
1639:
1636:
1634:
1631:
1629:
1626:
1624:
1621:
1619:
1616:
1614:
1611:
1609:
1606:
1604:
1601:
1600:
1599:
1596:
1592:
1589:
1587:
1584:
1582:
1579:
1577:
1574:
1572:
1569:
1567:
1566:with datapath
1564:
1563:
1562:
1559:
1557:
1554:
1552:
1549:
1548:
1546:
1544:
1540:
1536:
1529:
1524:
1522:
1517:
1515:
1510:
1509:
1506:
1500:
1497:
1495:
1492:
1491:
1471:
1467:
1466:
1461:
1454:
1439:
1438:
1433:
1427:
1409:
1403:
1388:
1382:
1374:
1368:
1354:on 2020-09-29
1350:
1343:
1337:
1330:
1324:
1316:
1314:9781351831567
1310:
1307:. p. 2.
1306:
1302:
1301:
1293:
1279:on 2016-03-04
1278:
1274:
1273:The Institute
1270:
1263:
1248:
1244:
1238:
1230:
1224:
1220:
1216:
1212:
1205:
1198:
1191:
1186:
1179:
1174:
1167:
1162:
1155:
1150:
1144:
1139:
1125:on 2020-02-17
1124:
1120:
1114:
1107:
1102:
1095:
1094:
1088:
1070:
1063:
1061:
1045:
1044:
1039:
1032:
1030:
1028:
1012:
1008:
1004:
998:
996:
994:
992:
990:
981:
977:
971:
967:
966:
958:
950:
946:
942:
938:
934:
930:
926:
922:
918:
912:
908:
904:
900:
896:
892:
885:
881:
872:
869:
867:
864:
861:
858:
856:
853:
851:
848:
846:
843:
841:
838:
836:
833:
832:
826:
824:
819:
814:
812:
807:
805:
804:ARM Cortex-A8
801:
796:
794:
789:
785:
783:
779:
775:
773:
769:
767:
763:
759:
755:
752:
748:
744:
742:
738:
734:
730:
726:
722:
718:
714:
710:
706:
702:
698:
696:
692:
687:
683:
680:
676:
673:
669:
667:
664:), use VLIW (
663:
658:
655:produces the
654:
650:
647:
637:
633:
629:
627:
623:
618:
615:
610:
607:
605:
601:
597:
593:
589:
585:
581:
577:
572:
570:
567:released the
566:
561:
559:
558:Motorola 6800
555:
551:
547:
543:
538:
536:
533:
530:
526:
522:
518:
514:
500:
489:
486:
484:
480:
475:
474:
473:
469:
464:
460:
459:
457:
456:
455:
453:
449:
445:
435:
434:is employed.
433:
428:
426:
422:
404:
400:
397:
393:
390:
387:
385:
381:
378:
377:
368:
364:
361:
358:
355:
352:
351:
342:
338:
335:
331:
328:
325:
321:
317:
313:
309:
306:
304:
301:
300:
298:
291:
288:
286:
283:
282:
280:
276:
273:
271:
268:
266:for filtering
265:
262:
261:
259:
255:
254:
252:
248:
245:
244:
238:
236:
233:
232:assembly-code
228:
226:
222:
206:
202:
200:
196:
192:
188:
184:
179:
175:
173:
168:
159:
150:
148:
144:
143:mobile phones
140:
135:
133:
129:
125:
124:mobile phones
121:
117:
113:
109:
105:
101:
97:
93:
89:
85:
81:
77:
73:
69:
61:
57:
53:
48:
41:
37:
36:Akai samplers
32:
19:
3795:Coprocessors
3694:Neuromorphic
3557:Cryptography
3541:
3513:Applications
3436:Chip carrier
3374:Clock gating
3293:Mixed-signal
3190:Write buffer
3167:Control unit
2979:Clock signal
2758:
2718:accelerators
2700:Cypress PSoC
2357:Simultaneous
2174:Out-of-order
1806:Neuromorphic
1687:Architecture
1645:Belt machine
1638:Zeno machine
1571:Hierarchical
1474:. Retrieved
1470:the original
1463:
1453:
1441:. Retrieved
1435:
1426:
1415:. Retrieved
1402:
1391:. Retrieved
1381:
1356:. Retrieved
1349:the original
1336:
1323:
1299:
1292:
1281:. Retrieved
