Idea Transcript
The Collection
CPUs by Manufacturer CPUs by Class CPUs by Production Year Complete List Trade List Information
Package Types Socket / Slot Types News Archive Communication
Contact
Home
Search
Configure
Links
Imprint
News Archive 2003 News Archive 2006 News Archive 2005 News Archive 2004 December 2003 November 2003 October 2003 September 2003 August 2003 July 2003 June 2003 May 2003 April 2003 March 2003 February 2003 January 2003 December 2003 updates and changes 2003-12-08 This will be the last update for this year, I'm on vacation till end of January. I wish you all a merry Xmas and a happy New Year! The site will have its first birthday on January 14th - feel free to celebrate this in my absence and drink to cpucollection.de :-) Updates and changes: Several of you wanted to also see the chips listed in chronological order, so I implemented the possibility to list the collection content by manufacturing date. Additionally you can adjust the date format used in the production week overviews to your preference. Please use 'Configure' to set. Because the navigation menu was getting kind of unstructured with its 15 top level entries I added subsections and moved some links to the top of the page. I hope you like it. And there are 8 additions to the collection today: several math coprocessors Intel C8087 The 8087 was the first coprocessor that Intel made available for the 80x86 family. It was introduced in 1980 and therefore does not have full compatibility with the IEEE-754 standard for floating-point arithmetic (which was finally released in 1985). It complements the 8088 and 8086 CPUs and can also be interfaced to the 80188 and 80186 processors. The 8087 is implemented using NMOS. It comes in a 40-pin CERDIP and is available in 5 MHz, 8 MHz, and 10 MHz versions. Power consumption is rated at max. 2400 mW. A neat trick to enhance the processing power of the 8087 for computations that use only the basic arithmetic operations (+,-,*,/) and do not require high precision is to set the precision control to single-precision. This gives one a performance increase of up to 20%. With the help of an additional chip, the 8087 can in theory be interfaced to an 80186 CPU. The 80186 was used in some PCs (e.g. from Philips, Siemens) in 1982/1983. Intel D80C187-16 The 80187 is a rather late coprocessor designed to support the 80C186 embedded processor. It was introduced in 1989 and implements the complete 80387 instruction set. It is available for 12.5 and 16 MHz operation. Intel C80287-3 Intel D80287-6 The 80287 is the original Intel coprocessor for the 80286 and was introduced in 1983. It uses the same execution unit as the 8087 and therefore has the same speed (sometimes slower due to additional overhead in CPU coprocessor communication). As the 8087, it does not provide full compatibility with the IEEE-754 floating point standard. There are 6 MHz, 8 MHz, and 10 MHz versions. The 80287 has been replaced by its CMOS successor, the Intel 287XL, which has been introduced in 1990. It is based on the 387 core and therefore much faster than the 80287. IIT 2C87-12 The IIT 2C87 was the first Intel 287 clone. It was introduced in 1989 and has about the same speed as the Intel 287XL. The 2C87 implements the full 387 instruction set. It is available for speeds of up to 20 MHz. Intel A80387DX-20 The Intel 387DX is the second generation Intel 387 and was introduced in 1989. This version is done in a more advanced CMOS process than the 80387 that enables the coprocessor to run at a maximum frequency of 33 MHz, while the 80387 had a maximum frequency of 25 MHz. The 387DX is about 20% faster than the 80387 at the same clock frequency. The Intel 387DX (and its predecessor 80387) are the only 387 coprocessors that support asynchronous operation of CPU and FPU. The 387 consists of a bus interface unit and a numerical execution unit. The bus interface unit always runs at the speed of the CPU clock. IIT 3C87-20 This IIT chip was introduced in 1989, about the same time as the Cyrix 83D87. Both coprocessors are faster than Intel's 387DX coprocessor. The IIT 3C87 also provides extra functions not available on any other 387 chip. It has 24 user-accessible floating-point registers organized into three register banks. Three additional instructions (FSBP0, FSBP1, FSBP2) allow switching from one bank to another. The register bank's main purpose is to aid the fourth additional instruction the 3C87 has (F4X4), which does a full multiply of a 4x4 matrix by a 4x1 vector, an operation common in 3D-graphics applications. There are very few applications that make use of it when an IIT coprocessor is detected at run time (among them Silver Screen and Fast-CAD 3-D). The 3C87 is implemented in an advanced CMOS process and has low power requirements, typically about 600 mW. Like the 387 'clones' from Cyrix and ULSI, the 3C87 does not support asynchronous operation of the CPU and the coprocessor, but always runs at the full speed of the CPU. It is available in 16, 20, 25, 33, and 40 MHz versions. Cyrix 387DX+ The Cyrix 387DX+ is a CX-83D87 for the European marked, relabeled 387DX+. It is the fastest 387compatible coprocessor and provides up to 50% more performance than the Intel 387DX. It also offers the most accurate transcendental functions of all coprocessors and is the 387 clone with the highest degree of compatibility to the Intel 387DX. -- Most of the information provided here is from Norbert Juffa's excellent FPU reference "Everything you always wanted to know about math coprocessors", please refer to 'copro16a.txt' for more information.
November 2003 updates and changes 2003-11-22 22 new additions to the collection today: 80286 processors The 80286 was introduced by Intel on February 1, 1982. As the 80186/80188 CPUs were not really significant to personal computing, the 80286 was Intel's next step processor for micro computers. Intel added four more address lines to the 8086/80186 design. The 8086, 8088, 80186, and 80188 all contained 20 address lines, giving these processors one megabyte of addressibility (2^20 = 1MB). The 80286, with its 24 address lines, gives 16 megabytes of addressibility (2^24 = 16 MB). The most substantial difference between the 80286 and the 8086/8088 is the addition of a protected mode. In protected mode, segment registers became pointers into a table of memory descriptors rather than being a direct part of the address. Among other things, protected mode allows safe execution of multiple programs at once by protecting each program in memory. DOS normally operates in real mode, in which segment registers act just as they do in the 8086/8088. Protected mode is used by Microsoft Windows, IBM's OS/2 and UNIX. (For an introduction to protected mode please refer to this source) The 80286 is a much more powerful CPU than the 8086, offering 3-6 times the performance of it. The 6 MHz 80286 is the CPU of the IBM AT (Advanced Technology), which also introduced a 16-bit motherboard and 16bit expansion bus to the PC world. The IBM AT was introduced in 1985 - three years after introduction of the 80286. With the 80286, the first "chipsets" were introduced. The computer chipset is a set of chips that replaced dozens of other peripheral chips while maintaining identical functionality. Chips and Technologies became one of the first popular chipset companies. Intel second-sourced the 80286 to ensure an adequate supply of chips to the computer industry. AMD, IBM, and Harris were known to produce 80286 chips as OEM products; while Siemens, Fujitsu, and Kruger either cloned it or were also second-sources. Between these various manufacturers, the 80286 was offered in speeds ranging from 6 MHz to 25 MHz: Intel: 6 - 12.5 MHz Siemens: 8 - 16 MHz AMD: 8 - 20 MHz Harris: 10 - 25 MHz The 80286 was typically made in 3 package versions, each with 68 contacts: a PGA-, CLCC- and a PLCCpackage. New items in the collection: Intel CG80286-6 C: 6 MHz, PGA Intel R80286-8: 8 MHz, CLCC Intel N80286-10: 10 MHz, PLCC (1 item for trade) Intel R80286-12: 12 MHz, CLCC Intel SAB 80286-8R: A Siemens OEM made Intel 80286 with SAB markings AMD R80286-8/C2: 8 MHz, CLCC AMD R80286-10/C2H: 10 MHz, CLCC AMD R80286-10/S: 10 MHz, CLCC AMD N80L286-10/S: 10 MHz, PLCC AMD N80L286-10/C2H: 10 MHz, PLCC AMD N80L286-16/S: 16 MHz, PLCC If you got any information on the /C2, /C2H or /S suffixes, please let me know. Fujitsu MBL80286-6: 6 MHz, PGA Harris CG80C286-16: 16 MHz, PGA Harris CS80C286-20: 20 MHz, PLCC (1 item for trade) Harris CS80C286-25: 25 MHz, PLCC IBM INT:286/12 A strange 12 MHz IBM made PLCC 80286 with fixed 'Lego'-style cooler on top Siemens SAB 80286-R: 10 MHz, CLCC Siemens SAB 80286-A: 10 MHz, PGA Siemens SAB 80286-1-A: 10 MHz, PGA Siemens SAB 80286-1-N: 10 MHz, PLCC Siemens SAB 80286-1-N/S: 10 MHz, PLCC Siemens SAB 80286-12-N: 12 MHz, PLCC
October 2003 updates and changes Due to a power loss at my site hoster the root filesystem on cpu-collection.de has been damaged and could not be restored. The site therefore was offline for about 1 week. Thanks to Grampa, webmaster at cpu-museum.de, cpu-collection.de has been transfered onto the cpu-museum.de server and went online again. There have been no updates in October.
