Hubbry Logo
Haswell (microarchitecture)Haswell (microarchitecture)Main
Open search
Haswell (microarchitecture)
Community hub
Haswell (microarchitecture)
logo
8 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Haswell (microarchitecture)
Haswell (microarchitecture)
Haswell (microarchitecture)
View on Wikipedia
from Wikipedia
Haswell
A Haswell wafer with several dies, with a pin for scale
General information
LaunchedJune 4, 2013; 12 years ago (June 4, 2013)
Marketed byIntel
Designed byIntel
Common manufacturer
  • Intel
CPUID code0306C3h
Product code
Performance
Max. CPU clock rate800 MHz to 4.4 GHz
QPI speeds9.6 GT/s
DMI speeds4 GT/s
Cache
L1 cache64 KB per core (32 KB instructions + 32 KB data)
L2 cache256 KB per core
L3 cache2–45 MB (shared)
L4 cache128 MB of eDRAM (Iris Pro models only)
Architecture and classification
Technology node22 nm (Tri-Gate)
MicroarchitectureHaswell
Instruction setx86-16, IA-32, x86-64
Extensions
Physical specifications
Cores
    • 2–4 (mainstream)
    • 6–8 (enthusiast)
    • 2–18 (Xeon)
GPUs
  • HD Graphics 4200
  • HD Graphics 4400
  • HD Graphics 4600
  • HD Graphics 5000
  • Iris 5100
  • Iris Pro 5200
Sockets
Products, models, variants
Models
  • Haswell-DT
  • Haswell-ULT
  • Haswell-ULX
  • Haswell-H
  • Haswell-MB
  • Haswell-E
  • Haswell-EP
  • Haswell-EX
Brand name
    • Core i3
    • Core i5
    • Core i7
    • Xeon E3 v3
    • Xeon E5 v3
    • Xeon E7 v3
    • Pentium
    • Celeron
History
PredecessorsSandy Bridge (tock)
Ivy Bridge (tick)
SuccessorsBroadwell (tick/process)
Skylake (tock)
Support status
Unsupported

Haswell is the codename for a processor microarchitecture developed by Intel as the "fourth-generation core" successor to the Ivy Bridge (which is a die shrink/tick of the Sandy Bridge microarchitecture). Intel officially announced CPUs based on this microarchitecture on June 4, 2013, at Computex Taipei 2013,[1] while a working Haswell chip was demonstrated at the 2011 Intel Developer Forum.[2] Haswell was the last generation of Intel processor to have socketed processors on mobile. With Haswell, which uses a 22 nm process,[3] Intel also introduced low-power processors designed for convertible or "hybrid" ultrabooks, designated by the "U" suffix. Haswell began shipping to manufacturers and OEMs in mid-2013, with its desktop chips officially launched in September 2013.

Haswell CPUs are used in conjunction with the Intel 8 Series chipsets, 9 Series chipsets, and C220 series chipsets.

At least one Haswell-based processor was still being sold in 2022 — the Pentium G3420.[4][5] Windows 7 through Windows 10 were released for the Haswell microarchitecture.

Design

[edit]

The Haswell architecture is specifically designed[6] to optimize the power savings and performance benefits from the move to FinFET (non-planar, "3D") transistors on the improved 22 nm process node.[7]

Haswell has been launched in three major forms:[8]

  • Desktop version (LGA 1150 socket and the LGA 2011-v3 socket): Haswell-DT
  • Mobile/Laptop version (PGA socket): Haswell-MB
  • BGA version:
    • 47 W and 57 W TDP classes: Haswell-H (for "All-in-one" systems, Mini-ITX form factor motherboards, and other small footprint formats)
    • 13.5 W and 15 W TDP classes (MCP): Haswell-ULT (for Intel's UltraBook platform)
    • 10 W TDP class (SoC): Haswell-ULX (for tablets and certain UltraBook-class implementations)

Notes

[edit]
  • ULT = Ultra Low TDP; ULX = Ultra Low eXtreme TDP
  • Only certain quad-core variants and BGA R-series stock keeping units (SKUs) receive GT3e (Intel Iris Pro 5200) integrated graphics. All other models have GT3 (Intel HD 5000 or Iris Pro 5100), GT2 (Intel HD 4200, 4400, 4600, P4600 or P4700) or GT1 (Intel HD Graphics) integrated graphics.[9] See also Intel HD and Iris Graphics for more details.
  • Due to the low power requirements of tablet and Ultrabook platforms, Haswell-ULT and Haswell-ULX are only available in dual-core configurations. All other versions come as dual- or quad-core variants.

Performance

[edit]

Compared to Ivy Bridge:

  • Approximately 8% faster vector processing[10]
  • Up to 5% higher single-threaded performance
  • 6% higher multi-threaded performance
  • Desktop variants of Haswell draw between 8% and 23% more power under load than Ivy Bridge.[10][11][12]
  • A 6% increase in sequential CPU performance (eight execution ports per core versus six)[10]
  • Up to 20% performance increase over the integrated HD4000 GPU (Haswell HD4600 vs Ivy Bridge's built-in Intel HD4000)[10]
  • Total performance improvement on average is about 3%[10]
  • Around 15 °C hotter than Ivy Bridge, while clock frequencies of over 4.6 GHz are achievable[13][14][15][16][17][18]

Technology

[edit]
See also: Intel HD Graphics

Features carried over from Ivy Bridge

[edit]

New features

[edit]
Haswell featured a Fully Integrated Voltage Regulator.

CPU

[edit]
Translation lookaside buffer sizes[43][44]
Cache Page size
Name Level 4 KB 2 MB 1 GB
DTLB 1st 64 32 4
ITLB 1st 128 8 / logical core none
STLB 2nd 1024 none

GPU

[edit]
  • Hardware graphics support for Direct3D 11.1 and OpenGL 4.3.[45][46][47] Intel 10.18.14.5180 driver is the last planned driver release on Windows 7/8.1.[48]
  • Four versions of the integrated GPU: GT1, GT2, GT3 and GT3e, where GT3 version has 40 execution units (EUs). Haswell's predecessor, Ivy Bridge, has a maximum of 16 EUs. GT3e version with 40 EUs and on-package 128 MB of embedded DRAM (eDRAM), called Crystalwell, is available only in mobile H-SKUs and desktop (BGA-only) R-SKUs. Effectively, this eDRAM is a Level 4 cache; it is shared dynamically between the on-die GPU and CPU, and serving as a victim cache to the CPU's Level 3 cache.[49][50][51][52][53]

I/O

[edit]

Server processors features

[edit]
  • Haswell-EP variant, released in September 2014, with up to 18 cores and marketed as the Xeon E5-1600 v3 and Xeon E5-2600 v3 series.[64]
  • Haswell-EX variant, released in May 2015, with 18 cores and functioning TSX.[34][65][66]
  • A new cache design.
  • Up to 35 MB total unified cache (last level cache, LLC) for Haswell-EP[67] and up to 40 MB for Haswell-EX.
  • LGA 2011-v3 socket replaces LGA 2011 for the Haswell EP; the new socket has the same number of pins, but it is keyed differently due to electrical incompatibility.[68][69][70]
  • The already launched Xeon E3 v3 Haswells will get a refresh in spring 2014,[71] together with a refreshed Intel C220 series PCH chipset.[72]
  • TDP up to 160 W for Haswell-EP.[73]
  • Haswell-EP models with ten and more cores support cluster on die (COD) operation mode,[74] allowing CPU's multiple columns of cores and last level cache (LLC) slices to be logically divided into what is presented as two non-uniform memory access (NUMA) CPUs to the operating system. By keeping data and instructions local to the "partition" of CPU which is processing them, therefore decreasing the LLC access latency, COD brings performance improvements to NUMA-aware operating systems and applications.[75]

Haswell Refresh

[edit]

Around the middle of 2014, Intel released a refresh of Haswell, simply titled Haswell Refresh. When compared to the original Haswell CPUs lineup, Haswell Refresh CPUs offer a modest increase in clock frequencies, usually of 100 MHz.[76] Haswell Refresh CPUs are supported by Intel's 9 Series chipsets (Z97 and H97, codenamed Wildcat Point), while motherboards with 8 Series chipsets (codenamed Lynx Point) usually require a BIOS update to support Haswell Refresh CPUs.[77]

The CPUs codenamed Devil's Canyon, covering the i5 and i7 K-series SKUs, employ a new and improved thermal interface material (TIM) called next-generation polymer thermal interface material (NGPTIM). This improved TIM reduces the CPU's operating temperatures and improves the overclocking potential, as something that had been problematic since the introduction of Ivy Bridge.[78] Other changes for the Devil's Canyon CPUs include a TDP increase to 88 W, additional decoupling capacitors to help smooth out the outputs from the fully integrated voltage regulator (FIVR), and support for the VT-d that was previously limited to non-K-series SKUs.[79] TSX was another feature brought over from the non-K-series SKUs, until August 2014 when a microcode update disabled TSX due to a bug that was discovered in its implementation.[34][35]

Windows XP and Vista support

[edit]

While Ivy Bridge is the last Intel processor to fully support all versions of Windows XP, Haswell includes limited driver support for certain XP editions such as POSReady2009. People have modified the graphics driver for these versions to adapt to normal Windows XP to varying degrees of success.

Windows Vista support is also dropped with this processor as well. People who have installed x64 version of Vista have reported various problems such as services not starting automatically. The KB4493471 update (officially intended only for Windows Server 2008, but can be installed on Vista) contains a HAL driver that allegedly fixes these issues; however, upon several tests it's been confirmed - it doesn't fix any of the issues. Windows XP and earlier, and all x86 versions and editions of Vista are unaffected by this bug.[citation needed]

List of Haswell processors

[edit]

Desktop processors

[edit]
Intel Haswell i7-4771 CPU, sitting atop its original packaging that contains an OEM fan-cooled heatsink
  • All models support: MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, F16C, Enhanced Intel SpeedStep Technology (EIST), Intel 64, XD bit (an NX bit implementation), Intel VT-x, and Smart Cache.
    • Core i3, i5 and i7 support AVX, AVX2, BMI1, BMI2, FMA3, and AES-NI.[80]
    • Core i3 and i7, as well as the Core i5-4570T and i5-4570TE, support Hyper-Threading (HT).[80]
    • Core i5 and i7 support Turbo Boost 2.0.[80]
    • Although it was initially supported on selected models, since August 2014 desktop variants no longer support TSX due to a bug that was discovered in its implementation; as a workaround, a microcode update disabled the TSX feature.[32][34][35][80]
    • SKUs below 45xx as well as R-series and K-series SKUs do not support Trusted Execution Technology or vPro.[80]
    • Intel VT-d, which is Intel's IOMMU, is supported on all i5 and i7 SKUs except the i5-4670K and i7-4770K.[80][81][82] Support for VT-d requires the chipset and motherboard to also support VT-d.
    • Models i5-4690K and i7-4790K, codenamed Devil's Canyon, have a better internal thermal grease to help heat escape and an improved internal voltage regulator ("FIVR"), to help deliver cleaner power in situations like overclocking.
  • Transistors: 1.4 billion[83][84]
  • Die size: 177 mm2[83]
  • Intel HD and Iris Graphics in following variants:
    • R-series desktop processors feature Intel Iris Pro 5200 graphics (GT3e).[85]
    • The i3-41xxx processors include HD 4400 graphics (GT2).
    • All other i3, i5 and i7 desktop processors include Intel HD 4600 graphics (GT2).[86]
    • Celeron and Pentium processors contain Intel HD Graphics (GT1).
  • Pentium G3258, also known as the Pentium Anniversary Edition, has an unlocked multiplier. Its release marks 20 years of "Pentium" as a brand.[87]

The following table lists available desktop processors.

