Sandy bridge processor graphics: Intel HD Graphics (Sandy Bridge)

283 Comments

IntroductionThe LineupOverclocking: Effortless 4.4GHz+ on AirThe 6-series Platform & 6Gbps SATA PerformanceUEFI Support: 3TB Drives & Mouse Support Pre-BootThe Future: Z68 Chipset in Q2, LGA-2011 in Q4A Near-Perfect HTPCIntel’s Quick Sync TechnologyQuick Sync: The Best Way to TranscodeIntel’s Gen 6 GraphicsIntel HD Graphics 2000/3000 PerformanceResolution Scaling with Intel HD Graphics 3000Overclocking Intel’s HD GraphicsThe TestSYSMark 2007 & Photoshop PerformanceVideo Encoding Performance3D Rendering PerformanceFile Compression/Decompression PerformanceVisual Studio 2008, Flash Video Creation, & Excel PerformanceGaming PerformancePower ConsumptionFinal Words

Intel’s Gen 6 Graphics

All 2nd generation Core series processors that fit into an LGA-1155 motherboard will have one of two GPUs integrated on-die: Intel’s HD Graphics 3000 or HD Graphics 2000. Intel’s upcoming Sandy Bridge E for LGA-2011 will not have an on-die GPU. All mobile 2nd generation Core series processors feature HD Graphics 3000.

The 3000 vs. 2000 comparison is pretty simple. The former has 12 cores or EUs as Intel likes to call them, while the latter only has 6. Clock speeds are the same although the higher end parts can turbo up to higher frequencies. Each EU is 128-bits wide, which makes a single EU sound a lot like a single Cayman SP.

Unlike Clarkdale, all versions of HD Graphics on Sandy Bridge support Turbo. Any TDP that is freed up by the CPU running at a lower frequency or having some of its cores shut off can be used by the GPU to turbo up. The default clock speed for both HD 2000 and 3000 on the desktop is 850MHz; however, the GPU can turbo up to 1100MHz in everything but the Core i7-2600/2600K. The top-end Sandy Bridge can run its GPU at up to 1350MHz.

Mobile is a bit different. The base GPU clock in all mobile SNB chips is 650MHz but the max turbo is higher at 1300MHz. The LV/ULV parts also have different max clocks, which we cover in the mobile article.

As I mentioned before, all mobile 2nd gen Core processors get the 12 EU version—Intel HD Graphics 3000. The desktop side is a bit more confusing. In desktop, the unlocked K-series SKUs get the 3000 GPU while everything else gets the 2000 GPU. That’s right: the SKUs most likely to be paired with discrete graphics are given the most powerful integrated graphics. Of course those users don’t pay any penalty for the beefier on-die GPU; when not in use the GPU is fully power gated.

Despite the odd perk for the K-series SKUs, Intel’s reasoning behind the GPU split does makes sense. The HD Graphics 2000 GPU is faster than any desktop integrated GPU on the market today, and it’s easy to add discrete graphics to a desktop system if the integrated GPU is insufficient. The 3000 is simply another feature to justify the small price adder for K-series buyers.

On the mobile side going entirely with 3000 is simply because of the quality of integrated or low-end graphics in mobile. You can’t easily add in a discrete card so Intel has to put its best foot forward to appease OEMs like Apple. I suspect the top-to-bottom use of HD Graphics 3000 in mobile is directly responsible for Apple using Sandy Bridge without a discrete GPU in its entry level notebooks in early 2011.

I’ve been careful to mention the use of HD Graphics 2000/3000 in 2nd generation Core series CPUs, as Intel will eventually bring Sandy Bridge down to the Pentium brand with the G800 and G600 series processors. These chips will feature a version of HD Graphics 2000 that Intel will simply call HD Graphics. Performance will be similar to the HD Graphics 2000 GPU, however it won’t feature Quick Sync.

Image Quality and Experience

Perhaps the best way to start this section is with a list. Between Jarred and I, these are the games we’ve tested with Intel’s on-die HD 3000 GPU:

Assassin’s Creed
Batman: Arkham Asylum
Borderlands
Battlefield: Bad Company 2
BioShock 2
Call of Duty: Black Ops
Call of Duty: Modern Warfare 2
Chronicles of Riddick: Dark Athena
Civilization V
Crysis: Warhead
Dawn of War II
DiRT 2
Dragon Age Origins
Elder Scrolls IV: Oblivion
Empire: Total War
Far Cry 2
Fallout 3
Fallout: New Vegas
FEAR 2: Project Origin
HAWX
HAWX 2
Left 4 Dead 2
Mafia II
Mass Effect 2
Metro 2033
STALKER: Call of Pripyat
Starcraft II
World of Warcraft

This is over two dozen titles, both old and new, that for the most part worked on Intel’s integrated graphics. Now for a GPU maker, this is nothing to be proud of, but given Intel’s track record with game compatibility this is a huge step forward.

