Pentium Processor

In: Computers and Technology

Submitted By blackgai
Words 348
Pages 2
“The Pentium Microprocessor Flaw”
Jamaal Bonner
April 25, 2014
NT1110/Computer Structure and Logic
Professor Marcus Price

The Pentium Microprocessor Flaw The Pentium microprocessor is the CPU (central processing unit) developed by Intel. Pentium chips include a floating-point unit (FPU) that has integrated instructions that tell the chip how to compute integer arithmetic, making them much faster [for heavy numerical calculations], more complex, and more expensive. The problem for Intel is that all Pentiums manufactured until sometime in fall of 1994 had errors in the on-chip FPU instructions for division. This caused the Pentium's FPU to incorrectly divide certain floating-point numbers.

Many software packages don't actually use a computer's FPU. These packages don't show the error and only certain numbers divide incorrectly. Thomas Nicely, a math professor at Lynchburg College, discovered the flaw in the Pentium’s FPU in summer/fall 1994. He computed the sum of the reciprocals of a large collection of prime numbers on his Pentium-based computer and found the result differed largely from theoretical values. Nicely posted a general notice of the flaw on the Internet and asked others to confirm his findings after receiving no response from Intel, which ultimately led to magazine and television interviews.

Intel publicly announced that the subtle error would only occur to an average spreadsheet user once in every 27,000 years of use. Critics noted that the Pentium’s output for bad inputs were wrong every time and others suggested that some bad inputs might occur with disproportionate frequency in common calculations. Within a month, IBM halted shipment on Pentium-based computers and announced that common spreadsheet programs could produce Pentium-related errors as often as once every 24 days.

Intel publicly admitted the problem around November 28,…...

Similar Documents

Intel Processors

...years Intel innovations have continuously created new possibilities in the lives of people around the world. 1975 The Altair 8800 microcomputer, based on the Intel® 8080 microprocessor, was the first successful home or personal computer. 1972 The Intel® 4004 processor, Intel’s first microprocessor, powered the Busicom calculator and paved the way for the personal computer. 1994 Intel chips powered almost 75 percent of all desktop computers. 1976 An operator in an early bunnysuit shows how a 4-inch wafer is prepared for a positive acid spin. 1982 Within 6 years of its release, an estimated 15 million 286-based personal computers were installed around the world. 2001 The Itanium® processor is the first in a family of 64-bit products from Intel and is designed for high-end, enterprise-class servers and workstations. 2005 Dual-core technology was introduced. The Revolution Continues 2006 Intel launched four processors for servers under the Xeon 5300 brand, and another processor under the Core 2 Extreme series for high performance computing. These "quad-core" processors show improved performance over others with just one or two processing cores. 1995 Released in the fall of 1995, the Intel® Pentium® Pro processor was designed to fuel 32-bit server and workstation applications, enabling fast computer-aided design, mechanical engineering and scientific computation. Intel continues to deliver on the promise of Moore’s Law with the introduction of powerful......

Words: 1484 - Pages: 6

Pentium Flaw

...Unit 5 Analysis 1 : Pentium Flaw The Pentium microprocessor is the CPU (central processing unit) for the widest selling personal computers. Unlike previous CPUs made by Intel, the 486DX and Pentium chips came along with a floating-point unit (FPU) also known as the math co-processor. Previous Intel CPUs did all their arithmetic using integers; programs that used floating-point numbers (non-integers like 2.5 or 3.14) needed to tell the chip how (for example) to divide them using integer arithmetic. The 486DX and Pentium chips have these instructions built into the chip, in their FPUs. This makes them much faster for intense numerical calculations, more complex, and more expensive. The problem for Intel is that all Pentiums manufactured until sometime this fall had errors in the on-chip FPU instructions for division. This caused the Pentium's FPU to incorrectly divide certain floating-point numbers. Many software packages, including many that do use floating-point numbers, don't actually use a computer's FPU. These packages don't show the error. Also, only certain numbers (whose binary representation show specific bit patterns) divide incorrectly. Consequently many users may never encounter the division error. The most famous example and the worst well-known case is 4195835/3145727, discovered by Tim Coe of Vitesse Semiconductors. The correct value is 1.33382 to 6 sig. figs, while the flawed Pentium's floating-point unit computed 1.33374 to 6 sig figs, a relative error of......