Target
segment 		Cores
(threads) 		Processor
branding and model 		GPU model CPU clock rate GPU clock rate Cache TDP PCIe 3.0 lane
configurations[a] 		VT-d[b] Release
date 		Release
price
(USD) 		Motherboard
Base Turbo Base Turbo L3 L4[a] Socket Interface Memory

Enthusiast / High-End 8 (16) Core i7
Extreme 		5960X — 3.0 GHz 3.5 GHz — — 20 MB — 140 W 2×16 + 1×8 Yes August 29, 2014 (2014-08-29)[88] $999 LGA 2011-v3 DMI 2.0
PCIe 3.0 		Up to quad
channel
DDR4-2133
6 (12) 5930K 3.5 GHz 3.7 GHz 15 MB $583
5820K 3.3 GHz 3.6 GHz 1×16 + 1×8 + 1×4 $389
Performance 4 (8) Core i7 4790K HD 4600 
(GT2) 		4.0 GHz 4.4 GHz 350 MHz[89] 1.25 GHz 8 MB 88 W 1×16
2×8
1×8 + 2×4 		June 2, 2014 (2014-06-02) $339 LGA
1150 		Up to dual
channel
DDR3-1600[90]
4790 3.6 GHz 4.0 GHz 1.2 GHz 84 W May 11, 2014 (2014-05-11) $303
4790S 3.2 GHz 65 W
4790T 2.7 GHz 3.9 GHz 45 W
4785T 2.2 GHz 3.2 GHz 35 W
4771 3.5 GHz 3.9 GHz 84 W September 1, 2013 (2013-09-01) $320
4770K 1.25 GHz No June 2, 2013 (2013-06-02)[91] $339
4770 3.4 GHz 1.2 GHz Yes $303
4770S 3.1 GHz 65 W
4770R Iris Pro 5200 
(GT3e) 		3.2 GHz 200 MHz 1.3 GHz 6 MB 128 MB $392 BGA
1364
4770T HD 4600 
(GT2) 		2.5 GHz 3.7 GHz 350 MHz[89] 1.2 GHz 8 MB — 45 W $303 LGA
1150
4770TE 2.3 GHz 3.3 GHz 1 GHz
4765T 2.0 GHz 3.0 GHz 1.2 GHz 35 W
Mainstream 4 (4) Core i5 4690K 3.5 GHz 3.9 GHz 6 MB 88 W June 2, 2014 (2014-06-02) $242
4690 84 W May 11, 2014 (2014-05-11) $213
4690S 3.2 GHz 65 W
4690T 2.5 GHz 3.5 GHz 45 W
4670K 3.4 GHz 3.8 GHz 84 W No June 2, 2013 (2013-06-02) $242
4670 Yes $213
4670S 3.1 GHz 65 W
4670R Iris Pro 5200 
(GT3e) 		3.0 GHz 3.7 GHz 200 MHz 1.3 GHz 4 MB 128 MB $310 BGA
1364
4670T HD 4600 
(GT2) 		2.3 GHz 3.3 GHz 350 MHz[89] 1.2 GHz 6 MB — 45 W $213 LGA
1150
4590 3.3 GHz 3.7 GHz 1.15 GHz 84 W May 11, 2014 (2014-05-11) $192
4590S 3.0 GHz 65 W
4590T 2.0 GHz 3.0 GHz 35 W
4570 3.2 GHz 3.6 GHz 84 W June 2, 2013 (2013-06-02)
4570S 2.9 GHz 65 W
4570R Iris Pro 5200 
(GT3e) 		2.7 GHz 3.2 GHz 200 MHz 4 MB 128 MB $288 BGA
1364
2 (4) 4570T HD 4600 
(GT2) 		2.9 GHz 3.6 GHz — 35 W $192 LGA
1150
4570TE 2.7 GHz 3.3 GHz 350 MHz[89] 1 GHz
4 (4) 4460 3.2 GHz 3.4 GHz 1.1 GHz 6 MB 84 W May 11, 2014 (2014-05-11) $182
4460S 2.9 GHz 65 W
4460T 1.9 GHz 2.7 GHz 35 W
4440 3.1 GHz 3.3 GHz 84 W September 1, 2013 (2013-09-01)
4440S 2.8 GHz 65 W
4430 3.0 GHz 3.2 GHz 84 W June 2, 2013 (2013-06-02)[91]
4430S 2.7 GHz 65 W
2 (4) Core i3 4370 3.8 GHz — 1.15 GHz 4 MB 54 W No July 20, 2014 (2014-07-20) $149
4360 3.7 GHz May 11, 2014 (2014-05-11)
4350 3.6 GHz $138
4340 September 1, 2013 (2013-09-01) $149
4330 3.5 GHz $138
4370T 3.3 GHz 200 MHz 35 W March 30, 2015 (2015-03-30)
4360T 3.2 GHz July 20, 2014 (2014-07-20)
4350T 3.1 GHz May 11, 2014 (2014-05-11)
4330T 3.0 GHz September 1, 2013 (2013-09-01)
4340TE 2.6 GHz 350 MHz 1 GHz May 11, 2014 (2014-05-11) $138
4330TE 2.4 GHz September 1, 2013 (2013-09-01) $122
4170 HD 4400 
(GT2) 		3.7 GHz 1.15 GHz 3 MB 54 W March 30, 2015 (2015-03-30) $117
4160 3.6 GHz July 20, 2014 (2014-07-20)
4150 3.5 GHz May 11, 2014 (2014-05-11)
4130 3.4 GHz September 1, 2013 (2013-09-01) $122
4170T 3.2 GHz 200 MHz 35 W March 30, 2015 (2015-03-30) $117
4160T 3.1 GHz July 20, 2014 (2014-07-20)
4150T 3.0 GHz May 11, 2014 (2014-05-11)
4130T 2.9 GHz September 1, 2013 (2013-09-01) $122
Budget 2 (2) Pentium G3470 HD Graphics (GT1) 3.6 GHz 350 MHz 1.1 GHz 53 W March 30, 2015 (2015-03-30) $86
G3460 3.5 GHz July 20, 2014 (2014-07-20)
G3450 3.4 GHz May 11, 2014 (2014-05-11)
G3440 3.3 GHz $75
G3430 December 1, 2013 (2013-12-01) $86
G3420 3.2 GHz $75
G3460T 3.0 GHz 200 MHz 1.1 GHz 35 W March 30, 2015 (2015-03-30)
G3450T 2.9 GHz July 20, 2014 (2014-07-20)
G3440T 2.8 GHz May 11, 2014 (2014-05-11)
G3420T 2.7 GHz December 1, 2013 (2013-12-01)
G3320TE 2.3 GHz 350 MHz 1 GHz Up to dual
channel
DDR3-1333
G3260 3.3 GHz 1.1 GHz 53 W March 30, 2015 (2015-03-30) $64
G3258[c] 3.2 GHz June 2, 2014 (2014-06-02) $72
G3250 July 20, 2014 (2014-07-20) $64
G3240 3.1 GHz May 11, 2014 (2014-05-11)
G3220 3.0 GHz December 1, 2013 (2013-12-01)
G3260T 2.9 GHz 200 MHz 35 W March 30, 2015 (2015-03-30)
G3250T 2.8 GHz July 20, 2014 (2014-07-20)
G3240T 2.7 GHz May 11, 2014 (2014-05-11)
G3220T 2.6 GHz December 1, 2013 (2013-12-01)
Celeron G1850 2.9 GHz 350 MHz 1.05 GHz 2 MB 53 W May 11, 2014 (2014-05-11) $52
G1840 2.8 GHz $42
G1830 December 1, 2013 (2013-12-01) $52
G1820 2.7 GHz $42
G1840T 2.5 GHz 200 MHz 35 W May 11, 2014 (2014-05-11)
G1820T 2.4 GHz December 1, 2013 (2013-12-01)
G1820TE 2.2 GHz 1 GHz
a Some of these configurations could be disabled by the chipset. For example, H-series chipsets disable all PCIe 3.0 lane configurations except 1×16.
b This feature also requires a chipset that supports VT-d like the Q87 chipset or the X99 chipset.
c This is called 20th Anniversary Edition and has an unlocked multiplier.

SKU suffixes to denote:

  • K – unlocked (adjustable CPU multiplier up to 63x)
    • The Pentium G3258 CPU is unlocked despite not having the K-suffix.
  • S – performance-optimized lifestyle (low power with 65 W TDP)
  • T – power-optimized lifestyle (ultra low power with 35–45 W TDP)
  • R – BGA packaging / High-performance GPU (Iris Pro 5200 (GT3e))
  • X – extreme edition (adjustable CPU ratio with no ratio limit)

Server processors

[edit]
Intel Xeon E3-1241 v3 CPU, on top of its original packaging with an OEM fan-cooled heatsink
Intel Xeon E5-1650 v3 CPU; its retail box contains no OEM heatsink

The first digit of the model number designates the largest supported multi-socket configuration; thus, E5-26xx v3 models support up to dual-socket configurations, while the E7-48xx v3 and E7-88xx v3 models support up to quad- and eight-socket configurations, respectively. Also, E5-16xx/26xx v3 and E7-48xx/88xx v3 models have no integrated GPU.

Lists of launched server processors are below, split between Haswell E3-12xx v3, E5-16xx/26xx v3 and E7-48xx/88xx v3 models.

