Tag Archives: Tech Report

Precision Boost Overdrive 2 : Adaptive Undervolting Comes To Ryzen!

AMD just introduced Precision Boost Overdrive 2 (PBO2), which introduces adaptive undervolting for Ryzen processors, amongst other improvements.

Find out what Precision Boost Overdrive 2 offers, and what effect it has on performance of AMD Ryzen 5000 series processors!

 

Precision Boost Overdrive 2 : Improved Overclocking + Adaptive Undervolting!

Precision Boost Overdrive is the dynamic overclocking capability built into AMD Ryzen processors, with these features :

  • raises socket power limit
  • raises motherboard VRM limit
  • can override max boost frequency
  • primarily designed to improve multi-thread performance

Precision Boost Overdrive 2 (PBO2) comes with all of the original PBO features, as well as these new features :.

  • improved single-thread performance
  • adds adaptive undervolting capability
  • adds standalone undervolting capability for AMD Ryzen 5000 Series processors
  • improved performance per watt.
  • can lower operating temperatures.
  • ability to customise voltages and frequencies to a specific processor.

 

Precision Boost Overdrive 2 : Curve Optimizer

The new adaptive undervolting capability in Precision Boost Overdrive 2 (PBO2) is achieved through the Curve Optimizer.

The new Curve Optimizer tells the firmware what window of undervolting is desired, and adapts to both light and heavy workloads.

Undervolting, however, will not be done with raw millivolts. Instead, it will be applied with “counts“.

  • 1 Count is approximately 3 mV ~ 5 mV. This range means less undervolting during high loads, and more undervolting when the load is lower.
  • You can adjust the voltage by +/- 30 Counts. 30 Counts would be a range of 90~150 mV.
  • The adjustments can be applied per core, or on all cores.

The Curve Optimizer algorithm is adaptive – so it will automatically underplot when the circumstances are suitable, and restore voltage when it needs to.

 

Precision Boost Overdrive 2 : Performance Boost

According to AMD, PBO2 offers better single-threaded performance than the original PBO.

Enabling it will let the Ryzen 7 5800X deliver 2.6% better 1T performance in CINEBENCH R20, while the Ryzen 9 5900X delivers 2% better 1T performance.

Hardly earth-shattering performance, but hey, it’s free!

Multi-threaded performance also gets a small but nice boost from PBO2.

AMD says the Ryzen 7 5800X should get a 2.2% boost, while the Ryzen 9 5900X gets a very nice 10% boost in multi-threaded performance!

 

Precision Boost Overdrive 2 : System Requirements

Here are the system requirements for Precision Boost Overdrive 2 to work :

  • AMD Ryzen 5000 Series desktop processor
  • AMD 500 or 400 Series motherboard
  • Motherboard BIOS based on AMD AGESA 1.1.8.0 or later

The PBO2 update is available for all AMD Ryzen 5000 Series desktop processors through a motherboard BIOS upgrade available in December 2020.

 

AMD Ryzen 5000 Series Processors : Where To Buy?

AMD Ryzen 9 5950X

AMD Ryzen 9 5900X

AMD Ryzen 7 5800X

AMD Ryzen 5 5600X

 

Recommended Reading

Go Back To > Computer | Home

 

Support Tech ARP!

If you like our work, you can help support us by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


AMD RDNA 2 Architecture : Tech Highlights!

The success of the AMD Radeon RX 6000 series graphics cards is entirely due to the new RDNA 2 architecture.

Take a look at what’s new in the AMD RDNA 2 architecture!

 

AMD RDNA 2 Architecture : Tech Highlights!

The AMD Radeon RX 6000 series graphics cards are built on the new AMD RDNA 2 architecture, which features an enhanced Compute Unit, a new visual pipeline with Ray Accelerators, and the new AMD Infinity Cache.

AMD Infinity Cache

The AMD Infinity Cache is a new and very large 128 MB data cache. Think of it as an L3 cache for the GPU.

AMD added it to dramatically increase memory bandwidth, which reducing memory latency and power consumption.

They claim it delivers up to 3.25X the bandwidth of the 256-bit GDDR6 memory, and up to 2.4X more effective bandwidth per watt.

Recommended :AMD Infinity Cache Explained : L3 Cache Comes To The GPU!

New Ray Accelerator

RDNA 2 introduces a new Ray Accelerator – one for each Compute Unit.

The Ray Accelerator is a fixed-function ray tracing acceleration engine to deliver real-time lighting, shadow and reflection realism through DirectX Raytracing (DXR).

It will calculate the intersections of the rays with the scene geometry as represented in a Bounding Volume Hierarchy, sort them, and return the information to the shaders for further scene traversal or result shading.

Each Ray Accelerator can calculate up to 4 rays per box intersections or 1 ray per triangle intersection per clock cycle.

Variable Rate Shading

Variable rate shading allows the GPU to better use its limited processing capability by focusing on the most important parts of the frame.

AMD RDNA 2 has variable rate shading built throughout the entire pixel pipeline, with 1 x 1, 2 x 1, 1 x 2 and 2x 2 shading rates.

RDNA 2 also allows for very fine granularity – a different shading rate can to be selected for every 8 x 8 pixels region.

Hardware Decoding + Encoding

RDNA. 2 also introduces support for hardware 8K and AV1 decoding, allowing you to stream and watch video with virtually no performance impact.

Codecs Decode Encode
VP9 4K @ 90 fps
8K @ 24 fps
H.264 1080p @ 600 fps
4K @ 150 fps
1080p @ 360 fps
4K @ 90 fps
H.265 1080p @ 360 fps
4K @ 90 fps
8K @ 24 fps
1080p @ 360 fps
4K @ 60 fps
AV1 8K @ 30 fps

 

Recommended Reading

Go Back To > Computer | GamingHome

 

Support Tech ARP!

If you like our work, you can help support us by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


AMD CDNA Architecture : Tech Highlights!

In addition to the gaming-centric RDNA architecture, AMD just introduced a new CDNA architecture that is optimised for compute workloads.

Here are some key tech highlights of the new AMD CDNA architecture!

 

AMD CDNA Architecture : What Is It?

Unlike the fixed-function graphics accelerators of the past, GPUs are now fully-programmable accelerators using what’s called the GPGPU (General Purpose GPU) Architecture.

GPGPU allowed the industry to leverage their tremendous processing power for machine learning and scientific computing purposes.

Instead of continuing down the GPGPU path, AMD has decided to introduce two architectures :

  • AMD RDNA : optimised for gaming to maximise frames per second
  • AMD CDNA : optimised for compute workloads to maximise FLOPS per second.

Designed to accelerate compute workloads, AMD CDNA augments scalar and vector processing with new Matrix Core Engines, and adds Infinity Fabric technology for scale-up capability.

This allows the first CDNA-based accelerator – AMD Instinct MI100 – to break the 10 TFLOPS per second (FP64) barrier.

The GPU is connected to its host processor using a PCI Express 4.0 interface, that delivers up to 32 GB/s of bandwidth in both directions.

 

AMD CDNA Architecture : Compute Units

The command processor and scheduling logic receives API-level commands and translates them into compute tasks.

These compute tasks are implemented as compute arrays and managed by the four Asynchronous Compute Engines (ACE), which maintain their independent stream of commands to the compute units.

Its 120 compute units (CUs) are derived from the earlier GCN architecture, and organised into four compute engines that execute wavefronts that contain 64 work-items.

The CUs are, however, enhanced with new Matrix Core Engines, that are optimised for matrix data processing.

Here is the block diagram of the AMD Instinct MI100 accelerator, showing how its main blocks are all tied together with the on-die Infinity Fabric.

Unlike the RDNA architecture, CDNA removes all of the fixed-function graphics hardware for tasks like rasterisation, tessellation, graphics caches, blending and even the display engine.

CDNA retains the dedicated logic for HEVC, H.264 and VP9 decoding that is sometimes used for compute workloads that operate on multimedia data.

The new Matrix Core Engines add a new family of wavefront-level instructions – the Matrix Fused Multiply-Add (MFMA). The MFMA instructions perform mixed-precision arithmetic and operates on KxN matrices using four different types of input data :

  • INT8 – 8-bit integers
  • FP16 – 16-bit half-precision
  • bf16 – 16-bit brain FP
  • FP32 – 32-bit single-precision

The new Matrix Core Engines has several advantages over the traditional vector pipelines in GCN :

  • the execution unit reduces the number of register file reads, since many input values are reused in a matrix multiplication
  • narrower datatypes create opportunity for workloads that do not require full FP32 precision, e.g. machine learning – saving energy.

