|December 2005 updates and changes
|...::: cpu-collection.de now has more than 800 different chips online :::...
23 additions to the collection today:
MIPS Rx000 Processors
MIPS (Microprocessor without interlocked pipeline stages) is a RISC microprocessor architecture
developed by MIPS Computer Systems Inc in the early eighties. MIPS processors were used in high-end servers from Siemens and DEC and
especially in SGI's computer product line, and have found broad application in embedded systems, Windows CE devices, and Cisco routers.
The Nintendo 64 and Sony PlayStation 2 consoles also use MIPS processors. The architecture was very successful, about one third of all RISC chips produced in mid 1990s were MIPS based
designs. In 1999 the ARM/StrongARM architecture took over rather decisively in thanks to cell phone and PDA usage.
The MIPS design is licensed to many 3rd party vendors, among them IDT, Siemens, NEC, HP and Toshiba. It is one of the most popular licensed architectures. This is not
because the MIPS architecture is particularly good in a specific task. It isn't. MIPS is just a convenient, clean, and easily scaled architecture around which
special-purpose network processors or protocol engines can be added. It is one of the cleanest and most generic processor designs around, finely tuned for absolutely nothing.
MIPS architecture at Wikipedia
Number one RISC architecture (1997)
MIPS architecture documentation
MIPS R series tech docs
Comprehensive SGI information at SGIstuff
23 new MIPS processors in the collection:
The first commercial MIPS CPU model, the R2000, was announced in 1985 as a 32-bit implementation. It was followed by the R3000, the first successful MIPS design in the marketplace with more than 1 million processors made. The R3000 was used in high-end UNIX computers by Siemens and DEC and in the 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.
The R3000 has an interface to handle 3 coprocessors. Each coprocessor has a flag line connected with the CPU that can be tested and a conditional branch executed
dependent on its value. Coprocessor instructions can be executed directly from the instruction stream.
Coprocessor 0 (CP0) is incorporated on the CPU chip and supports the virtual memory system and exception handling. It is also referred to as the System
Control Coprocessor. CP1 is reserved for the floating point coprocessor. An FPU is mandatory for most R3000 systems. CP2 is available for specific
implementations and is often used to accelerate memory access by connecting it to an R3020 memory buffer or
an R3220 read/write buffer chip. Later versions of the R3000 (R3000A) have built in memory buffer circuitry.
R3000 at SGIstuff
R3000 information (German)
15 new R3000 processors in the collection:
NEC D30310R-33 VR3000A-33
Performance Semi PR3000-33PGC
Performance Semi PR3000A-33MQ160C
IDT R3000 FPU 79R3010L-33G
NEC D30311R-33 VR3010A-33
Performance Semi R3000 FPU PR3010A-25PGC
Performance Semi R3000 FPU PR3010-33PGC
Siemens R3000 FPU R3010-25-A
Siemens R3000 FPU R3010A-25-QF
Early implementations of R2000/R3000 processors had no write buffer circuit inside to isolate the CPU from memory subsystem.
There were companion chips for R2000/R3000 to implement write buffering, connected via the CP2 coprocessor interface of the R3000 CPU.
Newer designs of the R3000 (R3000A) have built in memory buffer circuitry.
R3020 MB (Memory Buffer):
IDT R3000 MB 79R3020-20G
IDT R3000 MB 79R3020-33G
R3220 RWB (Read/Write Buffer):
LSI Logic R3000 RWB LR3220GC-25
The R3500 is an R3000 with integrated R3010 FPU, MMU and memory buffers.
Military versions of the R3500 got quite famous for their usage in the 1st generation
F-16 Modular Mission Computer (MMC).
IDT 79R3500 data sheet
The R4000 series, released in 1991, extended the MIPS instruction set to a full 64-bit architecture and moved the FPU onto the main die to create a single-chip system.
The design was so important to SGI, at the time MIPS' major customer, that SGI bought the company in 1992.
The R4000 processors have been available and used in different versions. PC (as in R4000PC) denotes primary cache only and SC denotes secondary cache.
The MC versions contain special support for cache architectures in multiprocessor systems.