September 2003 updates and changes 2003-09-30 14 new additions to the collection today: a variety of Cyrix processors and math units Cyrix Cx486DLC-33GP (black version) The Cyrix 486DLC is pin compatible to the Intel i386DX. While the internals of the 486DLC are roughly equivalent to those in the i486SX, the bus interface is identical to that of the Intel 386DX, respectively to allow easy replacement of the Intel CPUs by the Cyrix chips. The overall execution speed is about same as with an i486SX, but if you add a matching Cyrix FPU, e.g. a Cyrix 83D87, the speed is about 50% faster than an i486DX running at the same frequency. For more information on the Cyrix 486DLC/SLC processors please refer to Norbert Juffa's excellent DLC/SLC info summary and performance comparison. I have some of these processors for trade, if you you are interested, please make an offer. Cyrix Cx486DLC-33GP OEM with golden logo A special OEM version of the above chip with golden Cyrix logo. I have some of these processors for trade, if you you are interested, please make an offer. Cyrix Cx486DLC-40GP The 40 MHz version of the above processor. Cyrix Cx486DRx225/50GP Cyrix DRx2 chips are clock-doubled versions of the Cyrix DLC processors and thus the most effective way to upgrade an i386DX based system. Cyrix Cx486DX-33GP The Cyrix Cx486DX was designed to be completely software-compatible with Intel's i486DX processor. It is not just an Intel clone but a completely new design by Cyrix. Compared to the i486DX the Cx486DX had an improved cache algorithm and therefore was faster in specific applications. The price for a Cx486DX was a little lower than for Intel's 486DX and made the processor a very common i486DX alternative. Cyrix Cx486DX-V33QP A low voltage version of the Cx486DX-33 (3.3V compared to 5V in the standard version). This processor was intended to be used in mobile devices and comes in a surface-mount PQFP version. This item here is soldered on a small adaptor with standard socket 2/3 pinout. Cyrix Cx486DX-40 w/ Cooler, newer Logo Another Cx486DX, this time with 40 MHz core frequency and a fixed green cooler. The logo on the cooler is the second and most common version of Cyrix' DX logos. The chips below and above in the list are labeled with the first and the final version of this DX logo. Cyrix Cx486DX2-80 w/ Cooler The clock doubled Cx486DX-2 with 80 MHz core frequency and green cooler on top. Cyrix Cx486DX2-V80GP A low voltage version of the Cx486DX2-80 (3.3V compared to 5V in the standard version). Cyrix M II-300GP, 66MHz Bus, Black A National Semiconductors made Cyrix MII with black heat spreader. After Cyrix was purchased by National Semiconductor in 1998 they terminated their foundry agreement with IBM. National fabricated their own chips and the 6x86MX was re-named MII. Compared to early 6x86MX CPUs the MII had some enhancements to its processor core. Heat output was reduced allowing it to rise to clock speeds over 220MHz. The MII also required non-standard bus speeds at 75 or 83MHz on socket 7 boards and therefore had some troubles with stability. Cyrix MediaGX GXm-200GP The 200 MHz version of Cyrix'/National's MediaGX chip. National's strategy to concentrate on the budget market after aquiring Cyrix lead to the introduction of the MediaGX. It is a highly integrated chip based on the original 5x86 core, featuring graphics, sound and RAM controllers and a PCI bus interface. For this reason, it was called a "PC on a chip". It requires a special motherboard and is not pin compatible with the Pentium, but it was the cheapest route into a Pentium class system available on the market. Cyrix FPU 387DX-33 The predecessor of the CX-83D87. The Cyrix 387DX was soon replaced by the much more advanced CX83D87: Cyrix FPU FasMath CX-83D87-33-GP, black Cyrix FPU FasMath CX-83D87-40-GP The CX-83D87 was introduced in 1989. It is the fastest 387-compatible coprocessor and provides up to 50% more performance than the Intel 387DX. The 83D87 also offers the most accurate transcendental functions of all coprocessors. It is the 387 clone with the highest degree of compatibility to the Intel 387DX. Unlike the Intel 387DX, the 83D87 (and all other 387-compatible chips as well) does not support asynchronous operation of CPU and coprocessor. To reduce power consumption the 83D87 features advanced power saving features. Those portions of the coprocessor that are not needed are automatically shut down. If no coprocessor instructions are being executed, all parts except the bus interface unit are shut down.
2003-09-13 ...::: cpu-collection.de now has more than 400 chips online :::... 11 new additions to the collection today: some Eastern Bloc processors ZMD U830Cp An 8-bit slice processor for an East German PDP-11 clone. The chip was made by ZMD (Zentrum für Mikroelektronik Dresden) and used in SKR K 1600 (K1610/K1620/K1630) machines. If you got some information on this processor please contact me VEB Funkwerk Erfurt U808D The U808D is an Intel 8008 clone made in former German Democratic Republic by VEB Funkwerk Erfurt. It was used in one of GDR's first computers, a Robotron K 1510 from 1977. My U808D came in its original box, which is pretty much like a matchbox with metal coated paper inlay. Here's a picture of it. I have one U808D for trade, make a good offer if you are interested. MME UB880D MME UA880D The 880 series of East German processors are clones of Zilog Z80 processors. The UB880 is a Z80 clone, the UA880 is a Z80 A equivalent and the VB880 is a Z80 version with enhanced temperature range (-25° to +85° C) for industrial and military purposes. They were produced by VEB Mikroelektronik "Karl Marx" in Erfurt in the 80s. The 880 CPUs were used in several GDR microcomputers like the RFT KC 85 series, the Robotron KC 87 and the Z 9001. Tesla MHB8080A An Intel 8080A clone produced by Tesla in Czechoslovakia. It was used in Tesla PMD 85 computers. UNITRA CEMI MCY 7880 An i8080 clone, made by Naukowo-Produkcyjne Centrum Polprzewodników UNITRA CEMI, Warsaw, Poland. It was designed and produced in 1980. The MCY 7880 was the only microprocessor manufactured in Poland. In the early 80s UNITRA CEMI designed clones of i8086 CPUs and i8051 microcontrollers, but these chips have never been produced. KP580BM80A The KP580BM80A is an i8080 clone produced in the former Soviet Union. It comes in a pretty heavy package with strange steel inserts on the back. I don't know why it had to be enforced like this...
some Intel 960 processors Intel i960 N80960SA20 Intel i960 A80960CF-30 Intel i960 A80960HD66 Intel i960 GC80960RM100 The i960 is a superscalar 32-bit RISC microprocessor intended for high end embedded devices. It features a very clean embedded architecture, not designed for high level applications, but very effective and scalable. The i960 is used in a wide variety of devices like graphics controllers, network switches, X-terminals (like the HP Envizex series), SCSI- and RAID-controllers. One BGA i960 GC80960RM100 is for trade, please make an offer if you are interested.