Words: 678 - Pages: 3

Pentium Flaw

...Pentium Flaw The Pentium Microprocessor is the widest selling personal computers on the market. Intel created the 486DX and the Pentium chip that included Floating point unit (FPU). This FPU was known as a math co-processor. The problem arose in the fall of 1994, all the new Pentium chips were manufactured with an error in the FPU system. This error was in the FPU instructions for the division. This caused the chip to divide certain numbers incorrectly. Intel engineers discovered a flaw in the floating-point math subsection of the Pentium processor. But instead of assessing this problem they simply ignored it and thought that the customers would not ever notice the flaw. Thomas Nicely discovered this problem also when he was computing the sum of large reciprocals of prime numbers on his Pentium base computer. Later on Thomas emailed Intel and told them that there was a flaw in the co-processor. Intel never responded back so he took it to his own hands and posted the flaw online, asking other people to check for the flaw also on other users computers. Many people responded back and the news got involved sooner or later Intel started receiving more and more emails about the flaw. A common Known error was 4195825/3145727 dividing those numbers gave an incorrect answer to the problem. Intel’s response to this flaw was to replace Pentium chip microprocessor to whoever could explain the flaw to them. This angered man of the customers so Intel responded by replacing......

Words: 410 - Pages: 2

Pentium Flaw

...The Pentium flaw was that the math coprocessor would have errors when dividing certain floating numbers. “The most famous example and the worst well-known case is 4195835/3145727, discovered by Tim Coe of Vitesse Semiconductors.” (Mark Janeba 1995). This error was originally found by Intel designer shortly after the release of the processor, but it was thought to be such a minor error that there was no need to worry about it. Later the problem was found by a college math professor, named Thomas Nicely at Lynchburg College, who was computing high level equations when he notices that some of the answer that the computer was giving him did not match what his own work said the answer should be. By trying a few tests he was able to determine that the problem was with the Pentium processor in the computer. He then notified Intel about this problem and got no response back. When he got no response from Intel he posted his findings online and asked others to verify his findings, which led to the flaw being exposed on national TV and magazine. Intel’s first reaction was to ignore the problem stating that it would not be a problem for most users, then they offered replacement processors for those who could prove that they needed it for work. Finally after much negative publicity they did a full factory recall. Intel handled this problem by first refusing to acknowledge its existence, and when they could not ignore it any longer they did the bare minimum to try and fix it. At......

Words: 601 - Pages: 3

The Pentium Flaw

...Collin Pannell Warwick NT 1110 October 22, 2013 The Pentium Flaw During the year of 1994 one of Intel’s processors called the Pentium microprocessor chip had a small flaw inside the chip. These Pentium processors included something new called a floating-point unit, which is also know as a math coprocessor. These Pentium processors had this floating-point unit built into the processors, which makes them much faster. These chips that were made had errors on the floating-point unit that corrupted instructions for division. This cased the Floating-point unit to divide certain floating-point numbers incorrectly. This Pentium flaw was most know from an equation solved by Tim Coe of Vitesse Semiconductors. The number he tried to divide was 4195835 by 3145727. The value that was outputted by the computer was incorrect by 0.006%. Thomas Nicely, a math professor at Lynchburg College was computing the sum of the reciprocals of a large collection of prime numbers on one of Intel’s Pentium-based computers. He calculated the same equation on a different program and then traced the problem back to the Pentium itself. Nicely contacted Intel and didn’t get any real response to the problem. He then posted on the Internet asking others to confirm the problem he had found. Nicely was interviewed with a magazine and CNN aired a story about by his findings. After Nicely’s findings, Intel publicly announced “an error is only likely to occur [about] once in nine billion random floating...

