Author:
Manufacturer: Intel

System Overview

Announced at Intel's Developer Forum in 2012, and launched later that year, the Next Unit of Computing (NUC) project was initially a bit confusing to the enthusiast PC press. In a market that appeared to be discarding traditional desktops in favor of notebooks, it seemed a bit odd to launch a product that still depended on a monitor, mouse, and keyboard, yet didn't provide any more computing power.

Despite this criticism, the NUC lineup has rapidly expanded over the years, seeing success in areas such as digital signage and enterprise environments. However, the enthusiast PC market has mostly eluded the lure of the NUC.

Intel's Skylake-based Skull Canyon NUC was the company's first attempt to cater to the enthusiast market, with a slight stray from the traditional 4-in x 4-in form factor and the adoption of their best-ever integrated graphics solution in the Iris Pro. Additionally, the ability to connect external GPUs via Thunderbolt 3 meant Skull Canyon offered more of a focus on high-end PC graphics. 

However, Skull Canyon mostly fell on deaf ears among hardcore PC users, and it seemed that Intel lacked the proper solution to make a "gaming-focused" NUC device—until now.

8th Gen Intel Core processor.jpg

Announced at CES 2018, the lengthily named 8th Gen Intel® Core™ processors With Radeon™ RX Vega M Graphics (henceforth referred to as the code name, Kaby Lake-G) marks a new direction for Intel. By partnering with one of the leaders in high-end PC graphics, AMD, Intel can now pair their processors with graphics capable of playing modern games at high resolutions and frame rates.

DSC04773.JPG

The first product to launch using the new Kaby Lake-G family of processors is Intel's own NUC, the NUC8i7HVK (Hades Canyon). Will the marriage of Intel and AMD finally provide a NUC capable of at least moderate gaming? Let's dig a bit deeper and find out.

Click here to continue reading our review of the Intel Hades Canyon NUC!

Manufacturer: Microsoft

O Rayly? Ya Rayly. No Ray!

Microsoft has just announced a raytracing extension to DirectX 12, called DirectX Raytracing (DXR), at the 2018 Game Developer's Conference in San Francisco.

microsoft-2015-directx12-logo.jpg

The goal is not to completely replace rasterization… at least not yet. This effect will be mostly implemented for effects that require supplementary datasets, such as reflections, ambient occlusion, and refraction. Rasterization, the typical way that 3D geometry gets drawn on a 2D display, converts triangle coordinates into screen coordinates, and then a point-in-triangle test runs across every sample. This will likely occur once per AA sample (minus pixels that the triangle can’t possibly cover -- such as a pixel outside of the triangle's bounding box -- but that's just optimization).

microsoft-2018-gdc-directx12raytracing-rasterization.png

For rasterization, each triangle is laid on a 2D grid corresponding to the draw surface.
If any sample is in the triangle, the pixel shader is run.
This example shows the rotated grid MSAA case.

A program, called a pixel shader, is then run with some set of data that the GPU could gather on every valid pixel in the triangle. This set of data typically includes things like world coordinate, screen coordinate, texture coordinates, nearby vertices, and so forth. This lacks a lot of information, especially things that are not visible to the camera. The application is free to provide other sources of data for the shader to crawl… but what?

  • Cubemaps are useful for reflections, but they don’t necessarily match the scene.
  • Voxels are useful for lighting, as seen with NVIDIA’s VXGI and VXAO.

This is where DirectX Raytracing comes in. There’s quite a few components to it, but it’s basically a new pipeline that handles how rays are cast into the environment. After being queued, it starts out with a ray-generation stage, and then, depending on what happens to the ray in the scene, there are close-hit, any-hit, and miss shaders. Ray generation allows the developer to set up how the rays are cast, where they call an HLSL instrinsic instruction, TraceRay (which is a clever way of invoking them, by the way). This function takes an origin and a direction, so you can choose to, for example, cast rays only in the direction of lights if your algorithm was to, for instance, approximate partially occluded soft shadows from a non-point light. (There are better algorithms to do that, but it's just the first example that came off the top of my head.) The close-hit, any-hit, and miss shaders occur at the point where the traced ray ends.

To connect this with current technology, imagine that ray-generation is like a vertex shader in rasterization, where it sets up the triangle to be rasterized, leading to pixel shaders being called.

microsoft-2018-gdc-directx12raytracing-multibounce.png

Even more interesting – the close-hit, any-hit, and miss shaders can call TraceRay themselves, which is used for multi-bounce and other recursive algorithms (see: figure above). The obvious use case might be reflections, which is the headline of the GDC talk, but they want it to be as general as possible, aligning with the evolution of GPUs. Looking at NVIDIA’s VXAO implementation, it also seems like a natural fit for a raytracing algorithm.

Speaking of data structures, Microsoft also detailed what they call the acceleration structure. Each object is composed of two levels. The top level contains per-object metadata, like its transformation and whatever else data that the developer wants to add to it. The bottom level contains the geometry. The briefing states, “essentially vertex and index buffers” so we asked for clarification. DXR requires that triangle geometry be specified as vertex positions in either 32-bit float3 or 16-bit float3 values. There is also a stride property, so developers can tweak data alignment and use their rasterization vertex buffer, as long as it's HLSL float3, either 16-bit or 32-bit.

As for the tools to develop this in…

microsoft-2018-gdc-PIX.png

Microsoft announced PIX back in January 2017. This is a debugging and performance analyzer for 64-bit, DirectX 12 applications. Microsoft will upgrade it to support DXR as soon as the API is released (specifically, “Day 1”). This includes the API calls, the raytracing pipeline resources, the acceleration structure, and so forth. As usual, you can expect Microsoft to support their APIs with quite decent – not perfect, but decent – documentation and tools. They do it well, and they want to make sure it’s available when the API is.

ea-2018-SEED screenshot (002).png

Example of DXR via EA's in-development SEED engine.