 

AMD CDNA Architecture : L2 Cache + Memory

Most scientific and machine learning data sets are gigabytes or even terabytes in size. Therefore L2 cache and memory performance is critical.

In CDNA, the L2 cache is shared across the entire chip, and physically partitioned into multiple slices.

The MI100, specifically, has an 8 MB cache that is 16-way set-associative and made up of 32 slices. Each slice can sustain 128 bytes for an aggregate bandwidth of over 6 TB/s across the GPU.

The CDNA memory controller can drive 4- or 8-stacks high of HBM2 memory at 2.4 GT/s for a maximum throughput of 1.23 TB/s.

The memory contents are also protected by hardware ECC.

 

AMD CDNA Architecture : Communication + Scaling

CDNA is also designed for scaling up, using the high-speed Infinity Fabric technology to connect multiple GPUs.

AMD Infinity Fabric links are 16-bits wide, and operate at 23 GT/s, with three links in CDNA to allow for full connectivity in a quad-GPU configuration.

While the last generation Radeon Instinct MI50 GPU only uses a ring topology, the new fully-connected Infinity Fabric topology boosts performance for common communication patterns like all-reduce and scatter / gather.

Unlike PCI Express, Infinity Fabric links support coherent GPU memory, which lets multiple GPUs share an address space and tightly work on a single task.

 

Recommended Reading

Go Back To > Enterprise ITComputer HardwareHome

 

Support Tech ARP!

If you like our work, you can help support us by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


AMD Zen 3 Tech Report : What’s New, What’s Unchanged?

Take a look at what’s new in the AMD Zen 3 microarchitecture, and what they borrowed from the last-gen Zen 2 microarchitecture!

 

AMD Zen 3 Architecture

Codename Vermeer, Zen 3 is the next evolution of the AMD Zen architecture.

While it obviously borrowed considerably from the existing Zen 2 architecture, AMD says it is a ground-up redesign that with major performance and functionality improvements in every area of the core.

These architectural changes allow Zen 3-based processors, like the Ryzen 5000 series, to deliver a 19% improvement in instructions per clock (IPC).

Front-End Enhancements

  • Faster fetching, especially for branchy and large-footprint code
  • L1 branch target buffer doubled in size to 1024 entries for better prediction latency
  • Improved branch predictor bandwidth
  • Faster recovery from misprediction
  • “No bubble” prediction capabilities to make back-to-back predictions more quickly and better handle branchy code
  • Faster sequencing of op-cache fetches
  • Finer granularity in switching of op-cache pipes

Execution Engines

  • Reduce latency and enlarge structures to extract higher instruction-level parallelism (ILP)
  • New dedicated branch and st-data pickers for integer, now at 10 issues per cycle (+3 vs. Zen 2)
  • Larger integer window at +32 vs. Zen 2
  • Reduced latency for select float and int operations
  • Floating point has increased bandwidth by +2 for a total of 6-wide dispatch and issue
  • Floating point FMAC is now 1 cycle faster

Load Store

  • Larger structures and better prefetching to support the enhanced execution engine bandwidth
  • Overall higher bandwidth to feed the appetite of the larger/faster execution resources
  • Higher load bandwidth vs. Zen 2 by +1
  • Higher store bandwidth vs. Zen 2 by +1
  • More flexibility in load/store operations
  • Improved memory dependence detection
  • +4 table walkers in the TLB

SOC Architecture

  • Reduce dependency on main memory accesses, reduce core-to-core latency, reduce core-to-cache latency.
  • Unify all cores in a CCD into a single unified complex consisting of 4, 6, or 8 contiguous cores
  • Unify all L3 cache in a CCD into a single contiguous element of up to 32 MB
  • Rearchitect core/cache communication into a ring system

 

AMD Zen 3 SoC Design

In addition to micro architectural improvements, Zen 3 (Vermeer) also features SoC design changes.

In Zen 2, each CCD (Compute Die) is made up of two CCX (core complexes), each with a 16 MB L3 cache.

Zen 3 uses a unified complex, in which each CCD now contains a single CCX with a unified 32 MB L3 cache.

This unified CCD design completely eliminates CCX-to-CCX communication, greatly improving core-to-core latency.

On the other hand, AMD reused the chiplet design, with one or two CCDs (fabricated on 7 nm) paired with a 12 nm IOD (I/O Die).

Reads from CCD to IO are still 2X write, to conserve die area and transistor budget. And it uses the same IOD from Matisse (Zen 2).

 

AMD Zen 3 Transistor Count + Die Size

The new Zen 3 CCD has 4.15 billion transistors, with a die size of 80.7 mm². That’s up from the 3.8 billion transistors and a die size of 74 mm² for the Zen 2 CCD.

The Matisse-era IOD remains the same – 2.09 billion transistors, with a die size of 125 mm².

They will both be manufactured using the same 7 nm TSMC process for CCD, and 12 nm Global Foundries process for IOD.

Core Die Zen 3 Zen 2
Process 7 nm TSMC
Transistors 4.15 billion 3.8 billion
Die Size 80.7 mm² 74 mm²
I/O Die Zen 3 Zen 2
Process 12 nm GoFlo
Transistors 2.09 billion
Die Size 125 mm²

 

AMD Zen 3 Precision Boost : No Change

Precision Boost 2 in Zen 3 remains the same as that of Zen 2, just with higher frequencies to “play with”.

It is an opportunistic boost algorithm that drives the loaded cores to the highest possible frequency, until it hits any one of these limits :

  • socket power
  • VRM thermal limit
  • VRM current limit
  • maximum clock speed

Precision Boost 2 will dynamically analyse and boost or dither the core clock speeds every 1 ms using the Infinity Fabric command and control functions.

In this example of the new Ryzen 9 5900X with a base clock of 3.7 GHz and a boost clock of 4.8 GHz, Precision Boost 2 will typically result in frequencies of 4.4 GHz to 4.6 GHz, even with 24 threads running at the same time.

 

AMD Zen 3 Voltage Range : No Change

Despite the changes in the microarchitecture and SoC design, Zen 3 processors will continue to be engineered with the same voltage range as Zen- and Zen 2-based processors.

The typical voltage range will vary according to usage, but basically, Zen 3 processors, like the Ryzen 5000 series, will support 0.2 V to 1.5 V.

 

AMD Zen 3 Temperature Range : No Change

AMD Zen 3-based processors will also have the same temperature ranges as Zen- and Zen 2-based processors.

Note : The temperature range below assumes an enclosed chassis, and an air-conditioned room.

 

AMD Zen 3 Memory Overclocking : No Change

Only the Zen 3 core chiplet die (CCD) is new in the Ryzen 5000 series processors. They continue to use the same IO die (IOD) as the 3rd Gen Ryzen processors.

Therefore, the relationship between Infinity Fabric Clock (fclk), Memory Controller Clock (uclk) and Memory Clock (mclk) remains the same.

For optimal performance, they are synchronous – in a 1:1:1 ratio. But users can choose a 1:1:2 ratio if they have trouble overclocking memory.

 

Recommended Reading

Go Back To > Computer HardwareHome

 

Support Tech ARP!

If you like our work, you can help support us by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


AMD Radeon RX 5500 XT Tech Report + Video Briefing!

AMD just announced the Radeon RX 5500 XT graphics card – an AIB-only model targeted at 1080p and esports gamers!

Here is our quick primer on the two AMD Radeon RX 5500 XT variants, including their reference specifications and suggested retail prices!

 

AMD Radeon RX 5500 XT : The Official Tech Briefing

Let’s start with the official tech briefing by Mithun Chandrasekhar, AMD Senior Product Manager and Brian Kong, AMD Product Manager at Radeon Technologies Group.

 

AMD Radeon RX 5500 XT : A Quick Primer

The AMD Radeon RX 5500 XT is basically the desktop version of the Radeon RX 5500M, using the same GPU with 6.4 billion transistors fabricated on the 7 nm process technology.

This Navi GPU has 22 Compute Units with 1408 stream processors and a 128-bit memory interface. The only difference is the Radeon RX 5500 XT has higher clock speeds and TDP, with up to 8 GB of GDDR6 memory.

These cards are designed to deliver 1080p gaming at 60+ fps for AAA titles, and 90+ fps for esport games.

To reduce costs, AMD opted to use a PCI Express 4.0 x8 interface, instead of the PCIe 4.0 x16 interface. But if paired with a PCIe 4.0 chipset, it will deliver the same bandwidth as PCIe 3.0 x16.

Recommended : PowerColor Radeon RX 5500 XT (Red Dragon) Review!