Words: 462 - Pages: 2

Pentium Flaw

...The Pentium flaw was discovered in June 1994. The Pentium floating point flaw occurred in some models of the original Pentium microprocessor chip. Pentium processors with a clock speed of 120 MHz or higher are not affected by the flaw. The Intel Processor Frequency ID Utility checks for the floating point flaw on the affected microprocessors. Professor Thomas Nicely discovered the Pentium floating point unit flaw and emailed various contacts describing the flaw. Professor Nicely requested testing for the flaw on the 486-DX4s, and the Pentium clones. The flaw in in the Pentium floating point unit was verified and became known as the FDIV bug. The issue was a mathematical problem in the Floating Point (FPU), the math coprocessor. The Pentium chip had glitches when calculation large divisions. Testing on the 486 pin pointed the error to the Pentium Chip. Intel confirmed the error but said they had no other reports of it. Intel the considered the flaw as a minor problem and did not let the industry know about the flaw. However word of the flaw got out through the internet and articles and Intel started getting negative publicly. Intel handled did not handle the problem with the flaw correctly. By considering the flaw minor to avoid fixing the problem and negative publicly, Intel showed poor judgment and integrity as a company. Intel only attempted to fix the flaw after receiving negative publicly. To satisfy customers concerned about the flaw Intel fix the......

Words: 353 - Pages: 2


...Processor | Transistor count | Date of introduction | Manufacturer | Process | Area | Intel 4004 | 2,300 | 1971 | Intel | 10 µm | 12 mm² | Intel 8008 | 3,500 | 1972 | Intel | 10 µm | 14 mm² | Intel 8080 | 4,500 | 1974 | Intel | 6 μm | 20 mm² | Intel 8085 | 6,500 | 1976 | Intel | 3 μm | 20 mm² | Intel 8086 | 29,000 | 1978 | Intel | 3 μm | 33 mm² | Intel 8088 | 29,000 | 1979 | Intel | 3 μm | 33 mm² | Intel 80186 | 55,000 | 1982 | Intel | 3 μm | | Intel 80286 | 134,000 | 1982 | Intel | 1.5 µm | 49 mm² | Intel 80386 | 275,000 | 1985 | Intel | 1.5 µm | 104 mm² | Intel i960 | 250,000[7] | 1988 | Intel | 0.6 µm | | Intel 80486 | 1,180,235 | 1989 | Intel | 1 µm | 173 mm² | Pentium | 3,100,000 | 1993 | Intel | 0.8 µm | 294 mm² | Pentium Pro | 5,500,000[10] | 1995 | Intel | 0.5 µm | 307 mm² | Pentium II Klamath | 7,500,000 | 1997 | Intel | 0.35 µm | 195 mm² | Pentium II Deschutes | 7,500,000 | 1998 | Intel | 0.25 µm | 113 mm² | Pentium III Katmai | 9,500,000 | 1999 | Intel | 0.25 µm | 128 mm² | Pentium III Coppermine | 21,000,000 | 2000 | Intel | 180 nm | 80 mm² | Pentium III Taulatin | 45,000,000 | 2001 | Intel | 130 nm | 81 mm² | Pentium 4 Willamette | 42,000,000 | 2000 | Intel | 180 nm | 217 mm² | Pentium 4 Northwood | 55,000,000 | 2002 | Intel | 130 nm | 145 mm² | Pentium 4 Prescott | 112,000,000 | 2004 | Intel | 90 nm | 110 mm² | Pentium 4 Prescott-2M | 169,000,000 | 2005 | Intel | 90 nm | 143 mm² | Pentium 4 Cedar Mill | 184,000,000...