In short, raytracing is here, but it’s not taking over rasterization. It doesn’t need to. Microsoft is just giving game developers another, standardized mechanism to gather supplementary data for their games. Several game engines have already announced support for this technology, including the usual suspects of anything top-tier game technology:

  • Frostbite (EA/DICE)
  • SEED (EA)
  • 3DMark (Futuremark)
  • Unreal Engine 4 (Epic Games)
  • Unity Engine (Unity Technologies)

They also said, “and several others we can’t disclose yet”, so this list is not even complete. But, yeah, if you have Frostbite, Unreal Engine, and Unity, then you have a sizeable market as it is. There is always a question about how much each of these engines will support the technology. Currently, raytracing is not portable outside of DirectX 12, because it’s literally being announced today, and each of these engines intend to support more than just Windows 10 and Xbox.

Still, we finally have a standard for raytracing, which should drive vendors to optimize in a specific direction. From there, it's just a matter of someone taking the risk to actually use the technology for a cool work of art.

If you want to read more, check out Ryan's post about the also-announced RTX, NVIDIA's raytracing technology.

Manufacturer: Microsoft

It's all fun and games until something something AI.

Microsoft announced the Windows Machine Learning (WinML) API about two weeks ago, but they did so in a sort-of abstract context. This week, alongside the 2018 Game Developers Conference, they are grounding it in a practical application: video games!

microsoft-2018-winml-graphic.png

Specifically, the API provides the mechanisms for game developers to run inference on the target machine. The training data that it runs against would be in the Open Neural Network Exchange (ONNX) format from Microsoft, Facebook, and Amazon. Like the initial announcement suggests, it can be used for any application, not just games, but… you know. If you want to get a technology off the ground, and it requires a high-end GPU, then video game enthusiasts are good lead users. When run in a DirectX application, WinML kernels are queued on the DirectX 12 compute queue.

We’ve discussed the concept before. When you’re rendering a video game, simulating an accurate scenario isn’t your goal – the goal is to look like you are. The direct way of looking like you’re doing something is to do it. The problem is that some effects are too slow (or, sometimes, too complicated) to correctly simulate. In these cases, it might be viable to make a deep-learning AI hallucinate a convincing result, even though no actual simulation took place.

Fluid dynamics, global illumination, and up-scaling are three examples.

Previously mentioned SIGGRAPH demo of fluid simulation without fluid simulation...
... just a trained AI hallucinating a scene based on input parameters.

Another place where AI could be useful is… well… AI. One way of making AI is to give it some set of data from the game environment, often including information that a player in its position would not be able to know, and having it run against a branching logic tree. Deep learning, on the other hand, can train itself on billions of examples of good and bad play, and make results based on input parameters. While the two methods do not sound that different, the difference between logic being designed (vs logic being assembled from an abstract good/bad dataset) someone abstracts the potential for assumptions and programmer error. Of course, it abstracts that potential for error into the training dataset, but that’s a whole other discussion.

The third area that AI could be useful is when you’re creating the game itself.

There’s a lot of grunt and grind work when developing a video game. Licensing prefab solutions (or commissioning someone to do a one-off asset for you) helps ease this burden, but that gets expensive in terms of both time and money. If some of those assets could be created by giving parameters to a deep-learning AI, then those are assets that you would not need to make, allowing you to focus on other assets and how they all fit together.

These are three of the use cases that Microsoft is aiming WinML at.

nvidia-2018-deeplearningcarupscale.png

Sure, these are smooth curves of large details, but the antialiasing pattern looks almost perfect.

For instance, Microsoft is pointing to an NVIDIA demo where they up-sample a photo of a car, once with bilinear filtering and once with a machine learning algorithm (although not WinML-based). The bilinear algorithm behaves exactly as someone who has used Photoshop would expect. The machine learning algorithm, however, was able to identify the objects that the image intended to represent, and it drew the edges that it thought made sense.

microsoft-2018-gdc-PIX.png

Like their DirectX Raytracing (DXR) announcement, Microsoft plans to have PIX support WinML “on Day 1”. As for partners? They are currently working with Unity Technologies to provide WinML support in Unity’s ML-Agents plug-in. That’s all the game industry partners they have announced at the moment, though. It’ll be interesting to see who jumps in and who doesn’t over the next couple of years.

Author:
Manufacturer: AMD

Overview

It's clear by now that AMD's latest CPU releases, the Ryzen 3 2200G and the Ryzen 5 2400G are compelling products. We've already taken a look at them in our initial review, as well as investigated how memory speed affected the graphics performance of the internal GPU but it seemed there was something missing.

Recently, it's been painfully clear that GPUs excel at more than just graphics rendering. With the rise of cryptocurrency mining, OpenCL and CUDA performance are as important as ever.

Cryptocurrency mining certainly isn't the only application where having a powerful GPU can help system performance. We set out to see how much of an advantage the Radeon Vega 11 graphics in the Ryzen 5 2400G provided over the significantly less powerful UHD 630 graphics in the Intel i5-8400.

DSC04637.JPG

Test System Setup
CPU AMD Ryzen 5 2400G
Intel Core i5-8400
Motherboard Gigabyte AB350N-Gaming WiFi
ASUS STRIX Z370-E Gaming
Memory 2 x 8GB G.SKILL FlareX DDR4-3200
(All memory running at 3200 MHz)
Storage Corsair Neutron XTi 480 SSD
Sound Card On-board
Graphics Card AMD Radeon Vega 11 Graphics
Intel UHD 630 Graphics
Graphics Drivers AMD 17.40.3701
Intel 23.20.16.4901
Power Supply Corsair RM1000x
Operating System Windows 10 Pro x64 RS3

 

GPGPU Compute

Before we take a look at some real-world examples of where a powerful GPU can be utilized, let's look at the relative power of the Vega 11 graphics on the Ryzen 5 2400G compared to the UHD 630 graphics on the Intel i5-8400.

sisoft-screen.png

SiSoft Sandra is a suite of benchmarks covering a wide array of system hardware and functionality, including an extensive range of GPGPU tests, which we are looking at today. 

sandra1.png

Comparing the raw shader performance of the Ryzen 5 2400G and the Intel i5-8400 provides a clear snapshot of what we are dealing with. In every precision category, the Vega 11 graphics in the AMD part are significantly more powerful than the Intel UHD 630 graphics. This all combines to provide a 175% increase in aggregate shader performance over Intel for the AMD part. 