Haswell E7-48xx/88xx v3 SKUs[93][94]
Target
segment 		Cores
(threads) 		Processor
branding and model 		CPU clock rate L3
cache 		TDP Release
date 		Release
price
(USD)
 Motherboard
Normal Turbo Socket Interface Memory

Server 4 (8) Xeon E7 v3 E7-8893v3 3.2 GHz 3.5 GHz 45 MB 140 W May 2015 $6,841 LGA
2011-1 		QPI (up to
9.6 GT/s[b])
DMI 2.0
PCIe 3.0 		Up to DDR4-1866 or DDR3-1600
10 (20) E7-8891v3 2.8 GHz 165 W
18 (36) E7-8890v3 2.5 GHz 3.3 GHz $7,174
E7-8880v3 2.3 GHz 3.1 GHz 150 W $5,895
E7-8880Lv3 2.0 GHz 2.8 GHz 115 W $6,063
E7-8870v3 2.1 GHz 2.9 GHz 140 W $4,672
16 (32) E7-8867v3 2.5 GHz 3.3 GHz 165 W
E7-8860v3 2.2 GHz 3.2 GHz 40 MB $4,061
14 (28) E7-4850v3 2.8 GHz 35 MB 115 W $3,003
12 (24) E7-4830v3 2.1 GHz 2.7 GHz 30 MB $2,170
10 (20) E7-4820v3 1.9 GHz — 25 MB $1,502
8 (16) E7-4809v3 2.0 GHz
Haswell E5-16xx/26xx v3 SKUs
Target
segment 		Cores
(threads) 		Processor
branding and model 		CPU clock rate CPU
AVX clock rate[95] 		L3
cache 		TDP Release
date 		Release
price
(USD)
tray / box 		Motherboard
Normal Turbo Normal Turbo Socket Interface Memory

Server 18 (36) Xeon E5 v3 2699v3 2.3 GHz 3.6 GHz 1.9 GHz 3.3 GHz 45 MB 145 W September 9, 2014 (2014-09-09) — LGA
2011-3 		QPI (up to
9.6 GT/s[b])
DMI 2.0
PCIe 3.0 		up to DDR4-2133
16 (32) 2698v3 40 MB 135 W
2698Av3[96] 2.8 GHz 3.2 GHz 2.3 GHz 2.9 GHz 165 W November 2014 OEM
14 (28) 2697v3 2.6 GHz 3.6 GHz 2.2 GHz 3.3 GHz 35 MB 145 W September 9, 2014 (2014-09-09) $2,702 / $2,706
2695v3 2.3 GHz 3.3 GHz 1.9 GHz 3.0 GHz 120 W $2,424 / $2,428
12 (24) 2690v3 2.6 GHz 3.5 GHz 2.3 GHz 3.2 GHz 30 MB 135 W $2,090 / $2,094
14 (28) 2683v3 2.0 GHz 3.0 GHz 1.7 GHz 2.7 GHz 35 MB 120 W $1,846 / —
12 (24) 2680v3 2.5 GHz 3.3 GHz 2.1 GHz 3.1 GHz 30 MB $1,745 / $1,749
2673v3[c] 2.4 GHz 3.1 GHz 105 W
2670v3 2.3 GHz 3.1 GHz 2.0 GHz 2.9 GHz 120 W $1,589 / $1,593
8 (16) 2667v3 3.2 GHz 3.6 GHz 2.7 GHz 3.5 GHz 20 MB 135 W $2,057 / —
10 (20) 2660v3 2.6 GHz 3.3 GHz 2.2 GHz 3.1 GHz 25 MB 105 W $1,445 / $1,449
12 (24) 2650Lv3 1.8 GHz 2.5 GHz 1.5 GHz 2.3 GHz 30 MB 65 W $1,329 / —
2658v3 2.2 GHz 2.9 GHz 1.9 GHz 3.0 GHz 105 W $1,832 / —
10 (20) 2650v3 2.3 GHz 3.0 GHz 2.0 GHz 2.8 GHz 25 MB $1,166 / $1,171
12 (24) 2648Lv3 1.8 GHz 2.5 GHz 1.5 GHz 2.2 GHz 30 MB 75 W $1,544 / —
6 (12) 2643v3 3.4 GHz 3.7 GHz 2.8 GHz 3.6 GHz 20 MB 135 W $1,552 / —
8 (16) 2640v3 2.6 GHz 3.4 GHz 2.2 GHz 3.4 GHz 20 MB 90 W $939 / $944 up to DDR4-1866
4 (8) 2637v3 3.5 GHz 3.7 GHz 3.2 GHz 3.6 GHz 15 MB 135 W $996 / — up to DDR4-2133
8 (16) 2630v3 2.4 GHz 3.2 GHz 2.1 GHz 3.2 GHz 20 MB 85 W $667 / $671 up to DDR4-1866
2630Lv3 1.8 GHz 2.9 GHz 1.5 GHz 2.9 GHz 55 W $612 / —
10 (20) 2628Lv3 2.0 GHz 2.5 GHz 1.7 GHz 2.4 GHz 25 MB 75 W $1,364 / —
4 (8) 2623v3 3.0 GHz 3.5 GHz 2.7 GHz 3.5 GHz 10 MB 105 W $444 / —
6 (12) 2620v3 2.4 GHz 3.2 GHz 2.1 GHz 3.2 GHz 15 MB 85 W $417 / $422
8 (16) 2618Lv3 2.3 GHz 3.4 GHz 1.9 GHz 3.4 GHz 20 MB 75 W $779 / —
6 (6) 2609v3 1.9 GHz — 1.9 GHz — 15 MB 85 W $306 / $306 up to DDR4-1600
6 (12) 2608Lv3 2.0 GHz 1.7 GHz 52 W $441 / — up to DDR4-1866
6 (6) 2603v3 1.6 GHz 1.3 GHz 85 W $213 / $217 up to DDR4-1600
Workstation 10 (20) 2687Wv3 3.1 GHz 3.5 GHz 2.7 GHz 3.5 GHz 25 MB 160 W $2,141 / $2,145 up to DDR4-2133
8 (16) 1680v3 3.2 GHz 3.8 GHz 2.9 GHz 3.4 GHz 20 MB 140 W $1,723 / — DMI 2.0
PCIe 3.0
1660v3 3.0 GHz 3.5 GHz 2.7 GHz 3.5 GHz $1,080 / —
6 (12) 1650v3 3.5 GHz 3.8 GHz 3.2 GHz 3.7 GHz 15 MB $583 / $586
4 (8) 1630v3 3.7 GHz 3.8 GHz 3.4 GHz 3.7 GHz 10 MB $372 / —
1620v3 3.5 GHz 3.6 GHz 3.2 GHz 3.5 GHz $294 / $297
4 (4) 1607v3 3.1 GHz — 2.8 GHz — $255 / — up to DDR4-1866
1603v3 2.8 GHz 2.5 GHz $202 / —
Haswell E3-12xx v3 SKUs
Target
segment 		Cores
(threads) 		Processor
branding and model 		GPU model CPU clock rate Graphics clock rate L3
cache 		GPU
eDRAM 		TDP Release
date 		Release
price
(USD)
tray / box 		Motherboard
Normal Turbo Normal Turbo Socket Interface Memory

Server 4 (8) Xeon E3 v3 1286v3 HD P4700 (GT2) 3.7 GHz 4.1 GHz 350 MHz 1.3 GHz 8 MB — 84 W May 11, 2014 (2014-05-11) $662 / — LGA
1150 		DMI 2.0
PCIe 3.0 		up to dual
channel
DDR3-1600
with ECC
1286Lv3 3.2 GHz 4.0 GHz 1.25 GHz 65 W $774 / —
1285v3 3.6 GHz 1.3 GHz 84 W June 2, 2013 (2013-06-02) $662 / —
1285Lv3 3.1 GHz 3.9 GHz 1.25 GHz 65 W $774 / —
1284Lv3 Iris Pro 5200 (GT3e) 1.8 GHz 3.2 GHz 750 MHz 1 GHz 6 MB 128 MB 47 W February 18, 2014 (2014-02-18) OEM BGA
1364
1281v3 — 3.7 GHz 4.1 GHz — 8 MB — 82 W May 11, 2014 (2014-05-11) $612 / — LGA
1150
1280v3 3.6 GHz 4.0 GHz June 2, 2013 (2013-06-02)
1276v3 HD P4600 (GT2) 350 MHz 1.25 GHz 84 W May 11, 2014 (2014-05-11) $339 / $350
1275v3 3.5 GHz 3.9 GHz June 2, 2013 (2013-06-02) $339 / $350
1275Lv3 HD (GT1) 2.7 GHz 1.2 GHz 45 W May 11, 2014 (2014-05-11) $328 / —
1271v3 — 3.6 GHz 4.0 GHz — 80 W $328 / $339
1270v3 3.5 GHz 3.9 GHz June 2, 2013 (2013-06-02)
1268Lv3 HD P4600 (GT2) 2.3 GHz 3.3 GHz 350 MHz 1 GHz 45 W $310 / —
1265Lv3 HD (GT1) 2.5 GHz 3.7 GHz 1.2 GHz $294 / —
1246v3 HD P4600 (GT2) 3.5 GHz 3.9 GHz 84 W May 11, 2014 (2014-05-11) $276 / $287
1245v3 3.4 GHz 3.8 GHz June 2, 2013 (2013-06-02)
1241v3 — 3.5 GHz 3.9 GHz — 80 W May 11, 2014 (2014-05-11) $262 / $273
1240v3 3.4 GHz 3.8 GHz June 2, 2013 (2013-06-02)
1240Lv3 2.0 GHz 3.0 GHz 25 W May 11, 2014 (2014-05-11) $278 / —
1231v3 3.4 GHz 3.8 GHz 80 W $240 / $250
1230v3 3.3 GHz 3.7 GHz June 2, 2013 (2013-06-02)
1230Lv3 1.8 GHz 2.8 GHz 25 W $250 / —
4 (4) 1226v3 HD P4600 (GT2) 3.3 GHz 3.7 GHz 350 MHz 1.2 GHz 84 W May 11, 2014 (2014-05-11) $213 / $224
1225v3 3.2 GHz 3.6 GHz June 2, 2013 (2013-06-02)
1220v3 — 3.1 GHz 3.5 GHz — 80 W $193 / $203
2 (4) 1220Lv3 1.1 GHz 1.5 GHz 4 MB 13 W September 1, 2013 (2013-09-01) $193 / —

SKU suffixes to denote:

  • L – low power

Mobile processors

[edit]
  • All models support: MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, F16C, Enhanced Intel SpeedStep Technology (EIST), Intel VT-x, Intel 64, XD bit (an NX bit implementation), and Smart Cache.
    • Core i3, i5 and i7 support AVX, AVX2, BMI1, BMI2, FMA3, and hyper-threading (HT).
    • Core i3, i5 and i7 except the Core i3-4000M support AES-NI.[97]
    • Core i5 and i7 except the Core i5-4410E, i5-4402EC, i7-4700EC, and i7-4702EC support Turbo Boost 2.0.
  • Haswell-ULT and ULX: Platform Controller Hub (PCH) integrated into the CPU package, slightly reducing the amount of space used on motherboards.[98]
  • Transistors: 1.3 billion[99]
  • Die size: 181 mm2[99]

The following table lists available mobile processors.