 

AMD Radeon RX 5500 XT Specifications

In this table, we compare its specifications against its more powerful brethren – the RX 5700 and RX 5700 XT.

Please note that this is an AIB-only model, and its specifications are only for reference.

Specifications Radeon RX
5700 XT
Radeon RX
5700
Radeon RX
5500 XT
Transistors 10.3 billion 10.3 billion 6.4 billion
Fab Process 7 nm 7 nm 7 nm
Die Size 251 mm² 251 mm² 158 mm²
Total Graphics Power 225 W 185 W 130 W
Stream Processors 2560 2340 1408
Game Clock 1605 MHz 1465 MHz 1717 MHz
Boost Clock 1905 MHz 1725 MHz 1845 MHz
TMUs 160 144 88
Max. Texture Rate 304.8 GT/s 248.4 GT/s 162.4 GT/s
ROPs 64 64 32
Max. Pixel Rate 121.9 GP/s 110.4 GP/s 59.0 GP/s
Memory Size 8 GB 8 GB 4 GB / 8 GB
Memory Type GDDR6 GDDR6 GDDR6
Bus Width 256-bit 256-bit 128-bit
Bandwidth 448 GB/s 448 GB/s 224 GB/s
PCIe Interface PCIe 4.0 x16 PCIe 4.0 x16 PCIe 4.0 x8

 

AMD Radeon RX 5500 XT Performance Comparison

AMD shared these performance results comparing the 4 GB variant against the GeForce GTX 1650 Super (4 GB) in AAA games at 1080p.

Based on their tests, the RX 5500 XT delivers anywhere from 4% to 30% better performance than the GTX 1650 Super.

 

AMD Radeon RX 5500 XT : 4GB vs 8GB

AMD’s AIB partners will be offering their cards in two variants – with either 4 GB or 8 GB of GDDR6 memory.

While this may seem to be a minor difference, AMD says that opting for the 8 GB Variant can offer a significant performance boost, with high quality settings.

 

AMD Radeon RX 5500 XT : Price + Bundle

As we mentioned earlier, this is exclusively an AIB-only model. There will be no reference model, so specifications and prices will vary significantly.

However, AMD shared with us their Suggested Retail Price (SEP) for both variants :

  • AMD Radeon RX 5500 XT 4GB : US$169 / ~£129 / ~A$245 / ~S$229 / ~RM704
  • AMD Radeon RX 5500 XT 8GB : US$199 / ~£152 / ~A$288 / ~S$270 / ~RM829

Instead of bundling every card with either Borderlands 3 or Tom Clancy’s Ghost Recon Breakpoint as announced earlier, AMD has decided to bundle every card with :

  • Iceborne Master Edition
  • 3 months Xbox Game Pass subscription

While a 3 month subscription is not particularly exciting, it’s still nice to pick up a free game with your purchase, especially at this price point!

Recommended : AMD Radeon Software Adrenalin 2020 Edition Tech Report!

 

AMD Radeon RX 5500 XT Slides

Here is the complete set of the official AMD presentation slides for your perusal.

 

Recommended Reading

Go Back To > Computer | Gaming | Home

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


The Exclusive NVIDIA DLSS Tech Briefing + Demos

One of the new features that the NVIDIA Turing architecture introduces is NVIDIA DLSSDeep Learning Super-Sampling.

With help from Jeff Yen and John Gillooly, we will explain what DLSS does and why it will be a game-changer for the GeForce RTX family of graphics cards.

2018-10-06 : Added the DLSS vs TAA demos and Q&A session with John Gillooly, and additional slides and NVIDIA DLSS details.

2018-09-27 : Originally posted as NVIDIA DLSS (Deep Learning Super-Sampling) Explained!

 

What Is NVIDIA DLSS?

DLSS is short for Deep Learning Super-Sampling. NVIDIA DLSS is a proprietary technology that uses deep learning to deliver better image quality at a lower cost than traditional AA techniques like TAA (Temporal Anti-Aliasing).

It is a new feature that leverages the Inference capability of the Tensor Cores of the new NVIDIA GeForce RTX graphics cards.

Where To Purchase The GeForce RTX?

Here are some GeForce RTX 2080 Ti and GeForce RTX 2080 purchase links in Malaysia :

  • GIGABYTE GeForce RTX 2080 Ti WindForce OC 11G : RM 5,699
  • GIGABYTE GeForce RTX 2080 Ti Gaming OC 11G : RM 5,799
  • ZOTAC GAMING GeForce RTX 2080 Ti Triple Fan : RM 5,379
  • ZOTAC GAMING GeForce RTX 2080 Ti AMP : RM 5,639
  • GIGABYTE GeForce RTX 2080 WindForce OC 8G : RM 3,899
  • GIGABYTE GeForce RTX 2080 Gaming OC 8G : RM 3,999
  • ZOTAC GAMING GeForce RTX 2080 Blower : RM 3,699
  • ZOTAC GAMING GeForce RTX 2080 Twin Fan : RM 3,719
  • ZOTAC GAMING GeForce RTX 2080 AMP : RM 3,859

Here are some GeForce RTX 2080 Ti and GeForce RTX 2080 purchase links in the US :

  • EVGA GeForce RTX 2080 Ti FTW3 DT Gaming 11GB : $819.99
  • NVIDIA GeForce RTX 2080 Founders Edition : $954.00
  • ASUS GeForce RTX 2080 08G Dual Fan OC Edition : $948.99
  • EVGA GeForce RTX 2080 XC Ultra Gaming 8G : $949.00
  • GIGABYTE GeForce RTX 2080 Gaming OC 8GB : $919.99
  • GIGABYTE GeForce RTX 2080 WindForce OC 8GB : $789.99
  • ZOTAC GAMING GeForce RTX 2080 AMP 8GB : $839.99

Here are some GeForce RTX 2080 Ti and GeForce RTX 2080 purchase links in the US :

[adrotate group=”2″]
  • KFA2 GeForce RTX 2080 Ti OC Black : £1,703.85
  • ASUS GeForce RTX 2080 Dual O8G : £845.87
  • GIGABYTE GeForce RTX 2080 GAMING OC 8G : £984.61
  • GIGABYTE GeForce RTX 2080 WindForce OC 8G : £983.71
  • PNY GeForce RTX 2080 XLR8 OC Twin Fan : £916.53
  • ZOTAC Gaming GeForce RTX 2080 AMP 8GB : £645.51

 

Why NVIDIA DLSS Matters

First, let’s start with why NVIDIA DLSS matters. It basically allows GeForce RTX graphics cards to deliver better image quality that is better than, or equivalent to TAA (Temporal Anti-Aliasing) with 33% better frame rates.

NVIDIA shared these benchmark results to demonstrate how much of a performance boost you’ll get with DLSS.

Now, remember that NVIDIA is comparing DLSS to TAA. So you are NOT going to get a 33% boost in frame rate if you don’t use any form of anti-aliasing at all.

But if you enjoy gaming with anti-aliasing enabled for better image quality, NVIDIA DLSS offers image quality that is better or equivalent to TAA, with a significantly lower performance penalty.

 

NVIDIA DLSS 2X = Ultra High Quality

NVIDIA also offers a DLSS 2X mode, which is designed to deliver ultra high quality renders that are very close to that of 64X super sampling.

 

NVIDIA DLSS (Deep Learning Super-Sampling) Explained!

At an exclusive NVIDIA tech briefing in Bangkok, NVIDIA Director of Technical Marketing for APAC, Jeff Yen, gave us two briefing sessions on NVIDIA DLSS, which we combined in this video for you. It offers you a quick 11-minute primer on DLSS.

 

NVIDIA DLSS vs TAA Demos + Q&A

Next, John Gillooly compared the performance of the NVIDIA DLSS against TAA (Temporal Anti-Aliasing). We also had a Q&A session with him about DLSS.

Now, if you are still skeptical about how NVIDIA DLSS can deliver better image quality with a much lower performance hit, let’s examine how NVIDIA DLSS actually works.

Next Page > How NVIDIA DLSS Works, The NVIDIA DLSS FAQ

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


How NVIDIA DLSS Works

Deep Learning Using NVIDIA DNN

The first step in creating DLSS actually begins with NVIDIA generating thousands of “ground truth” reference images that are supersampled or “jittered” 64 times. Basically, the same image is rendered with 64 different sub-pixel offsets, to create an incredibly high quality 64xSS (64x supersampled) sample.

Then NVIDIA trains a DLSS deep neural network (DNN) that uses DGX SATURNV supercomputers to generate images equivalent to the 64xSS targets. By repeatedly generating an output and comparing it to the 64xSS target, the DLSS network learns to create images that closely approximates the quality of 64xSS.