R4400 processors where used in many Silicon Graphics computers, e.g. the Indigo2 workstations,
Onyx supercomputers and
MIPS R4000 Product Information
R4000 at SGIstuff
IDT 79R4700 Documents
4 new R4x00 processors in the collection:
NEC D30401RJ-50 VR4000SC-50
The R4700 'Orion' was designed by Quantum Effect Devices (QED), a company started by former MIPS employees. R4700 processors where used in Silicon Graphics
Indy workstations and are still used in Cisco series 4000 routers.
IDT R4700 79RV4700-100G 'Orion'
IDT R4700 79RV4700-133G
The R5000, also a QED design, replaced the R4600 in 1996. The R5000 FPU had more flexible single precision floating-point scheduling than the R4x00,
and as a result, R5000-based SGI Indys had much better graphics
performance than similarly clocked R4400 Indys with the same graphics hardware. SGI gave the old graphics board a new name (XL8 -> XGE8) when it was combined with R5000
in order to emphasize the improvement.
MIPS R5000 introduction
MIPS R5000: Fast, Affordable 3-D
IDT 79RV5000 Documents
R5000 at SGIstuff
1 new R5000 processor in the collection:
NEC VR5000 D30500RJ-180
The RM7000 is a higher performance successor of the R5000 with built-in 256 KB L2 cache and a controller for optional L3 cache.
It was primarily targeted at high end embedded designs such as internetworking, high-performance image manipulation and high complexity printing.
Despite of being targeted at the embedded market SGI did use the RM7000 to upgrade it's O2 workstation to the O2+.
QED RM7000 data sheet
2 new R7000 processors in the collection:
QED RM7000-250S (old Logo)
QED RM7000-250S (new Logo)
The R10000 'T5' is the first single-chip 4-issue superscalar processor by MIPS, introduced in 1995.
It gave a 70 to 100% performance increase over the R4400 at 250 MHz, previously the most powerful MIPS processor.
The R10000 CPU was used in NEC supercomputers and various SGI machines like the
R10000 User Guide
R10000 at SGIstuff
NEC VR10000 announcement
T5: Brute Force (BYTE.com)
1 new R10000 processor in the collection:
|November 2005 updates and changes
12 additions to the collection today:
Intel 486 OverDrive Processors
Intel 486 OverDrive processors are a category of various Intel CPUs that were produced to upgrade 486 based computers.
ODP is the acronym for Over Drive Processor, however most OverDrives are ODPR (Over Drive Processor Replacement) chips: To upgrade a computer to
a new processor you just replace it with the new one. Their interfaces appear to the motherboard like an older chip, so you don't need to worry
about whether or not the motherboard supports the higher speeds. ODP chips (like the
DX2ODP66 or the
ODP486SX-25) in contrast were made for additional sockets on the mainboard, the
Performance Upgrade Socket.
On older versions Intel labeled the OverDrive chips with the external local bus speed they supported, like the
ODPR486DX-33. Since all these OverDrive chips have the clock doubling feature,
it was assumed by Intel that everyone knows that the CPU runs at twice the bus speed internally. Newer chips follow the DX2/DX4 naming scheme of the
5 new Intel 486 OverDrive processors in the collection:
old naming scheme (external bus speed):
Intel ODPR486DX-25 OverDrive
Upgrades 25 MHz Intel 486SX based systems to 50 MHz.
Intel ODPR486DX-25 OverDrive (V3.0)
Another version of the same processor; upgrades 25 MHz Intel 486SX based systems to 50 MHz.
Intel ODPR486DX-33 OverDrive
Upgrades 33 MHz Intel 486DX and SX based systems to 66 MHz.
new naming scheme (internal CPU clock speed):
Intel DX2ODP50 OverDrive
Upgrades 25 MHz Intel 486DX and SX based systems to 50 MHz.
Intel DX4ODP75 OverDrive
Upgrades 25 MHz Intel 486DX & 16/20/25 MHz 486SX based systems to 75 MHz. Both are OverDrive processors for an additional socket on the mainboard.
Intel Pentium OverDrive Processors
- Pentium OverDrive for 486 systems
This Pentium OverDrives are Intel Pentium processors for 486 Socket 3 and Socket 2 motherboards, provided as a means to give a Pentium performance-level
upgrade option for owners of 486 computer systems. It was however criticised for being more expensive and slower than competing CPU-upgrade options
such as the AMD Am5x86 and
Cyrix 5x86, and being too late to the market. Mainboard compatibility was also a problem, it turned
out that many boards didn't support this new chip.