Target
segment 		Cores
(threads) 		Processor
branding and model 		GPU model Programmable TDP[100]: 69–72  CPU Turbo
(single core) 		GPU clock rate L3
cache 		GPUeDRAM Release
date 		Release
price
(USD)
SDP[101][102]: 71  cTDP down[a] Nominal TDP[b] cTDP up[c] Base Turbo

Performance  4 (8)  Core i7 4940MX HD 4600 (GT2) — — 57 W / 3.1 GHz 65 W / 3.8 GHz  4.0 GHz 400 MHz 1.35 GHz 8 MB — January 21, 2014[103] $1096
4930MX 57 W / 3.0 GHz 65 W / 3.7 GHz  3.9 GHz June 2, 2013[104]
4980HQ Iris Pro 5200 (GT3e) 47 W / 2.8 GHz —  4.0 GHz 200 MHz 1.3 GHz 6 MB 128 MB[50] July 21, 2014[105] $623
4960HQ 47 W / 2.6 GHz 55 W / 3.6 GHz  3.8 GHz September 1, 2013[106]
4950HQ 47 W / 2.4 GHz 55 W / 3.4 GHz  3.6 GHz June 2, 2013[104]
4910MQ HD 4600 (GT2) 47 W / 2.9 GHz 55 W / 3.7 GHz  3.9 GHz 400 MHz 8 MB — January 21, 2014[103] $568
4900MQ 47 W / 2.8 GHz 55 W / 3.6 GHz  3.8 GHz June 2, 2013[104] $570
4870HQ Iris Pro 5200 (GT3e) 47 W / 2.5 GHz —  3.7 GHz 200 MHz 1.2 GHz 6 MB 128 MB July 21, 2014[105] $434
4860EQ 47 W / 1.8 GHz  3.2 GHz 750 MHz 1 GHz August 2013 $508
4860HQ 47 W / 2.4 GHz 55 W / 3.4 GHz  3.6 GHz 200 MHz 1.2 GHz January 21, 2014[103] $434
4850EQ 47 W / 1.6 GHz —  3.2 GHz 650 MHz 1 GHz August 2013 $466
4850HQ 47 W / 2.3 GHz 55 W / 3.3 GHz  3.5 GHz 200 MHz 1.2 GHz June 2, 2013[104] $434
4810MQ HD 4600 (GT2) 47 W / 2.8 GHz 55 W / 3.6 GHz  3.8 GHz 400 MHz 1.3 GHz — January 21, 2014[103] $378
4800MQ 47 W / 2.7 GHz 55 W / 3.5 GHz  3.7 GHz June 2, 2013[104] $380
4770HQ Iris Pro 5200 (GT3e) 47 W / 2.2 GHz  3.4 GHz 200 MHz 1.2 GHz 128 MB July 21, 2014[105] $434
4760HQ 47 W / 2.1 GHz 55 W / 3.1 GHz  3.3 GHz April 14, 2014 $434
4750HQ 47 W / 2.0 GHz 55 W / 3.0 GHz  3.2 GHz June 2, 2013[104] $440
4720HQ HD 4600 (GT2) 47 W / 2.6 GHz —  3.6 GHz 400 MHz 1.2 GHz — January 2015 $378
4712MQ 37 W / 2.3 GHz 45 W / 3.1 GHz  3.3 GHz 1.15 GHz April 14, 2014
4712HQ
4710MQ 47 W / 2.5 GHz 55 W / 3.3 GHz  3.5 GHz
4710HQ 1.2 GHz
4702MQ 37 W / 2.2 GHz 45 W / 2.9 GHz  3.2 GHz 1.15 GHz June 2, 2013[104] $383
4702HQ
4700MQ 47 W / 2.4 GHz 55 W / 3.2 GHz  3.4 GHz
4700HQ 1.2 GHz
4701EQ 1 GHz September 1, 2013 $415
4700EQ June 2, 2013[104] $378
4702EC — 27 W / 2.0 GHz — — — — 8 MB April 2014 $459
4700EC 43 W / 2.7 GHz —
Mainstream  2 (4) 4650U HD 5000 (GT3) — 11.5 W / 800 MHz 15 W / 1.7 GHz  3.3 GHz 200 MHz 1.1 GHz 4 MB June 2, 2013[104] $454
4610Y HD 4200 (GT2) 6 W / 800 MHz 9.5 W / 800 MHz 11.5 W / 1.7 GHz  2.9 GHz 850 MHz September 1, 2013 $393
4610M HD 4600 (GT2) — — 37 W / 3.0 GHz  3.7 GHz 400 MHz 1.3 GHz January 21, 2014[103] $346
4600M 37 W / 2.9 GHz  3.6 GHz September 1, 2013
4600U HD 4400 (GT2) 11.5 W / 800 MHz 15 W / 2.1 GHz  3.3 GHz 200 MHz 1.1 GHz $398
4578U Iris 5100 (GT3) 23 W / 800 MHz 28 W / 3.0 GHz  3.5 GHz 1.2 GHz July 20, 2014 —
4558U 28 W / 2.8 GHz  3.3 GHz 1.2 GHz June 2, 2013[104] $454
4550U HD 5000 (GT3) 11.5 W / 800 MHz 15 W / 1.5 GHz  3.0 GHz 1.1 GHz
4510U HD 4400 (GT2) 15 W / 2.0 GHz  3.1 GHz April 2014 $393
4500U 15 W / 1.8 GHz 25 W / 3.0 GHz  3.0 GHz June 2, 2013[104] $398
 Core i5 4402EC — — 27 W / 2.5 GHz — — — — April 2014 $324
4422E HD 4600 (GT2) 25 W / 1.8 GHz  2.9 GHz 400 MHz 900 MHz 3 MB April 14, 2014 $266
4410E 37 W / 2.9 GHz — 1 GHz
4402E 25 W / 1.6 GHz  2.7 GHz 900 MHz September 1, 2013
4400E 37 W / 2.7 GHz  3.3 GHz 1 GHz
4360U HD 5000 (GT3) 11.5 W / 800 MHz 15 W / 1.5 GHz  3.0 GHz 200 MHz 1.1 GHz January 21, 2014[103] $315
4350U 15 W / 1.4 GHz  2.9 GHz June 2, 2013[104] $342
4340M HD 4600 (GT2) — 37 W / 2.9 GHz  3.6 GHz 400 MHz 1.25 GHz January 21, 2014[103] $266
4330M 37 W / 2.8 GHz  3.5 GHz September 1, 2013
4310M HD 4600 (GT2) 37 W / 2.7 GHz  3.4 GHz 400 MHz 1.25 GHz January 21, 2014[103] $225
4310U HD 4400 (GT2) 11.5 W / 800 MHz 15 W / 2.0 GHz  3.0 GHz 200 MHz 1.1 GHz $281
4302Y HD 4200 (GT2) 4.5 W / 800 MHz 11.5 W / 1.6 GHz  2.3 GHz 200 MHz 850 MHz September 1, 2013 —
4300Y 6 W / 800 MHz 9.5 W / 800 MHz $304
4300M HD 4600 (GT2) — — 37 W / 2.6 GHz  3.3 GHz 400 MHz 1.25 GHz $225
4300U HD 4400 (GT2) 11.5 W / 800 MHz 15 W / 1.9 GHz  2.9 GHz 200 MHz 1.1 GHz $287
4288U Iris 5100 (GT3) 23 W / 800 MHz 28 W / 2.6 GHz  3.1 GHz 1.2 GHz June 2, 2013[104] $342
4258U 28 W / 2.4 GHz  2.9 GHz 1.1 GHz
4308U 28 W / 2.8 GHz  3.3 GHz 1.2 GHz July 20, 2014[107] $315
4260U HD 5000 (GT3) 11.5 W / 800 MHz 15 W / 1.4 GHz  2.7 GHz 1 GHz April 14, 2014 $315
4250U 15 W / 1.3 GHz  2.6 GHz June 2, 2013[104] $342
4210H HD 4600 (GT2) — 47 W / 2.9 GHz  3.5 GHz 400 MHz 1.15 GHz July 20, 2014 $225
4210M 37 W / 2.6 GHz  3.2 GHz April 14, 2014
4210U HD 4400 (GT2) 11.5 W / 800 MHz 15 W / 1.7 GHz  2.7 GHz 200 MHz 1 GHz $287
4220Y HD 4200 (GT2) 6 W / 800 MHz 9.5 W / 800 MHz 11.5 W / 1.6 GHz  2.0 GHz 850 MHz $281
4210Y 11.5 W / 1.5 GHz  1.9 GHz September 1, 2013 $304
4202Y 4.5 W / 800 MHz 11.5 W / 1.6 GHz  2.0 GHz —
4200Y 6 W / 800 MHz 11.5 W / 1.4 GHz  1.9 GHz June 2, 2013[104] $304
4200U HD 4400 (GT2) — 11.5 W / 800 MHz 15 W / 1.6 GHz 25 W / ?  2.6 GHz 1 GHz $287
4200H HD 4600 (GT2) — 47 W / 2.8 GHz —  3.4 GHz 400 MHz 1.15 GHz September 1, 2013 $257
4200M 37 W / 2.5 GHz  3.1 GHz $240
 Core i3 4158U Iris 5100 (GT3) 23 W / 800 MHz 28 W / 2.0 GHz — 200 MHz 1.1 GHz June 2, 2013[104] $342
4120U HD 4400 (GT2) 11.5 W / 800 MHz 15 W / 2.0 GHz 1 GHz April 14, 2014 $281
4112E HD 4600 (GT2) — 25 W / 1.8 GHz 400 MHz 900 MHz $225
4110E 37 W / 2.6 GHz
4102E 25 W / 1.6 GHz September 1, 2013
4100E 37 W / 2.4 GHz
4110M 37 W / 2.6 GHz 1.1 GHz April 14, 2014
4100M 37 W / 2.5 GHz September 1, 2013
4100U HD 4400 (GT2) 11.5 W / 800 MHz 15 W / 1.8 GHz 200 MHz 1 GHz June 2, 2013[104] $287
4030Y HD 4200 (GT2) 6 W / 800 MHz 9.5 W / 800 MHz 11.5 W / 1.6 GHz 850 MHz April 14, 2014 $281
4020Y 11.5 W / 1.5 GHz September 1, 2013 $304
4012Y 4.5 W / 800 MHz —
4010Y 6 W / 800 MHz 9.5 W / 800 MHz 11.5 W / 1.3 GHz June 2, 2013[104]
4030U HD 4400 (GT2) — 11.5 W / 800 MHz 15 W / 1.9 GHz 1 GHz April 14, 2014 $281
4025U 950 MHz $275
4010U 15 W / 1.7 GHz 1 GHz September 1, 2013 $287
4005U 950 MHz $281
4000M HD 4600 (GT2) — 37 W / 2.4 GHz 400 MHz 1.1 GHz $240
Budget  2 (2)  Pentium 3561Y HD Graphics (GT1) 6 W / 800 MHz 11.5 W / 1.2 GHz 200 MHz 850 MHz 2 MB December 2013 $161
3560Y September 1, 2013 OEM
3558U — 15 W / 1.7 GHz 1 GHz December 2013 $161
3556U September 1, 2013 OEM
3560M 37 W / 2.4 GHz 400 MHz 1.1 GHz April 14, 2014 $134
3550M 37 W / 2.3 GHz September 1, 2013
 Celeron 2981U 15 W / 1.6 GHz 200 MHz 1 GHz December 2013 $137
2980U September 1, 2013
2957U 15 W / 1.4 GHz December 2013 $132
2955U September 1, 2013
2970M 37 W / 2.2 GHz 400 MHz 1.1 GHz April 14, 2014 $75
2950M 37 W / 2.0 GHz September 1, 2013 $86
2961Y 6 W / 800 MHz 11.5 W / 1.1 GHz 200 MHz 850 MHz December 2013 OEM
  1. When a cooler or quieter mode of operation is desired, this mode specifies a lower TDP and lower guaranteed frequency versus the nominal mode.[100]: 71–72 
  2. This is the processor's rated frequency and TDP.[100]: 71–72 
  3. When extra cooling is available, this mode specifies a higher TDP and higher guaranteed frequency versus the nominal mode.[100]: 71–72 