August 2003 updates and changes 2003-08-14 23 new additions to the collection today: some classic processors from several manufacturers Intel P4040 The Intel 4040 is the successor to the Intel 4004. It was introduced in 1974. The 4040 added 14 instructions, larger stack (8 level), 8K program space, 8 more registers, and interrupt abilities. It was used primarily in games, instrumentation and point of sale terminals, development- and control equipment. The 4040 family is also referred to as the MCS-40. Intel D8008-1 The Intel 8008 was the first 8-bit microprocessor, introduced in April, 1972. It was designed for use as a terminal controller for Control Terminal Corporation (CTC) to replace their TTL logic chips cpu design. The chip has a single 8-bit bus and requires a very large amount of external logic to support it. It has a little less processing speed (in terms of MIPs) than the Intel 4004 and Intel 4040, but it processes 8-bits at a time and can access significantly more RAM. These facts give it 3 to 4 times the true processing power of the 4-bit chips. The 8008 was an acceptable design for controller and CRT terminal use, but it was too difficult to use for most other tasks. A few early computer designs (e.g. the Mark-8, the Scelbi 8H and the Micral) were based on it, but most would use the later and greatly improved Intel 8080 instead. AMD AM9080ADC / C8080A Mitsubishi M5L8080AP NEC D8080AFC The Intel 8080 was the successor to the Intel 8008. It provides a 16-bit address bus and an 8-bit data bus. Because of its increased function and power over the 4004 and 8008, the 8080 was the first widely accepted processor and is generally considered to be the first truly usable microprocessor design. It was used in the Altair 8800, the first personal computer, and many other early computers. The Intel 8080 was introduced in April 1974. New in my collection are the listed clones or second source versions by AMD, Mitsubishi and NEC. Mitsubishi M5M80C85AP-2 NEC D8085AC-2 OKI M80C85A OKI M80C85A-2 Siemens SAB 8085 A-P Siemens SAB 8085AH2P Intel made the 8085 as a significant improvement on the 8080, both in performance and handling issues. It has improved hardware by only using +5V power (the 8080 required +5V, -5V and +12V), and clock generator and bus controller circuits on-chip. The 8085 was a transition design on the way to the 16-bit 8086. Above are some second source 8085s from various manufacturers. Intel D8086-1 AMD D8086-2 NEC V30 D70116D-5 The 8086, officially called iAPX 86, was Intel's first venture in 16-bit computing in 1978. It has a totally different instruction set, different hardware architecture, and much greater capability. It was in no direct way related to any of the previous processors, it just happened to be the next number in sequence. The 8086 gave rise to the x86 architecture. The first commercial microcomputer built on the basis of the 8086 was the Mycron 2000 (from Norway). This computer was used by Digital Research as the development platform for the CP/M-86 operating system. The 8086-1 is an enhanced version of the 8086 with 10 MHz, the 8086-2 with 8 MHz. AMD D8088 AMD P8088 AMD P8088-1 Fujitsu MBL8088-2 Harris CP80C88 Intel P8088-2 NEC D8088D-2 NEC V25 D70325L-10 (Japan) Siemens SAB 8088-P The 8088 is a reduced and cheaper version of the 8086 for the 8-bit market. It is an 8086 internally with an 8-bit data bus - still a 16-bit processor but with an 8-bit external bus to make connection to hardware less difficult. It was introduced in 1979 and used as the CPU in the original IBM PC and XT. The 8088-1 is an enhanced version of the 8088 with 10 MHz, the 8088-2 with 8 MHz. The NEC V25 is an 8088 with some additional instructions and integrated peripherals.
July 2003 updates and changes 2003-07-19 17 new additions to the collection today: Cyrix 6x86 and MediaGX processors Cyrix 6x86-P150+GP Cyrix 6x86-PR166+GP Cyrix entered the fifth generation processor market with the 6x86 processor, formerly projected as the M1. As a successful and cheaper (often less than half the cost) alternative to the Intel Pentium, it is pin- and voltagecompatible with it. Cyrix gave it the 6x86 name in reference to some of its more advanced features, which it calls "sixth generation". In reality, the processor is comparable in power and architecture to the fifth-generation Pentium. The 6x86 is not a Pentium clone. Clones are exact or near-exact copies, usually being reversed engineered or based on licensed code. The 6x86 is based on an original Cyrix design. It incorporates several advanced architectural features that allow it to outperform a Pentium of equal clock speed. For this reason Cyrix helped invent the "P rating" system. This was an advantage to help people make a valid comparism but also caused some confusion when setting the appropriate clock speed on the motherboard. The Cyrix 6x86 range has the most powerful processor core of any x86 processor of its generation. Unfortunately, like the AMD K5, it also has a very slow floating point math capability and so was a very poor games and 3D performer. The processor is designed for Socket 7 and is available in several clock speeds, some of them rather unusual. The 6x86 PR200 (150 Mhz) processor introduced non-standard bus speeds up to 75 MHz - many motherboards and PCI cards did not support this speeds: Since the PCI bus runs at half of the mainboard bus speed, you are increasing the PCI bus to 37.5 MHz with a bus speed of 75 MHz. Several PCI cards, especially some graphics cards, did not run correctly with this overclocked bus. The Cyrix 6x86 is manufactured by IBM and SGS Thomson, because Cyrix doesn't have its own manufacturing facilities. IBM and ST also sold the 6x86 under their own name. IBM labeled 6x86 processors: IBM 6x86 P166+, 3.5V - a higher voltage 6x86 P166 (3.3V with the standard 6x86 P166) IBM 6x86 w/o P-Rating - a 6x86 P166 with different marking, maybe an OEM version and with SGS Thomson label: SGS-Thomson ST6x86-P120+
Cyrix 6x86L-PR166+GP The earlier versions of the 6x86 had problems with heat and power consumption. Responding to customer complaints, Cyrix came out with a lower powered version called the 6x86L. This processor reduces power consumption by 25% or more compared to the original 6x86. It uses a smaller, 0.35 micron circuit size and split voltage much like the Pentium MMX (3.3V external, 2.8V internal).
Cyrix 6x86MX-PR166 Cyrix 6x86MX-PR200 with 66MHz and with 75MHz bus speed Cyrix 6x86MX-PR233 The successor to the 6x86 and 6x86L is the 6x86MX (M2), introduced in June 1997 and intended to compete with AMD's K6 and Intel's Pentium MMX. The 6x86MX is an evolutionary step from the 6x86 and is very similar to it in internal function. The 6x86MX supports the MMX extension and also features several other improvements over the original 6x86 chip. The internal cache was increased from 16 to 64 KB, a 256 byte additional pre-cache was added to help improve efficiency in how the regular level 1 cache is operated, the Branch Prediction was improved and the internals of the chip have been optimized for 32-bit operation. IBM labeled 6x86MX processors: IBM 6x86MX PR233 IBM 6x86MX PR300 IBM 6x86MX PR300 w/ black heat spreader
Cyrix M II-233GP, 66MHz Bus Cyrix M II-333GP, 83MHz Bus, Engineering Sample After Cyrix was purchased by National Semiconductor in 1998 they terminated their foundry agreement with IBM. National fabricated their own chips and the 6x86MX was re-named MII. Compared to early 6x86MX CPUs the MII had some enhancements to its processor core. Heat output was reduced allowing it to rise to clock speeds over 220MHz. The MII also required non-standard bus speeds at 75 or 83MHz on socket 7 boards and therefore had some troubles with stability. The ES marked M II-333GP in the collection is an Engineering Sample chip.
Cyrix MediaGX GXm-180GP Cyrix MediaGX GXm-233GP National's strategy to concentrate on the budget market lead to the introduction of the MediaGX. It is a highly integrated chip based on the original 5x86 core, featuring graphics, sound and RAM controllers and a PCI bus interface. For this reason, it was called a "PC on a chip". It requires a special motherboard and is not pin compatible with the Pentium, but it was the cheapest route into a Pentium class system available on the market. There is one 180 MHz MediaGX in my trade list.