We did of course run into some issues. Fallout 3 (but not New Vegas) requires a DLL hack to even run on Intel integrated graphics, and we saw some shadow rendering issues in Mafia II, but for the most part the titles—both old and new—worked.

Modern Warfare 2 in High Quality

Now the bad news. Despite huge performance gains and much improved compatibility, even the Intel HD Graphics 3000 requires that you run at fairly low detail settings to get playable frame rates in most of these games. There are a couple of exceptions but for the most part the rule of integrated graphics hasn’t changed: turn everything down before you start playing.

Modern Warfare 2 the way you have to run it on Intel HD Graphics 3000

This reality has been true for more than just Intel integrated graphics however. Even IGPs from AMD and NVIDIA had the same limitations, as well as the lowest end discrete cards on the market. The only advantage those solutions had over Intel in the past was performance.

Realistically we need at least another doubling of graphics performance before we can even begin to talk about playing games smoothly at higher quality settings. Interestingly enough, I’ve heard the performance of Intel’s HD Graphics 3000 is roughly equal to the GPU in the Xbox 360 at this point. It only took six years for Intel to get there. If Intel wants to contribute positively to PC gaming, we need to see continued doubling of processor graphics performance for at least the next couple generations. Unfortunately I’m worried that Ivy Bridge won’t bring another doubling as it only adds 4 EUs to the array.

Quick Sync: The Best Way to Transcode
Intel HD Graphics 2000/3000 Performance
IntroductionThe LineupOverclocking: Effortless 4.4GHz+ on AirThe 6-series Platform & 6Gbps SATA PerformanceUEFI Support: 3TB Drives & Mouse Support Pre-BootThe Future: Z68 Chipset in Q2, LGA-2011 in Q4A Near-Perfect HTPCIntel’s Quick Sync TechnologyQuick Sync: The Best Way to TranscodeIntel’s Gen 6 GraphicsIntel HD Graphics 2000/3000 PerformanceResolution Scaling with Intel HD Graphics 3000Overclocking Intel’s HD GraphicsThe TestSYSMark 2007 & Photoshop PerformanceVideo Encoding Performance3D Rendering PerformanceFile Compression/Decompression PerformanceVisual Studio 2008, Flash Video Creation, & Excel PerformanceGaming PerformancePower ConsumptionFinal Words

PRINT THIS ARTICLE

New Intel Sandy Bridge architecture processors

Processors of the new
Sandy architecture
Bridge
(«sand bridge») replaced the current
Nehalem
(45 nm) and Westmere
(32 nm). Under the release rule
tick-tock, Sandy
Bridge
is «such» — a new microarchitecture,
released on an already developed technical process
(32 nm). However, at the end of 2011
another tick is expected — a little
updated 22nm version of Sandy
Bridge
called Ivy
Bridge
(ivy bridge).

Short list
real advantages compared to
current generation of Intel CPU:

• Into the chip now
  chipset northbridge integrated together
  with video core.

• Self graphic
  core (GPU) updated but kept
  Possibility to connect external
  graphics.

• The core of the common
  destinations support expansion
  AVX (Advanced)
  vector
  extensions).
  Being an evolutionary development of a set
  SSE commands, this innovation will allow
  improve processor performance
  at the same clock frequency. Besides,
  new processors will inherit the mechanism
  support for AES encryption, for the first time
  implemented in Westmere.

• L3 cache increased
  throughput through
  high-speed connection to cores
  (including to the GPU).

• Technology
  auto-acceleration TurboBoost
  improved and now allows automatically
  raise/lower x86 core frequencies and
  graphics, and even higher level
  nominal heat pack (TDP), if
  general temperature conditions are
  allow. There was also an additional
  technology within Turbo
  boost,
  which for maximum responsiveness
  systems in interactive tasks can
  for a short time, but very aggressively
  increase the frequency of the processor.

Here on these and more
small points and go through.

TO

large-block
the CPU core has not changed. It’s all about the details.