After many iterations, the DLSS network will “know” how to generate images close to the quality of 64xSS, while avoiding the blurring, disocclusion and transparency that affects traditional anti-aliasing methods like TAA. This example of DLSS 2X shows how it avoids the blurring of the semi-transparent overlay.

After the deep neural network generates the DLSS algorithms for that game, it’s ready for use in the actual game. This is where the GeForce RTX’s Tensor Cores come in.

Tensor Cores

A key component of NVIDIA DLSS are the Tensor Cores in the GeForce RTX graphics cards. These are new processing cores designed to accelerate large matrix operations.

They allow the GeForce RTX graphics cards to deliver far superior AI inference performance than the previous Pascal CUDA cores, or a CPU for the matter.

Using the DLSS algorithms generated specifically for that particular game, the Tensor Cores can now infer how the frame can be rendered with an image quality that is equal to, or better than, TAA with about half the shading work. Note that this is using the standard DLSS mode, and not the higher-quality DLSS 2X mode.

[adrotate group=”1″]

 

The NVIDIA DLSS FAQ

Here are the answers to some common questions that we believe you may be wondering about DLSS.

  1. Does NVIDIA DLSS work on Pascal or earlier GPUs?
    No. DLSS requires the use of the new Tensor Cores, that are not available in Pascal or earlier GPUs.
  2. Does NVIDIA DLSS automatically work in all games?
    No. NVIDIA DLSS requires two components to work :
    a) the game must support DLSS – NVIDIA says it’s simple for any game developer to add it to their game
    b) the graphics driver must support DLSS for that particular game – essentially, the driver should have the pre-generated DLSS algorithms for the game

  1. How big are the DLSS algorithms?
    According to NVIDIA, the algorithms for each game are small in size – in the tens of MB.
  2. Where are the DLSS algorithms stored?
    They are stored in the NVIDIA GeForce drivers. You will need to update your driver to add algorithms for more games, as support expands.
  3. Do we need to install NVIDIA GeForce Experience?
    No, GeForce Experience is not necessary to support DLSS. You only need to install the NVIDIA GeForce driver.

  1. Are game developers charged to have their DLSS algorithms generated?
    No, NVIDIA does the DLSS training on their Deep Neural Network for free.
  2. What’s the speed bump if I don’t use anti-aliasing at all?
    DLSS does not give you a performance boost. It only reduces the performance penalty, while delivering image quality equal to, or better than, Temporal Anti-Aliasing.
  3. What’s the difference between DLSS and DLSS 2X?
    DLSS offers image quality equal to, or better than TAA, with lower performance penalty.
    DLSS 2X offers image quality close to that of 64X supersampling – the gold standard of image quality. NVIDIA does not mention the performance penalty of DLSS 2X.

Next Page > The NVIDIA DLSS Presentation Slides, Where To Buy, Recommended Reading

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


The NVIDIA DLSS Presentation Slides

Here is our curated NVIDIA DLSS presentation slides for your reference.

 

Where To Purchase The GeForce RTX?

Here are some GeForce RTX 2080 Ti and GeForce RTX 2080 purchase links in Malaysia :

  • GIGABYTE GeForce RTX 2080 Ti WindForce OC 11G : RM 5,699
  • GIGABYTE GeForce RTX 2080 Ti Gaming OC 11G : RM 5,799
  • ZOTAC GAMING GeForce RTX 2080 Ti Triple Fan : RM 5,379
  • ZOTAC GAMING GeForce RTX 2080 Ti AMP : RM 5,639
  • GIGABYTE GeForce RTX 2080 WindForce OC 8G : RM 3,899
  • GIGABYTE GeForce RTX 2080 Gaming OC 8G : RM 3,999
  • ZOTAC GAMING GeForce RTX 2080 Blower : RM 3,699
  • ZOTAC GAMING GeForce RTX 2080 Twin Fan : RM 3,719
  • ZOTAC GAMING GeForce RTX 2080 AMP : RM 3,859

Here are some GeForce RTX 2080 Ti and GeForce RTX 2080 purchase links in the US :

  • EVGA GeForce RTX 2080 Ti FTW3 DT Gaming 11GB : $819.99
  • NVIDIA GeForce RTX 2080 Founders Edition : $954.00
  • ASUS GeForce RTX 2080 08G Dual Fan OC Edition : $948.99
  • EVGA GeForce RTX 2080 XC Ultra Gaming 8G : $949.00
  • GIGABYTE GeForce RTX 2080 Gaming OC 8GB : $919.99
  • GIGABYTE GeForce RTX 2080 WindForce OC 8GB : $789.99
  • ZOTAC GAMING GeForce RTX 2080 AMP 8GB : $839.99

Here are some GeForce RTX 2080 Ti and GeForce RTX 2080 purchase links in the US :

  • KFA2 GeForce RTX 2080 Ti OC Black : £1,703.85
  • ASUS GeForce RTX 2080 Dual O8G : £845.87
  • GIGABYTE GeForce RTX 2080 GAMING OC 8G : £984.61
  • GIGABYTE GeForce RTX 2080 WindForce OC 8G : £983.71
  • PNY GeForce RTX 2080 XLR8 OC Twin Fan : £916.53
  • ZOTAC Gaming GeForce RTX 2080 AMP 8GB : £645.51

 

Recommended Reading

[adrotate group=”2″]

Go Back To > First PageComputer Hardware + Systems | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


The Official HUAWEI Kirin 980 Tech Briefing + Demos!

We were recently invited to an exclusive HUAWEI Kirin 980 tech briefing session.

Here is everything you need to know about the HUAWEI Kirin 980, with videos of the demos and its development board.

Now you can pretty much guess how the upcoming HUAWEI Mate 20 will perform before its official launch on 16 October 2018!

 

The HUAWEI Kirin 980

Although Apple was the first to launch smartphones with a 7nm mobile SoC, the HUAWEI Kirin 980 is technically the first 7nm mobile SoC in the world.

Take a look at the development board, and how small the Kirin 980 really is.

The HUAWEI Kirin 980 is also the first to feature the ARM Cortex-A76 processor core, and Mali-G76 GPU, as well as the first to come with two NPUs, a Cat.21 modem and support for faster LPDDR4X-2133 memory.

It took HUAWEI and TSMC 36 months of joint research and co-engineering to develop this mobile SoC.

The process, which started in 2015, involved over 1000 semiconductor design and process experts.

Thanks to their efforts, the 7 nm process technology delivered a 20% speed improvement, with a 40% boost in power efficiency, and a 60% greater transistor density. This allowed HUAWEI to cram 6.9 billion transistors into the Kirin 980.

Now, let’s take a closer look at what’s new in the HUAWEI Kirin 980!

[adrotate group=”1″]

 

The Kirin 980 CPU Architecture

The HUAWEI Kirin 980 pairs four new ARM Cortex-A76 high-performance cores, with four ARM Cortex-A55 cores.

However, instead of the traditional big.LITTLE 4+4 clusters, HUAWEI uses a 2+2+4 cluster architecture.

The four ARM Cortex-A55 cores, running at 1.8 GHz, are used for low loads and background tasks, while two ARM Cortex-A76 cores, running at 1.92 GHz, will be used for most loads, together with some of the A55 cores.

But when you really need a big boost in performance, the last two Cortex-A76 cores kick in.

With a higher clock speed of 2.6 GHz, they offer a big boost in performance. But to conserve battery life, they are meant to be used in short bursts.

Here is a table that illustrates the kinds of loads that will trigger each cluster :

 

New Mali-G76 GPU

The HUAWEI Kirin 980 is the world’s first mobile SoC to deploy the ARM Mali-G76 GPU.

HUAWEI did not reveal how many cores it has, but they say that it offers 46% better performance than the Mali-G72 GPU used in the Kirin 970, with 178% better power efficiency.

Go Back To > Dual NPUs + ISPs, Faster Connectivity, Dual Frequency GPS

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


Kirin 980 Has Dual NPUs

The HUAWEI Kirin 980 retains the same NPU that was first introduced in the Kirin 970.

However, HUAWEI double its AI performance by using two of them, instead of just one.

The boost in AI performance allows HUAWEI to introduce new AI-related technologies like :

  • the HUAWEI AI Loading Prediction technology, and
  • the Real-Time Multi-Person Gesture Recognition technology.

We will take a closer look at these two technologies below.

 

HUAWEI AI Loading Prediction Technology

The HUAWEI Kirin 980 manages its CPU and GPU performance using an AI Loading Prediction Technology.