Now that we've taken a look at the theoretical power of these GPUs, let's see how they perform in real-world applications.

Continue reading our look at the GPU compute performance of the Ryzen 5 2400G!

Author:
Manufacturer: AMD

Memory Matters

Memory speed is not a factor that the average gamer thinks about when building their PC. For the most part, memory performance hasn't had much of an effect on modern processors running high-speed memory such as DDR3 and DDR4.

With the launch of AMD's Ryzen processors, last year emerged a platform that was more sensitive to memory speeds. By running Ryzen processors with higher frequency and lower latency memory, users should see significant performance improvements, especially in 1080p gaming scenarios.

However, the Ryzen processors are not the only ones to exhibit this behavior.

Gaming on integrated GPUs is a perfect example of a memory starved situation. Take for instance the new AMD Ryzen 5 2400G and it's Vega-based GPU cores. In a full Vega 56 or 64 situation, these Vega cores utilize blazingly fast HBM 2.0 memory. However, due to constraints such as die space and cost, this processor does not integrate HBM.

DSC04643.JPG

Instead, both the CPU portion and the graphics portion of the APU must both depend on the same pool of DDR4 system memory. DDR4 is significantly slower than memory traditionally found on graphics cards such as GDDR5 or HBM. As a result, APU performance is usually memory limited to some extent.

In the past, we've done memory speed testing with AMD's older APUs, however with the launch of the new Ryzen and Vega based R3 2200G and R5 2400G, we decided to take another look at this topic.

For our testing, we are running the Ryzen 5 2400G at three different memory speeds, 2400 MHz, 2933 MHz, and 3200 MHz. While the maximum supported JEDEC memory standard for the R5 2400G is 2933, the memory provided by AMD for our processor review will support overclocking to 3200MHz just fine.

Continue reading our look at memory speed scaling with the Ryzen 5 2400G!

Author:
Manufacturer: ASUS

Specifications and Design

With all of the activity in both the GPU and CPU markets this year, it's hard to remember some of the launches in the first half of the year—including NVIDIA's GTX 1080 Ti. Maintaining the rank of fastest gaming GPU for the majority of the year, little has challenged NVIDIA's GP102-based offering, making it the defacto choice for high-end gamers.

Even though we've been giving a lot of attention to NVIDIA's new flagship TITAN V graphics card, the $3000 puts it out of the range of almost every gamer who doesn't have a day job involving deep learning.

IMG_5011.JPG

Today, we're taking a look back to the (slightly) more reasonable GP102 and the one of the most premiere offerings to feature it, the ASUS ROG Strix GTX 1080 Ti.

Hardware Specifications

While the actual specifications of the GP102 GPU onboard the ASUS Strix GTX 1080 Ti hasn't changed at all, let's take a moment to refresh ourselves on where it sits in regards to the rest of the market.

  RX Vega 64 Liquid RX Vega 56 GTX 1080 Ti GTX 1080 GTX 1070 Ti GTX 1070
GPU Cores 4096 3584 3584 2560 2432 1920
Base Clock 1406 MHz 1156 MHz 1480 MHz 1607 MHz 1607 MHz 1506 MHz
Boost Clock 1677 MHz 1471 MHz 1582 MHz 1733 MHz 1683 MHz 1683 MHz
Texture Units 256 256 224 160 152 120
ROP Units 64 64 88 64 64 64
Memory 8GB 8GB 11GB 8GB 8GB 8GB
Memory Clock 1890 MHz 1600 MHz 11000 MHz 10000 MHz 8000 MHz 8000 MHz
Memory Interface 2048-bit HBM2 2048-bit HBM2 352-bit G5X 256-bit G5X 256-bit 256-bit
Memory Bandwidth 484 GB/s 410 GB/s 484 GB/s 320 GB/s 256 GB/s 256 GB/s
TDP 345 watts 210 watts 250 watts 180 watts 180 watts 150 watts
Peak Compute 13.7 TFLOPS 10.5 TFLOPS 11.3 TFLOPS 8.2 TFLOPS 7.8 TFLOPS 5.7 TFLOPS
MSRP (current) $699 $399 $699 $499 $449 $399

If you'd like some additional details on the NVIDIA GTX 1080 Ti, or it's GP102 GPU, take a look at our review of the reference Founder's edition.

The GTX 1000 series of products from NVIDIA has marked a consolidation in ASUS's GPU offerings. Instead of having both Strix and Matrix products available, the Strix has supplanted everything to be the most premium option from ASUS for any given GPU, and the Strix GTX 1080 Ti doesn't disappoint.

IMG_5018.JPG

While it might not be the largest graphics card we've ever seen, the ASUS Strix GTX 1080 Ti is more massive in all dimensions compared to both the NVIDIA Founder's Edition card, as well as the EVGA ICX option we took a look at earlier this year. Compared to the Founder's Edition, the Strix GTX 1080 Ti is 1.23-in longer, 0.9-in taller, and takes up an extra PCIe slot in width.

Continue reading our review of the ASUS ROG Strix GTX 1080 Ti!!

How deep is your learning?

Recently, we've had some hands-on time with NVIDIA's new TITAN V graphics card. Equipped with the GV100 GPU, the TITAN V has shown us some impressive results in both gaming and GPGPU compute workloads.