SKU suffixes to denote:

  • M – dual-core mobile (Socket G3)
  • MQ – quad-core mobile (Socket G3)
  • U – ultra-low power (BGA1168 packaging)
  • MX – quad-core extreme mobile (Socket G3)
  • Y – extreme low-power (BGA1168 packaging)
  • H – dual-core BGA1364 packaging
  • HQ – quad-core BGA1364 packaging
  • E  – embedded version of H
  • EQ  – embedded version of HQ

See also

[edit]

Notes

[edit]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Haswell (microarchitecture)

View on Grokipedia
from Grokipedia
Haswell is Intel's codename for the powering its fourth-generation Core processor family, introduced in June 2013 as a significant redesign—known internally as a "tock"—over the preceding Ivy Bridge . Built on an optimized 22 nm tri-gate (FinFET) process technology, it emphasizes power efficiency for mobile and tablet applications while supporting desktop and server variants, with typical transistor counts around 1.4 billion for quad-core configurations with integrated . The design adopts a system-on-chip (SoC) approach, integrating CPU cores, , memory controllers, and I/O in a modular fashion to enable lower power envelopes down to 10 W for ultrabooks and tablets, alongside improvements in instructions-per-cycle (IPC) performance of up to 10-15% in key workloads compared to Ivy Bridge. At its core, Haswell employs a dual-threaded, engine capable of decoding up to five instructions per cycle, issuing four fused micro-operations (uops), and dispatching eight uops to execution ports, building on Ivy Bridge's pipeline with enhancements for better branch prediction and reduced latency in the front-end. The execution resources include four integer units, four floating-point units supporting 256-bit AVX2 operations, and dedicated ports for load/store operations, enabling high-throughput vector processing with new fused multiply-add (FMA3) instructions for scientific computing and media workloads. Notable additions include AVX2 for 256-bit integer SIMD with gather/scatter support, for hardware-accelerated lock-free programming, bit manipulation instructions (BMI1/BMI2), and enhanced features like an integrated (FIVR) and the S0ix active idle state to minimize standby power in mobile scenarios. The also refines the subsystem with a last-level cache (LLC) configurable up to 8 MB for desktop quad-cores (shared among cores), dual-channel DDR3/LPDDR3 support, and improved prefetchers for bandwidth efficiency, while the integrated HD Graphics 4600 (GT2) represents a major upgrade with 20 execution units and support for DirectX 11.1, 4K output, and Quick Sync Video encoding. Overall, Haswell's focus on converged SoC design and efficiency paved the way for subsequent architectures like Broadwell, enabling 's expansion into thinner, longer-battery-life devices without sacrificing desktop performance.

Introduction

Background and Development

Development of the Haswell microarchitecture began around 2010 as Intel's planned successor to the Ivy Bridge architecture, representing the "tock" phase in Intel's tick-tock model where a new core design is introduced on an existing process node. The primary aims included a implementation to maintain compatibility with Ivy Bridge's manufacturing while emphasizing extended battery life for mobile platforms and enhanced integrated graphics capabilities for both mobile and desktop variants, aligning with Intel's push toward ultrathin laptops and all-in-one systems. Key engineering objectives for Haswell centered on achieving a 5-15% improvement in (IPC) compared to Ivy Bridge through optimizations in the execution and , alongside advanced power management techniques such as enhanced to minimize leakage current in idle components. Additionally, the incorporated a fully integrated (FIVR) on the die to enable finer-grained control over multiple power domains, including the CPU cores, integrated GPU, and I/O subsystems, thereby improving overall energy efficiency without relying on external regulators. Among the major challenges overcome during Haswell's design were managing the increased density, with the typical quad-core die featuring approximately 1.4 billion transistors to support expanded functionality while controlling thermal output. Engineers also targeted significant reductions in (TDP) for configurations, dropping to as low as 11.5 for certain mobile SKUs to enable thinner designs and longer battery —a 25% decrease in platform-level power compared to prior generations. Furthermore, the was prepared to introduce support for AVX2 vector extensions, doubling the width of certain operations to 256 bits and enhancing parallel processing capabilities for future workloads. Haswell was first publicly announced at the Intel Developer Forum (IDF) in September 2011, where prototypes were demonstrated emphasizing its low-power attributes, such as up to 20x lower standby power for mobile systems. The codename "Haswell" was derived from the small town of Haswell in , selected from a database of locations near 's design teams in to continue the company's tradition of geography-inspired naming.

Release and Specifications

The Haswell microarchitecture was officially launched by on June 4, 2013, with the initial desktop release featuring the Core i7-4770K processor as the flagship model. Mobile variants targeted at ultrabooks followed in June 2013, enabling thinner designs with improved battery life, while tablet and 2-in-1 device configurations, including low-power Haswell-Y parts, became available starting in September 2013. Haswell processors were fabricated using Intel's 22 nm tri-gate FinFET process technology, which enhanced transistor density and power efficiency compared to prior nodes. Die sizes varied by configuration, with the standard quad-core desktop variant measuring 177 mm² and containing approximately 1.4 billion transistors. Client-oriented Haswell processors supported up to 4 cores and 8 threads via Hyper-Threading, while server implementations under the Xeon E5 v3 (Haswell-EP) and E7 v3 (Haswell-EX) families scaled to a maximum of 18 cores per socket. Clock speeds ranged from a base of around 1.9 GHz in low-power mobile SKUs to turbo boosts up to 4.0 GHz in high-end desktop models like the i7-4770K. Thermal design power (TDP) spanned from 10 W for ultralow-voltage mobile parts to 84 W for unlocked desktop variants, balancing performance and efficiency across form factors. Memory support included dual-channel DDR3L-1600, with a maximum capacity of 32 GB, optimized for low-voltage operation in mobile devices while maintaining compatibility with standard DDR3 in desktops. Packaging differed by platform: desktop models used the Flip-Chip () socket, whereas low-power mobile and embedded variants employed (BGA) packages for direct to motherboards, while higher-power mobile variants used socketed packages.
SpecificationDesktop (e.g., Core i7-4770K)Mobile (e.g., Core i7-4700MQ)Server (e.g., E5-2699 v3)
Cores/Threads4/84/8Up to 18/36
Base Clock3.5 GHz2.4 GHz2.3 GHz
Max Turbo3.9 GHz3.4 GHz3.6 GHz
TDP84 W37 W145 W
MemoryDDR3-1600, up to 32 GBDDR3L-1600, up to 32 GBDDR4-2133, up to 768 GB
Socket/PackageFC-PGA946LGA 2011-1

Core Design

Microarchitecture Overview

The Haswell microarchitecture features a superscalar, out-of-order core design that builds upon the Ivy Bridge foundation with refinements to enhance instruction throughput and resource utilization. The front-end supports a decode stage capable of processing up to five , which feeds into a instruction decoder for efficient handling of complex x86 instructions. Branch is handled by a sophisticated unit including a tournament-style predictor that selects between global and local history tables, augmented by a loop stream detector to optimize for repetitive loops by buffering up to 56 micro-operations (uops) from small loops, reducing fetch and decode overhead. This setup allows for better of in typical workloads. The core maintains a unified physical register file with 168 entries for integer operations and 168 for floating-point and SIMD operations, providing ample renaming capacity to sustain without frequent stalls; this represents an increase from Ivy Bridge's 160 integer and 144 FP/SIMD registers, supporting more aggressive reordering of up to 192 micro-operations in the reorder buffer. The memory subsystem employs a per-core consisting of a 32 KB 8-way set-associative L1 instruction cache, a 32 KB 8-way L1 data cache with support for 64-byte cache lines, and a 256 KB 8-way L2 cache that serves as a victim cache for L1. These private caches connect to a shared last-level L3 cache of up to 8 MB (16-way set-associative), which is inclusive of L1 contents to simplify coherence and enable faster inter-core data sharing. The out-of-order engine includes a 60-entry unified scheduler and dedicated load/store queues, enabling up to 2 loads and 1 store per cycle through ports 2, 3, and 7, with the reorder buffer managing retirement of up to 4 instructions per cycle.

Execution Units and Pipeline

The Haswell microarchitecture features a 14-stage pipeline designed to balance instruction throughput and latency, with a branch misprediction penalty of 19 cycles that underscores the importance of accurate prediction in performance-critical workloads. This pipeline structure allows for efficient handling of out-of-order execution while maintaining compatibility with prior Core architectures. The front-end of the pipeline fetches up to 16 bytes of instructions per cycle from the L1 instruction cache and decodes up to 5 instructions per cycle when they fit within that fetch window, enabling sustained high instruction bandwidth under favorable code alignment conditions. Haswell's execution backend employs 8 dispatch to allocate micro-operations (μops) to specialized units, supporting up to 8 μops per cycle for peak throughput. 0 and 1 handle loads, allowing up to 2 loads per cycle, while 2 is dedicated to store generation and simple operations. ALU operations are distributed across 0, 1, 5, and 6, enabling up to 4 ALU instructions per cycle, with multiply and add operations available on 0, 1, and 5 for balanced scalar and vector workloads. The execution resources include dual 64-bit ALUs on 2 and 3 for calculations and basic arithmetic, complemented by a dedicated shifter on 1 that natively supports BMI2 instructions such as PEXT and PDEP for efficient in algorithms like sparse set operations. The in Haswell introduces full 256-bit AVX2 support, extending vector widths beyond Ivy Bridge's capabilities to process wider SIMD data sets. Dedicated fused multiply-add (FMA) units on Ports 0 and 1 deliver up to 16 double-precision FLOPs per cycle through two independent 256-bit pipelines, each capable of 8 FLOPs (4 multiplies fused with 4 adds), significantly boosting compute-intensive applications like scientific simulations and kernels. This design prioritizes throughput for vectorized code while integrating seamlessly with the scalar integer pipeline for mixed workloads.

