Celeron (P6)
Covington
The first Celeron (codenamed Covington) was essentially a 266 MHz Deschutes Pentium II manufactured without any secondary cache at all.[4] Covington also shared the 80523 product code of Deschutes. Although clocked at 266 or 300 MHz (frequencies 33 or 66 MHz higher than the desktop version of the Pentium w/MMX), the cacheless Celerons were a good deal slower than the parts they were designed to replace.[3] 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.[5] 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. Nevertheless the first Celerons were quite popular among some overclockers, for their flexible overclockability and reasonable price.[3] Covington was only manufactured in slot 1 SEPP format.
Mendocino
The Mendocino Celeron, launched 24 August 1998, was the first mass-market CPU to use on-die L2 cache.[6] Whereas Covington had no secondary cache at all, Mendocino included 128 KiB of L2 cache running at full clock speed. The first Mendocino-core Celeron was clocked at a then-modest 300 MHz but offered almost twice the performance of the old cacheless Covington Celeron at the same clock speed. To distinguish it from the older Covington 300 MHz, Intel called the Mendocino core Celeron 300A.[6] Although the other Mendocino Celerons (the 333 MHz part, for example) did not have an "A" appended, some people call all Mendocino processors "Celeron-A" regardless of speed.
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. Overclockers soon discovered that, given a high-end motherboard, the Celeron 300A could run reliably at 450 MHz. This was achieved by simply increasing the Front Side Bus (FSB) speed from the stock 66 MHz to the 100 MHz spec of the Pentium II. At this speed, the Mendocino Celeron rivaled the fastest x86 processors available.[6]
At the time on-die cache was difficult to manufacture; especially L2 as more of it is needed to attain an adequate level of performance. A benefit of on-die cache is that it operates at the same clock frequency as the CPU. All other CPUs at that time used motherboard mounted or slot mounted secondary L2 cache, which was very easy to manufacture, cheap, and simple to enlarge to any desired size (typical cache sizes were 512 KiB or 1 MiB), but they carried the performance penalty of slower cache speed, typically running at FSB speed (60 to 100 MHz) for motherboard mounted L2 cache. The implementation of the Pentium II's 512 KiB of L2 cache was unique at the time (and later copied by AMD's Athlon), comprising moderately high-speed L2 cache chips mounted on a special-purpose board alongside the processor itself, running at half processor speed and communicating with the CPU through a special backside bus. This method of cache placement was expensive and imposed practical cache-size limits, but allowed the Pentium II to be clocked faster and avoided front side bus RAM/L2 cache contention typical with motherboard-placed L2 cache configurations.[7]
Over time, newer Mendocino processors were released at 333, 366, 400, 433, 466, 500, and 533 MHz. The "Mendocino" Celeron CPU came only designed for a 66 MHz frontside bus, but this would not be a serious performance bottleneck until clock speeds reached higher levels.
The Mendocino Celerons also introduced new packaging. When the Mendocinos debuted they came in both a Slot 1 SEPP 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. (Third-party manufacturers made motherboard slot-to-socket adapters (nicknamed Slotkets) available for a few dollars, which allowed, for example, a Celeron 500 to be fitted to a Slot 1 motherboard.) One interesting note about the PPGA Socket 370 Mendocinos is that SMP (symmetric multiprocessing) mode was available, and there was at least one motherboard released (the ABIT BP6) which took advantage of this fact.
The Mendocino also came in a mobile variant, with speeds from 266, 300, 333, 366, 400, 433, and 466, 500, 533, 566, 600 MHz.
In Intel's "Family/Model/Stepping" scheme, Mendocino CPUs are family 6, model 6 and their Intel product code is 80524. These identifiers are shared with the related Dixon Mobile Pentium II variant.
Coppermine-128
The next generation Celeron was the Coppermine-128 (sometimes known as the "Celeron II"). These were a derivative of Intel's Coppermine Pentium III and were released on March 29, 2000.[8] Like the Mendocino, the Celeron-128 used 128 KiB of on-chip L2 cache and was (initially) restricted to a 66 MHz bus speed, but the big news was the addition of SSE instructions, due to the new Coppermine core. Other than half the L2 cache (128 KiB instead of 256 KiB) and a slower FSB (66 to 100 MHz instead of 100 to 133 MHz), the Coppermine Celeron was identical to the Coppermine Pentium III.