However, one of the most interesting areas that NVIDIA has been touting for GV100 has been deep learning. With a 1.33x increase in single-precision FP32 compute over the Titan Xp, and the addition of specialized Tensor Cores for deep learning, the TITAN V is well positioned for deep learning workflows.

In mathematics, a tensor is a multi-dimensional array of numerical values with respect to a given basis. While we won't go deep into the math behind it, Tensors are a crucial data structure for deep learning applications.

07.jpg

NVIDIA's Tensor Cores aim to accelerate Tensor-based math by utilizing half-precision FP16 math in order to process both dimensions of a Tensor at the same time. The GV100 GPU contains 640 of these Tensor Cores to accelerate FP16 neural network training.

It's worth noting that these are not the first Tensor operation-specific hardware, with others such as Google developing hardware for these specific functions.

Test Setup

  PC Perspective Deep Learning Testbed
Processor AMD Ryzen Threadripper 1920X
Motherboard GIGABYTE X399 AORUS Gaming 7
Memory 64GB Corsair Vengeance RGB DDR4-3000 
Storage Samsung SSD 960 Pro 2TB
Power Supply Corsair AX1500i 1500 watt
OS Ubuntu 16.04.3 LTS
Drivers AMD: AMD GPU Pro 17.50
NVIDIA: 387.34

For our NVIDIA testing, we used the NVIDIA GPU Cloud 17.12 Docker containers for both TensorFlow and Caffe2 inside of our Ubuntu 16.04.3 host operating system.

AMD testing was done using the hiptensorflow port from the AMD ROCm GitHub repositories.

For all tests, we are using the ImageNet Large Scale Visual Recognition Challenge 2012 (ILSVRC2012) data set.

Continue reading our look at deep learning performance with the NVIDIA Titan V!!

Author:
Manufacturer: NVIDIA

Looking Towards the Professionals

This is a multi-part story for the NVIDIA Titan V:

Earlier this week we dove into the new NVIDIA Titan V graphics card and looked at its performacne from a gaming perspective. Our conclusions were more or less what we expected - the card was on average ~20% faster than the Titan Xp and about ~80% faster than the GeForce GTX 1080. But with that $3000 price tag, the Titan V isn't going to win any enthusiasts over.

What the Titan V is meant for in reality is the compute space. Developers, coders, engineers, and professionals that use GPU hardware for research, for profit, or for both. In that case, $2999 for the Titan V is simply an investment that needs to show value in select workloads. And though $3000 is still a lot of money, keep in mind that the NVIDIA Quadro GP100, the most recent part with full-performance double precision compute from the Pascal chip, is still selling for well over $6000 today. 

IMG_5009.JPG

The Volta GV100 GPU offers 1:2 double precision performance, equating to 2560 FP64 cores. That is a HUGE leap over the GP102 GPU used on the Titan Xp that uses a 1:32 ratio, giving us just 120 FP64 cores equivalent.

  Titan V Titan Xp GTX 1080 Ti GTX 1080 GTX 1070 Ti GTX 1070 RX Vega 64 Liquid Vega Frontier Edition
GPU Cores 5120 3840 3584 2560 2432 1920 4096 4096
FP64 Cores 2560 120 112 80 76 60 256 256
Base Clock 1200 MHz 1480 MHz 1480 MHz 1607 MHz 1607 MHz 1506 MHz 1406 MHz 1382 MHz
Boost Clock 1455 MHz 1582 MHz 1582 MHz 1733 MHz 1683 MHz 1683 MHz 1677 MHz 1600 MHz
Texture Units 320 240 224 160 152 120 256 256
ROP Units 96 96 88 64 64 64 64 64
Memory 12GB 12GB 11GB 8GB 8GB 8GB 8GB 16GB
Memory Clock 1700 MHz MHz 11400 MHz 11000 MHz 10000 MHz 8000 MHz 8000 MHz 1890 MHz 1890 MHz
Memory Interface 3072-bit
HBM2
384-bit G5X 352-bit G5X 256-bit G5X 256-bit 256-bit 2048-bit HBM2 2048-bit HBM2
Memory Bandwidth 653 GB/s 547 GB/s 484 GB/s 320 GB/s 256 GB/s 256 GB/s 484 GB/s 484 GB/s
TDP 250 watts 250 watts 250 watts 180 watts 180 watts 150 watts 345 watts 300 watts
Peak Compute 12.2 (base) TFLOPS
14.9 (boost) TFLOPS
12.1 TFLOPS 11.3 TFLOPS 8.2 TFLOPS 7.8 TFLOPS 5.7 TFLOPS 13.7 TFLOPS 13.1 TFLOPS
Peak DP Compute 6.1 (base) TFLOPS
7.45 (boost) TFLOPS
0.37 TFLOPS 0.35 TFLOPS 0.25 TFLOPS 0.24 TFLOPS 0.17 TFLOPS 0.85 TFLOPS 0.81 TFLOPS
MSRP (current) $2999 $1299 $699 $499 $449 $399 $699 $999

The current AMD Radeon RX Vega 64, and the Vega Frontier Edition, all ship with a 1:16 FP64 ratio, giving us the equivalent of 256 DP cores per card.

Test Setup and Benchmarks

Our testing setup remains the same from our gaming tests, but obviously the software stack is quite different. 

  PC Perspective GPU Testbed
Processor Intel Core i7-5960X Haswell-E
Motherboard ASUS Rampage V Extreme X99
Memory G.Skill Ripjaws 16GB DDR4-3200
Storage OCZ Agility 4 256GB (OS)
Adata SP610 500GB (games)
Power Supply Corsair AX1500i 1500 watt
OS Windows 10 x64
Drivers AMD: 17.10.2
NVIDIA: 388.59

Applications in use include:

  • Luxmark 
  • Cinebench R15
  • VRay
  • Sisoft Sandra GPU Compute
  • SPECviewperf 12.1
  • FAHBench

Let's not drag this along - I know you are hungry for results! (Thanks to Ken for running most of these tests for us!!)