To perform properly, the Pentium OverDrive was dependent on a high amount of secondary cache ram being present on the motherboard; without it the
chip was only a trivial amount faster than a DX4.
1 new Intel Pentium OverDrive for 486 systems in the collection:
Intel PODP5V63 Pentium OverDrive
Upgrades 25 MHz Intel 486 based systems to Pentium technology at 63MHz.
- Pentium OverDrive for Pentium 60/66 systems
The original Pentium P5 chips were different than later P54C
versions, in terms of voltage, socket size and power consumption. Intel made a clock-doubling OverDrive for these chips, which is sold as one 120/133 MHz
chip: when replacing a Pentium 60 it runs at 120 MHz, and when replacing a 66 it runs at 133. This is a true Pentium chip since it is on a Pentium
motherboard, although it still benchmarks below the real Pentium 120/133, most likely due to the older design of the Pentium 60/66 motherboards.
1 new Intel Pentium OverDrive for Pentium 60/66 systems in the collection:
Intel PODP5V133 Pentium OverDrive
Upgrades classic Pentium P5 60/66 MHz systems to 120/133 MHz.
- Pentium OverDrive for Pentium P54C systems
Pentium 75, 90 and 100 MHz run with a clock multiplier of 1.5 on system buses of 50, 60 and 66 MHz respectively. Intel has made OverDrives for
these running at 125, 150 and 166 MHz by using a clock multiplier of 2.5.
3 new Intel Pentium OverDrive for Pentium P54C systems in the collection:
Intel PODP3V125 Pentium OverDrive
Upgrades Pentium 75 Mhz to Pentium 125 MHz.
Intel PODP3V150 Pentium OverDrive
Upgrades Pentium 90 Mhz to Pentium 150 MHz.
Intel PODP3V166 Pentium OverDrive
Upgrades Pentium 100 Mhz to Pentium 166 MHz.
- Pentium MMX OverDrives
With the introduction of the Intel Pentium with MMX Technology, Intel also created OverDrive processors to upgrade existing Pentium motherboards
to the new MMX chip. Most older Pentium motherboards cannot handle the new Pentium with MMX because of its requirement for a 2.8V core. Keeping
with the tradition of Intel's OverDrive line, the Pentium with MMX OverDrive includes a converter that lets it run in Socket 5 motherboards
(except for the 200) and Socket 7s that do not have 2.8V support. Otherwise the chip is identical to the standard Pentium with MMX.
2 new Intel Pentium MMX OverDrives in the collection:
Intel PODPMT66X166 Pentium MMX OverDrive
Upgrades original Pentium 75 to 125, 90 to 150 MHz and 100 MHz (66 MHz bus) to Pentium MMX 166 MHz.
Intel PODPMT60X180 Pentium MMX OverDrive
Upgrades original Pentium 75 to 150, 90/120 to 180 MHz and 100 MHz (50 MHz bus) to Pentium MMX 150 MHz.
|September 2005 updates and changes
13 additions to the collection today:
Cyrix MediaGX Processors
In 1996 Cyrix released the MediaGX CPU, which integrated all of the major discrete components of a PC, including memory controller, graphics,
sound and PCI controller, onto a single chip. Initially based on the old 5x86 technology and running at 120 or 133
MHz, its performance was widely criticized but its low price made it quite successful. Though it required a special motherboard and was not pin compatible
with the Pentium, it was the cheapest route into a Pentium class system available on the market. Later
versions of the MediaGX ran at speeds of up to 333 MHz and added MMX support. A second chip was added to extend its video capabilities.
The MediaGX led to Cyrix's first big win, when Compaq used it in its lowest-priced Presario computers. This led to further MediaGX sales to Packard Bell
and also seemed to give Cyrix legitimacy, as 6x86 sales to Packard Bell and eMachines quickly followed.
But because it seemed to have so much potential in the low-cost market, it dragged Cyrix's attention away from the main market - high-performance desktop
parts - and attracted the interest of other companies, notably National Semiconductor, which bought Cyrix in July 1997
largely on the strength of the MediaGX design, and over the next year or so proceeded to mismanage the company into oblivion.
MediaGX Processor FAQ
4 new Cyrix MediaGX processors in the collection:
Cyrix Gx86 GX-133GP
Before Cyrix officially introduced
the MediaGX in February 1997, some early versions - still called Gx86 before they were relabeled MediaGX - went to OEM manufacturers.