Words: 578 - Pages: 3

Pentium Flaw giant, Intel which is now the largest CPU manufacturer in the world. In 1993 Intel introduced the Pentium series CPU. The Pentium architecture became the most successful processor of all time. As successful as the Pentium was it had issues. Most can and could not be seen by the typical user but in 1994 it had a flaw that was noticed by a mathematician from Lynchburg collage in West Virginia by the name of Thomas Nicely while doing some calculations on his PC. The Pentium flaw was the flaw in the Float-point unit would cause incorrect answers when preforming double precision arithmetic, which is a common operation. Intel’s first response to the problem was to deny that the problem did not exist. Then would later acknowledge the flaw. However state that the flaw was insignificant and most users would not notice. Intel would replace the defected processor if a user could demonstrate the need for a processor that was not flawed. After IBM cancelled sales of its computers that had the flawed processer in it and lots of negative publicity Intel replaced the flawed processor with an unflawed processor to any customer who asked for it. In conclusion I believe that Intel acted irresponsibly and unethically to the problem. They should have admitted the problem from the get go and replaced it for all that purchased a computer using the flawed processor. I think would have looked better on their image as a company. Denying the problem and then taking the......

Words: 402 - Pages: 2


...Task 1: LGA 1156 Connector- Uses I5, I7, Xeon, Celeron and Pentium processors. Its purpose is to allow PCI, DMI and FDI devices to connect to the rest of the system. Intel H55 Chipset-( Uses Pentium, I3, I5, I7 and Xeon processors. Its purpose us to allow PCI, USB, Network and serial devices to connect to the rest of the system Gigabyte GA-H55M-UD2H Motherboard- Supports Lynnfield I5,I3 and I7 CPU’s Task 2: (AMD.COM) AMD 770 Northbridge- Uses Phenom processors and it is for systems used for high end gaming. AMD SB710 Southbridge- Uses Athlon processors and it is for systems used for high end embedded client applications. Socket AM3 Connector- Uses Phenom II processors and was used to support DDR3 SDRAM. Task 3: ( Processor | Clock Speed | Internal Cache | Bus Speed | Archictecture | I7 | 3.7 Ghz | 1.5 MB | 4.8 GT/s | Bloomfield | I7 Mobile | 1.6 Ghz | 6 MB | 2.5 Gt/s | Clarksfield | I5 | 3.46 Ghz | 8 MB | 4.8 GT/s | Lynnfield | I5 Mobile | 2.40 Ghz | 3 MB | 2.5 GT/s | Arrandale | I3 | 2.93 Ghz | 4 MB | 2.5 GT/s | Clarkdale | I3 Mobile | 2.13 Ghz | 3 MB | 2.5 GT/s | Nehalem | Vpro | 2.90 GHz | 3 MB | 2.5 GT/s | Clarksfield | Vpro Mobile | 3.40 GHZ | 4 MB | 5 Gt/s | Nehalem | Xeon 7000 | 1.73 Ghz | 4 MB | 4 GT/s | Nehalem | Xeon 5000 | 1.87 Ghz | 4 MB | 2 GT/s | Gainestown | Xeon 300 | 2 Ghz | 4 MB | DMI | Clarkdale | Itanium 9000 | 733 Mhz | 2 MB | 133 Mhz | Merced | Pentium | 50 Mhz | 2 MB | 60 MT/s | P5 | Celeron | 267 Mhz |......

Words: 366 - Pages: 2

Pentium Flaw

...Pentium Flaw The Pentium Flaw, also known as the Pentium FDIV bug, was a bug located in the FPU, and certain floating point division operations performed would produce incorrect results, which includes decimals, addition and subtraction. These are mostly used in math and some science work. Intel was heavily criticized for the flaw, and the handling of this flaw, which caused them to recall every one of the processors. Dr. Thomas Nicely discovered the bug in 1994. After emailing Intel and receiving no response, Dr. Nicely went ahead and posted a message on the internet. Word of the bug quickly spread to industry press. Intel responded by saying it was a minor issue, this was not accepted by computer users. Intel later changed its position and offered to replace every chip, which resulted in a $500 million dollar loss for the company. This also actually propelled the company from being a tech supplier to now becoming a house hold name. Next, they in turn used the negative publicity to change business practices and become an overall better company. The Pentium Flaw made Intel into the most successful processor making company that it is today. Intel achieved a position of dominance in the market as a supplier. While the Pentium Flaw hurt the company short term, it actually helped the company in the long run. Gaining popularity and ensuring consistency within its products. Without the Pentium Flaw no one knows if Intel would be as successful as they are today, or if any of...