2003-07-06 13 new additions to the collection today: processor modules, mobile CPUs and a little mobile package history: HP PA-RISC 7200 An HP CPU module from a J200 series HP server. The PA-RISC 7200 was used in C- and D-class servers and J200 and 210 workstations. It is leveraged from the original PA7100 design, big parts of the core were just shrunk for the new 0.55 micron process. The FPU was taken over completely unchanged, retaining the same latencies for addition and multiplication even at a higher clock rate. It also acquired the cache design, e.g. had (for the time) big external caches clocked at full CPU speed. This chip was aimed at high-performance general-purpose applications but also on specialized applications that used large working sets which could take advantage of the high-bandwidth bus interface. Intergraph Clipper C4 with Intergraph Clipper C4 FPU The Clipper architecture was originally developed at Fairchild Semiconductor in Oct. 1985, and began shipping in 1986. The architecture was acquired by Intergraph Corporation in 1987. Intergraph is a manufacturer of highend graphics workstations. The C400 processor combined two key architectural techniques to achieve a new level of performance. The first – superscalar instruction dispatch – boosted performance by executing more than one operation at a time. The second technique – superpipelined operation – used high clock rates to boost performance. Since complex tasks such as floating-point multiplications could not be done in a single, short clock cycle, the C400 would break long operations into many short pieces called stages. Many 2nd generation RISC processors used one or the other of these techniques, but only the C400 Clipper used both. Intel i486 SX SB80486SX-33 Mobile Module Intel i486 DX FC80486DX4-100 Mobile Module Two PQFP 486 processors on mobile modules. Intel Pentium P54C PP100 Mobile Module Intel Pentium P54C PP133 Mobile Module Intel Pentium P54C PP133 Mobile Module, different version TCP (Tape Carrier Package) Pentiums on adaptor boards to be plugged onto notebook connectors. Before the introduction of the MMC-1 (Mobile Module Connector 1) there were many methods to connect the CPU to a notebook. Some of them were soldered on directly, making it almost impossible to upgrade them, some used standard desktop size sockets (which took a lot of space), some were fixed on a small daughter board and attached to the mainboard. A variety of proprietary module connectors was introduced, providing a possibility to upgrade to a higher clocked version of the initial processor. Intel Pentium MMX TT233 Tillamook MMC-1 This module is much more than just a CPU. The Intel Mobile Module contains CPU, Northbridge, L2 cache and voltage regulator. Intel introduced these mobile modules with the Pentium Processor including MMX and TX Chipset. Later Intel upgraded the MMC-1 (Mobile Module Connector 1) to a Pentium II Processor with additional level two cache and BX Chipset. The notebook itself had the Southbridge and all the peripheral chips like VGA card, sound card, PCMCIA controller, charger, etc. mounted on the notebook motherboard. The MMC-1 connector has 4 rows and a total of 280 pins. Intel Pentium II 266MHz MMC-2 Intel Pentium II 333MHz MMC-2 While the MMC-1 module used a PCI interface to connect to the notebook motherboard the MMC-2 (Mobile Module Connector 2) uses an AGP/PCI interface. The MMC-2 has 10 rows and 400 pins. It is not possible to use an MMC-1 module in an MMC-2 slot or vice versa. The BX chipset used with the mobile modules always supports AGP, but the connector of the MMC-1 doesn't allow it's use. The MMC-2 processors kicked-off AGP graphics inside of a notebook for 3D graphics and better DVD playback. On the 266MHz version the L2 cache is not included on the processor die. You see 2 cache RAMs and a controller chip on the backside of the module. The 333MHz version CPU includes the L2 cache. Intel Pentium II KP 400/256 Mobile Intel P III Celeron KP 400/128 Mobile To reduce costs and provide a more easy way for processor upgrades, the mobile modules were superseded by just the processors packed in small packages. The BGA (Ball Grid Array) was introduced for mobile CPUs and in low-cost notebooks these BGA processors were soldered directly on the mainboard. To make a notebook upgradeable, the BGA processors had to be mounted on adaptors to fit into standardized micro-sockets on notebook mainboards. This CPUs use a BGA package and are soldered on small PCBs with pins to fit into a standard notebook socket. The complete package is called Micro-PGA1 and has 615 pins (243 of them are not connected). The second generation Micro-PGA - also a BGA chip on a PCB with pin array, the Micro-PGA2, is reduced to 495 pins with just 24 of them not connected. Most all-in-one notebooks in the Pentium III era used a MicroPGA2 processor.
June 2003 updates and changes 2003-06-23 13 new additions to the collection today: filling some gaps in the AMD line of processors: AMD Am386DX/DXL-25 AMD Am386DX-33 AMD 386 processors were offering approximately equal performance like Intel i386 and were significantly less priced. AMD Am486DX4-120SV8B The last AMD 486DX CPU, featuring 120 MHz clock frequency and made for 40 MHz boards. It was introduced in September 1994. This item here was made in January 1996. AMD5k86-P90 SSA/5-90ABQ AMD-K5-PR100ABR AMD-K5-PR133ABQ AMD's 5th generation x86 implementation, introduced in March 1996. The processor, later renamed to K5, was eagerly awaited and it was hoped that it would provide a viable alternative to the Pentium early in the Pentium's life cycle. Unfortunately, AMD delivered the processor over a year late and at much lower clock speeds than had been originally anticipated. As a result instead of being the "Pentium killer" AMD had hoped for, the K5 was positioned as a low-cost Pentium alternative. The K5 is, internally, a very advanced processor, the most advanced of the fifth-generation chips. Internally it is more comparable to the Pentium Pro. It is an x86 translation/emulation processor, decoding x86 instructions into RISC-like microinstructions and executing them on a 6-pipeline internal core. This allows the K5 to achieve higher performance than a Pentium of the same speed. AMD-K5-PR200ABX Officially the K5 line of processors went up to a 116MHz version (PR166) and was discontinued when the K6 was released in April 1997. However, there are some K5 PR200 around, probably as sample chips for mainboard manufacturers or for testing purposes. The K5 PR200 never has been officially released and is not mentioned in AMDs technical documentation (the fastest K5 in the Data Sheets is an AMD-K5-PR166ABX). AMD-K6-233ANR AMD-K6/300AFR After AMD's plans with the K5 turned into a debacle, AMD again claimed that they had the solution to make them more than just discount alternatives to Intel: A chip with the power of a Pentium Pro, with MMX support and intended to fit into a standard socket 7. AMD delivered its much anticipated K6 chip in April of 1997, beating Intel's Pentium II to the market by a month. AMD showed their commitment to establishing themselves in the market when they purchased NexGen in 1996 and with it, the design for the NX686 processor. NexGen had been intending to market this chip in its own socket, but AMD changed the design to fit the standard socket 7, added MMX support, and renamed it the K6. Despite the name implying a design evolving from the K5, it is in fact a totally different design that was created by the NexGen team and adapted after the AMD purchase. AMD Athlon K7800CPRBBA The Athlon made its debut on August 21, 1999. The original Athlon core revision, code-named "K7", was available in speeds of 500 to 650 MHz at its introduction and was later sold at speeds up to 1000 MHz. Internally, the Athlon was essentially a major reworking of the K6 processor core designed for compatibility with the EV6 bus protocol (first used on DEC's Alpha 21264 RISC processor). AMD dramatically improved the floating-point unit from the K6 and put a large 128K level 1 cache on the chip. Like on the Intel Pentium II there was 512k of secondary cache, mounted on the CPU module and running at a lower speed than the core. The resulting processor was the fastest x86 in the world. Various different versions of the Athlon held this distinction continuously from August 1999 until January 2002. AMD Duron D700AUT1B AMD Duron D750AUT1B AMD Duron D800AUT1B The AMD Duron was released in the summer of 2000 as a low-cost alternative to the Athlon processor and the Pentium III and Celeron processor lines from Intel. The Duron is pin-compatible with the Athlon and operating on the same motherboards. It has the same 128K of level 1 cache as the Athlon, but only 64K of level 2 cache, as compared to 256K on the more expensive chip. Because of this, the Duron generally lags behind the Athlon on business applications, but keeps up in floatingpoint operations thanks to its powerful FPU, which is identical to the Athlon's. The original Duron was limited to operating on a 100 MHz front-side bus speed, while the Athlon at the time could run on a bus clock of 133 MHz. Later Athlons supported a 200MHz bus.