Continue reading part 2 of our Titan V review on compute performance!!

Author:
Manufacturer: NVIDIA

A preview of potential Volta gaming hardware

This is a multi-part story for the NVIDIA Titan V:

As a surprise to most of us in the media community, NVIDIA launched a new graphics card to the world, the TITAN V. No longer sporting the GeForce brand, NVIDIA has returned the Titan line of cards to where it began – clearly targeted at the world of developers and general purpose compute. And if that branding switch isn’t enough to drive that home, I’m guessing the $2999 price tag will be.

Today’s article is going to look at the TITAN V from the angle that is likely most interesting to the majority of our readers, that also happens to be the angle that NVIDIA is least interested in us discussing. Though targeted at machine learning and the like, there is little doubt in my mind that some crazy people will want to take on the $3000 price to see what kind of gaming power this card can provide. After all, this marks the first time that a Volta-based GPU from NVIDIA has shipped in a place a consumer can get their hands on it, and the first time it has shipped with display outputs. (That’s kind of important to build a PC around it…)

IMG_4999.JPG

From a scientific standpoint, we wanted to look at the Titan V for the same reasons we tested the AMD Vega Frontier Edition cards upon their launch: using it to estimate how future consumer-class cards will perform in gaming. And, just as we had to do then, we purchased this Titan V from NVIDIA.com with our own money. (If anyone wants to buy this from me to recoup the costs, please let me know! Ha!)

  Titan V Titan Xp GTX 1080 Ti GTX 1080 GTX 1070 Ti GTX 1070 RX Vega 64 Liquid Vega Frontier Edition
GPU Cores 5120 3840 3584 2560 2432 1920 4096 4096
Base Clock 1200 MHz 1480 MHz 1480 MHz 1607 MHz 1607 MHz 1506 MHz 1406 MHz 1382 MHz
Boost Clock 1455 MHz 1582 MHz 1582 MHz 1733 MHz 1683 MHz 1683 MHz 1677 MHz 1600 MHz
Texture Units 320 240 224 160 152 120 256 256
ROP Units 96 96 88 64 64 64 64 64
Memory 12GB 12GB 11GB 8GB 8GB 8GB 8GB 16GB
Memory Clock 1700 MHz MHz 11400 MHz 11000 MHz 10000 MHz 8000 MHz 8000 MHz 1890 MHz 1890 MHz
Memory Interface 3072-bit
HBM2
384-bit G5X 352-bit G5X 256-bit G5X 256-bit 256-bit 2048-bit HBM2 2048-bit HBM2
Memory Bandwidth 653 GB/s 547 GB/s 484 GB/s 320 GB/s 256 GB/s 256 GB/s 484 GB/s 484 GB/s
TDP 250 watts 250 watts 250 watts 180 watts 180 watts 150 watts 345 watts 300 watts
Peak Compute 12.2 (base) TFLOPS
14.9 (boost) TFLOPS
12.1 TFLOPS 11.3 TFLOPS 8.2 TFLOPS 7.8 TFLOPS 5.7 TFLOPS 13.7 TFLOPS 13.1 TFLOPS
MSRP (current) $2999 $1299 $699 $499   $399 $699 $999

The Titan V is based on the GV100 GPU though with some tweaks that lower performance and capability slightly when compared to the Tesla-branded equivalent hardware. Though our add-in card iteration has the full 5120 CUDA cores enabled, the HBM2 memory bus is reduced from 4096-bit to 3072-bit and it has one of the four stacks on the package disabled. This also drops the memory capacity from 16GB to 12GB, and memory bandwidth to 652.8 GB/s.

Continue reading our gaming review of the NVIDIA Titan V!!

Author:
Manufacturer: AMD

The flower, not the hormone

It was way back in December of 2014 that AMD and the Radeon group first started down the path of major driver updates on an annual cadence. The Catalyst Omega release marked the beginning of a recommitment to the needs of gamers (and now professionals) with more frequent, and more dramatic, software updates and improvements. Cognizant of the previous reputation the company had with drivers and software, often a distant second to the success that NVIDIA had created with it GeForce drivers, Radeon users were promised continuous increases.

And make no mistake, the team at AMD had an uphill battle. But with releases like Omega, Crimson, ReLive, and now Adrenalin, it’s clear that the leadership has received the message and put emphasis on the portion of its product that can have the most significant impact on experience.

AMD joins us at the PCPer offices to talk through all the new features and capabilities!

Named after the adrenalin rose, rather than the drug that flows through your body when being chased by feral cats, this latest major software release for Radeon users includes a host of new features and upgraded ones that should bring a fresh coat of paint to any existing GPU. Two big features will steal the show, the new Radeon Overlay and a mobile app called AMD Link. But expansions to ReLive, Wattman, Enhanced Sync, and Chill are equally compelling.

Let’s start with what I think will get the most attention and deservedly so, the Radeon Overlay. As the name would suggest, the overlay can be turned out through a hotkey in-game, and allows the gamer to access graphics card monitoring tools and many driver settings without leaving the game, having to alt-tab, or having to close the game to apply. By hitting Alt-R, a screen will show up on the right-hand side of the display, with the game continuing to run in the background. The user will be able to interact with the menu via mouse or keyboard, and then hit the same hotkey or Esc to return.

adrenalin-49.jpg

Continue reading our look at the new AMD Radeon Softare Adrenalin Edition driver!!

A New Frontier

Console game performance has always been an area that we've been interested in here at PC Perspective but has been mostly out of our reach to evaluate with any kind of scientific tilt. Our Frame Rating methodology for PC-based game analysis relies on having an overlay application during screen capture which is later analyzed by a series of scripts. Obviously, we can not take this approach with consoles as we cannot install our own code on the consoles to run that overlay. 