It basically predicts when a game will require more power, or less; and pre-emptively boosts or reduces the CPU and GPU clock speeds accordingly. It also determines when the CPU’s turbo cluster (of two Cortex-A76 cores running at 2.6 GHz) should be activated for a boost in performance.

The combination of much faster CPU and GPU cores, with AI Loading Prediction Technology, allows the Kirin 980 to hit and sustain 60 fps even with strenuous games.

Note that while it can go beyond 60 fps in this demo of NBA 2K18, the AI Loading Prediction Technology caps its frame rate to reduce power consumption and extend battery life.

 

Real-Time Multi-Person Gesture Recognition

Thanks to its greater AI processing capabilities, the HUAWEI Kirin 980 can perform real-time multi-person gesture recognition.

It can also determine their distance using a single camera, allowing it to know when someone has gone in front of, or behind, someone else.

Most impressively, it accomplished this using a single camera. Imagine how much more accurate it can be using two cameras which would offer much better depth perception.

 

Kirin 980 Has Dual ISPs

It has two image signal processors which offer a 46% boost in performance, a 33% reduction in recording latency, and 23% better power efficiency.

The boost in performance greatly expands its ability to “massage” the raw images to produce better photos.

Although HUAWEI did not detail how they did it, their sample photos above appear to be using HDR to bring out details in the shadows while correcting for overblown highlights.

For sharp images, it will needs to be paired with a fast image sensor that can take multiple photos with different exposures in quick succession.

[adrotate group=”1″]

 

Kirin 980 Has Faster LTE + WiFi Speeds

The HUAWEI Kirin 980 boasts a new LTE Cat.21 modem, which allows it to deliver download speeds of up to 1.4 Gbps! Upload speed remains at 200 Mbps.

When paired with the HUAWEI Hi1103 chip, it can offer WiFi transfer rates of up to 1.7 Gbps.

 

Kirin 980 Introduces Dual-Frequency GPS

This is arguably its most underrated feature – support for dual-frequency GPS.

By using both L1 and L5 frequencies, it can deliver 10X better positioning accuracy. You won’t ever have to worry taking the wrong turn!

Go Back To > HUAWEI Kirin 980 Performance Comparisons + Summaries

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


How Fast Is The HUAWEI Kirin 980?

These HUAWEI performance comparisons show the performance differences between the Kirin 980 and its predecessor, the Kirin 970 and the Qualcomm Snapdragon 845.

CPU : 75% Better Performance, 58% Better Power Efficiency Than Kirin 970

CPU : 37% Better Performance, 32% Better Power Efficiency Than Snapdragon 845

Memory : 20% More Bandwidth, 22% Lower Latency Than Snapdragon 845

GPU : 46% Better Performance, 178% Better Power Efficiency Than Kirin 970

Image Recognition : 89% Faster Than Snapdragon 845, 3X Faster Than Apple A11

App Launch Time : 6% to 23% Faster Than Snapdragon 845

Game Performance : 22% Better Performance, 32% Lower Power Consumption Than Snapdragon 845

AI Performance : 2X Better Performance, 87% Better Power Efficiency Than Snapdragon 845

[adrotate group=”1″]

 

Kirin 980 vs Kirin 970 Summarised

Here is a HUAWEI comparison of the Kirin 980 against the Kirin 970.

 

Kirin 980 vs Snapdragon 845 Summarised

And this is a HUAWEI table summarising the differences between the Kirin 980 and the Snapdragon 845.

Go Back To > HUAWEI Kirin 980 Keynote Slides, Recommended Reading

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


The Complete HUAWEI Kirin 980 Keynote Slides

Here is the complete set of the official HUAWEI keynote slides for your perusal :

[adrotate group=”1″]

 

Suggested Reading

[adrotate group=”2″]

Go Back To > First PageMobile Devices | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!


AMD B450 Chipset Explained with Official AMD Tech Briefing!

We had earlier shared with you our previews of the GIGABYTE B450 AORUS  PRO WIFI motherboard, as well as the MSI B450 Tomahawk and B450A-Pro motherboards. The AMD B450 chipset powering those motherboards is officially launched today, so yes, you can purchase these motherboards starting today!

In this article, we will share with you the official AMD B450 tech briefing video, as well as the full details and slides on the AMD B450 chipset. Don’t forget to share this with your friends!

 

The Official AMD B450 Tech Briefing

Let’s start with the official AMD B450 tech briefing, conducted by Don Woligroski, AMD Business Development and Desktop Processor Marketing.

 

The AMD B450 Chipset Explained!

Needless to say, the AMD B450 chipset uses Socket AM4. That means it will support every AM4 processor from their 7th Gen APUs all the way to the highest-end 2nd Gen Ryzen processors.

Key message : Future AMD desktop processors will use Socket AM4 until 2020, so they will run on your AMD B450 motherboard with a BIOS update.

Because the new AMD B450 chipset can be used with all AMD Ryzen and 2nd Gen Ryzen processors, it can be used in everything from a compact HTPC, all the way to an enthusiast-class gaming PC.

Key message : You can build an AMD gaming PC with a B450 motherboard for as low as US$350 / €300 / RM 1,400!

This table shows you all of the AMD AM4 processor families that are fully compatible with the new AMD B450 chipset. They have “suggested chipsets” for each family, but every AM4 processor will work with the new B450 chipset as well as older AM4 chipsets.

Key message : You will be able to upgrade to multiple generations of processors with a B450 motherboard.

This comparison shows you how much performance headroom you have with the B450 chipset. You can start with a low-cost AMD Ryzen 3 2200G APU and eventually upgrade to an AMD Ryzen 7 2700X, using the same motherboard.

Key message : Upgrading your AMD processor does not require you to buy a new motherboard!

For a mid-range chipset, the AMD B450 has a number of impressive features. It allows for better memory overclocking and the AMD StoreMI Technology.

Key message : The AMD B450 chipset offers better memory overclocking and a free AMD StoreMI license.

This slide just shows how AMD B450 chipset communicates with the AMD Ryzen processor through four PCI Express 3.0 lanes.

Key message : The AMD B450 chipset offers 24 PCIe Gen 3 lanes with all AMD Ryzen and 2nd Gen Ryzen processors.

This slide reinforces AMD’s messaging that they will continue to use Socket AM4 until 2020.

Key message : The AMD B450 chipset should be able to support another 3 generations of AMD processors, with the necessary BIOS updates.

[adrotate group=”1″]

Next Page > The AMD B450 Chipset Explained Part 2

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The AMD B450 Chipset Explained Part 2

Compared to the X470 chipset, the only thing you are really giving up is CPU PCIe Bifurcation. This is only necessary if you plan to use a M.2 or PCI Express expansion card, like the ASUS Hyper M.2 X16 card we showcased recently… or get maximum performance from two graphics cards.

Key message : The AMD B450 chipset is great value!

This slide shows how AMD offers unlocked processor overclocking, while Intel does not.

Key message : If you want to overclock, get AMD!

This slide shows just how much more affordable the AMD B450 platform is, when paired with the AMD Ryzen 5 2400G APU. If you top up your budget to the same price point, you get the Ryzen 5 2600 CPU, which can process twice the number of threads simultaneously!

Key message : AMD offers better value for comparable performance.

The AMD B450 chipset comes with AMD StoreMI, which combines your solid state drive and hard disk drive with a RAM cache, to create an incredibly fast storage solution.

Key message : AMD B450 motherboards offer a big boost to storage performance.

StoreMI works by intelligently rearranging your files so that the most-frequently used blocks are store on the fastest drive. Best of all – it’s FREE with B450 motherboards!

Key message : AMD B450 motherboards offer a FREE boost in storage performance.

This is not quite a fair comparison, since Optane Memory is a dedicated solid state cache, while StoreMI uses your own SSD. It’s not like AMD is providing you with a fast cache with the B450 motherboards.

Key message : StoreMI gives you a free boost in storage performance… if you have your own SSD.

This slide summarises the price-performance advantage of the AMD B450 platform.

Key message : AMD B450 offers incredible price-performance over competing Intel platforms.

[adrotate group=”1″]

Next Page > Complete AMD B450 Presentation Slides, Recommended Reading

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The Complete AMD B450 Presentation Slides

Here is the complete set of AMD B450 presentation slides for your perusal.

 

Recommended Reading

Go Back To > First PageComputer Hardware + Systems | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

AMD Ryzen Mobile APUs – Everything You Need To Know!

AMD has officially announced the Ryzen Mobile APUs!. Formerly codenamed Raven Ridge, they combine the AMD Zen CPU cores with AMD Vega graphics to deliver new heights in CPU and graphics performance for ultra-thin notebooks. Let’s dive into the details of the new AMD Ryzen Mobile with Radeon Vega Graphics APUs!