Performance

IPC and Clock Improvements

Haswell delivered an approximate 10% increase in (IPC) for integer workloads relative to Ivy Bridge, driven by expanded execution resources and refined instruction fusion techniques that allowed for greater throughput in the backend pipeline. This uplift stemmed from architectural enhancements such as a wider capable of handling more operations simultaneously and improved macro-fusion, which combined compatible instructions like compare-and-branch sequences into single micro-operations to reduce dispatch bandwidth demands. In floating-point domains, the IPC gain reached about 15%, largely due to the introduction of fused multiply-add (FMA) units that enabled higher parallelism in vector computations while maintaining low latency comparable to standard floating-point operations. These changes, combined with broader support for AVX2 instructions, allowed Haswell to execute up to 16 double-precision floating-point operations per cycle, doubling Ivy Bridge's capability in optimized workloads. Clock speeds saw incremental advances, with desktop models like the Core i7-4770 featuring a base frequency of 3.4 GHz and single-core turbo boost up to 3.9 GHz, matching the Ivy Bridge Core i7-3770's base but benefiting from more efficient thermal management for sustained higher boosts under load. In standardized benchmarks such as SPECint 2006, Haswell achieved roughly 20% higher scores than Ivy Bridge at equivalent power levels, thanks to optimizations that minimized stalls in the rename and allocate pipeline stages, enabling smoother instruction flow and reduced . Multi-threaded performance scaling improved through enhancements to , yielding 25-30% uplifts in threaded applications compared to prior generations, facilitated by a larger 192-entry reorder buffer (versus Ivy Bridge's 168 entries) and better sharing of execution resources between logical threads. This allowed for more effective utilization of idle slots, particularly in workloads with mixed integer and floating-point demands.

Power Efficiency Enhancements

Haswell introduced the Fully Integrated Voltage Regulator (FIVR), an on-die power delivery system that independently regulates voltage for the CPU cores, integrated GPU, and uncore components, eliminating the need for multiple external voltage regulators on the motherboard. This design reduces package pin count by up to 18 and board space, while enabling rapid voltage transitions and lower transient power spikes, contributing to overall system cost savings and improved efficiency in mobile platforms. FIVR supports a wide range of configurations, from 3W fanless devices to higher-power systems, and facilitates up to 50% longer battery life in ultrabook scenarios through precise power allocation. Advanced mechanisms in Haswell enhance idle and light-load efficiency, including per-core C6 states that power-gate individual cores while saving architectural state to SRAM for quick restoration upon wakeup. The C7 state further deepens this by gating additional logic, achieving package power as low as 0.5W during extended idle periods. The , including the ring interconnect and last-level cache, operates in a separate power domain that can be independently gated or frequency-scaled, allowing cores to enter low-power modes without halting system agent activity. These features result in a 20-fold reduction in idle power compared to Ivy Bridge platforms. Refinements to and Turbo Boost 2.0 in Haswell include dynamic core parking for inactive threads and adaptive power limits that allow short bursts beyond base TDP, such as up to 22W for dual-core operation in 15W TDP ultrathin laptops. This enables higher performance during bursty workloads while maintaining thermal envelopes, with core parking reducing active core count to minimize unnecessary power draw in multi-threaded scenarios. The adoption of an optimized 22 nm tri-gate transistor process in Haswell improves leakage currents compared to Ivy Bridge and provides reductions in active power at iso-performance alongside improved static power efficiency through refinements in gate control and reduced short-channel effects. This enables race-to-halt optimizations, where the processor rapidly completes tasks and enters low-power states, ideal for intermittent workloads in battery-constrained environments.

Technology and Features

Inherited from Ivy Bridge

The Haswell microarchitecture retained the fundamental core pipeline structure from Ivy Bridge, featuring a 4-wide superscalar engine capable of issuing up to four micro-operations per cycle. The front-end fetch and decode stages remained largely unchanged, with a 16-byte fetch width and dual decode clusters that could process up to four instructions per cycle when drawing from the micro-op cache. Additionally, technology was carried over unchanged, allowing each physical core to support two logical threads for improved throughput on multithreaded workloads. Cache hierarchy continuity from Ivy Bridge ensured seamless compatibility, with the L1 data cache maintaining its 32 KB size and 8-way set associativity per core, while the private L2 cache stayed at 256 KB and 8-way associative. The shared L3 cache adhered to principles, meaning all L1 and L2 data was mirrored in L3 for coherence. These structures preserved the overall latency and bandwidth characteristics of Ivy Bridge's caching system. To maintain broad software compatibility, Haswell inherited the full baseline from Ivy Bridge, including complete support for SSE4.2 extensions for string processing and CRC operations, as well as 256-bit AVX vector instructions without any modifications. components were directly carried over to support multi-core scalability, including the QuickPath Interconnect (QPI) protocol at 8 GT/s for server variants to enable inter-processor communication. The on-die ring bus interconnect, operating at 25 GB/s bandwidth per direction for quad-core configurations, was retained to facilitate efficient data sharing and coherence across cores and the integrated GPU.

CPU-Specific Innovations

Haswell introduced several CPU-specific instruction set extensions that enhanced compute capabilities, particularly for vector processing, , , and security primitives. These innovations built upon prior architectures by expanding vector widths, enabling hardware-accelerated transactions, and improving arithmetic operations for specialized workloads. The 2 (AVX2) extension marked a significant advancement in SIMD processing, extending 256-bit vector operations to integer data types alongside floating-point support already present in AVX. AVX2 added instructions for gather operations, which allow non-contiguous memory loads into vector registers, improving efficiency in scattered data access patterns common in scientific computing and image processing. Additionally, the inclusion of Fused Multiply-Add 3-operand (FMA3) instructions enabled simultaneous multiplication and addition in a single operation, effectively doubling the floating-point throughput compared to standard AVX implementations by leveraging dedicated FMA units in the execution pipeline. This resulted in up to 2x performance gains for floating-point intensive applications on Haswell cores. Transactional Synchronization Extensions (TSX) provided hardware support for through lock elision, simplifying parallel programming by allowing transactions to execute speculatively without explicit locks. TSX comprises two modes: Restricted Transactional Memory (RTM), which uses XBEGIN, XEND, and XABORT instructions for explicit transaction management and abort handling, and Hardware Lock Elision (HLE), a lighter-weight fallback using prefixes on existing lock instructions. In contended lock scenarios, such as database transactions or multi-threaded applications, TSX could deliver up to 40% performance improvements by reducing synchronization overhead and retry costs. However, due to a hardware (HSD29), TSX was disabled by default via updates starting in 2014, though it can be re-enabled in software. The Bit Manipulation Instructions 2 (BMI2) set introduced advanced bit-level operations, including Parallel Bits Deposit (PDEP) and Parallel Bits Extract (PEXT), which rearrange sparse bit patterns efficiently without conditional branching. These instructions, along with MULX for flagless unsigned , facilitated faster multi-precision arithmetic by preserving the and enabling independent high- and low-word results. BMI2 proved particularly beneficial for cryptographic algorithms and data compression, where bit packing and are frequent, reducing instruction counts and improving throughput in libraries like . Note that while MULX is part of BMI2 in Haswell, related extensions like ADCX for carry-extended addition appeared in subsequent architectures.

GPU and Graphics Advancements

The Haswell microarchitecture integrates the fourth-generation HD Graphics core, codenamed Generation 7.5 (Gen7.5), which represents a substantial evolution from the Ivy Bridge's Gen7 integrated graphics. Available in GT1 and GT2 configurations, the GPU features 10 execution units (EUs) in GT1 setups and up to 20 EUs in GT2 variants, such as the Intel HD Graphics 4600. These EUs operate at clock speeds reaching 1.15 GHz in the HD 4600, enabling higher computational throughput for rendering tasks. The architecture incorporates a 128-bit sampler and 256-bit data ports across its dual slices, delivering approximately 2x the texture throughput compared to Ivy Bridge's equivalent hardware. Key API advancements include full support for 11.1 and 4.0, with hardware-accelerated for detailed geometry processing and compute shaders for general-purpose GPU (GPGPU) workloads. These features allow developers to leverage advanced shading models and parallel processing without discrete graphics. Additionally, Quick Sync Video receives significant enhancements, enabling H.264 encoding at (level 5.2) up to 60 frames per second, which improves efficiency for video and streaming applications. This builds on prior generations by expanding high-resolution media handling while maintaining low CPU overhead. In pixel , Haswell achieves a 2x pixels-per-cycle fill rate over Ivy Bridge through refined pipeline optimizations, paired with upgraded render output units (ROPs) that enhance (MSAA) performance for smoother edge rendering in games and visualizations. The geometry engine supports up to 1.25 giga-triangles per second (GT/s), facilitating faster primitive assembly and improved handling of complex scenes. These changes prioritize balanced rasterization without excessive numerical detail, focusing on real-world rendering . Power efficiency is bolstered by a dedicated GPU power domain, independent of the CPU, which integrates C5 and C6 idle states to minimize leakage and dynamic consumption. This results in idle draw power below 0.1W, a marked improvement that contributes to overall system-level savings in ultrabooks and desktops. The dual-pipe design further enables asynchronous execution of compute and workloads, allowing the GPU to handle mixed pipelines more without stalling rendering threads. These optimizations align with Haswell's emphasis on sustained battery life and thermal management in integrated SoCs.

I/O and Connectivity Upgrades

Haswell introduced significant enhancements to platform I/O capabilities, primarily through the integration of PCIe 3.0 directly into the CPU, providing up to 16 lanes operating at 8 GT/s for improved peripheral and graphics connectivity. These lanes support bifurcation configurations, such as x8/x8 splits, enabling multi-GPU setups like NVIDIA SLI or without performance bottlenecks in bandwidth-intensive scenarios. The 8-series (PCH), such as the Z87 chipset, contributes additional PCIe connectivity, with up to 8 lanes primarily at Gen 2 speeds routed through the DMI interface, augmenting the CPU's capabilities for onboard devices and expansion slots. USB 3.0 support was natively integrated into the 8-series PCH with dual xHCI controllers, enabling up to six ports at 5 Gbps each for a theoretical aggregate bandwidth of 30 Gbps, though practical throughput is shared and typically limited to around 10 Gbps total across controllers due to internal PCIe mapping. This upgrade from Ivy Bridge's four USB 3.0 ports improved data transfer rates for external storage and peripherals, reducing reliance on add-in cards. Accompanying this, 3.0 connectivity expanded to six native 6 Gb/s ports in the PCH, supporting configurations and high-speed SSDs with full AHCI compliance. The DMI 2.0 interface linking the CPU to the PCH operated at 5 GT/s per lane across four lanes, delivering approximately 2 GB/s of bidirectional bandwidth to handle aggregated I/O traffic without significant contention in typical desktop workloads. Display output capabilities advanced with dual independent display pipes in the integrated graphics, supporting Embedded DisplayPort (eDP) 1.3 for panel connections up to 3840x2160 at 60 Hz, alongside 1.4a for 4K at 30 Hz and 1.2 for higher refresh rates on external monitors. These pipes enable simultaneous multi-monitor setups, with HDCP 2.2 compliance ensuring protected 4K content playback over and DP interfaces, a key upgrade for media consumption. The 8-series chipsets, exemplified by the Z87, incorporated optional native 1 support via dedicated headers on select motherboards, allowing 10 Gbps bidirectional connectivity for daisy-chained peripherals and displays without additional controllers. Additionally, enhanced power delivery circuits in the PCH improved USB charging efficiency, supporting up to 900 mA at 5 V for mobile devices even in sleep states, reducing power draw compared to prior generations.