A few other publications such as Eurogamer with their Digital Foundry subsite have done fantastic work developing their internal toolsets for evaluating console games, but this type of technology has mostly remained out of reach of the everyman.

trdrop2.PNG_.png

Recently, we came across an open source project which aims to address this. Trdrop is an open source software built upon OpenCV, a stalwart library in the world of computer vision. Using OpenCV, trdrop can analyze the frames of ordinary gameplay (without an overlay), detecting if there are differences between two frames, looking for dropped frames and tears to come up with a real-time frame rate.

trdrop1.PNG

This means that trdrop can analyze gameplay footage from any source, be it console, PC, or anything in-between from which you can get a direct video capture feed. Now that PC capture cards capable of 1080p60, and even 4K60p are coming down in price, software like this is allowing more gamers to peek at the performance of their games, which we think is always a good thing.

It's worth noting that trdrop is still listed as "alpha" software on it's GitHub repo, but we have found the software to be very stable and flexible in the current iteration.

  Xbox One S Xbox One X PS4 PS4 Pro
CPU 8x Jaguar
1.75 Ghz
8x Jaguar
2.3 Ghz
8x Jaguar
1.6 Ghz
8x Jaguar
2.1 Ghz
GPU CU 12x GCN
914 Mhz
40x Custom
1172 Mhz
18x GCN
800 Mhz
36x GCN
911 Mhz
GPU
Compute
1.4 TF 6.0 TF 1.84 TF 4.2 TF
Memory 8 GB DDR3
32MB ESRAM
12 GB GDDR5 8 GB GDDR5 8 GB GDDR5
Memory
Bandwidth
219GB/s 326GB/s 176GB/s 218GB/s

Now that the Xbox One X is out, we figured it would be a good time to take a look at the current generation of consoles and their performance in a few games as a way to get our feet wet with this new software and method. We are only testing 1080p here, but we now have our hands on a 4K HDMI capture card capable of 60Hz for some future testing! (More on that soon.)

Continue reading our look at measuring performance of the Xbox One X!

Author:
Manufacturer: Intel

The Expected Unexpected

Last night we first received word that Raja had resigned from AMD (during a sabbatical) after they had launched Vega.  The initial statement was that Raja would come back to resume his position at AMD in a December/January timeframe.  During this time there was some doubt as to if Raja would in fact come back to AMD, as “sabbaticals” in the tech world would often lead the individual to take stock of their situation and move on to what they would consider to be greener pastures.

raja_ryan.JPG

Raja has dropped by the PCPer offices in the past.

Initially it was thought that Raja would take the time off and then eventually jump to another company and tackle the issues there.  This behavior is quite common in Silicon Valley and Raja is no stranger to this.  Raja cut his teeth on 3D graphics at S3, but in 2001 he moved to ATI.  While there he worked on a variety of programs including the original Radeon, the industry changing Radeon 9700 series, and finishing up with the strong HD 4000 series of parts.  During this time ATI was acquired by AMD and he became one of the top graphics guru at that company.  In 2009 he quit AMD and moved on to Apple.  He was Director of Graphics Architecture at Apple, but little is known about what he actually did.  During that time Apple utilized AMD GPUs and licensed Imagination Technologies graphics technology.  Apple could have been working on developing their own architecture at this point, which has recently showed up in the latest iPhone products.

In 2013 Raja rejoined AMD and became a corporate VP of Visual Computing, but in 2015 he was promoted to leading the Radeon Technology Group after Lisu Su became CEO of the company. While there Raja worked to get AMD back on an even footing under pretty strained conditions. AMD had not had the greatest of years and had seen their primary moneymakers start taking on water.  AMD had competitive graphics for the most part, and the Radeon technology integrated into AMD’s APUs truly was class leading.  On the discrete side AMD was able to compare favorably to NVIDIA with the HD 7000 and later R9 200 series of cards.  After NVIDIA released their Maxwell based chips, AMD had a hard time keeping up.  The general consensus here is that the RTG group saw its headcount decreased by the company-wide cuts as well as a decrease in R&D funds.

Continue reading about Raja Koduri joinging Intel...

Author:
Manufacturer: NVIDIA

Here comes a new challenger

The release of the GeForce GTX 1070 Ti has been an odd adventure. Launched into a narrow window of a product stack between the GTX 1070 and the GTX 1080, the GTX 1070 Ti is a result of the competition from the AMD RX Vega product line. Sure, NVIDIA might have speced out and prepared an in-between product for some time, but it was the release of competitive high-end graphics cards from AMD (for the first time in forever it seems) that pushed NVIDIA to launch what you see before us today.

With MSRPs of $399 and $499 for the GTX 1070 and GTX 1080 respectively, a new product that fits between them performance wise has very little room to stretch its legs. Because of that, there are some interesting peculiarities involved with the release cycle surrounding overclocks, partner cards, and more.

IMG_4944.JPG

But before we get into that concoction, let’s first look at the specifications of this new GPU option from NVIDIA as well as the reference Founders Edition and EVGA SC Black Edition cards that made it to our offices!

GeForce GTX 1070 Ti Specifications

We start with our classic table of details.