IN

O
front, the main innovation is the cache on
1536 uops buffering results
decoder work. Also redesigned
branch predictor storing
now twice as many target addresses and
a longer history of command behavior.
Connected to it is a block that defines
hits in the mop-cache, and in these cases
L1I cache, length predecoder and decoders
turn off to save energy. Also
performance increases for
by improving the response to incorrect
predicted transition (correct branch
may end up in the mop cache) and release
from having to decode long
commands for several cycles at each
their performance.

Rear of the conveyor
changed more. First obvious change:
transition from the full ROB buffer (storing
mops and the values ​​of the registers they change)
to a combination of abbreviated ROB (storing
only mops in original order) and
two physical register files
(PRFs that store the contents of the renamed
registers that uops refer to)
– integer and vector-real.
The meaning of the transition is that instead of forwarding
mops are sent along with the data
only links to registers, and mops themselves
and the data stays in place until
resignations — as a result of the removal
unnecessary transfers saves energy.
Responsible for performance
duty increase in the size of everything
variety of buffers and queues.

H

A
functional unit level main
news — support for AVX, a new addition
x86-64 command systems. Unlike the next
SSE just adding new commands,
AVX allows:

• refers to
  256-bit ymm vector registers,
  extending the usual xmm for SSE;

• use during
  all vector commands 3-4 operands in
  non-destructive form when the result
  operation does not destroy the original
  arguments;

• save on
  size of vector instructions due to
  code compaction (especially for SSSE3 and
  Further).

R

of course
there are also new instructions — for the distribution
elements in vector and masked
memory operations.

To good
beginnings did not remain only at the level
AVX command recognition, FU block redesigned
to handle ymm registers with the same
speed, like xmm. For this 128-bit
adder, multiplier, ALU and shufflers
doubled the bit depth, respectively
boosting peak performance
kernels.

L1D cache increased by 50%
full bandwidth — now
that’s 32 bytes of read + 16 bytes of write per
tact. To do this, in LSU (L1D controller)
a second 16-byte read port is supplied.
This will allow you to saturate at the right pace.
FU data when executing AVX code.

Out of the core (so
Intel
names the rest of the CPU not included
in general purpose kernels) list of changes
opens, as it were, a new (appeared yet
in March 2010 in 8-core server chips
Nehalem EX)
32-byte ring bus linking
L3 cache, compute and graphics
nuclei and «system agent» (former northern
bridge with ICP for DDR3 memory). L3 Cache,
designated as LLC (last
level
cache,
last level cache), consists of
several banks (according to the number of cores of the total
destination) with an access port in each.
The ring bus that connects it all
operates at a frequency comparable to the frequency
cores, and ideally transmits
the amount of data equal to the product
32 for the number of banks. In this manner
shared L3 performance
Cache
increases in proportion to the number
cores, but the promised reduction in latency
(at least in cycles) doubtful
compared to centralized
controller L3.

Sandy
Bridge
is the first architecture
integrates on a single chip as a CPU,
as well as the GPU. chips
with Sandy architecture
Bridge
will be produced according to the standards of 32 nm
technology, however, unlike
existing analogues, integrated
the graphics core will be created in them
according to the same 32 nm standards and will be located on
one chip with computing
cores.

These two components
can share a common Cache
— memory integrated on a chip,
which improves overall work efficiency
systems thanks to the superiority of Cache
over RAM (Intel
speaks of a fourfold increase
throughput efficiency).
Since the GPU is now located on
processor chip, and not on a separate one,
as in current Core i3/i5, L3 Cache
can be used incl. and for
graphics, which alleviates the disadvantage
system memory bandwidth
compared to specialized
video memory, rarely available for
built-in GPU.

The video core is promised
double performance, and
only due to the fast L3 Cache,
but also from doubling the number of executive
blocks up to 12 — however, not in all models.
Support for DirectX 10 has also been introduced (and on
yard is already the 11th…) and new HD-video formats
for energy efficient hardware
speed up (de)coding.

The CPU will receive
triple power — for common cores
destination and L3, for the GPU, as well as for
system agent with controllers
all kinds of tires.

Controller
thermal mode and power implements
updated version of TurboBoost technology,
taking into account transient processes. She
allow in a short time
(20–60 sec) overclock cores (including GPU) to
frequencies at which heat generation even
above the nominal peak (TDP), if before
The CPU has been idle for a while. This
will provide more responsiveness to short
in run time, but computationally
intensive tasks.

Structurally,
processor generation change
associated with the replacement of the processor socket.
The LGA 1156 connector will be replaced by the LGA connector
1155. With external resemblance and identical
sizes, these connectors are completely
incompatible with each other. Processors
will be supplied with a wide range
TDP values ​​and frequencies in different (not
compatible with modern) cases
for new motherboards — LGA1155
(frequencies 2.3-3.4 GHz). At the same time, all produced
now motherboards will become morally
obsolete.