All Coppermine-128s were produced in the same FCPGA Socket 370 format that most Coppermine Pentium III CPUs used. These Celeron processors began at 533 MHz and continued through 566, 600, 633, 666, 700, 733, and 766 MHz. Because of the limitations of the 66 MHz bus, there were diminishing returns on performance as clock rate increased. On January 3, 2001, Intel switched to a 100 MHz bus with the launch of the 800 MHz Celeron, resulting in a significant performance-per-clock improvement.[9] All Celeron-128 CPUs from 800 MHz and faster use the 100 MHz front side bus. Various models were made at 800, 850, 900, 950, 1000, and 1100 MHz.
In Intel's "Family/Model/Stepping" scheme, Coppermine Celerons and Pentium IIIs are family 6, model 8 and their Intel product code is 80526.
A Tualatin core Celeron 1200 MHz (Tualeron) (FC-PGA2 package)
A Tualatin core Celeron 1200 MHz (Tualeron) (FC-PGA2 package)
Tualatin-256
These Celeron processors, released initially at 1200 MHz (1.2 GHz) on October 2, 2001,[10] were based on Pentium III Tualatin core and made with a 0.13 micrometer process for the FCPGA2 socket 370 . They were nicknamed "Tualeron" — a portmanteau of the words Tualatin and Celeron. Some software and users refer to the chips as "Celeron-S", referring to the chip's lineage with the Pentium III-S, but this is not an official designation. Intel later released 1000 MHz 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).[11] A 1,300 MHz chip, launched January 4, 2002,[12] and finally a 1,400 MHz chip, launched May 15, 2002 (the same day as the Netburst Willamette 1.7 GHz Celeron launch),[13] marked the end of the Tualatin-256 line.
With regards to core functionality, Tualatin-256 was again quite similar to its Pentium III sibling. The most significant differences were a slower 100 MHz bus and only 256 KiB of L2 cache (whereas the Pentium III had either 256 KiB or 512 KiB of L2 cache). Furthermore, the Tualeron's L2 cache had a higher latency which boosted manufacturing yields for this budget CPU.
Despite offering much improved performance over the Coppermine Celeron it superseded, the Tualatin Celeron still suffered stiff competition from AMD's Duron budget processor.[14] Intel later responded by releasing the Netburst Willamette Celeron, and for some time Tualatin Celerons were manufactured and sold in parallel with their replacement Pentium 4-based Celerons.
In Intel's "Family/Model/Stepping" scheme, Tualatin Celerons and Pentium IIIs are family 6, model 11 and their Intel product code is 80530.
Celeron (NetBurst)
Willamette-128
These Celerons were for socket 478 and were based on the Willamette Pentium 4 core, being a completely different design compared to the previous Tualatin Celeron. These are often known as the "Celeron 4". Their L2 cache (128 KiB) is half that of the Pentium 4 Willamette's 256 KiB of L2 cache, but otherwise the two are very similar. With the transition to the Pentium 4 core the Celeron now featured SSE2 instructions. The ability to share the same socket as the Pentium 4 meant that the Celeron now had the option to use RDRAM, DDR SDRAM, or traditional SDRAM. Willamette Celerons were launched May 15, 2002, initially at 1.7 GHz, and offered a noticeable performance improvement over the older Tualatin Celeron 1,300 MHz part, being able to finally beat the Duron 1.3 GHz, which at the time was AMD's top competing budget processor.[15] On June 12, 2002, Intel launched the last Willamette Celeron, a 1.8 GHz model.[16]
In Intel's "Family/Model/Stepping" scheme, Willamette Celerons and Pentium 4s are family 15, model 1, and their Intel product code is 80531.