  RX Vega 64 Liquid RX Vega 64 Air RX Vega 56 Vega Frontier Edition GTX 1080 Ti GTX 1080 GTX 1070 Ti GTX 1070
GPU Cores 4096 4096 3584 4096 3584 2560 2432 1920
Base Clock 1406 MHz 1247 MHz 1156 MHz 1382 MHz 1480 MHz 1607 MHz 1607 MHz 1506 MHz
Boost Clock 1677 MHz 1546 MHz 1471 MHz 1600 MHz 1582 MHz 1733 MHz 1683 MHz 1683 MHz
Texture Units 256 256 256 256 224 160 152 120
ROP Units 64 64 64 64 88 64 64 64
Memory 8GB 8GB 8GB 16GB 11GB 8GB 8GB 8GB
Memory Clock 1890 MHz 1890 MHz 1600 MHz 1890 MHz 11000 MHz 10000 MHz 8000 MHz 8000 MHz
Memory Interface 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 352-bit G5X 256-bit G5X 256-bit 256-bit
Memory Bandwidth 484 GB/s 484 GB/s 410 GB/s 484 GB/s 484 GB/s 320 GB/s 256 GB/s 256 GB/s
TDP 345 watts 295 watts 210 watts 300 watts 250 watts 180 watts 180 watts 150 watts
Peak Compute 13.7 TFLOPS 12.6 TFLOPS 10.5 TFLOPS 13.1 TFLOPS 11.3 TFLOPS 8.2 TFLOPS 7.8 TFLOPS 5.7 TFLOPS
MSRP (current) $699 $499 $399 $999 $699 $499 $449 $399

If you have followed the leaks and stories over the last month or so, the information here isn’t going to be a surprise. The CUDA core count of the GTX 1070 Ti is 2432, only one SM unit less than the GTX 1080. Base and boost clock speeds are the same as the GTX 1080. The memory system includes 8GB of GDDR5 running at 8 GHz, matching the performance of the GTX 1070 in this case. The TDP gets a bump up to 180 watts, in line with the GTX 1080 and slightly higher than the GTX 1070.

Continue reading our review of the GeForce GTX 1070 Ti!

Forza Motorsport 7 Performance

The first full Forza Motorsport title available for the PC, Forza Motorsport 7 on Windows 10 launched simultaneously with the Xbox version earlier this month. With native 4K assets, HDR support, and new visual features like fully dynamic weather, this title is an excellent showcase of what modern PC hardware can do.

forza7-screen.png

Now that both AMD and NVIDIA have released drivers optimized for Forza 7, we've taken an opportunity to measure performance across an array of different GPUs. After some significant performance mishaps with last year's Forza Horizon 3 at launch on PC, we are excited to see if Forza Motorsport 7 brings any much-needed improvements. 

For this testing, we used our standard GPU testbed, including an 8-core Haswell-E processor and plenty of memory and storage.

  PC Perspective GPU Testbed
Processor Intel Core i7-5960X Haswell-E
Motherboard ASUS Rampage V Extreme X99
Memory G.Skill Ripjaws 16GB DDR4-3200
Storage OCZ Agility 4 256GB (OS)
Adata SP610 500GB (games)
Power Supply Corsair AX1500i 1500 watt
OS Windows 10 x64 
Drivers AMD: 17.10.1 (Beta)
NVIDIA: 387.92

As with a lot of modern console-first titles, Forza 7 defaults to "Dynamic" image quality settings. This means that the game engine is supposed to find the best image settings for your hardware automatically, and dynamically adjust them so that you hit a target frame rate (adjustable between 30 and 60fps) no matter what is going on in the current scene that is being rendered.

While this is a good strategy for consoles, and even for casual PC gamers, it poses a problem for us trying to measure equivalent performance across GPUs. Luckily the developers of Forza Motorsport 7, Turn 10 Studios, still let you disable the dynamic control and configure the image quality settings as you desire.

One quirk however though is that in order for V-Sync to be disabled, the rendering resolution within the game must match the native resolution of your monitor. This means that if you are running 2560x1440 on your 4K monitor, you must first set the resolution within windows to 2560x1440 in order to run the game in V-Sync off mode.

forza7-settings.png

We did our testing with an array of three different resolutions (1080p, 1440p, and 4K) at maximum image quality settings. We tested both AMD and NVIDIA graphics cards in similar price and performance segments. The built-in benchmark mode for this game was used, which does feature some variance due to dynamic weather patterns. However, our testing within the full game matched the results of the benchmark mode closely, so we used it for our final results.

forza7-avgfps.png

Right off the bat, I have been impressed at how well optimized Forza Motorsport 7 seems to be on the PC. Compared to the unoptimized disaster that was Forza Horizon 3 when it launched on PC last year, it's clear that Turn 10 Studios and Microsoft have come a long way.

Even gamers looking to play on a 4K display at 60Hz can seemingly get away with the cheaper, and more mainstream GPUs such as the RX 580 or the GTX 1060 with acceptable performance in most scenarios.

Games on high-refresh-rate displays don't appear to have the same luxury. If you want to game at a resolution such as 2560x1440 at a full 144Hz, neither the RX Vega 64 or GTX 1080 will do this with maximum image quality settings. Although these GPUs appear to be in the margin where you could turn down a few settings to achieve your full refresh rate.

For some reason, the RX Vega cards didn't seem to show any scaling in performance when moving from 2560x1440 to 1920x1080, unlike the Polaris-based RX 580 and the NVIDIA options. We aren't quite sure of the cause of this and have reached out to AMD for clarification.

As far as frame times are concerned, we also gathered some data with our Frame Rating capture analysis system

Forza7_2560x1440_PLOT.png

Forza7_2560x1440_STUT.png

Taking a look at the first chart, we can see while the GTX 1080 frame times are extremely consistent, the RX Vega 64 shows some additional variance.

However, the frame time variance chart shows that over 95% of the frame times of the RX Vega 64 come in at under 2ms of variance, which will still provide a smooth gameplay experience in most scenarios. This matches with our experience while playing on both AMD and NVIDIA hardware where we saw no major issues with gameplay smoothness.

forza7-screen2.png

Forza Motorsport 7 seems to be a great addition to the PC gaming world (if you don't mind using the Microsoft store exclusively) and will run great on a wide array of hardware. Whether or not you have a NVIDIA or AMD GPU, you should be able to enjoy this fantastic racing simulator. 

Author:
Manufacturer: NVIDIA

Can you hear me now?