Cyrix MediaGX GXI-180BP
BGA version of the 180 MHz MediaGX, still without MMX.
Cyrix MediaGX GXm-233BP
BGA version of the 233 MHz MediaGX, with added MMX suport.
Cyrix MediaGX GXm-266GP
PGA version of the 266 MHz MediaGX with MMX suport. Fits into socket 5 and 7, but is not pin compatible with a Pentium and therefore needs a special board.
VIA Cyrix III / C3 Processors
The VIA Cyrix III, later renamed C3, is an x86 CPU based on a design by Centaur Technology, a former IDT subsidiary
and creator of the WinChip C6. VIA
bought Centaur from IDT in 1999.
While being much slower than x86 CPUs made by AMD and Intel, VIA designed the chip for the low power segment of the market, and use simplicity
in design, to deliver niche qualities which make it attractive to some buyers. Features of the C3 include the lowest manufacturing cost of all
modern x86 processors by far and low electrical consumption levels (about 10-30% of a Pentium 4 at comparable speed) and thus low heat output,
enabling the processor to run with a passive heatsink cooling solution. In later versions (Nehemiah) it also features a hardware random number
generator and very high performance AES encryption in hardware, all packed in a BGA package the size of a 1 cent coin.
VIA initially planned to develop the Cyrix III as a successor
to the Cyrix M II, codenamed "Joshua". In the meantime the Centaur team worked on the successor to that,
a WinChip 4 core based technology codenamed "Samuel". After many delays, the Joshua project was finally cancelled due to lack of performance and
the ahead of schedule Samuel was renamed and
released as the Cyrix III in June 2000.
However, VIA kept the name of "Cyrix III" in order to avoid wasting the money the company had already spent on the marketing campaign. From a
technical point of view, the Cyrix III has nothing to do with Cyrix anymore.
Introduction Press Release
VIA Cyrix III site 2000
VIA C3 site 2005
Cyrix III review on tecchannel.de
9 new VIA Cyrix III / C3 processors in the collection:
Cyrix III Samuel:
VIA Cyrix III-533 MHz
VIA Cyrix III-550 MHz
VIA Cyrix III-600 MHz
VIA Cyrix III-667 MHz
VIA Cyrix III-700 MHz
By 2001, the Samuel 2 was released, this time just as the VIA C3. Samuel 2 ran at a higher clock speed and added a level 2 cache,
required lower voltage and was also available in an Enhanced Ball Grid Array (EBGA) version. EBGA allowed the CPU to be mounted directly
onto a motherboard without using Intel's costly Socket 370.
VIA Cyrix III-750A MHz, Engineering Sample
This is a Samuel 2 Engineering Sample.
VIA C3-733 MHz
VIA C3-750A MHz, Engineering Sample
The next version C3 (internally named C5C) was the "Ezra" running at 1.35V and designed to run at 800 - 1000 MHz.
VIA C3 800 MHz
|July 2005 updates and changes
|...::: cpu-collection.de now has more than 700 chips online :::...
17 additions to the collection today:
Intel Celeron Processors
The Celeron family is a line of budget x86 processors based on Pentium P6 designs. It is marketed by Intel as a second line to complement their
more expensive but higher-performance Pentium CPUs. The first Celeron was introduced in April 1998 and based on the
Pentium II. Later versions are based on the Pentium III and
Pentium 4 designs.
The Celeron product concept was introduced by Intel in response to the company's loss of low-end market share, in particular to
Cyrix's 6x86 and AMD's K6, but also to other competers such as the
IDT Winchip. Intel's venerable Pentium MMX was no
longer performance competitive and although a faster Pentium MMX would be cheap to make and technically straightforward, Intel preferred
to move away from the industry standard Socket 7 platform (for which competitors made drop-in replacement CPUs) and produced a budget part
that was pin-compatible with their high-end Pentium II product (Slot 1). For both technical and legal reasons, competitors had difficulty
making Slot 1 parts.
Intel Celeron Processor Home
Intel Celeron (Pentium II based)
The first Celeron (codenamed Covington) is essentially a 266MHz Pentium II manufactured
without any secondary cache at all. Although clocked at 266 or 300MHz (substantially higher than the old Pentium MMX), the cacheless Celerons
are a good deal slower than the parts they were designed to replace. Substantial numbers were sold on first release, largely on the strength
of the Intel name, but the Celeron quickly achieved a poor reputation both in the trade press and among computer professionals. Many people
referred to it unkindly but accurately enough as the Slugeron, Celery or Deceleron. The initial market interest faded rapidly in the face of
its poor performance and with sales at a very low level, Intel felt obliged to develop a substantially faster replacement as soon as possible.