2003-06-10 11 new additions to the collection today: math processors Intel C8087-3 and Intel C8087-2 This was the first coprocessor that Intel made available for the 80x86 family. It was introduced in 1980 and therefore does not have full compatibility with the IEEE-754 standard for floating-point arithmetic (which was finally released in 1985). It complements the 8088 and 8086 CPUs and can also be interfaced to the 80188 and 80186 processors. The 8087 is implemented using NMOS. It comes in a 40-pin CERDIP and is available in 5 MHz, 8 MHz, and 10 MHz versions. Power consumption is rated at max. 2400 mW. A neat trick to enhance the processing power of the 8087 for computations that use only the basic arithmetic operations (+,-,*,/) and do not require high precision is to set the precision control to single-precision. This gives one a performance increase of up to 20%. With the help of an additional chip, the 8087 can in theory be interfaced to an 80186 CPU. The 80186 was used in some PCs (e.g. from Philips, Siemens) in 1982/1983. Both 8087 chips are contributed by Christian Lederer. Thanks a lot! Intel D80287-8 and Intel C80287-10 This is the original Intel coprocessor for the 80286, introduced in 1983. It uses the same internal execution unit as the 8087 and therefore has the same speed (actually, it is sometimes slower due to additional overhead in CPU-coprocessor communication). As with the 8087, it does not provide full compatibility with the IEEE-754 floating point standard released in 1985. The 80287 was manufactured in NMOS technology, and is packaged in a 40-pin CERDIP. There are 6 MHz, 8 MHz, and 10 MHz versions. Power consumption can be estimated to be the same as that for the 8087, which is 2400 mW max. The 80287 has been replaced in the Intel 80x87 family with its faster successor, the CMOS-based Intel 287XL, which was introduced in 1990 (see below). The C80287-10 was a donation by Christian Lederer, thanks again. Intel C80287XL This chip is Intel's second-generation 287, first introduced in 1990. Since it is based on the 80387 coprocessor core, it features full IEEE 754 compatibility and faster instruction execution. Intel claims about 50% faster operation than the 80287 for typical benchmark tests. Compared with benchmark results for the AMD 80C287, which is identical to the Intel 80287, the Intel 287XL performed 66% faster than the AMD 80C287 on a fractal benchmark and 66% faster on the Whetstone benchmark. Since the 287XL has all the additional instructions and enhancements of a 387, most software automatically identifies it as an 80387-compatible coprocessor and therefore can make use of extra 387-only features, such as the FSIN and FCOS instructions. The 287XL is manufactured in CMOS and therefore uses much less power than the older NMOS-based 80287. At 12.5 MHz, the power consumption is rated at max. 675 mW, about 1/4 of the 80287 power consumption. The 287XL is available in either a 40-pin CERDIP (ceramic dual inline package) or a 44 pin PLCC (plastic leaded chip carrier). (This latter version is called the 287XLT and intended mainly for laptop use.) The 287XL is rated for speeds of up to 12.5 MHz. Cyrix FasMath CX-83S87-20-KP The 83S87 is the SX version of the Cyrix 83D87. Just as the 83D87 is the fastest 387-compatible coprocessor, the Cyrix 83S87 is the fastest of the 387SX compatible coprocessors, as well as providing the most accurate transcendental functions. 83S87 chips manufactured after 1991 use the internals of the Cyrix 387+, the successor to the original 83D87. The Cyrix 83S87 is ideally suited to be used with the Cyrix Cx486SLC CPU, a 486SX compatible CPU which is a replacement chip for the Intel 386SX CPU. The 83S87 is packaged in a 68-pin PLCC and is available in 16, 20, 25, and 33 MHz versions. Due to the advanced power saving features of the Cyrix coprocessor, the typical power consumption of the 20 MHz version is only about 350 mW, while maximum power dissipation is 1.6 W. Cyrix Cx487DLC-33GP The 487DLC is a math processor for the Cyrix 486 DLC series of 386DX replacement processors. The Cx487DLC is an OEM only chip and has not been sold separately. Cyrix Cx87DLC-33QP Another co-processor for 386/486 SLC/DLC systems. This chip is a 40-pin PQFP package mounted on a 68pin PGA PCB to fit into a standard socket. Chips and Technologies Super Math J38700DX B Produced by Chips & Technologies, this was the latest entry into the 387-compatible marketplace. Originally announced in October, 1991, it had apparently not been available to end-users before the third quarter of 1992. The product was discontinued after only a few month since C&T stopped all work on their CPU and coprocessor development. Compatibility with Intel products is very good, even for the more arcane features of the 387DX and comparable to the coprocessors from Cyrix. Overall, accuracy of the transcendentals is slightly better than on the Intel 387DX. The C&T 38700DX showed performance at about 90-100% the level of the Cyrix 83D87, which is the 387 clone with the highest performance. For floating-point-intensive benchmarks, the C&T 38700DX provides up to 50% more computational performance than the Intel 387DX. However, as with all other 387 compatible coprocessors, the speed advantage over the Intel 387DX is far less significant in real applications. The SuperMATH 38700DX is implemented in 1.2 micron CMOS with on-chip power management, which makes for low power consumption. The 38700DX is packaged in a 68-pin ceramic PGA and available in speeds of 16, 20, 25, 33, and 40 MHz. IIT FPU 3C87-25 and IIT FPU 3C87-33 This IIT chip was introduced in 1989, about the same time as the Cyrix 83D87. Both coprocessors are faster than Intel's 387DX coprocessor. The IIT 3C87 also provides extra functions not available on any other 387 chip. It has 24 user-accessible floating-point registers organized into three register banks. Three additional instructions (FSBP0, FSBP1, FSBP2) allow switching from one bank to another. The register bank's main purpose is to aid the fourth additional instruction the 3C87 has (F4X4), which does a full multiply of a 4x4 matrix by a 4x1 vector, an operation common in 3D-graphics applications. There are very few applications that make use of it when an IIT coprocessor is detected at run time (among them Silver Screen and Fast-CAD 3-D). These IIT-specific instructions also work correctly when using a Chips & Technologies 38600DX or a Cyrix 486DLC CPU, which are both marketed as faster replacements for the Intel 386DX CPU. The 3C87 is implemented in an advanced CMOS process and has low power requirements, typically about 600 mW. Like the 387 'clones' from Cyrix and ULSI, the 3C87 does not support asynchronous operation of the CPU and the coprocessor, but always runs at the full speed of the CPU. It is available in 16, 20, 25, 33, and 40 MHz versions.
May 2003 updates and changes 2003-05-19 ...::: cpu-collection.de now has more than 300 chips online :::... 17 new additions to the collection today: some rare or uncommon processors Intel D4004 Finally I got an i4004 in my collection :-) The 4004 was the first single chip microprocessor, originally designed and manufactured by Intel for Busicom for use in their calculators. It was invented by Intel engineers Federico Faggin, Ted Hoff, and Stan Mazor and introduced in November 1971. For information on the i4004 please refer to http://www.intel4004.com/ or one of the other various online ressources. The item here was made in 1981. Intel 80186 A80C186CPICE Sample An 80186 sample chip - I do not know the background of this chip, but it seems to be a development sample for the 80186. It has a lot more pin contacts than a regular i80186, maybe some of them are unused and some provide signals for testing and evaluation. If you got more information on this chip please contact me. Intel 80186 FA80C186CP Engineering Sample An engineering sample of an embedded version of the i80C186 in a PQFP package with 156 contacts. I don't have any information on the clock rate or if this version ever made it into production - if you got more information on this chip please contact me. I have more of these, so if you're interested in getting one, I'd like to trade it for another sample chip. Intel Pentium P5 at 60MHz An early version of the Pentium 60 with golden heatspreader and 'PROCESSOR'-marking under the 'Pentium' logo. This CPU has the FDIV bug. 2 processors made in former USSR KM1810BM86 A soviet 8086 clone in white ceramic, made in 1986. KP1810BM86 Another 8086 clone in a strange, segmented plastic package. some 6800 CPUs 2 AMI 6800 processors A plastic version made in 1978 and a ceramic version from 1984. The 6800 processor was released by Motorola in 1975 and has been developed as an enhancement of the 8008 at the same time that Intel was developing the 8080A, also as an enhancement of the 8008. The 6800 was first used in cars by GM and Ford and later found its way into the computer market through several computer manufacturers (MITS, Systems Research, Sphere and Southwest Technical Products). I have some of the ceramic version chips for trade, if you are interested in one of them please contact me. Hitachi HD6803P and Motorola 6803 MC68B03P The 6803 is an enhanced 6800 with some new 16- and 8-bit instructions, 128 bytes of RAM and an internal clock oscillator added. Motorola MC6809P Third generation addition to the 6800 familiy, in competition to Intel 8088. The 6809 is an enhanced 6800 with RAM and clock added. some gaps filled in my AMD K6 and K6-2 division AMD-K6/266AFR The K6 was originally launched running at speeds of 166 and 200 MHz in April 1997. It was followed by a 233 MHz version later in the summer of 1997. The release of the 266 MHz version of this chip was not until spring 1998 when AMD were able to move to the 0.25 micron manufacturing process. The final iteration of the K6 design was released in May 1998 running at 300 MHz. AMD-K6-2/266AFR AMD-K6-2/333AFR AMD-K6-2/350AFR AMD-K6-2/450AFX AMD-K6-2/500AFX The K6-2 was originally manufactured in speeds of 266 & 300 MHz in May 1998. The 300 MHz chip saw the introduction of the 100 MHz bus over the conventional 66 MHz bus used by the 266 MHz chip. August 1998 a 333 MHz version on a 95 MHz bus has been released and was quickly followed by a 350 MHz version on the 100 MHz bus. November 1998 saw the release of the 366, 380 and 400 MHz versions of the chip. This has been followed by the release of a 450 MHz K6-2 in February 1999 and the 500 MHz version in August 1999.
April 2003 updates and changes 2003-04-27 *new* now you can subscribe to the cpu-collection newsletter to get informed about site news and updates :-) 14 new additions to the collection today: Some nice processors made by DEC: DEC Alpha AXP 21064-BB A233 and DEC Alpha AXP 21064-DB A275 The 233 and 275 MHz versions of Digital's Alpha AXP 21064A (EV45) processors. Designed as a successor to the VAX line of computers, it supported the VMS operating system, as well as the DEC flavour of UNIX. Later open source operating systems also ran on the Alpha, notably certain BSD systems. Microsoft supported the processor in earlier versions of Windows NT. The Alpha 21064A was used in DEC 3000/900 and DEC AlphaStation 255 systems. DEC StrongARM SA-110S A 32-bit RISC microprocessor featuring superior power efficiency, low cost, and high performance. The SA110 is an implementation of Advanced RISC Machines Ltd. (ARM) Version 4 architecture intended for embedded applications. It was used as a CPU in Apple's Newton MessagePad 2000 series and Psion organizers. And there's also an Intel manufactured version of this processor added to the collection, the Intel StrongARM SA-110 21281-EB, 233 MHz There are also updated DEC, Alpha AXP and StrongARM overview pages.