 

AMD Ryzen Mobile “Raven Ridge” APUs

The AMD Ryzen Mobile is not just a mobile processor, it is a mobile APU (Accelerated Processing Unit). Basically, it is a combination of a traditional processor, with a GPU, using a System-on-Chip design.

The new Ryzen Mobile APU combines the new AMD Zen cores used in desktop Ryzen processors with AMD Vega cores used in the new AMD Radeon RX Vega graphics cards.

Mobile APUs are not new. Even Intel mobile processors come with integrated graphics. But AMD is still the only manufacturer to integrate “premium CPU cores” with “premium graphics cores”.

The new CPU and GPU cores used in the new Ryzen Mobile APUs allow them to deliver 3X better CPU performance, and 2.3X better GPU performance over that of equivalent 7th Gen AMD APUs. Performance isn’t everything though. The Ryzen Mobile APUs also promise to use 58% less power!

[adrotate group=”1″]

 

The AMD Ryzen Mobile Specifications

AMD has announced two Ryzen Mobile models – the AMD Ryzen 7 2700U and the AMD Ryzen 5 2500U. Here are their specifications.

SpecificationsAMD Ryzen 7 2700UAMD Ryzen 5 2500U
CPU Cores4 Cores4 Cores
Simultaneous Threads8 Threads8 Threads
L1 Cache64 KB instruction cache
32 KB data cache
64 KB instruction cache
32 KB data cache
L2 Cache512 KB x 4512 KB x 4
L3 Cache4 MB shared4 MB shared
Base Clock Speed2.2 GHz2.0 GHz
Boost Clock Speed3.8 GHz3.6 GHz
GPU Cores10 Radeon Vega CUs8 Radeon Vega CUs
GPU ClockUp to 1.3 GHzUp to 1.1 GHz
Memory SupportUp to DDR4-2400 (Dual Channel)Up to DDR4-2400 (Dual Channel)
TDP15 W Nominal
12-25 W (Configurable)
15 W Nominal
12-25 W (Configurable)

Next Page > Precision Boost 2, Mobile XFR, Synergistic Power Rail Sharing

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

Precision Boost 2

The AMD Ryzen Mobile APU introduces a new Precision Boost 2 algorithm that is part of the new AMD SenseMI technology. Instead of switching between a simple 2-core and all-core boost option, it is designed to dynamically seek the maximum possible clock speed according to the current CPU temperature, current and load.

It is also now able to change frequencies in very fine granularity of just 25 MHz. This allows for a smoother boost curve, and better boost clocks for different kinds of loads on-the-fly.

 

Mobile XFR (mXFR)

Mobile XFR (eXtended Frequency Range) is the automatic overclocking feature that boosts the processor voltage and clock speed beyond the Precision Boost clock speed. It leverages the temperature awareness of Precision Boost 2 to determine how much faster it can overclock.

Like the desktop XFR feature, it is highly dependent on the processor temperature, so a good cooling solution is necessary. Notebooks must have coolers that meet AMD’s performance criteria if they want to offer mXFR.

[adrotate group=”1″]

 

Synergistic Power Rail Sharing

The new AMD Ryzen Mobile APUs have a unified VDD power rail with digital low-dropout (LDO) regulators. This synergistic power rail sharing allows the maximum current requirements to be reduced by 36%.

They also have multiple digital LDO regions for the CPU cores, graphics core and subregions. This allows those regions to be completely turned off when they are idle, greatly saving power.

The shared voltage regulators not only reduce the total current draw, it allows the Ryzen Mobile APU to gain a higher peak CPU and GPU current to boost performance.

Next Page > Per-Core Control, Enhanced Gate States, Power Gating & Faster Gate Exit

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

Per-Core Frequency & Voltage

Thanks to its fine-grained integrated voltage regulation, the AMD Ryzen Mobile APUs can dynamically adjust the frequency and voltage for each core. As each processing thread has different compute requirements, this allows the Ryzen Mobile to push power to the cores that require more performance on-the-fly.

This capability is not restricted to the CPU cores. The AMD Ryzen Mobile APUs can trade power and current between the CPU and GPU as well!

 

Enhanced Gate States

The AMD Ryzen Mobile APUs feature enhanced gate states. The CPU cores can enter CC6 power gating individually, and the L3 cache power can be lowered when all four cores are in CC6 mode. The enhanced gate states also allow the Ryzen Mobile to power down up to 95% of the Radeon Vega graphics core.

[adrotate group=”1″]

 

Dual Region Power Gating

The SoC elements of the AMD Ryzen Mobile APU are organised into two types of region. The Type A Region can remain gated during display refresh, while the Type B Region can become briefly active. This allows Type B regions to save more power during common battery life scenarios.

 

Faster Gate Exit

AMD designed the Ryzen Mobile APU to deliver 47% to 87% faster gate exits over the previous generation.

Next Page > How Fast Are The Ryzen Mobile & Radeon Vega Graphics?

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

How Fast Is The Ryzen Mobile CPU?

AMD shared the following performance results of the Ryzen Mobile CPU.

In this comparison, the AMD Ryzen 7 2700U is 8% faster than the Intel Core i7-8550U, and 10% faster than the Core i7-7500U in the single-core test.

When all four cores were enabled, the Ryzen 7 2700U was 44% faster than the Intel Core i7-8550U, and 2.2X faster than the Core i7-7500U!

Here are some additional benchmark results of the new AMD Ryzen 7 2700U and Ryzen 5 2500U APUs against the Intel Core i7-8550U and the previous-generation AMD FX-9800P APU.

 

Desktop-Class Performance

With the greatly-increased performance of the new Ryzen 7 2700U APU, AMD points out that it is slightly (7%) faster than the Intel Core i5-7600K – a quad-core desktop processor with 6X higher TDP!

[adrotate group=”1″]

 

How Fast Is Radeon Vega Graphics?

With just ten or eight Compute Units, the Radeon Vega GPU embedded into the Ryzen Mobile has nowhere near the performance of the Radeon RX Vega 64 or Vega 56 desktop graphics cards. However, they are still much faster than the mobile GPUs that both AMD and Intel have ever introduced in a mobile processor.

In the 3DMark Time Spy DirectX 12 test, the Ryzen 7 2700U with its 10 Compute Units was able to slightly beat the NVIDIA GeForce GTX 950M discrete graphics. That makes it about 2.5X faster than Intel HD Graphics.

What this means is the Radeon Vega Graphics in the AMD Ryzen 7 2700U is fast enough to deliver playable frame rates at 1080p resolution for most games.

Next Page > Battery Life, Thinner Profile, Performance Summary & Conclusion

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

Better Battery Life

As pointed out earlier, AMD claims that the Ryzen Mobile APUs will deliver 58% lower power consumption, compared to the 7th Generation AMD APUs. Their benchmark results show battery life improvements from 15% to 104%.

 

Thinner Profile

The AMD Ryzen Mobile is designed for ultra-thin notebooks. The AMD Ryzen Mobile APU itself will have a 24% thinner Max Z-Height. That may not mean a lot, but when it comes to mobile devices, every millimetre counts.

 

Performance Summary

Here is a spider chart that neatly summarises the performance advantage of the AMD Ryzen 7 2700U over the Intel Core i7-8550U.

[adrotate group=”1″]

 

Conclusion

There is no doubt that the AMD Ryzen Mobile APUs are AMD’s most advanced mobile processor. This isn’t hyperbole. It’s a fact that is backed by what we have seen with their Ryzen 7, Ryzen 5 and Ryzen 3 desktop processors, as well as their Radeon RX Vega desktop graphics cards.

Finally, after many years of producing mobile APUs with lacklustre performance and high power consumption, AMD may finally have a thoroughbred in the Ryzen Mobile. We can’t wait to see the ultra-thin notebooks that will use these new APUs!

Go Back To > First Page | Computer Hardware & Systems | Home

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The 8th Gen Intel Core Desktop CPU Tech Report

Intel fans have been waiting A LONG TIME for the 8th Gen Intel Core desktop CPUs, codenamed Coffee Lake. Thanks to the Ryzen Effect, they promise a new level of performance unseen in recent years from Intel. For the first time ever, the Core i7, Core i5 and Core i3 CPUs will come with two additional cores!

Earlier, we shared with you leaked details of the Coffee Lake CPUs – price, specifications and availability. Now, we will share with you the Intel presentation on the 8th Gen Intel Core desktop CPUs.