2 new Celeron Covington processors in the collection:
Intel Celeron 266 MHz
Intel Celeron 266 MHz Engineering Sample (one processor available for trade)
An engineering sample of the very first Celeron version. It was made in February 1998, some weeks before the official introduction of the
Celeron on April 15th, 1998. There is one Engineering Sample of the Celeron 266 MHz available for trade, get it while it's hot.
Intel was well aware of the poor reputation of the original Celeron and determined not to make the same mistake twice, with the result that
the new Mendocino core Celeron was a good performer from the outset. Indeed, most industry analysts regarded the first Mendocino-based Celerons as
too successful - performance was sufficiently high to not only compete strongly with rival parts, but also to attract buyers away from Intel's
high-profit flagship, the Pentium II.
The key to the new Celeron's performance is cache. Where the old model had no secondary cache at all, the new part includes 128KB of L2 cache as
part of the chip itself. Otherwise, it is identical. With a total of 19.2 million transistors (including cache) on a single chip, the Mendocino
Celeron was difficult and expensive to manufacture, but Intel managed a flawless execution of an ambitious project.
The first Mendocino-core Celeron was clocked at a then-modest 300MHz but was almost twice as fast as the old cacheless Celeron at the same
clockspeed. To distinguish it from the old model, Intel called it the 300A. Although the other Mendocino Celerons (this 333MHz part, for example)
did not have an A appended, some people call all Mendocino processors Celeron-A regardless of speed.
8 new Celeron Mendocino processors in the collection:
Intel Celeron 333 MHz (one processor available for trade)
Intel Celeron 366 MHz
The Mendocino Celerons also introduced new packaging. When the Mendocinos debuted they came in both a Slot 1 and Socket 370 PPGA package.
The Slot 1 form had been designed to accommodate the off-chip cache of the Pentium II and had mounting problems with motherboards.
Because all Celerons are a single-chip design, however, there was no reason to retain the slot packaging for L2 cache storage, and Intel
discontinued the Slot 1 variant: beginning with the 466 MHz part, only the PPGA Socket 370 form was offered.
Intel Celeron FV524RX366 (366 MHz)
Intel Celeron FV524RX400 (400 MHz)
Intel Celeron FV524RX433 (433 MHz; one chip available for trade)
Intel Celeron FV524RX466 (466 MHz)
Intel Celeron FV524RX500 (500 MHz)
The Celeron processors were also offered as mobile versions in different variants. This here is an MMC-2 module with a 400 MHz Celeron on it:
Intel Celeron 400 MHz MMC-2
Intel Celeron (Pentium III based)
The Pentium III generation Celeron is the Coppermine-128, also known as the Celeron II. These are a derivative of the Coppermine
Pentium III and were released in March 2000. The Celeron-128 used 128 KB of on-chip L2 cache and is
restricted to a 66MHz bus speed in early versions. It is identical to the Pentium III except for the smaller secondary cache and the slower bus.
6 new Celeron Coppermine-128 processors in the collection:
Intel Celeron 600/128/66/1.7V
Intel Celeron 633/128/66/1.7V
Intel Celeron 700/128/66/1.7V
Intel Celeron 950/128/100/1.75V
There were also mobile versions of the Coppermine-128:
Intel Celeron KP 600/128
Intel Celeron KP 650/128
The next series of Celerons was based on the Pentium III Tualatin core, and made with a 0.13 µ process. They were nicknamed Tualeron. The series
began with 1000 and 1100 MHz parts (which were given the extension "A" to their name to differentiate them from the Coppermine-128 of the same
speed they replaced) and the line continued with 1200, 1300, and 1400 MHz chips.
Tualerons are identical to their fully-fledged Pentium III sibling, except that they use a 100 MHz bus rather than a 133 MHz bus. They are excellent
overclockers, since they have higher multipliers and users could put them on a 133 MHz bus easily. The cache is the same as in the Pentium III;
256KB in both.