Even more SPARC additions: Weitek SPARC POWER µP This is a 'POWER µP' CPU replacement chip for a Sun SPARCstation IPX. It replaces the 40 MHz original IPX processor with a clock doubled 80 MHz version. LSI Logic SPARC L64811GC-40S A SPARC Integer Unit from a SPARCstation 2 (4/75) 'Calvin'. The SS2's CPU (like in all older SPARC systems) consists of three separate parts, an Integer Unit (IU), a Floating-Point Unit (FPU) and cache memory (high speed local memory for the IU). SPARC V7 architecture. Texas Instruments SPARC FPU TMX390C602AGA A math coprocessor from my Sun SPARCstation 2.
A bunch of additional 680x0 CPUs from Motorola, Mostek and SGS-Thomson: Motorola MC68000L8 Motorola MC68020RC12E Motorola MC68020RC16B Motorola MC68020RC16E Motorola XC68040HRC33E Mostek MK68000P-8A SGS-Thomson TS68000CP8 For more information on these processors please refer to my 68000, 68020 and 68040 overview pages.
2003-04-19 13 new additions to the collection today: SPARC additions and some other RISC processors: LSI Logic SPARC S1A0007 A SPARC CPU from my SPARCstation 1 (4/60) 'Campus'. SPARC V7 architecture. Fujitsu SPARC MB86903-40 This processor is from my Sun SPARCstation IPX (4/50) 'Hobbes' from 1991.
Intel i860 XP The i860 processor, modestly described by Intel as "a Cray on a Chip", achieves high levels of integer, floating point, and 3D graphics performance. Machines based on the i860 address simulation and modelling, animation, virtual reality, image processing, and other high end, computationally intensive applications. It was used in UNIX workstations and the Intel iPSC/860 Supercomputer. IDT R4600-133G A MIPS R4600 'Orion' CPU, manufactured by MIPS second source IDT. These processors where used in Silicon Graphics Indy workstations and are still used in Cisco series 4000 routers. The R4600 was designed by Quantum Effect Design (QED).
Some math coprocessors: Motorola MC68881RC20D Motorola MC68882FN20A Motorola MC68882RC33A Motorola FPUs for 68020 and 68030 processors. These math units were used in UNIX workstations and servers from Sun, Apollo and others. Also some models of the Commodore Amiga and Atari ST lines of computers and pre-PowerPC Apple Macintosh machines used this FPUs. Weitek 3170 A math copro for Sun SPARCstation 1 (4/60) computers. IIT 3C87-40 Cyrix CX-83D87-25 GP Intel A80387-16 FPUs for 386SX systems.
And 2 Z80 processors: Zilog Z0840004PSC and MME VB880D W3 The MME VB880D is an East German (former German Democratic Republic) Z80 clone for military purposes (enhanced temperature range).
2003-04-06 It's RISCy SPARCling SUNday! Today we have a closer look on what's inside these workstations and servers made by Sun Microsystems. SPARC (which stands for Scalable Processor ARChitecture) is an open set of technical specifications that any person or company can license and use to develop microprocessors and other semiconductor devices based on published industry standards. SPARC was invented in the labs of Sun Microsystems Inc., based upon pioneering research into Reduced Instruction Set Computing (RISC) at the University of California at Berkeley. The first standard product based on the SPARC architecture was produced by Sun and Fujitsu in 1986; Sun followed in 1987 with its first workstation based on a SPARC processor. If you want to see pictures of various Sun machines, http://www.black-cube.net/Sun/ or http://www.obsolyte.com/sunPICS/ have lots of them. For technical information please refer to the Sun website or the handbooks on SunSolve. Many of the items in the collection were contributed by Sunopsis - the Sun Museum - thank you very much! It is located at the Sun Benchmark Center in Langen/Germany. You can see some pictures from inside there at the schrotthal.de website. If you want to have a close look on old Sun hardware please contact them: Sunopsis - the Sun Museum c/o Sun Microsystems GmbH Amperestr. 6 63225 Langen Germany. Now let's have a look at the processors: Fujitsu SPARC MB86900 The very first SPARC processor, running at 14.28 MHz. It was used in one of the first SPARC driven Sun machines, a Sun 4/110 'Cobra' from 1987. A donation by Sunopsis. Fujitsu SPARC MB86902 A processor used in Sun SPARCstation IPC (4/40) 'Phoenix' machines from 1991. Donated by Desideriu, webmaster at cpu-museu.net. Thanks a lot! Weitek SPARC FPU 3172A The FPU for the Fujitsu SPARC MB86902, a Weitek math processor from the SPARCstation IPC. LSI Logic SPARC L1A3761 100-1808-01 A SPARC CPU from a SPARCstation 1 (4/60) 'Campus' early model. Another donation by Sunopsis. Sun Microsystems SuperSPARC TMS390-50 A SuperSPARC processor module for a Sun SPARCserver 10 or 20 (SPARC V8 architecture). This CPU is mounted together with a cache controller and 1MB cache memory on an SBus module (Sun part 501-2258) and is clocked with 41MHz. Donation by Sunopsis. Sun Microsystems microSPARC II STP1012PGA-70 and STP1012APGA-85 The microSPARC II was used in Sun SPARCstation 4, 5 and Voyager workstations. Sun Microsystems UltraSPARC STP1030BGA-200 A processor from a Sun Ultra workstation. The UltraSPARC is a high-performance, highly-integrated superscalar processor implementing the SPARC V9 64-bit RISC architecture. It supports 2D, 3D graphics, image processing, video compression and decompression and video effects through the VISual Instruction Set. SPARC V9 architecture. A donation by Sunopsis. Sun Microsystems UltraSPARC II STP1031LGA, 300MHz and 336MHz This CPU is used in Sun Ultra workstations, Netra- and Enterprise servers. Donation by Sunopsis. Sun Microsystems UltraSPARC III STP1031LGA The high end CPU class of Sun's UltraSPARC series, used in Sun Fire servers and Sun Blade workstations. The UltraSparc III features Scalable Shared Memory (SSM) and is able to scale to up to 1000 processors in a single system. Donated by Sunopsis. For an overview of SPARC processors in the collection go to the SPARC class page. March 2003 updates and changes 2003-03-26 10 new additions to the collection today: Some pretty old processors from the 70s and 80s: Intel P8085A The 8085 was designed to improve the performance of the 8080 by higher system speed. It was developed by Intel in 1976 as a transition to the 16-bit 8086. This item here was manufactured in 1979. Intel D8086-2 The 8086, officially called iAPX 86, is a 16-bit processor designed by Intel in 1978. It gave rise to the x86 architecture. The 8086-2 was an enhanced version with 8 MHz (compared to 8086 at 5 MHz). The first commercial microcomputer built on the basis of the 8086 was the Mycron 2000 (from Norway). This computer was used by Digital Research as the development platform for the CP/M-86 operating system. This item was made in 1984. Intel P8088 The 8088 is a reduced and cheaper version of the 8086 for the 8-bit market. It was introduced in 1979 and had 16-bit internal registers and an 8-bit data bus. This processor was used in the original IBM PC. It was manufactured in 1984. Siemens SAB 8086-2-C A Siemens manufactured version of the 8086-2. It was made in 1984. Siemens SAB 8088-2-P A Siemens manufactured version of the 8088-2, made in 1986. NEC V20 D70108D-8 The V20 is pin compatible with the 8088 processor but operates up to 40% faster at the same clock frequency. It is software compatible with V25 to V50 NEC microprocessors. The item in the collection was made in 1986. Motorola MC6800L The 6800 processor was released by Motorola in 1974, shortly after the Intel 8080. It was the basic for Motorola's entrance into the microprocessor market. Like the Intel 8080 it was designed as an enhancement of the Intel 8008. This chip here was made in 1977. Hitachi 6800 HD68A00P A slightly enhanced version of the 6800 by Hitachi. The clock frequency of the 68A00 was rised to 1.5 MHz compared to 1 MHz in the 6800. Motorola MC6802P The 6802 is an enhanced 6800 with 128 bytes of RAM and an internal clock oscillator added. It was also introduced in 1974; this item here is from 1989. SGS Z8001D1 A Zilog Z8001 clone by SGS. The Z8001 featured multiprocessor support, separate code, data and stack memory space and ran in two operation modes, system and normal. My SEG Z8001 was made in 1982. Next update will be a RISCy, SPARCling SUNday, so stay tuned :-) 2003-03-11 11 new additions to the collection today: Three 386 CPUs: AMD Am386DX/DXL-33 AMD 386 processors were offering approximately equal performance like Intel i386 and were significantly less priced. Intel i386 DX A80386DX-20 IV A 20MHz Intel i386. Intel i386 DX A80386DX-25 IV with DX logo One of the later i386 models with the 'new' DX logo. Intel i386 SX NG80386SX-20 An i386SX CPU with 20MHz. This item was made in 1991. Two TI 386/486 processors: Texas Instruments TX486DLC-33GA and Texas Instruments TX486DLC-40BGA These CPUs had a 486 instruction set and a 386 pinout and made 486 advantages available for 386 boards. Some clock doubled 486 CPUs: Intel i486 DX A80486DX2-50 The first chip to use "clock doubling" technology. 50MHz systems had stability problems these days (1992) so the board speed was reduced to 25MHz running at half the processor speed. SGS-Thomson ST486 DX2-66GS and SGS-Thomson ST486 DX2-80GS Two 486DX2 processors by ST with 66 and 80MHz. These were in fact Cyrix 486DX2-processors, made under license and labeled by ST. IBM 486-V580GA An IBM "Blue Lightning" 486 clock doubled CPU with 80MHz. Two UNIX machine processors: Sun Microsystems SuperSPARC STP1020APGA-60 A SuperSPARC processor module from my SPARCserver 10. The CPU was designed by Sun Microsystems and manufactured by Texas Instruments in 1994. IDT R3000A-12G A MIPS R3000 CPU, manufactured by MIPS second source IDT. These processors where used in Silicon Graphics SGI Personal IRIS 4D/20 graphic workstations. On this machines 3D sequences for movies like The Abyss, Jurrasic Park or Terminator 2 were rendered. This item was made in 1989.