List of chipsets
for motherboards with LGA 1155 socket
includes models P67, H67, H61 and Q67. Remarkably,
that support for SATA 6 Gb/s will be implemented
in the southbridge, but USB 3.0 is not. Won’t be there
and the system clock generator,
also migrated to the CPU chip is
means the impossibility of overclocking through
base frequency, because it’s fixed
at 100 MHz. However, Intel leaves
the possibility of overclocking due to the multiplier,
which will be unlocked in models
with the letter K and can reach the value
57. Speaking of models — their names are the same
remain as Core i3/i5/i7, but the model number
now 4 digits. In addition to the letter K, it
can complement S (low TDP), T (even more
economical) or M (mobile). A little bit later
server and budget options will be released.

successor
processor socket LGA 1366 will become LGA
2011. Significant increase in the number
contacts is explained by the presence of
processor 256-bit memory bus
(four DDR3 channels) and integration into
northbridge function processor.
PCI-Express 2.0 interface root block
will provide system developers
boards 32 lines (the LGA-1155 will have 16 of them).

What is the result? Intel
seriously updated Nehalem — without revolutionary
breakthroughs, but simply consistently
updating microarchitecture details and
introducing all the latest innovations
two years. As a result, we got more
faster, but remaining in the same
thermal framework, processors.

Sandy
bridge,
like the old Nehalem chips
and Westmere
still have integrated
crystal pci controllers
express,
2-channel DDR3, and now also
displayport.
Decision to use 2-channel
memory controller taken on volume
based on the fact that the majority of clients
Intel
3 memory channels are not needed, and the LGA market
1366 is too small.

Sandy
Bridge
are the first Intel processors
with AVX instruction set support
(advanced vector extensions),
improving performance in a number of
highly parallel simple operations,
commonly used in multimedia applications.

roundabout concept
tires also greatly simplifies the creation
based on Sandy
Bridge
various crystals. For example, despite
that the prototypes of new processors
have 4 x86 processing cores, technically
very easy to make a 2-core option
or add/remove any blocks in
the future.

Server Chips
Sandy
Bridge
will be based on the same roundabout design
tires and will be able to have in the foreseeable future
many cores (chips with 10
nuclei).

Every Sandy Core
Bridge
has 256 KB of second data cache
level. To access the cache
the third level you need 25 full
machine cycles. The processor is equipped
integrated PCI Express controller
2.0 and Dual Channel Memory Controller
with support for DDR3-1600 memory. Calculated
for production according to 32 nm standards,
quad-core microprocessor will take
on a silicon crystal with an area of ​​225 sq. mm.
Power consumption will be 85 watts.

These
design features are believed to
will improve performance
and efficiency of Sandy processors
Bridge.

Expected that
new processors along with the transition to
new architecture will get another
nomenclature: instead of three-digit
model marking after series indication
(Core i3, Core i5, or Core i7), a four-digit
type 2XX0. So Intel
will try to avoid confusion with names
processor models of Westmere architectures
and Sandy
Bridge.

As it was said
higher, initially a new generation of processors
Intel
will be represented by five models:

• Core i7-2600: quad-core
  technology-enabled processor
  Hyper-Threading, clock speed — 3. 2 GHz,
  third level cache — 8
  MB;

• Core i5-2400: quad-core
  processor without technology support
  Hyper-Threading, clock speed — 3.1 GHz,
  third level cache — 6
  MB;

• Core i5-2500: quad-core
  processor without technology support
  Hyper-Threading, clock speed — 3.3 GHz,
  third level cache — 6
  MB;

• Core i3-2100: dual core
  technology-enabled processor
  Hyper-Threading, clock speed — 3.1 GHz,
  third level cache — 3
  MB;

• Core i3-2120: dual core
  technology-enabled processor
  Hyper-Threading, clock speed — 3.3 GHz,
  third level cache — 3
  MB.

Thus,
first models on Sandy architecture
Bridge
will be represented by dual and quad-core
technology-enabled processors
Multithreading Hyper-Threading
(or without it), operating at a frequency
over 3 GHz. Then to them, obviously,
more powerful six- and
octa-core models with 16 and 24 MB cache
third level.

better chips, performance improvements and more

Intel processors have been at the forefront of technology for several decades, and the processors have gradually become stronger over the years. Perhaps you are looking to purchase an Intel processor and the seller is confusing you with its generation and performance. It can be difficult to understand the generations of Intel processors and their differences. Moreover, Intel Core i3, i5, and i7 are not processor generations, but Intel processor models.