1 new Celeron Tualatin processor in the collection:
Intel Celeron 1000A/256/100/1.475
|May - June 2005 updates and changes
12 additions to the collection today:
AMD Athlon K7
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
AMD Athlon Tech Docs
4 new Athlon K7 processors in the collection:
The original Athlon K7, code-named "Argon", was manufactured in a 0.25 µ process. In the year of the K7's introduction AMD switched their manufacturing process
to a more advanced aluminum 0.18-micron six-layer metal technology. The first Athlon to be built using the new technology was the K75 (codenamed "Pluto" for the
550-850 MHz versions and "Orion" for the 900 - 1000 MHz versions). K7 Athlons are marked with "C", K75 Athlons with "A" at the end of the product code.
AMD Athlon K7500MTR51B C (K7, 500 MHz)
AMD Athlon K7600MTR51B A (K75, 600 MHz)
AMD Athlon K7700MTR51B A (K75, 700 MHz)
AMD Athlon K7750MTR52B A (K75, 750 MHz)
AMD Athlon Thunderbird
On June 4, 2000 AMD introduced an enhanced version of the Athlon processor, codenamed "Thunderbird". Fabricated using AMD's 0.18-micron process technology,
the new core replaced the K75 chip's 512KB of off-die Level 2 cache by 256KB of cache integrated onto the die itself and running at the full clock speed of the
processor. This is in contrast to the original Athlons that operated their L2 cache at a certain fraction of the core clock speed.
As well as boosting performance, moving the cache on-die also allowed AMD to follow Intel's lead in moving from slot-based processors in favour of a socket
form factor - in AMD's case, a 462-pin format, named Socket A. Supporting PC133 memory, the enhanced Athlon processor was
initially available in six speeds, from 750 MHz to 1 GHz, in both Slot A (albeit available to OEMs only) and the new Socket A packaging.
The last Athlon processors based on the Thunderbird core were released in the summer of 2001, by which time speeds had reached 1.4 GHz. Hereafter,
the Athlon was to be replaced by the Athlon XP family based on the new Palomino core.
6 new Athlon Thunderbird processors in the collection:
The Athlon Thunderbird, like the K7/K75, used a 100 MHz mainboard bus.
AMD Athlon A0700APT3B (700 MHz)
AMD Athlon A0850AMT3B (850 MHz)
AMD Athlon A1100AMS3B (1100 MHz)
In October 2000 the Athlon "C" was introduced, raising the mainboard front side bus speed to 133 MHz (FSB 266) and providing roughly 10% extra performance over
the "B" model Thunderbird.
AMD Athlon A1200AMS3C (1200 MHz)
AMD Athlon A1333AMS3C (1333 MHz)
AMD Athlon A1400AMS3C (1400 MHz)
AMD Athlon XP
AMD released the Athlon XP ("eXtra Performance") on May 14, 2001, code-named "Palomino". This Athlon version, the first to
include the SSE instruction set from the Intel Pentium III as well as AMD's 3DNow! Professional, was introduced at speeds between 1333 and 1733 MHz.
The major changes were optimizations to the core design to increase efficiency by roughly 10% over a Thunderbird at the same clock-speed, and power
consumption reductions to allow it to be clocked faster.
The "Palomino" was first released as a mobile version, called the Mobile Athlon 4 (also code-named "Corvette"), after the fact that it was AMD's fourth core
to be called Athlon (after the original K7, the 0.18 µm K75, and the Thunderbird), but many people noted that the name was most likely a jab at the
then-brand-new Intel Pentium 4. The desktop Athlon XP followed a few months later, in October.
The Athlon XP was marketed using a PR rating system, which compared its performance to a Thunderbird. Because the XP has much higher IPC
(instructions per clock) than the Pentium 4 (and about 10% higher than a Thunderbird), it is more efficient and delivers the same level of performance at
a lower clock-speed, or higher performance at the same speed.
2 new Athlon XP processors in the collection:
AMD Athlon XP 1500+ AX1500DMT3C (1333 MHz)
AMD Athlon XP 1800+ AX1800DMT3C (1533 MHz)
|April 2005 updates and changes
14 additions to the collection today:
The K5 was developed by AMD to compete with Intel's Pentium microprocessor range. Introduced in 1995 more than 1 year late,
AMD's problems were compounded by bring unable to manufacture the chip at the clock speeds originally projected. In its favor, the K5 did at least offer good
All models had 4.3 million transistors on-chip. No K5 supported MMX instructions. Internally ambitious, it was closer to a
Pentium Pro than a Pentium, based upon an internal highly parallel RISC processor architecture with an x86 decoding front-end.