2003-03-03 18 new additions to the collection today: Some math coprocessors: Intel FPU A80387DX-25 and Intel FPU A80387DX-33 2 math coprocessors for i386 SX CPUs Intel FPU A80487SX Math coprocessor for i486 SX CPUs Some Intel processors: Intel C8080A An almost 30 years old 8080 CPU from Intel :-) This nice white ceramic chip was made in 1975 and is a donation by Christian Lederer. Thanks very much! Intel i486 DX A80486DX2-66 w/ Cooler A blue cooler is factory mounted on this version of the Intel i486 DX2-66. Intel i486 DX FC80486DX4-75 Mobile This CPU-module is the processor inside my IBM Thinkpad 755C notebook from 1995. To take a picture of it Grampa and I removed about 40 screws and lots of circuit boards, cables, plugs and shieldings inside the Thinkpad - hard work. But it was worth it, we have never seen a CPU module like this before. After reassembly I found out that we forgot to reattach the heat pad (which connects the CPU surface with the cooling metal part of the Thinkpad case) - we will have to dis- and reassemble it all again :-( Intel i486 SX A80486SX-33 The 33MHz version of Intel's i486 SX. Intel Pentium P5 A80501-66 Goldcap And finally I got the 66MHz Pentium P5. This item here is a D1 stepping without the FDIV bug. Intel ODP486SX-25 Overdrive and Intel ODPR486DX-33 Overdrive Two Intel Overdrive processors Some AMD 486 processors: AMD Am486DX2-50 AMD Am486DXL2-66 AMD Am486DX4-120 AMD Am5x86-P75 Kingston Turbochip A Kinston "TurboChip" 486 upgrade kit for 486 systems. It has a 133MHz AMD 5x86-P75 PQFP packed CPU mounted on a PGA curcuit board to fit into a standard 486 socket. A fan is mounted on top of the module. These kits were sold as upgrades for common i486 systems and came with an installation booklet and a chip puller. Two 286 processors: AMD 80286 N80L286-12/S 286 CPU from AMD with 12MHz. Harris 80286 CS80C286-16 A Harris 80286 CPU with 16MHz. Harris made 286 processors with clock frequencies up to 25MHz while Intel 286s just had frequencies between 6 and 12MHz. And two black Motorola 680x0 processors: Motorola MC68020RP20E and Motorola MC68030RP25B
February 2003 updates and changes 2003-02-15 11 new additions to the collection today: AMD Duron DHD1300AMT1B An AMD Duron processor with 1,3GHz clock frequency. Cyrix Cx486DX-40 w/ Cooler Cyrix Cx486 processor with cooler on top. This is a version with a newer style DX logo than the version I had in collection already. Cyrix M II-300GP, 66MHz FSB Another version of the Cyrix M II 300 for 66MHz system bus. The core/bus ratio here is 3.5 giving 225MHz core frequency. IBM 6x86L PR166+ Low power (2.8V compared to 3.3V in the standard version) version of the 6x86L. IDT WinChip C6-PSMF200GA A processor designed by Centaur Technologies and manufactured by Integrated Device Technologies. It was introduced in 1997 as an alternative to the Pentium MMX processor. The WinChip has a much simpler design than the Pentium and therefore the die is 27% smaller than the Pentium's and 46% smaller than AMD's K5. It only uses a single pipeline for example, processing each instruction one at a time as quickly as possible instead of executing multiple instructions per clock cycle. To compete with Intel and AMD in speed comparison Centaur doubled the L1 cache size to 32KB for instructions and 32KB for data. This made the WinChip about as fast as a Pentium MMX or a K5 with same clock frequency. Intel FPU A80387DX 16-33 An Intel math coprocessor for 386 systems from 16 to 33MHz. Intel i486 SX 486SX2-50/SA OEM A not so common because not very successful double clocked version of the Intel 486 SX. Intel Pentium P54C A80502100 Mobile A low power (2.9V compared to 3.15-3.6V standard voltage) version of the Pentium 100 made for mobile devices. Intel P8085AH An 8bit predecessor of the 8086 processor by Intel, designed to improve the 8080's performance by a higher system speed. It was 100% software compatible to the 8080A. Intel Pentium II A80523 350MHz The 2nd generation Pentium II 'Deschutes' with 350MHz and 512KB L2 cache. This P II is a slot 1 cartridge version, the images show what's inside the cartridge. Motorola MC68030RC20B 20MHz version of Motorola's 68030 processor.
2003-02-14 Added possibility to search the cpu-collection. You can search for markings printed on the chip and/or in CPU comments. Enjoy.
2003-02-08 Correction of 80286 information page. Thanks to a user comment by Martin I corrected the 80286 page and added information about Harris manufactured 25MHz versions of the 80286. Thank you, Martin!
And there are 18 new additions to the collection today: Some math coprocessors: ULSI 387 Math-Co DX 40MHz and ULSI 387 US83C87-C DX/DLC 40MHz FPUs by ULSI for Intel 386 and compatible systems. These chips had to be plugged in an additional FPU socket on 386 boards. Weitek 8701 This is a CPU from a Sun SPARCstation IPX. The IPX were shipped with either this Weitek CPU, a Weitek 8601 or a Fujitsu MB86903. Some UNIX machine processors: HP PA-RISC 7100LC Finally I took a picture of my PA-RISC 7100LC processor. It is built in my HP Apollo 9000 710 UNIX workstation from 1992. NEC VR12000, 300MHz The VR12000 is a high-end 64bit MIPS R12000 compatible RISC processor made for NEC supercomputers, Silicon Graphics O2 and O2+ graphics workstations and Origin 2000/3000 UNIX servers. It was introduced in 1998 and NEC claimed, the performance places it in the highest class of microprocessors in the world. It was sold for 300,000 Yen (~2400 €/$) when it was introduced. The VR12000 could use up to 8MB L2 cache. LSI Logic L5A8946 This CPU is said to be from a UNIX machine - I could not find any information about it. It seems to be a prototype chip ('PROTO' marking). If you got some info about it or the other unknown LSI Logic chip in my collection please let me know. Siemens S77-T120 Another CPU from a UNIX machine. Could not find any information about it either - please contact me if you know something about it. Some 8080A processors: AMD D8080A, Intel D8080A and NEC 8080AF The 8080 is an enhanced version of the 8008 CPU and was developed by Intel in 1974. It is said to be the technical initialization for the world-wide microprocessor market. Some Slot processors: AMD Athlon K7650MTR51B, 650MHz An Athlon A with 650MHz for AMD Slot A boards. It was introduced in 1999 and has a cartridge package with L2 cache chips mounted on a circuit board together with the CPU. I took pictures of the cartridge and the internals to give you an impression of what's inside the package. Intel Pentium III A80525 500MHz The first generation Intel P III Katmai in SECC-2 package. Like the Athlon A it has L2 cache chips and the CPU mounted together on a circuit board and is packed in a plastic cartridge. Also pictures of the cartridge and the board on the description page. And some more CPUs: AMD-K6-2, 300MHz and 400MHz The first two K6-2 processors in this collection. Intel A80286-8 A PGA version of Intel's 286 CPU running at 8MHz. Intel Pentium Pro 200MHz, 256KB Cache The 200MHz version of the Pentium Pro with 256KB L2 cache on board. I'm still looking for the black aluminum 1MB cache version of the 200MHz Pro, so if you want to trade one, please contact me. Texas Instruments TI486DX2-G80-GA A Texas Instruments 486DX2-80 with a beautiful colored Windows-Logo on top. The only color-labeled CPU I know of. UMC U5SX 486-A A pretty rare UMC 486 processor with SMD PQFP package mounted on a PGA board. It does not have any frequency markings printed on - if you know how fast it was clocked, please tell me.