Northwood-128
These socket 478 Celerons are based on the Northwood Pentium 4 core, and also have 128 KiB of L2 cache. The only difference between the Northwood-128 and the Willamette-128 Celeron is the fact that it was built on the new 0.13 micrometre process which shrunk the die size, increased the transistor count, and lowered the core voltage from 1.7 V on the Willamette-128 to 1.52 V for the Northwood-128. Despite these differences, they are functionally the same as the Willamette-128 Celeron, and perform largely the same clock-for-clock. The Northwood-128 family of processors were initially released as a 2.0 GHz Model (a 1.9 GHz model was announced earlier, but never launched[17]) on September 18, 2002.[18] Since that time Intel has released at total of 10 different speed grades ranging from 1.8 GHz to 2.8 GHz, before being surpassed by the Celeron D. Although the Northwood Celerons suffer considerably from their small L2 cache, some speed grades have been favored in the enthusiast market, because like the old 300A, they can run well above their rated speeds.[19]
In Intel's "Family/Model/Stepping" scheme, Northwood Celerons and Pentium 4s are family 15, model 2, and their Intel product code is 80532.
Celeron D
Prescott-256
Prescott-256 Celeron D processors, initially launched June 25, 2004,[20] feature double the L1 cache (16 KiB) and L2 cache (256 KiB) as compared to the previous Willamette and Northwood desktop Celerons, by virtue of being based on the Prescott Pentium 4 core.[21] It also features a 533 MT/s bus and SSE3, and a 3xx model number (compared to 5xx for Pentium 4s and 7xx for Pentium Ms). The Prescott-256 Celeron D was manufactured for socket 478 as well LGA 775, and they were released carrying model numbers of 355 (3.33 GHz), 350 (3.2 GHz), 345 (3.06 GHz) 340 (2.93 GHz), 335 (2.80 GHz), 330 (2.66 GHz), 325 (2.53 GHz), 320 (2.40 GHz), 315 (2.26 GHz), and 310 (2.13 GHz). They also have hardware-level support of Intel's Intel 64 technology by virtue of it also being built into the Prescott core, although the feature is disabled in all 3x0/3x5 models (with the exception of the Celeron D model 355). It has been activated in all 3x1 and 3x6 models. The Intel Celeron D processor works with the Intel 845 and 865 chipset families. It should be noted that the "D" suffix actually has no official designation although in fact is corresponding to the number "4" as the fourth letter of the alphabet. It is used simply to distinguish this line of Celeron from the previous, lower performing Northwood and Willamette series, and also from the mobile series, the Celeron M.[22] It should also be stated that unlike the Pentium D, the Celeron D is not a dual core processor.
The Celeron D was a major performance improvement over previous Netburst Celerons. A test using a variety of applications, run by Derek Wilson at Anandtech.com, showed that the new Celeron D architecture alone offered up performance improvements on average of >10% over a Northwood Celeron when both CPUs were run at the same bus and clock speed.[23] The addition of SSE 3 instructions and the higher FSB only added to this already impressive gain.
Despite its many improvements, the Prescott core of the Celeron D had at least one major drawback: heat. Unlike the fairly cool running Northwood Celeron, the Prescott-256 had a class-rated TDP of 73 watts, which prompted Intel to include a more intricate copper core/aluminum finned cooler to help handle the additional heat.[24]
In mid-2005, Intel refreshed the Celeron D with Intel 64 and XD Bit (eXecute Disable) enabled. Model numbers increase by 1 over the previous generation (e.g. 330 became 331). This only applied to LGA 775 Celeron Ds. There is no Socket 478 CPU with 64-bit or XD Bit capabilities.
In Intel's "Family/Model/Stepping" scheme, Prescott Celeron Ds and Pentium 4s are family 15, model 3 (up to stepping E0) or 4 (stepping E0 onwards), and their Intel product code is 80546 or 80547, depending on socket type.
Cedar Mill-512
Based on the Cedar Mill Pentium 4 core, this version of the Celeron D was launched May 28, 2006,and continued the 3xx naming scheme with the Celeron D 347 (3.06 GHz), 352 (3.2 GHz), 356 (3.33 GHz), 360 (3.46 GHz), and 365 (3.6GHz). The Cedar Mill Celeron D is largely the same as the Prescott-256, except with double the L2 cache (512KB) and based on a 65nm manufacturing process. The Cedar Mill-512 Celeron D is LGA 775 exclusive. The main benefits of the Cedar Mill Celerons over the Prescott Celerons are the slightly increased performance due to the larger L2 cache, higher clock speeds, and less heat dissipation, with several models having a TDP lowered to 65 watts from Prescott's lowest offering of 73W.[25]
In Intel's "Family/Model/Stepping" scheme, Cedar Mill Celeron Ds and Pentium 4s are family 15, model 6, and their Intel product code is 80552.