Variants and Extensions

Haswell Refresh

The Haswell Refresh, introduced by in the second quarter of 2014, extended the lifecycle of the original Haswell microarchitecture through higher clock speeds and additional processor models targeted at desktop and mobile platforms. This update arrived without substantive changes to the core , retaining the 22 nm manufacturing process, execution units, and design from the initial Haswell launch. The refresh primarily focused on performance uplifts via modest frequency increases—typically up to 10%—and enhanced binning to improve yield for scenarios, particularly in unlocked desktop variants. On the desktop side, the refresh highlighted the Devil's Canyon series, launched in June 2014, which included unlocked processors like the Core i7-4790K with a 4.0 GHz base clock and 4.4 GHz turbo, surpassing the original Core i7-4770K's 3.5 GHz base by approximately 14%. These models incorporated an improved thermal interface material between the die and integrated , along with adjusted power delivery limits, to facilitate sustained higher clocks and better headroom compared to prior Haswell chips. Mobile implementations, rolled out starting in April 2014, added 17 new stock keeping units (SKUs) across Core i7, i5, and lower-tier families, emphasizing higher base frequencies for and notebook designs while maintaining compatibility with existing platforms. Graphics enhancements in the refresh built on Haswell's integrated GPU capabilities, with new low-power GT3e variants incorporating the Iris Pro 5200 configuration that includes 128 MB of on-package cache, effectively doubling effective bandwidth over standard GT3 implementations without . These were particularly aimed at power-constrained mobile SKUs, enabling improved 4K video handling and light gaming performance in thin-and-light devices. The saw refinements for better stability at DDR3-1866 speeds in select configurations, though official specifications remained aligned with DDR3-1600 for most models. To bridge toward the subsequent Broadwell generation, the Haswell Refresh aligned with the introduction of Intel's 9-series chipsets in mid-2014, which supported socket compatibility for future Broadwell drop-in upgrades via updates, easing the transition for desktop users without requiring new motherboards. Mobile refreshes similarly prepared ecosystems for Broadwell's mobile debut later in the year, focusing on evolutionary rather than revolutionary shifts in power efficiency and integration.

Server and Xeon Modifications

The Haswell microarchitecture was adapted for server environments through the E5-2600 v3 product family, codenamed Haswell-EP, which supports dual-socket configurations optimized for enterprise workloads. These processors, fabricated on a , offer up to 18 cores and 36 threads per socket, paired with a shared 45 MB L3 cache to enhance multi-threaded performance in centers. Interconnects utilize (QPI) 2.0 at speeds up to 9.6 GT/s, enabling efficient communication between two sockets while maintaining low latency for scalable . Enterprise-grade reliability is bolstered by advanced (RAS) extensions, including error correction capabilities in the L1, L2, and L3 caches to detect and recover from transient errors without system disruption. and I/O are supported via Intel VT-d, which provides input-output (IOMMU) functionality for direct device assignment and secure DMA operations. Additionally, these processors enable SMI-free operation modes to minimize interrupt latency, facilitating real-time applications in server settings. Memory subsystem enhancements allow up to 768 GB of DDR4-2133 per socket across four channels, supporting high-bandwidth demands for and large-scale databases. Power and thermal management are tailored for cluster-scale deployments, with thermal design power (TDP) ratings ranging from 120 W to 150 W per socket to balance performance and efficiency. Intel Node Manager integration enables fine-grained power capping at the node level, allowing administrators to enforce cluster-wide power limits while optimizing resource utilization. NUMA-aware hardware topology, including cluster-on-die configurations for high-core-count SKUs, facilitates OS-level scheduling optimizations that localize access and reduce inter-socket . Launched in September 2014, the Haswell-EP lineup targets (HPC) and environments with its scalable design. A related variant, the E7 v3 (Haswell-EX), extends support to four-socket systems with up to 18 cores per socket, emphasizing mission-critical applications requiring enhanced and throughput.

Compatibility Considerations

Haswell processors maintain full with prior x86 instruction sets, ensuring seamless execution of legacy software developed for earlier architectures. However, the introduction of (TSX) in Haswell encountered hardware errata that led to instability under certain conditions, prompting to disable TSX via a update in 2014. This update affected all Haswell and early Broadwell CPUs, rendering TSX features unavailable without specialized reconfiguration, though it did not impact overall x86 compatibility. Hardware compatibility for Haswell centers on the socket, which is incompatible with previous Ivy Bridge processors on the socket, necessitating new motherboards based on Intel's 8-series chipsets (such as Z87, H87, Q87, B85, and H81). Initial 8-series motherboards required a update to support Haswell Refresh variants released in 2014, as these CPUs featured minor revisions that demanded updated firmware for stable operation. Failure to apply such updates could result in boot failures or reduced performance, though the process typically involved using an original Haswell CPU to flash the before installing a Refresh model. Official operating system support for Haswell is limited for legacy versions like and Vista, with ceasing graphics driver development for these OSes after the Ivy Bridge generation; the final driver (version 14.51.11.5437) was released in February 2013 and does not cover Haswell's HD Graphics. Similarly, no official HD Graphics drivers exist for on Haswell, leading to potential instability or lack of acceleration, though unofficial modifications have been reported in enthusiast communities—at the cost of and reliability. Post-2014, Microsoft's end-of-support for XP and Vista further exacerbates these issues, as no security updates address Haswell-specific interactions. For modern OSes, HD Graphics in Haswell requires or later for 11 support, with full 11.1 features available only on and newer due to OS-level limitations. On , optimal , including the intel_pstate driver for , necessitates kernel version 3.10 or higher, which introduced comprehensive Haswell support in June 2013.

Processor Lineup

Desktop Processors

The Haswell desktop processors, targeted at consumer and enthusiast systems using the socket, span the Core i7, Core i5, Core i3, , and families, offering a range of core counts, clock speeds, and thermal design powers (TDPs) from 53W to 84W. These processors integrate as standard, with select Haswell Refresh models featuring the more capable Iris Pro Graphics 5200 augmented by 128 MB of for improved in graphics-intensive tasks.

Core i7 Models

The flagship Core i7 desktop processors in the Haswell lineup provide four cores and eight threads, supporting for enhanced multitasking. Representative examples include the unlocked i7-4770K, with a base frequency of 3.5 GHz, maximum turbo frequency of 3.9 GHz, 8 MB of Smart Cache, and an 84W TDP, designed for enthusiasts. The i7-4790, part of the Haswell Refresh, offers a higher base frequency of 3.6 GHz and maximum turbo of 4.0 GHz while maintaining the same core/thread count, cache size, TDP, and integrated HD Graphics 4600.

Core i5 Models

Core i5 desktop processors feature four cores without , balancing performance and efficiency for mainstream users. The unlocked i5-4670K runs at a base frequency of 3.4 GHz with a maximum turbo of 3.8 GHz, 6 MB Smart Cache, and 84W TDP, paired with HD Graphics 4600. In the Refresh lineup, the i5-4690 improves to a 3.5 GHz base and 3.9 GHz turbo, retaining the four-core configuration, 6 MB cache, 84W TDP, and HD Graphics 4600.

Lower-End Models

Entry-level desktop options include the and series, both with two cores and two threads, suited for basic computing and budget builds with lower TDPs. The G3220 operates at a fixed 3.0 GHz frequency, 3 MB Smart Cache, and 53W TDP, integrating Intel HD Graphics for 4th Generation Intel Processors. Similarly, the G1820 has a 2.7 GHz clock speed, 2 MB Smart Cache, 53W TDP, and the same integrated graphics solution.
ModelCores/ThreadsBase FrequencyMax TurboCacheTDPGraphicsUnlocked
i7-4770K4/83.5 GHz3.9 GHz8 MB84WHD 4600Yes
i7-47904/83.6 GHz4.0 GHz8 MB84WHD 4600No
i5-4670K4/43.4 GHz3.8 GHz6 MB84WHD 4600Yes
i5-46904/43.5 GHz3.9 GHz6 MB84WHD 4600No
Pentium G32202/23.0 GHzN/A3 MB53WHD (4th Gen)No
Celeron G18202/22.7 GHzN/A2 MB53WHD (4th Gen)No

Mobile Processors

The Haswell microarchitecture powered a range of mobile processors optimized for laptops, ultrabooks, and tablets, with a focus on balancing performance and power efficiency through low (TDP) variants. These processors, part of Intel's 4th Generation Core family, targeted portable devices by incorporating ultra-low-voltage (ULV) designs in the U and Y series, which operated at TDPs of 15W or lower to extend battery life while supporting features like and integrated graphics. For ultrabooks, the Core i7-4650U served as a premium dual-core option with four threads, a base clock of 1.7 GHz, and a maximum turbo frequency of 3.3 GHz, paired with a 15W TDP and for efficient handling. Similarly, the Core i5-4200U provided a mid-range alternative with two cores and four threads, a 1.6 GHz base clock, up to 2.6 GHz turbo, 3 MB cache, and the same 15W TDP, enabling slim designs without sacrificing responsiveness. High-end mobile configurations, such as those in performance laptops, featured the Core i7-4940MX Extreme Edition, a quad-core processor with eight threads, a 3.1 GHz base frequency, turbo boost up to 4.0 GHz, and support for quad-channel DDR3L memory to handle demanding workloads like . At the entry level, the Core i3-4010U offered two cores and four threads at a fixed 1.7 GHz clock with a 15W TDP and HD Graphics 4400, suitable for basic productivity tasks. For even lower-power tablet applications, the 2955U delivered two cores and two threads at 1.4 GHz with a 15W TDP (configurable down to 7.5W in some systems), prioritizing longevity over peak performance. Certain variants integrated advanced graphics for gaming and creative laptops, such as the Core i7-4950HQ with quad cores, eight threads, a 2.4 GHz base up to 3.6 GHz turbo, and Iris Pro Graphics 5200 featuring 128 execution units and eDRAM cache, available in 47W TDP configurations that could scale down to 28W for better thermal management in mobile chassis.
Processor ModelCores/ThreadsBase/Turbo Frequency (GHz)Cache (MB)TDP (W)Integrated Graphics
Core i7-4650U2/41.7 / 3.3415HD 5000
Core i5-4200U2/41.6 / 2.6315HD 4400
Core i7-4940MX4/83.1 / 4.0857HD 4600
Core i3-4010U2/41.7 / N/A315HD 4400
Celeron 2955U2/21.4 / N/A215HD (Gen 4)
Core i7-4950HQ4/82.4 / 3.6647Iris Pro 5200