1277:the original
1272:
1262:
1251:. Retrieved
1246:
1237:
1210:
1197:
1185:
1173:
1161:
1149:
1138:
1127:. Retrieved
1123:the original
1113:
1101:
1091:
1087:
1076:. Retrieved
1047:. Retrieved
1041:
1014:. Retrieved
1006:
980:Google Books
978:– via
964:
957:
894:
884:
815:
808:
797:
790:
786:
776:
770:
745:
737:Nucleus RTOS
703:produce the
699:
684:
677:
670:
651:
643:
634:
630:
619:
611:
608:
573:
562:
539:
510:
441:
429:
423:or Modified
418:
391:architecture
374:Program flow
320:ring buffers
229:
217:
209:Architecture
203:
180:
176:
164:
136:
71:
67:
65:
42:arcade board
3221:Multiplexer
3185:Data buffer
2896:Single-core
2868:bit slicing
2726:Coprocessor
2581:Coprocessor
2462:performance
2384:Cooperative
2374:Speculative
2334:Distributed
2293:Superscalar
2278:Instruction
2246:Parallelism
2219:Speculative
2051:System/3x0
1923:Instruction
1700:Von Neumann
1613:Post–Turing
1413:. p. 6
1389:. p. 1
764:as well as
713:multipliers
640:Modern DSPs
622:tight loops
580:NEC μPD7720
507:Development
398:(MAC units)
285:FIR filters
270:dot product
264:convolution
260:operations
191:downlinking
128:disk drives
82:. DSPs are
3769:Categories
3725:chronology
3586:Networking
3341:management
3236:Multiplier
3097:Logic gate
3087:Sequential
2994:Functional
2974:Clock rate
2947:Data cache
2920:Components
2901:Multi-core
2889:Core count
2379:Preemptive
2283:Pipelining
2266:Bit-serial
2209:Wide-issue
2154:Structural
2076:Tilera ISA
2042:MicroBlaze
2012:ETRAX CRIS
1907:Comparison
1752:Load–store
1732:Endianness
1417:2023-12-01
1393:2023-11-13
1358:2019-02-17
1283:2012-03-02
1253:2012-03-02
1180:. p. 4-15.
1129:2020-03-03
1078:2017-06-13
1049:14 October
1016:14 October
877:References
871:Sound card
686:CEVA, Inc.
367:pipelining
167:algorithms
84:fabricated
3715:Processor
3669:Multicore
3275:Circuitry
3195:Microcode
3119:Registers
2962:coherence
2937:CPU cache
2795:Word size
2460:Processor
2104:Execution
2007:DEC Alpha
1985:Power ISA
1801:Cognitive
1608:Universal
1305:CRC Press
949:10070096M
941:858439915
925:0065-2458
672:Freescale
604:voiceband
563:In 1979,
540:In 1978,
452:processes
389:Pipelined
3674:Manycore
3659:Dataflow
3612:C to HDL
3213:Datapath
2906:Manycore
2878:variable
2716:Hardware
2352:Temporal
2032:OpenRISC
1727:Cellular
1717:Dataflow
1710:modified
1367:cite web
1331:. p. 48.
933:59015761
829:See also
751:TriMedia
725:Blackfin
614:TMS32010
584:AT&T
384:datapath
235:routines
187:uplinked
153:Overview
52:NeXTcube
38:and the
3703:Related
3530:DirectX
3389:Related
3320:Quantum
3310:Digital
3305:Boolean
3203:Counter
3102:Quantum
2863:512-bit
2858:256-bit
2853:128-bit
2696:(MPSoC)
2681:on chip
2679:Systems
2497:(FLOPS)
2310:Process
2159:Control
2141:Hazards
2027:Itanium
2022:Unicore
1980:PowerPC
1705:Harvard
1665:Pointer
1660:Counter
1618:Quantum
733:μCLinux
598:(PSTN)
521:TMS5100
494:History
461:Allows
403:looping
205:below.