Improvements and differences to the Intel Pentium include:
The K5 project represented an early chance for AMD to take technical leadership from Intel. Although the chip addressed the right design concepts, the actual
engineering implementation was weak. The low clock rates were due in part to AMD's deficiencies as a manufacturing company in the period. However, having a
branch prediction unit four times the size of the Pentium, yet reportedly not delivering superior performance, is an example of how the actual implementation
fell short of the project goals. Additionally, while the K5's floating point performance was better than that of the Cyrix 6x86,
it was weaker than that of the Pentium.
- Five integer units, which could process instructions out of order, one floating point unit, compared to two units of the Pentium
- The branch target buffer was four times the size of the Pentium's, although not reportedly more accurate
- Register renaming improved parallel performance of the pipelines
- Speculative execution of instructions reduced pipeline stall
- The instruction cache is 16 Kb, double the Pentium
- The primary cache is 4-way set associative instead of the Pentium's 2-way
Because it was late to market and did not meet performance expectations, the K5 never gained the acceptance among
large computer manufacturers that the Am486 and AMD K6 enjoyed. Overall, the chip failed
to deliver, both in terms of raw performance, and financially for AMD.
There were two sets of K5 processors, internally called the 5k86/SSA5 and the K5. The 5k86 line ran from 75 to 90 MHz;
the K5 line ran from 90 to 133 MHz. However, with later versions of the K5 AMD used what it called a PR rating, or performance rating, to label the chips according
to their equivalence to a Pentium of that clock speed. Thus, a 116 MHz chip from the second line was marketed as the AMD-K5 PR166.
5 new AMD K5 processors in the collection:
AMD-K5-PR75ABR (75 MHz)
AMD-K5-PR90ABQ (90 MHz)
The standard ceramic PGA package versions of the 100 MHz K5 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:
AMD-K5-PR100ABQ Goldcap (100 MHz)
With the K5 PR120 AMD changed the naming scheme to match an equivalent Pentium's clock speed, so the K5 PR100 was the last K5 to show its clock rate in the PR rating.
AMD-K5-PR120ABQ (90 MHz)
AMD-K5-PR120ABR (90 MHz)
Additionally, there are some AMD-K5-PR75ABR available for trade.
AMD K5 Tech Docs
AMD-K5 Processor Overview
Byte.com: Next generation: K5
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.
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 and continued with the K6-2.
5 new AMD K6 processors in the collection:
Initially, the AMD K6 processors used Pentium II Rating (PR2) to designate their speed. The PR2 rating was dropped because the rated frequency of the processor
was the same as the real frequency. This item was made in the month the K6 was released (week 17/1997 = April 21st - 27th, 1997) and has the initial PR2 marking:
This K6 is even younger, produced end of March 1997, but it does not have a PR2 marking. It still misses the I/O voltage marking later K6-166 have:
And this one has a strange engraved logo, not like the K6-2 chips have, but quite different:
AMD-K6-166ALR, Engraved Logo
This chip has the K6 marketing logo:
AMD-K6-200ALR, 'K6' Logo
Last one is a ALYD version of the K6-200:
AMD K6 Tech Docs
Byte.com: AMD K6 Takes On Intel P6
Intel's Enemy No. 1: The AMD K6 CPU
The K6-2 was a significant improvement over the K6. It built upon the K6's processing core, with the addition of 21 new instructions
called "3D Now!". These are SIMD (Single Instruction Multiple Data) instructions designed to enhance the 3D geometry capability of the chip's floating point unit.
This allowed the K6-2 to overcome the handicap of the slow (relative to Intel) CPU which the K6 owned.
The K6-2 also saw the introduction of a 100 MHz front side bus. This was only available to owners of new Super Socket 7 motherboards which also included features such
as AGP. All of these features helped to give the K6-2 performance fast enough to be a credible challenger to the dominant
Pentium II. Performance of the two chips was broadly similar: the K6-2 tending to be faster for general-purpose
computing, the Intel part clearly superior at floating-point tasks. The K6-2 was a very successful chip and provided AMD with the marketing base and the
financial stability it would need to introduce the Athlon.