 

Note On Availability

In our earlier article on the Intel Coffee Lake CPUs, we revealed that Intel would be launching them on 5 October 2017. In fact, tech publications privy to this presentation had to sign an NDA not to reveal this until 5 October 2017.

However, after Intel gave this presentation on 22 September 2017, it got leaked. So Intel basically cancelled their NDA, and here we are with this tech report.

Please note that the Coffee Lake CPUs are still only going to hit the market on 5 October 2017. It’s just the announcement that got brought up to 25 September 2017.

 

The 8th Gen Intel Core Desktop CPU

The processors being announced today are the premium desktop models. Intel had earlier released the 8th Gen Intel Core mobile CPUs in August. That was, confusingly, based on the previous-generation “Kaby Lake Refresh” microarchitecture.

As we revealed in our Intel Coffee Lake article, the new 8th Gen Intel Core desktop processors now have up to 6 processor cores that can handle up to 12 simultaneous threads. You can also see from this slide that they definitely planned to launch these processors on 5 October 2017.

[adrotate group=”1″]

 

Basically – MORE Cores!

What everyone is excited about is the fact that Intel is introducing two extra cores into each Coffee Lake CPU!

Intel Core i7 : 4 cores, 8 threads -> 6 cores, 12 threads

Intel Core i5 : 4 cores, 4 threads -> 6 cores, 6 threads

Intel Core i3 : 2 cores, 4 threads -> 4 cores, 4 threads

In addition to more cores, the 8th Gen Intel Core desktop processors have larger caches.

 

A New Chipset Too

This is no secret either – the 8th Gen Intel Core desktop processors will run on the new Intel 300 series chipset, the first of which is the Intel Z370 chipset.

Next Page > Gaming & Overclocking Performance, Specifications, Price & Availability

[adrotate group=”1″]

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

Better Gaming Performance

Intel promises that the 8th Gen Intel Core desktop CPUs will deliver significantly better performance. They paired the Core i7-8700K with the NVIDIA GeForce GTX 1080 Ti, and noted the following improvements over the Core i7-7700K :

  • 25% higher frame rate in Gears of War 4
  • 45% better performance while gaming + streaming + recording in PlayerUnknown’s Battleground
  • 32% faster 4K video editing in Adobe Premier Pro

There’s the inevitable comparison with older processors to provide a greater performance contrast. Interestingly, Intel only compared against a 3-year CPU (the Intel Core i7-4790K) this time, instead of a 5-year old CPU as they did in past years.

 

Better Overclocking

Unlike AMD whose Ryzen processors are all unlocked, Intel is still restricting the multiplier lock for all 8th Gen Intel Core desktop processors, except for their K-grade processors. But they promise better overclocking features like :

  • Per Core Overclocking
  • An increase in Memory Ratio up to 8,400 MT/s
  • Real-time memory latency control
  • Extended PLL trim controls
  • Enhanced package power delivery
  • Performance Tuning Protection Plan (optional)

[adrotate group=”1″]

 

8th Gen Intel Core Desktop CPU Specifications

Here is a table comparing the key specifications of the 8th Gen Intel Core desktop processors :

SpecificationsCore i7-8700KCore i7-8700Core i5-8600KCore i5-8400Core i3-8350KCore i3-8300Core i3-8100
Cores6666444
Threads121266444
Base Clock3.7 GHz3.2 GHz3.5 GHz2.8 GHz4.0 GHz4.0 GHz3.6 GHz
Boost Clock4.7 GHz4.6 GHz4.4 GHz3.9 GHzNoneNoneNone
L1 Cache384 KB384 KB384 KB384 KB256 KB256 KB256 KB
L2 Cache1.5 MB1.5 MB1.5 MB1.5 MB1 MB1 MB1 MB
L3 Cache12 MB12 MB9 MB9 MB8 MB6 MB6 MB
Multiplier UnlockedYesNoYesNoYesNoNo
TDP95 W65 W95 W65 W91 W65 W65 W

 

8th Gen Intel Core Price & Availability

As we revealed in our Intel Coffee Lake article earlier, Intel is only hiking the price of their Core i7 K-grade CPUs by $20 and their Core i5 K-grade CPUs by $15. All other 8th Gen Intel Core desktop CPU SKUs will maintain price parity with the 7th Gen models they replace.

8th Gen Intel Core 7th Gen Intel Core Difference
Model Launch Price Model Launch Price
Core i7-8700K $359 Core i7-7700K $339 +$20
Core i7-8700 $303 Core i7-7700 $303
Core i5-8600K $257 Core i5-7600K $242 +$15
Core i5-8400 $182 Core i5-7400 $182
Core i3-8350K $168 Core i3-7350K $168
Core i3-8100 $117 Core i3-7100 $117

While Intel announced the details of the 8th Gen Intel Core desktop CPUs on 25 September, they are only making these processors available on 5 October 2017, as originally planned.

Next Page > Summary & Full Set Of Presentation Slides

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The 8th Gen Intel Core Desktop CPUs Summarised

Basically, the 8th Gen Intel Core desktop CPUs are pretty much the 7th Gen Intel Core desktop CPUs with two additional cores, and larger caches. There are no major architectural changes, just some tweaks here and there.

That said, the additional cores and larger caches alone will give the 8th Gen Intel Core desktop CPUs a BIG boost in performance. That’s what got so many people hyped about it.

[adrotate group=”1″]

 

The Complete 8th Gen Intel Core Presentation Slides

Here is the complete set of presentation slides for the 8th Gen Intel Core desktop CPUs for your perusal :

Go Back To > First PageArticles | Home

[adrotate group=”1″]

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The Complete AMD Ryzen 7 Tech Report Rev. 2.0

You read our AMD Ryzen Tech Briefing, and saw how it matched and even beat the performance of the Intel Core i7-6900K processor, with lower power consumption! Well, the AMD Ryzen is here, and the first AMD Ryzen family will be known as AMD Ryzen 7.

In this article, we will look at the key specifications, performance and value proposition of the new AMD Ryzen 7 processors – the Ryzen 7 1800X, the Ryzen 7 1700X and the Ryzen 7 1700.

Originally Posted @ 2017-02-23

Updated @ 2017-05-06 : Added a new “The Balance Of Power Has Shifted” section, and updated other sections of the article.

 

Introducing The AMD Ryzen 7

It has been a long journey for AMD. As Dr. Lisa Su shared at the 2016 AMD Tech Summit, they spent 4 years and over two million engineering hours on the AMD Zen program. When they started AMD Zen with a clean slate, they set an ambitious goal of achieving 40% more instructions per clock over the previous generation Excavator core.

With AMD Ryzen, they not only achieved that goal, they actually beat it by a significant percentage. Instead of a 40% performance advantage over Excavator, the final AMD Zen silicon delivered more than 52% more instructions per clock.

If you read our AMD Ryzen Tech Briefing, you already know why AMD chose the name Ryzen, and that it is an 8-core, 16 thread processor. But now, the final AMD Ryzen silicon is revealed, together with the first Ryzen family of processors – the AMD Ryzen 7.

The new AMD Ryzen 7 family of processors consists of three models – the Ryzen 7 1800X, the Ryzen 7 1700X and the Ryzen 7 1700. Let’s take a closer look at the three AMD Ryzen 7 CPUs.

[adrotate banner=”5″]

 

The Ryzen 7 1800X Processor

At the 2016 AMD Tech Summit, AMD showed off an AMD Ryzen running at 3.4 GHz without boost, and promised that the final AMD Ryzen CPU would do better than that. They delivered on that with the Ryzen 7 1800X. This top-of-the-line 95W Ryzen CPU boasts a base clock speed of 3.6 GHz, with a boost clock of 4.0 GHz.

In the Cinebench R15 benchmark, the AMD Ryzen 7 1800X actually beat the Intel Core i7-6900K by 9% in the multi-threaded test. This dovetails with the recent AMD Ryzen performance demo. This may not seem like much, but we should point out that the same AMD Ryzen demo showed that the 3.4 GHz Ryzen consumed less power than the Core i7-6900K.

But we think the AMD Ryzen 7 1800X‘s main advantage is its value proposition. It now offers the same performance of the Intel Core i7-6900K processor at less than half the price!

Next Page > The Ryzen 7 1700X & 1700 Processors, Market Adoption

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The Ryzen 7 1700X Processor

Right below the Ryzen 7 1800X is the AMD Ryzen 7 1700X. This is a 95 W processor with a 3.4 GHz base clock, and a 3.8 GHz boost clock. According to AMD’s tests, it is just a tad slower than the Intel Core i7-6900K, and 39% faster than the Intel Core i7-6800K.