Server Processors

The Haswell microarchitecture powered a range of processors optimized for server environments, emphasizing high core counts, error-correcting code ( support, and scalability for multi-socket configurations to handle demanding enterprise workloads such as , , and . These server variants introduced enhancements like larger last-level caches and improved compared to prior generations, enabling better performance in parallel processing tasks while maintaining compatibility with existing server infrastructure. The Xeon E5-2600 v3 series targeted dual-socket servers, offering up to 18 cores and 36 threads per socket with a focus on balanced performance and power efficiency. For instance, the E5-2699 v3 model features 18 cores, 36 threads, a base frequency of 2.3 GHz, a maximum turbo frequency of 3.6 GHz, 45 MB of L3 cache, and a 145 W TDP, supporting DDR4-2133 memory with ECC for reliable data integrity in mission-critical applications. Similarly, the E5-2650 v3 provides 10 cores, 20 threads, a base frequency of 2.3 GHz, a maximum turbo frequency of 3.0 GHz, 25 MB of L3 cache, and a 105 W TDP, making it suitable for mid-range servers requiring cost-effective scalability. This series utilized Intel's QuickPath Interconnect (QPI) for multi-socket communication, with up to three links per processor to support configurations up to two sockets. For single-socket server designs, the E3-1200 v3 series delivered workstation-like performance with server-grade features, including ECC support for DDR3 memory. The E3-1280 v3, for example, includes 4 cores, 8 threads, a base of 3.6 GHz, a maximum turbo of 4.0 GHz, 8 MB of L3 cache, an 80 W TDP, and compatibility with DDR3-1600 , positioning it for entry-level servers and embedded systems where space and power constraints are key. High-end multi-socket systems were served by the E7-4800 v3 series, which supported up to quad-socket configurations and introduced DDR4 memory support for higher bandwidth. Models in this lineup scaled to up to 18 cores per socket, with the series rated up to 165 W TDP, enabling robust RAS (reliability, availability, serviceability) features like advanced error correction and hot-swappable components for large-scale data centers.

References

  1. https://pages.cs.wisc.edu/~rajwar/papers/ieee_micro_haswell.pdf
  2. https://old.hotchips.org/wp-content/uploads/hc_archives/hc25/HC25.80-Processors2-epub/HC25.27.820-Haswell-Hammarlund-Intel.pdf
  3. https://www.realworldtech.com/haswell-cpu/
  4. https://umairmohsin.wordpress.com/2009/12/23/beyond-the-core-intel-roadmap-2010/
  5. https://www.intc.com/news-events/press-releases/detail/627/personal-computing-transformed-ultrabook-ushers-in-the
  6. https://www.servethehome.com/intel-xeon-e3-1230-v3-review-haswell-xeon-benchmarks/
  7. https://www.wired.com/story/intel-haswell-cpu/
  8. https://www.denverpost.com/2013/06/02/intel-names-computer-chip-after-tiny-colorado-town/
  9. https://www.engadget.com/2013-06-01-intel-haswell-launch.html
  10. https://pcper.com/2013/09/idf-2013-intel-shows-off-haswell-y-and-14nm-broadwell-chips-in-new-devices/
  11. https://pcper.com/2013/06/the-haswell-review-intel-core-i7-4770k-performance-and-architecture/
  12. https://www.intel.com/content/www/us/en/ark/products/codename/42174/products-formerly-haswell.html
  13. https://www.intel.com/content/www/us/en/products/sku/75117/intel-core-i74700mq-processor-6m-cache-up-to-3-40-ghz/specifications.html
  14. https://www.agner.org/optimize/microarchitecture.pdf
  15. https://www.realworldtech.com/haswell-cpu/6/
  16. https://www.realworldtech.com/haswell-cpu/2/
  17. https://cdrdv2-public.intel.com/814199/356477-Optimization-Reference-Manual-V2-002.pdf
  18. https://en.wikichip.org/wiki/intel/microarchitectures/haswell_(client)
  19. https://www.realworldtech.com/haswell-cpu/5/
  20. https://inria.hal.science/hal-01259139/document
  21. https://arstechnica.com/gadgets/2013/05/a-look-at-haswell/
  22. https://www.realworldtech.com/haswell-cpu/4/
  23. https://prace-ri.eu/wp-content/uploads/Best-Practice-Guide_Haswell.pdf
  24. https://ntrs.nasa.gov/api/citations/20170000291/downloads/20170000291.pdf
  25. https://ark.intel.com/content/www/us/en/ark/products/75122/intel-core-i7-4770-processor-8m-cache-up-to-3-90-ghz.html
  26. https://www.researchgate.net/publication/271416878_FIVR_-_Fully_integrated_voltage_regulators_on_4th_generation_IntelR_Core_SoCs
  27. https://www.pugetsystems.com/labs/articles/technology-primer-intel-haswell-desktop-cpus-457/
  28. https://www.theverge.com/2011/09/13/525778/intel-announces-haswell-processor-2013-improved-power-efficiency
  29. https://download.intel.com/newsroom/kits/22nm/pdfs/22nm-Details_Presentation.pdf
  30. https://cdrdv2-public.intel.com/821614/356477-Software-Optimization-Manual_V2.pdf
  31. https://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-software-developer-vol-2a-manual.pdf
  32. https://www.intel.com/content/www/us/en/developer/articles/community/exploring-tsx-with-software-development-emulator.html
  33. https://www.intel.com/content/dam/develop/external/us/en/documents/hpca-tsx-slides-393551.pdf
  34. https://ieeexplore.ieee.org/document/6877294
  35. https://www.techpowerup.com/gpu-specs/intel-haswell-gt2.g591
  36. https://www.intel.com/content/www/us/en/products/sku/75045/intel-core-i54570t-processor-4m-cache-up-to-3-60-ghz/specifications.html
  37. http://www.diva-portal.org/smash/get/diva2:1137252/FULLTEXT01.pdf
  38. https://www.notebookcheck.net/Intel-HD-Graphics-Haswell.93341.0.html
  39. https://community.intel.com/t5/Media-Intel-oneAPI-Video/Full-4k-HW-h264-encode-on-Haswell/td-p/943381
  40. https://www.intel.com/content/www/us/en/products/sku/80806/intel-core-i74790-processor-8m-cache-up-to-4-00-ghz/specifications.html
  41. https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/8-series-chipset-pch-datasheet.pdf
  42. https://arstechnica.com/gadgets/2013/06/haswell-is-here-intel-announces-first-4th-generation-core-cpus/
  43. https://www.nag.co.za/2014/09/16/a-note-on-intels-lingering-dmi-bottleneck-and-storage/
  44. https://cdrdv2-public.intel.com/328901/4th-gen-core-family-mobile-m-h-processor-lines-vol-1-datasheet.pdf
  45. https://www.cpu-world.com/news_2012/2012082302_New_video_and_graphics_features_of_Haswell_CPUs.html
  46. https://www.thunderbolttechnology.net/product/asrock-z87-extreme4tb4
  47. https://www.techpowerup.com/199861/intel-starts-shipping-haswell-refresh-processors
  48. https://bit-tech.net/reviews/tech/intel-core-i7-4790k-devil-s-canyon-review/1/
  49. https://www.legitreviews.com/hands-on-with-devils-canyon-intel-haswell-refresh_137746
  50. https://pcper.com/2014/06/the-intel-core-i7-4790k-devils-canyon-review-and-overclocking/
  51. https://pcper.com/2014/04/refreshed-intel-haswell-processors-coming-next-month-for-desktop-and-mobile/
  52. https://www.intel.com/content/www/us/en/products/sku/80807/intel-core-i74790k-processor-8m-cache-up-to-4-40-ghz/specifications.html
  53. https://pcper.com/2014/03/intel-brings-iris-pro-graphics-to-broadwell-in-lga-sockets/
  54. https://www.techpowerup.com/187987/intel-broadwell-cpu-lineup-arrives-in-2h-2014
  55. https://www.intel.com/content/www/us/en/products/sku/81061/intel-xeon-processor-e52699-v3-45m-cache-2-30-ghz/specifications.html
  56. https://www.intel.com/content/www/us/en/products/sku/81713/intel-xeon-processor-e52690-v3-30m-cache-2-60-ghz/specifications.html
  57. https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/xeon-e5-v3-datasheet-vol-1.pdf
  58. https://www.intel.com/content/dam/www/public/us/en/documents/datasheets/xeon-e5-v3-datasheet-vol-2.pdf
  59. https://www.intel.com/content/www/us/en/products/sku/83352/intel-xeon-processor-e52620-v3-15m-cache-2-40-ghz/specifications.html
  60. https://www.microway.com/hardware/intel-xeon-e5-2600-v3-haswell-processor-review/
  61. https://www.intel.com/content/dam/www/public/us/en/documents/technical-specifications/intel-power-node-manager-v3-spec.pdf
  62. https://forum.level1techs.com/t/numa-support-for-up-xeon-e5-26xx-v3-cluster-on-die/166446
  63. https://www.techpowerup.com/cpu-specs/xeon-e5-2640-v3.c2868
  64. https://www.sqlskills.com/blogs/glenn/intel-xeon-e7-v3-product-family-released/
  65. https://www.kitguru.net/components/cpu/anton-shilov/intel-finds-erratum-in-haswell-and-broadwell-chips-disables-tsx-instructions/
  66. https://www.techpowerup.com/201046/ecs-8-series-motherboards-now-support-intel-haswell-refresh-processors
  67. https://www.intel.com/content/www/us/en/download/16543/intel-hd-graphics-driver-for-windows-xp-exe.html
  68. https://www.intel.com/content/www/us/en/download/16652/intel-hd-graphics-driver-for-windows-vista-64-exe.html
  69. https://www.phoronix.com/review/intel_haswell_linux/3
  70. https://www.notebookcheck.net/Intel-Iris-Pro-Graphics-5200.90965.0.html
  71. https://ark.intel.com/content/www/us/en/ark/products/75608/intel-celeron-processor-2955u-2m-cache-1-40-ghz.html
  72. https://www.intel.com/content/www/us/en/support/products/78583/processors/intel-xeon-processors/intel-xeon-processor-e5-v3-family.html
Add your contribution
Related Hubs
User Avatar
No comments yet.