172:latency
3720:design
3579:scrypt
3486:Theory
3325:Switch
3315:Analog
3053:(IMC)
3024:(MMU)
2873:others
2848:64-bit
2843:48-bit
2838:32-bit
2833:24-bit
2828:16-bit
2823:15-bit
2818:12-bit
2655:Mobile
2571:Stream
2566:Barrel
2561:Vector
2550:(GPU)
2509:(SUPS)
2477:(IPC)
2329:Memory
2322:Vector
2305:Thread
2288:Scalar
2090:Others
2037:RISC-V
2002:SuperH
1971:Power
1967:MIPS-X
1942:PDP-11
1791:Fabric
1543:Models
1476:2 July
1443:2 July
1311:
1249:. IEEE
1225:
972:
947:
939:
931:
923:
913:
866:OpenCL
709:MFLOPS
450:among
316:modulo
258:matrix
90:(MOS)
3537:Audio
3525:GPGPU
3381:(PPW)
3339:Power
3231:Adder
3107:Array
3074:Logic
3035:(TLB)
3018:(FPU)
3012:(AGU)
3006:(ALU)
2996:units
2932:Cache
2813:8-bit
2808:4-bit
2803:1-bit
2767:(TPU)
2761:(DSP)
2755:(PPU)
2749:(VPU)
2738:(GPU)
2707:(NoC)
2690:(SoC)
2625:(PoP)
2619:(SiP)
2613:(MCM)
2554:GPGPU
2544:(CPU)
2534:Types
2515:(PPW)
2503:(TPS)
2491:(IPS)
2483:(CPI)
2254:Level
2065:S/390
2060:S/370
2055:S/360
1997:SPARC
1975:POWER
1858:TRIPS
1826:Types
1458:王珏玢.
1411:(PDF)
1352:(PDF)
1345:(PDF)
1207:(PDF)
1072:(PDF)
811:ASOCS
800:OMAP3
731:like
705:SHARC
657:C6000
565:Intel
292:(FFT)
112:sonar
108:radar
3629:CPLD
3624:ASIC
3619:FPGA
3591:Data
3359:ACPI
3092:Glue
2984:FIFO
2927:Core
2665:ASIP
2606:CPLD
2601:FPOA
2596:FPGA
2591:ASIC
2444:SPMD
2439:MIMD
2434:MISD
2427:SWAR
2407:SIMD
2402:SISD
2317:Data
2300:Task
2271:Word
2017:M32R
1962:MIPS
1925:sets
1892:ZISC
1887:NISC
1882:OISC
1877:MISC
1870:EPIC
1865:VLIW
1853:EDGE
1843:RISC
1838:CISC
1747:HUMA
1742:NUMA
1478:2018
1445:2018
1437:科技日报
1373:link
1309:ISBN
1223:ISBN
1051:2019
1018:2019
970:ISBN
937:OCLC
929:LCCN
921:ISSN
911:ISBN
860:MDSP
754:VLIW
721:SIMD
717:ALUs
715:and
691:MACs
679:XMOS
588:DSP1
569:2920
554:VMOS
532:PMOS
529:7 μm
308:VLIW
303:SIMD
130:and
114:and
50:The
3562:TLS
3520:GPU
3354:APM
3349:PMU
3241:CPU
3198:ROM
2969:Bus
2586:PAL
2261:Bit
2047:LMC
1952:ARM
1947:x86
1937:VAX
1215:doi
1043:EDN
903:doi
772:CSR
758:SoC
646:DMA
586:'s
578:'s
432:DMA
324:FFT
225:x86
223:or
221:ARM
195:SES
86:on
72:DSP
3771::
3288:3D
1462:.
1434:.
1369:}}
1365:{{
1303:.
1271:.
1245:.
1221:.
1209:.
1059:^
1040:.
1026:^
1009:.
1005:.
988:^
945:OL
943:.
935:.
927:.
919:.
909:.
825:.
719:,
628:.
537:.
126:,
110:,
106:,
102:,
98:,
66:A
3470:e
3463:t
3456:v
1527:e
1520:t
1513:v
1480:.
1447:.
1420:.
1396:.
1375:)
1361:.
1317:.
1286:.
1256:.
1231:.
1217::
1192:.
1168:.
1156:.
1132:.
1108:.
1081:.
1053:.
1020:.
982:.
951:.
905::
660:(
369:.
70:(
20:)
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.