Celeron (Core)
These are single-core processors built on the Intel Core microarchitecture and thus are clocked much lower than the Cedar Mill Celerons, but still outperform them. They use the 65 nm Conroe-L core,[26] and use a 400 model number sequence.[27] The FSB has been increased to 800 MHz from 533 MHz and the TDP has decreased from 65W to 35W. As is traditional with Celerons, they do not have Intel VT-x instruction support or SpeedStep. They are single-core processors for the value segment of the market, much like the AMD K8-based Sempron. The product line was launched on June 5, 2007.[28] Intel plans to launch new dual core Celeron E1000 processor on January 2008,also base on 65nm Conore-L. The dual core Celeron will have 512KiB L2, 800MB FSB, 1.6GHz starting frequency,65W TDP.[29]
Celeron Models launched on June 5, 2007 (with list prices per unit, for 1,000 units) Model Clock rate FSB speed L2 cache Price
Celeron 420 1.6 GHz 200 MHz QDR 512 KiB US$39
Celeron 430 1.8 GHz 200 MHz QDR 512 KiB US$49
Celeron 440 2.0 GHz 200 MHz QDR 512 KiB US$59
Mobile Celeron and Celeron M
Tualatin-256
These were the first Mobile Celerons, based on the Tualatin core. These differed from their desktop counterparts in that the Mobile series were offered in both 100 MHz and 133 MHz FSB. Like the desktop Tualatins, these chips had 256KB of L2 cache.
Northwood-256
These are the Mobile Celeron range used in laptops. Also based on the Northwood core, they feature a 256 KiB L2 Cache. These Celeron processors were a good deal faster than the desktop counterparts because of their larger L2 cache sizes.[30] They were eventually replaced by the Celeron M brand which is built around the Pentium M processor design.
Banias-512
This Celeron (sold under the Celeron M brand) is based on the Banias Pentium M, and differs from its parent in that it has half the L2 cache, and does not support the clock-varying SpeedStep technology. It performs reasonably well compared to the Pentium M, but battery life is noticeably shorter on a Celeron M–based notebook than it is on a comparable Pentium M notebook.
A system based on the Celeron M processor may not use the Centrino brand name, regardless of what chipset and Wi-Fi components are used.
In Intel's "Family/Model/Stepping" scheme, Banias Celeron Ms and Pentium Ms are family 6, model 9 and their Intel product code is 80535.
Shelton
The Shelton core is a Banias core without any L2 cache, and without SpeedStep. It is used in Intel's small form factor D845GVSH motherboard, intended for Asian and South American markets. The processor identifies itself as a "Intel Celeron 1.0B GHz", to differentiate it from the previous Coppermine-128 and "Tualeron" 1.0 GHz processors.
Dothan-1024
A 90 nm Celeron M with half of the L2 cache of the 90 nm Dothan Pentium Ms (twice the L2 cache of the 130nm Celeron Ms, though), and, like its predecessor, lacking SpeedStep. Revisions of the Dothan-1024 Celeron Ms were released in the first quarter of 2005 that added support for the XD bit.
In Intel's "Family/Model/Stepping" scheme, Dothan Celeron Ms and Pentium Ms are family 6, model 13 and their Intel product code is 80536.
Yonah-1024
The Celeron M 400-series is a 65 nm Celeron M based on the single-core Yonah Core Solo CPU. Like its predecessors in the Celeron M series, this Celeron M has half of the L2 cache (1 MiB) of Core Solo and lacks SpeedStep. This core also brings new features to Celeron M including a faster front side bus (533 MT/s), SSE3 instructions. September 2006 and January 4, 2008 mark a discontinuation of many Celeron M branded CPUs.[31]
Merom-1024
The new Celeron M 520 is a single-core 65 nm CPU based on the Merom Core 2 CPU. It has half of the L2 cache (1 MiB) of the low end Core 2 Duo (2 MiB cache) and lacks both SpeedStep and Virtualization Technology. It also features Intel 64 technology and XD-bit support with 533 MHz FSB. January 4, 2008 marks a discontinuation of Merom-1024 CPU