4 new AMD K6-2 processors in the collection:
AMD K6-2 Tech Docs
|January - March 2005 updates and changes
|...::: cpu-collection.de now has more than 600 chips online :::...
23 new additions to the collection today:
Intel 8080 processors and second source versions
The Intel 8080 is an early 8-bit CPU, released in April 1974 running at 2 MHz, and is generally considered to be the first truly usable microprocessor CPU design.
It is the successor to the Intel 8008, with which it was assembly language source-compatible because it used the same
instruction set developed by Computer Terminal Corporation. The 8080's 40 pin DIP packaging permitted it to provide a 16-bit address bus and an 8-bit data bus,
allowing access to 64 kilobytes of memory. It has seven 8-bit registers (six of which could be combined into three 16-bit registers),
a 16-bit stack pointer to memory (replacing the 8008's internal stack), and a 16-bit program counter.
The 8080 had 256 I/O ports allowing I/O devices to be connected without the need to allocate memory space - as is required for memory mapped devices -
but at the expense of having programmers deal with separate I/O instructions. The first single-board microcomputer was built on the basis of the 8080.
The 8080 was used in many early computers, such as the MITS Altair 8800 and
IMSAI 8080, forming the basis for machines running the CP/M operating
system (the later, fully compatible and more capable, Zilog Z80 processor would capitalize on this, with Z80 & CP/M becoming the dominant CPU & OS combination
of the period much like x86 & MS-DOS for the PC a decade later).
Shortly after the launch of the 8080, the Motorola 6800 competing design was introduced, and after that,
the MOS Technology 6502 clone of the 6800.
At Intel, the 8080 was followed by the compatible and electrically more elegant 8085, and later by the assembly language compatible 16-bit
8086 and then the 8/16-bit 8088, which was selected by IBM for its new PC to be launched in 1981. The 8080, via its ISA, thus got a lasting impact on computer history.
The 8080 was very popular and was second-sourced by various manufacturers. Clones of the 8080 were also made in
former Eastern Block countries like USSR, Poland, CSSR, Hungary and Romania.
Versions and speed:
8080/8080A: 2.0 MHz
8080A-1: 3.125 MHz
8080A-2: 2.67 MHz
All 8080 chips on cpu-collection.de
8 new Intel and second source 8080 chips in the collection:
An 8080A-1 in ceramic DIP, running at 3.125 MHz.
AMD AM9080ACC / D8080A (Black Ceramic)
AMD AM9080ACC / D8080A
The AMD 9080A was a plug-in replacement for Intel's 8080, an unauthorized second source. As a result of that AMD negotiated a technology cross-license and
patent cross-license with Intel in 1975 and renamed the chip 8080A.
Siemens SAB 8080 A-C
Texas Instruments TMS8080A
Eastern Block 8080 peripheral KP580BB55A (USSR)
The KP580BB55A is a programmable peripheral interface chip for use with Soviet i8080 clones KP580BM80A.
It is a clone of the Intel i8255.
Intel 8080A/8080A-1/8080A-2 datasheet [pdf]
Early Intel chip development and naming scheme [pdf]
Technical information and differences between 8080 and 8080A at cpu-museum.com
Intel 8085 processors and second source versions
The Intel 8085 is an 8-bit processor made by Intel and introduced in 1976. It is binary compatible with the Intel 8080 but requires less supporting hardware,
thus allowing simpler and less expensive microcomputer systems to be built.
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 "5" in the model number came from the fact that the 8085
required only a 5-volt power supply. The 8085 was a transition design on the way to the 16-bit 8086.
There are multiple versions of Intel 8085 processor: The original 8085, the 8085A with bug fixes and the 8085AH HMOS version. Some second-source manufactures also
produced CMOS version of the 8085 microprocessor (80C85).
The 8085 was produced at speeds ranging from 3 MHz to 6 MHz:
8085/8085A(H): 3 MHz
8085A(H)-1: 6 MHz
8085A(H)-2: 5 MHz
All 8085 chips on cpu-collection.de
15 new Intel and second source 8085 chips in the collection:
AMD AM8085APC / P8085A
AMD AM8085ADCB / C8085AB
Toshiba TMP8085AP, old Logo
Intel 8085A/8085A-2 datasheet (kind of...)
Following processors are a donation by Jorge Segovia:
AMD AM9080ACC / D8080A (Black Ceramic),
Texas Instruments TMS8080A,
Thank you very much for your support, Jorge!