This article discusses important Intel processors and their features from 2017 to the latest generations.

Explanation of Intel processor names and numbers

Before we get into the Intel processor history, let’s find out the exact information that an Intel processor model can provide. give. The model name has a countless number of numbers and letters, so it can be confusing.

The Intel processor name contains the following information:

• Brand: indicates the entire product line, such as Core, Pentium, Xeon and Celeron.
• Brand Modifier: Indicates the CPU performance for this specific brand. The effectiveness of a brand modifier increases with its value. Example: i3, i5, i7, i9.
• Generation indicator. In the processor number, the first or first two digits indicate the generation of the processor. If it is a 12th generation processor, then the first two digits are the indicator. Similarly, if the number is 9th generation or any number starting with a single whole number, the indicator being the first digit. Example: 7XXX, 11XXX, etc.
• Part Number: These are the last three digits of the processor number. The higher the SKU, the more features are likely to be included in the processor.
• Product line suffix: is the alphabet at the end. It indicates the critical characteristics of the processor.

Here is a list of all Intel Core product line suffixes and their features:

Intel processor generations

The following are all generations of Intel processors, their code names and release dates:

2017

Intel processor generations: from Coffee Lake to Raptor Lake

Recently, Intel has released advanced and powerful Core processors. However, since the release of Coffee Lake in 2017, the world has experienced a generational revolution; after this generation of Intel processors are gradually improved.

Let’s discuss all about Intel processors, their performance and improvements since the release of Coffee Lake.

9th Generation Intel CPU — Coffee Lake

Intel Core i-9 processors first appeared after the release of Coffee Lake in 2017. These processors break the usual limit of four cores per CPU. It can support up to eight cores.

However, Intel knew that increasing the number of cores would produce more heat, so they added an integrated heat spreader (IHS) to the processor instead of regular thermal paste. It supports more than 4 cores and up to 16MB L3 cache as it uses 1151 LGA sockets along with a modified pinout.

Some of the best 9th generation chips:

• Intel Core i9 9900K — best for gaming and overclocking
• Intel Core i7 9700K — best i7 processor
• Intel Core i5 9600K — the best budget processor

10th Gen Intel CPU — Cannon Lake/Ice Lake

The 10th Gen Intel CPU, Canon Lake, was duly introduced in 2018 using brand new 10nm technology. Ice Lake was also produced as the second generation of 10nm processors.

These processors use BGA1526 sockets and support DDR4 3200 and LPDDR4X 3733. In addition, they were the first processors with built-in Wi-Fi 6 and Thunderbolt 3 support. 10700K is the best overall

11th Gen Intel CPU — Tiger Lake

The third generation of the 11th Gen Intel 10nm CPU was released in September 2020. Its code name is Tiger Lake and it has a performance boost of up to 19% compared to previous Ice Lake processors. In addition, the L4 cache has been introduced in these processors to further improve performance.

Notably, the famous Intel Xe graphics were introduced with the release of the 11th generation. This feature has greatly increased FPS in games.

Some of the best 11th gen processors:

• Intel Core i9 11900K top score
• Intel Core i5 11400F is the best budget processor for gaming
• Intel Core i7 11700K — best i7 processor

12th Gen Intel CPU — Alder Lake

The 12th Gen CPUs are known as Alder Lake and boast 10nm transistor technology with hybrid capabilities. It was released in November 2021. These processors are designed for speed and power with P and E cores. P stands for power cores and E stands for efficiency cores.

These processors support DDR5 memory, Wi-Fi 6E (Gig+), and Thunderbolt 4. Intel claims that the 12th Gen processor can deliver an average 13% performance boost in games.

Some of the best 12th generation chips:

• Intel Core i9 12900K — best overall
• Intel Core i7 12700KF — best i7 processor
• Intel Core i5 12600KF — Best Budget Processor

13th Gen Intel CPU — Raptor Lake

Intel’s 13th Gen CPU is known as Raptor Lake. This is the most recent version released in October 2022. It boasts improved 10nm SuperFin technology. The improvements are astronomical over its predecessor.

Core i9 13900K is the only model with 24 cores and 32 threads. Alternatively, the Core i5 13400 will have 6 P-cores and 4 E-cores.

Some of the best 12th generation chips:

• The Intel Core i9 13900K is the best processor overall.