Words: 266 - Pages: 2

Pentium Flaw

...Pentium Flaw The Pentium microprocessor flaw was basically that the CPU didn’t know how to do arithmetic or math calculations, so it failed at calculating any type of math. The Pentium flaw was found in the summer/fall of 1994 by Thomas Nicely according to Mark Janeba from the article I read. Mark Janeba stated that Thomas Nicely was a math professor at Lynchburg College and that Thomas was computing the sum of the reciprocals of a large collection of prime numbers on his Pentium-based computer. That’s when the problem started showing up, the CPU gave Thomas different results from his theoretical values so Thomas did the same problem on a different computer, the computer gave Thomas the same answer he got theoretically. He finally tracked the problem to the Pentium chip itself. According to the article by Mark Janeba, Intel admitted the problem publically around November of 1994. Intel was going to replace the Pentium chips but only for the people that could explain their need for complicated calculations. In the end of December Intel stated that a free replacement Pentium CPU would be given for anyone that asks for one. Basically Intel was going to do a total recall, replacement and they were going to destroy any of the flawed Pentium microprocessors. I honestly don’t think Intel handled the problem very well because at first they ignored it and said that it wasn’t happening to every Pentium CPU but people were saying that the Pentium CPU was not giving them the right......

Words: 452 - Pages: 2

Processor Model

...Processor Model Year Transistor Count Intel 4004 1971 2,300 Intel 8008 1972 2,500 Intel 8080 1974 4,500 Intel 8085 1976 8,500 Intel 8086 1978 29,000 Intel 8088 1979 29,000 Intel 80186 1982 55,000 Intel 286 1982 134,000 Intel 386 1985 275,000 Intel 486 1989 1,200,000 Intel Pentium I 1993 3,100,000 Intel Pentium II 1997 7,500,000 Intel Pentium III 1999 9,500,000 Intel Pentium 4 2000 42,000,000 Intel Itanium 2001 25,000,000 Intel Itanium 2 2003 220,000,000 Intel Itanium 2 (9MB Cache) 2004 592,000,000 Core 2 Duo 2006 291,000,000 Core i7 (quad) 2008 731,000,000 Quad-core Itanium 2010 2,000,000,000 Six-Core i7/8 2011 2,270,000,000 8-Core Itanium 2012 3,100,000,000 62-Core Xeon Phi 2010 5,000,000,000 The processor chip to hold 2 billion transistors is the Intel Itanium 9300 Processor announced February 8, 2010. The growth is reasonable, computing power has grown every few years and will continue to rise in the future just as it has done since its first Intel Processor in 1971. My opinion on the growth moving fast or slow I would say it’s moving very fast every two years a new processor is out and technology is evolving with it every day. Reaching 100 billion even a trillion transistors will probably be around 2022 since looking at history every chip doubles every two years. References:

Words: 261 - Pages: 2


...Processors and Their Characteristics Three of the latest and fastest processors available are the 1 Intel Core i7-980x. This is the leader in speed. For under $1,000 the “3.33GHz Core i7-980x (with overclocking to 3.6 GHz) is Intel’s flagship.” (Murphy, 2010) It's the first desktop-grade CPU with six physical cores, but hyper threading yields 12 virtualized ones. The six cores share 12MB of integrated L3 cache. The end result is a measurable performance boost for optimized applications--and a score of 147 on our World Bench 6 tests.” (Murphy, 2010) Core i7-870 processor Intel’s 2.93GHz (with automatic overclocking to 3.6GHz) Core i7-870. The differences between this $564 processor and the Core i7-980X, though important, result in only a 14 percent decrease in performance from that of the Core i7-980X. The i7-870 drops its internal L3 cache to 8MB. In addition, it carries a 45-nanometer designation (the size of the space between transistors). The smaller this number, the smaller and more numerous the transistors that can be packed onto a processor. Also, this midrange CPU limits you to four memory slots on your motherboard instead of six. BENCHMARK SCORE: 127 AMD has a brand-new, six,-core, 3.2GHz Phenom II X6 1090T processor. Codenamed "Thuban," this $299, 45-nanometer CPU can hit speeds of up to 3.6GHz as a result of Turbo Core, the AMD equivalent of Intel's automatic overclocking technology (which is called Turbo Boost)...

Words: 524 - Pages: 3


...Intel 4004 2,300 1971 Motorola 6800 4,100 1974 Intel 8080 4,500 1974 RCA 1802 5,000 1974 Intel 8085 6,500 1976 Zilog Z80 8,500 1976 Motorola 6809 9,000 1978 ARM 1 25,000 1985 ARM 2 25,000 1986 Intel 8086 29,000 1978 Intel 8088 29,000 1979 ARM 6 30,000 1991 Intel 80186 55,000 1982 Intel 80286 134,000 1982 Intel 80386 275,000 1985 ARM 7 578977 1994 Intel 80486 1,180,235 1989 R4000 1,350,000 1991 Pentium 3 100,000 1993 AMD K5 4,300,000 1996 Pentium Pro 5 500,000 1995 Pentium II 7,500,000 1997 AMD K68 800,000 1997 Pentium III 9,500,000 1999 AMD K6-III 21,300,000 1999 AMD K7 22,000,000 1999 ARM Cortex-A9 26,000,000 2007 Pentium 4 42,000,000 2000 Atom 47,000,000 2008 Barton 54,300,000 2003 AMD K8 105,900,000 2003 Itanium 2 McKinley 220,000,000 2002 Celll 241,000,000 2006 Core 2 Duo 291,000,000 2006 Itanium 2 Madison 6M 410,000,000 2003 AMD K10 quad-core 463,000,005 2007 Itanium 2 with 9MB 592,000,000 2004 Core i7 (Quad) 731,000,000 2008 AMD K10 quad-cor 758,000,000 2008 POWER 6 789,000,000 2007 Six-Core Opteron 2400 904,000,000 2009 16-Core SPARC T3 1,000,000,000 2010 Quad-Core + GPU Core i7 1,160,000,000 2011 Six-Core Core i7 (Gulftown) 1,170,000,000 2010 8-Core AMD Bulldozer 1,200,000,000[8] 2012 8-core......

Words: 436 - Pages: 2

Computer Processors

...Harrison Intro to Computer Hardware SHJC Assignment # 8 1) State how many Generation of Processor there are. There are six generations of Processor. 2) Explain each of the Generation of Processor. P1 (086) Processor - Intel introduced the 8086 back in June 1978. The 8086 was one of the first 16-bit processor chips on the market; at the time, virtually all other processors were 8-bit designs. The 8086 had 16-bit internal registers and could run a new class of software using 16- bit instructions. [pic][pic] P2 (286) Processor - In 1982, Intel introduced the Intel 80286 processor, normally abbreviated as 286. The first CPU behind the original IBM PC AT (Advanced Technology), it did not suffer from the compatibility problems that damned the 80186 and 80188. [pic][pic] P3 (386) Processor - The third generation represents perhaps the most significant change in processors since the first PC. The big deal was the migration from processors that handled 16- bit operations to true 32-bit chips. [pic] [pic] P4 (486) Processor - Although fourth-generation processors were more about refinement than redesign, the Intel 80486 was another major leap forward in the race for speed. The additional power available in the 486 fueled tremendous growths in the software industry. [pic][pic] P5 (586) Processor - After the fourth-generation chips such as the 486, Intel and other chip manufacturers......

Words: 758 - Pages: 4