I also added pictures of the cartridged Intel Pentium II PE Mobile OEM 266MHz to show you what a mobile mini-cartridge looks like.
January 2003 updates and changes 2003-01-30 13 new additions to the collection today: AMD Am486DX-40 Another version of the AMD Am486DX-40, with Windows-Logo. Cyrix M II-300GP After Cyrix was purchased by National in 1998 the 6x86MX series chips were re-named MII. Compared to early 6x86MX CPUs the MII had some enhancements to its processor core. Heat output was reduced allowing it to rise to clock speeds over 220MHz. The MII required non-standard bus speeds at 75 or 83MHz on socket 7 boards and therefore had some troubles with stability. Especially some graphics cards had problems with nonstandard (66MHz) PCI bus speeds. Intel Pentium P54C A80502-90 Goldcap A version of the Pentium 90MHz with golden heat spreader. Intel Pentium II 266 Mobile OEM The Pentium II 266 Mobile is a low power notebook version of the Pentium II. This item here has a Mini Cartridge package type and is encased in a metal sheet. The picture shows the cartridge with removed sheet to give you a look at the internals. 2 Motorola 68881 FPUs Two 68881 math coprocessors for the Motorola 680x0 series CPUs with 16 and 25MHz. SGS-Thomson ST6x86-P150+ An SGS Thomson manufactured version of the Cyrix 6x86 CPU with Pentium Rating 150 running at 120MHz. SGS-Thomson ST6x86-P166+ w/o Cyrix Logo A Cyrix 6x86 P166+ from SGS Thomson. This processor has no Cyrix markings printed on the top side and no Cyrix copyright remark on the back. SGS-Thomson ST486 DX2-80 And another SGS Thomson CPU, a clock doubled 486 with 80MHz core frequency. Toshiba TMPZ84C00AP Two Z80 CPUs made by Toshiba in the late eighties, a TMPZ84C00AP with 4MHz and a TMPZ84C00AP-10 with 10MHz. WDC 6502 W65C02SP-6 A CMOS 6502 processor with 6MHz manufactured by MOS second source Western Design Center. Those CPUs were used by Apple in their IIc and III series computers in the late eighties. Zilog Z8001APS A Z8001 16bit CPU from Zilog. The Z8001 featured multiprocessor support, separate code, data and stack memory space and ran in two operation modes, system and normal.
2003-01-26 4 new additions to the collection today: NexGen Nx586-P90 Finally I got a NexGen CPU in this collection :-) This here is a P-90 with 84MHz core- and 42MHz busfrequency. The Nx586 was a 586 CPU with a RISC core and translated x86 instructions into efficient RISC code before execution. This processor also featured an extra bus for L2 cache running at core speed - other 586 CPUs at that time accessed their L2 cache at external bus speed. It therefore did not run in common Socket 5 or 7 Pentium-boards but required a special board design. AMD Am386DX/DXL-40 A 386 AMD CPU running at 40MHz. AMD K5 PR150 The PR150 version of AMD's K5 CPUs, running at 105MHz core frequency. Intel Pentium MMX 200 A 200MHz Pentium with MMX technology.
2003-01-25 a new section about CPU sockets and slots was added to the site. I hope you find this information useful.
2003-01-21 7 new additions to the collection today: AMD Am486DX2-66 A double clocked 486DX CPU from AMD with 66MHz core frequency. Found this item at a local recycling yard. Cyrix 6x86-P166+GP The 133MHz Cyrix socket 7 CPU with Pentium Rating 166 and dual voltage 3.3 or 3.52V. Fujitsu MBL8086-2 A 8086 clone from Fujitsu with 2 MHz. This processor was built in an old computer from Schneider. Got this also from the recycling yard. Intel R80286-6 and Intel N80286-8 Two 286 Intel CPUs with 6 and 8MHz from old Siemens and Nixdorf computers I found. Intel i486 DX2-50 This is an uncommon i486 DX2 with 50MHz, manufactured by IBM, not Intel. Too bad it is not in a good condition :-( Intel Pentium 166 The missing 166MHz version of the Pentium P54C. Now I just need a rare 200MHz version to get the standard P54C line complete... 2003-01-21 added user comment on Cyrix Cx487S-40QP and an answer to the 'FasCache'-question from four days ago on Cyrix Cx486S-33GP. 2003-01-20 Sascha Reißner contributed an IBM 6x86L P150+, a low power version (2.8V) of the 6x86 P150+ (3.3V). Thank you very much! 2003-01-17 added some Commodore/MOS CPUs to the collection today: MOS 6502 and MOS 6502 A processors from CBM 1541 floppy disk drives and a MOS 6510 CPU from a Commodore C64. These machines, all together two C64s, three 1541 drives, a CBM MPS 801 printer and loads of software were offered to me for 10 Euros - just couldn't resist :-) and a comment on Cyrix Cx486S-33GP was added - a question about 'FasCache'. Maybe you can help? 2003-01-16 Claus Bettag from SOFTWERK contributed 3 CPUs to the collection: A Cyrix Cx486SLC/e-33MP, an AT motherboard with an Intel R80286-10 and an Intel N80186. Thank you very much! 2003-01-16 new additions to the collection today: Cyrix Cx486DX2-66GP I only had the low voltage version of this chip (Cx486DX2-V66GP) in my collection, now here's the regular one. IBM 6x86MX PR166 The 133MHz version of IBM's licensed Cyrix 6x86 with Pentium rating 166. It was introduced in May 1997 and features MMX technology. Intel i486 DX-50 The last i486 with core frequency equal to bus frequency, introduced in June 1991. It was followed 1992 by the DX2 with clock doubled core frequency. Intel Pentium P5, 60MHz without heat spreader One of the earlier Pentium 60s, a B1 stepping model. These models suffered from heat troubles and where therefore enhanced with a heat spreader, a flat metal plate designed to distribute heat evenly across the processor surface, in later models of the P5. The item in the collection had a huge cooler mounted on top (which was not easy to unmount ;-). Of course it features the infamous FDIV bug. Intel Pentium P54C, 100MHz with heat spreader A not-so common version of the Pentium 100MHz - with head spreader. This was AFAIK the last ceramic PGA Pentium to use a head spreader. Intel i960 A superscalar 32-bit RISC microprocessor from Intel intended for embedded applications. I found two items today in a recycling container. One was the processor on a broken SCSI RAID controller card, the other one a processor inside a network switch. Intel N80186 Another item from the container - an 80186 processor. This 80186 was used as a micro-controller on an old network card. NEC V50 D70216L-10 The NEC V50 is an 8086 compatible processor, generally used as an embedded controller. I found 2 pieces today, also in the recycling container. They were inside 2 dumped network switches. NEC D8085AHC A NEC 8085 CPU. This processor works inside my Taylorix System 4 CP/M computer from the early eighties. 2003-01-15 added user comments on Cyrix Cx486DLC-33GP, IBM 5352 and Zilog Z84C3004PEC CTC 2003-01-14 added the first 2 user comments on Intel i386 DX-20 and IBM 5352 2003-01-14 cpu-collection.de is finally online :-) © 2003 - 2017 cpu-collection.de