The performance pwnage is not surprising since the Intel Core i7-6800K is a quad-core processor, while the Ryzen 7 1700X is an octa-core processor. So what the Ryzen 7 1700X offers you is twice the number of cores and threads with a large boost in performance for slightly less than what you pay for an Intel Core i7-6800K processor.

That is a really enticing value proposition!

 

The Balance Of Power Has Shifted With The Ryzen 7 1700X New!

AMD compared the performance of the Ryzen 7 1700X to the Intel Core i7-7700K in various PC usage scenarios, including 4K gaming, game streaming, content creation and video encoding. Here were their key findings :

  • Compared to the Ryzen 7 1700X, the Core i7-7700K achieved 67% the video encoding performance in Handbrake and 82% the performance in Adobe Premiere CC.
  • In simultaneous game streaming via OBS, the Ryzen 7 1700X dropped less than 1% of frames, while the Core i7-7700K dropped 18% of frames.
  • Compared to the Ryzen 7 1700X, the Core i7-7700K achieved 65% the performance in POVRay, 69% the performance in Blender, and 63% the performance in Cinebench, for an average content creation performance of 66% relative to the Ryzen 7 1700X.

They even created this radar chart (also known as a spider chart) of their findings.

 

The Ryzen 7 1700 Processor

If you want even better power efficiency, the AMD Ryzen 7 1700 might be the answer. With a TDP of only 65 W, the Ryzen 7 1700 is an octa-core processor with a base clock of 3.0 GHz, and a boost clock of 3.7 GHz. According to AMD, the Ryzen 7 1700 is 46% faster than the new Intel Core i7-7700K Kaby Lake processor!

Again, the performance pwnage is due to the fact the Intel Core i7-7700K Kaby Lake processor is a quad-core processor, while the Ryzen 7 1700 is an octa-core processor. So you get twice as many cores and threads, with a large boost in performance, for a slightly lower price. Not bad at all.

Okay, that was arguably the understatement of the year, especially since it comes with a free Wraith Spire LED cooler! 😀

[adrotate banner=”5″]

 

Market Adoption

One of the worries about the success of the AMD Ryzen is support from motherboard manufacturers. AMD has declared that there will be over 82 motherboard options when the AMD Ryzen 7 series hits the market. We will show you the current AMD motherboards that are already available for pre-order with the AMD Ryzen 7 processors in an upcoming article.

Right now, you can pre-order AMD Ryzen desktops from 19 smaller PC builders and 185 retailers and boutique OEMs. AMD expects another 200 expected to offer similar desktops in Q1, 2017. The major brands though will only start offering AMD Ryzen desktops later in the first half of this year.

Next Page > Prices & Availability, Other Ryzen Articles, The Presentation Slides

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

Prices & Availability

The AMD Ryzen 7 CPUs will hit the retail shelves on March 2, 2017. As you saw earlier, the Ryzen 7 1800X will be priced at US$ 499, while the Ryzen 7 1700X will cost US$ 399 and the Ryzen 7 1700 priced at US$ 329. For worldwide pricing, please take a look at our article – The AMD Ryzen 7 Prices & Availability Revealed!

However, you can now pre-order your AMD Ryzen 7 CPU and systems from 185 retailers, online stores and boutique OEMs right now! Here are direct links to the AMD Ryzen 7 CPU and bundles on sale at Amazon :

[adrotate banner=”4″]

 

Other AMD Ryzen Articles

Don’t forget to also read our other AMD Ryzen-related articles :

 

The Complete AMD Ryzen 7 Presentation Slides

For those of you who want to peruse Dr. Lisa Su’s full presentation slides for the AMD Ryzen 7 processors, here they are :

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

The AMD Ryzen 5 Processor Tech Report

We have all been blown away by the raw processing power of the AMD Ryzen 7 processors. Even its gaming performance proved to be quite competitive against the Intel Core i7 processors. But as exciting as the Ryzen 7 processors are, they may be out-of-reach for many people. This is where the AMD Ryzen 5 processors come in.

 

A Bigger Slice Of The Pie

AMD may have priced the Ryzen 7 CPUs very competitively against the Intel Core i7 processors, but the sub-$300 market is twice as large. So they are quickly introducing the AMD Ryzen 5 processors to try and win a bigger slice of the pie.

 

The AMD Ryzen 5 Key Features

The AMD Ryzen 5 processors will feature the same AMD Zen features as the Ryzen 7 CPUs. This includes the AMD SenseMI technology that we first revealed in The Complete AMD Ryzen “Summit Ridge” Tech Briefing.

Here is a primer on AMD SenseMI by Mark Papermaster, Senior Vice President & CTO of AMD.

The AMD Ryzen 5 processors will also be multiplier-unlocked, allowing for easier overclocking. In fact, AMD has announced that all Ryzen processors will be multiplier-unlocked. Awesome!

The AMD Ryzen 5 processors will continue to support two threads per core. The only real difference is the number of processor cores they offer.

The first slew of AMD Ryzen 5 processors will offer either 6 processor cores, or 4 processor cores :

  • the Ryzen 5 1600X – 6 cores, 12 threads, 3.6 GHz core clock, 4.0 GHz boost clock
  • the Ryzen 5 1600 – 6 cores, 12 threads, 3.2 GHz core clock, 3.6 GHz boost clock
  • the Ryzen 5 1500X – 4 cores, 8 threads, 3.5 GHz core  clock, 3.7 GHz boost clock
  • the Ryzen 5 1400 – 4 cores, 8 threads, 3.2 GHz core clock, 3.6 GHz boost clock

Next Page > How Fast Is The Ryzen 5?, The AMD AM4 Platform, Bundled Coolers”

[adrotate banner=”5″]

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

How Fast Is The AMD Ryzen 5 Processor?

Until the NDA expires on launch day (11 April 2017), we won’t have any hard numbers. But AMD shared a single Cinebench R15 benchmarking result with us. They compared their flagship AMD Ryzen 5 processor – the Ryzen 5 1600X, against the Intel Core i5-7600K.

The AMD Ryzen 5 1600X is a 6-core, 12-thread processor, while the Intel Core i5-7600K is only a 4-core, 4-thread processor. Even though the Intel processor has higher clock speeds (4 GHz core, 4.2 GHz boost), the Ryzen 5 processor has twice as many processor cores.

Hence, it is hardly surprising to see the the Ryzen 5 1600X beat the Core i5-7600K by 69%. The AMD Ryzen 5 1600X scored 1195, while the Intel Core i5-7600K only scored 655, in this test.

This is obviously a grossly unfair comparison… until you consider the very likely possibility that AMD intends to price the Ryzen 5 1600X against the Core i5-7600K. In fact, considering how they priced the Ryzen 7 processors, the Ryzen 5 1600X will likely be cheaper than the Core i5-7600K.

 

The AMD AM4 Platform

Like the Ryzen 7 CPUs, the AMD Ryzen 5 CPUs can be paired with any Socket AM4 motherboard. This allows for easier and cheaper upgrades if you already have an existing Socket AM4 motherboard.

However, if you wish to overclock, and take advantage of the AMD Ryzen’s unlocked multiplier, you need to use a motherboard built around the AMD X370, X300 or B350 chipset. And only the X370 and X300 chipsets allows for dual PCI Express graphics cards.

For more information on the new AMD Ryzen motherboards, please read our article – The 34 New AMD Ryzen 7 Motherboards Revealed

[adrotate banner=”5″]

 

AMD Ryzen 5 Bundled Coolers

The AMD Ryzen 5 1600X does not come with a bundled cooler, but the other Ryzen 5 models will come with a bundled AMD Wraith cooler. They are quieter, feature a new spring-screw clamping system and RGB LED lighting.

For more information on these AMD Wraith coolers, and other Ryzen-compatible coolers, check out our article – The 15 New AMD Ryzen 7 CPU Coolers Revealed.

Next Page > Price & Availability, Presentation Slides, Other Ryzen Articles

 

Support Tech ARP!

If you like our work, you can help support our work by visiting our sponsors, participating in the Tech ARP Forums, or even donating to our fund. Any help you can render is greatly appreciated!

AMD Ryzen 5 Price & Availability

All four AMD Ryzen 5 models will be available worldwide on 11 April 2017, at the following prices :

If you are interested in their worldwide prices, please check out our article – The Worldwide AMD Ryzen 5 Price & Availability.

And if you are waiting for the Ryzen 3 processors, they are coming soon… in the second half of 2017.

[adrotate banner=”5″]

 

The AMD Ryzen 5 Presentation Slides

If you are interested in checking out the full set of AMD Ryzen 5 presentation slides, here they are.