One of the more significant downsides to modern gaming notebooks is noise. These devices normally have small fans that have to spin quickly to cool the high-performance components found inside. While the answer for loud gaming desktops might be a nice set of headphones, for notebooks that may be used in more public spaces, that's not necessarily a good solution for friends or loved ones.

Attempting to address the problem of loud gaming notebooks, NVIDIA released a technology called WhisperMode. WhisperMode launched alongside NVIDIA's Max-Q design notebooks earlier this year, but it will work with any notebook enabled with an NVIDIA GTX 1060 or higher. This software solution aims to limit noise and power consumption of notebooks by restricting the frame rate of your game to a reasonable compromise of performance, noise, and power levels. NVIDIA has profiled over 400 games to find this sweet spot and added profiles for those games to WhisperMode technology.

WhisperMode is enabled through the NVIDIA GeForce Experience application.

GFE-whisper.PNG

From GFE, you can also choose to "Optimize games for WhisperMode." This will automatically adjust settings (in-game) to complement the frame rate target control of WhisperMode.

NVCP_whisper.PNG

If you want to adjust the Frame Rate Target, that must be done in the traditional NVIDIA Control Panel and is done on a per app basis. The target can be set at intervals of 5 FPS from 30 to the maximum refresh of your display. Having to go between two pieces of software to tweak these settings seems overly complex and hopefully some upcoming revamp of the NVIDIA software stack might address this user interface falacy. 

To put WhisperMode through its paces, we tried it on two notebooks - one with a GTX 1070 Max-Q (the MSI GS63VR) and one with a GTX 1080 Max-Q (the ASUS ROG Zephyrus). Our testing consisted of two games, Metro: Last Light and Hitman. Both of these games were run for 15 minutes to get the system up to temperature and achieve sound measurements that are more realistic to extended gameplay sessions. Sound levels were measured with our Extech 407739 Sound Level Meter placed at a distance of 6 inches from the given notebooks, above the keyboard and offset to the right.

Continue reading our review of the new NVIDIA WhisperMode technology!

Author:
Manufacturer: Intel

A surprise twist from Intel

Any expectations I had of a slower and less turbulent late summer and fall for the technology and hardware segments is getting shattered today with the beginning stages of Intel’s 8th Generation Core Processors. If you happen to think that this 8th generation is coming hot on the heels of the 7th generation that only just released to the consumer desktop market in January of this year, you’d be on the same page as me. If you are curious how Intel plans to balance Kaby Lake, Coffee Lake, and Cannon Lake, all releasing in similar time frames and still use terms like “generation,” then again, we are on the same page.

badge.jpg

Today Intel launches the 15-watt version of its 8th Generation Core Processors, based on a refresh of the Kaby Lake CPU design. This not a new architecture nor is this is not a new process node, though Intel does talk about slight changes in design and manufacturing that make it possible. The U-series processors that make up the majority of the thin and light and 2-in-1 designs for consumers and businesses are getting a significant upgrade in performance with this release. The Core i7 and Core i5 processors being announced will all be quad-core, HyperThreaded designs, moving us away from the world of dual-core processors in the 7th generation. Doubling core and thread count, while remaining inside the 15-watt thermal envelope for designs, is an incredible move and will strengthen Intel’s claim to this very important and very profitable segment.

Let’s look at the specifications table first. After all, we’re all geeks here.

  Core i7-8650U Core i7-8550U Core i5-8350U Core i5-8250U Core i7-7600U Core i7-7500U
Architecture Kaby Lake Refresh Kaby Lake Refresh Kaby Lake Refresh Kaby Lake Refresh Kaby Lake Kaby Lake
Process Tech 14nm+ 14nm+ 14nm+ 14nm+ 14nm+ 14nm+
Socket BGA1356 BGA1356 BGA1356 BGA1356 BGA1356 BGA1356
Cores/Threads 4/8 4/8 4/8 4/8 2/4 2/4
Base Clock 1.9 GHz 1.8 GHz 1.7 GHz 1.6 GHz 2.8 GHz 2.7 GHz
Max Turbo Clock 4.2 GHz 4.0 GHz 3.8 GHz 3.6 GHz 3.9 GHz 3.5 GHz
Memory Tech DDR4/LPDDR3 DDR4/LPDDR3 DDR4/LPDDR3 DDR4/LPDDR3 DDR4/LPDDR3 DDR4/LPDDR3
Memory Speeds 2400/2133 2400/2133 2400/2133 2400/2133 2133/1866 2133/1866
Cache (L4 Cache) 8MB 8MB 6MB 6MB 4MB 4MB
System Bus DMI3 - 8.0 GT/s DMI3 - 8.0 GT/s DMI2 - 6.4 GT/s DMI2 - 5.0 GT/s DMI2 - 5.0 GT/s DMI2 - 5.0 GT/s
Graphics UHD Graphics 620 UHD Graphics 620 UHD Graphics 620 UHD Graphics 620 HD Graphics 620 HD Graphics 620
Max Graphics Clock 1.15 GHz 1.15 GHz 1.1 GHz 1.1 GHz 1.15 GHz 1.05 GHz
TDP 15W 15W 15W 15W 15W 15W
MSRP $409 $409 $297 $297 $393 $393

The only differences between the Core i7 and Core i5 designs will be in cache size (Core i5 has 6MB, Core i7 has 8MB) and the clock speeds of the processors. All of them feature four true Kaby Lake cores with HyperThreading enabled to support 8 simultaneous threads in a notebook. Dual channel memory capable of speeds of 2400 MHz in DDR4 and 2133 MHz in LPDDR3 remain. The integrated graphics portion offers the same performance as the 7th generation designs, though the branding has moved from Intel HD Graphics to Intel UHD Graphics. Because Ultra.

8th Gen Intel Core U-series front.jpg

But take a gander at the clock speeds. The base clock on the four new CPUs range from 1.6 GHz to 1.9 GHz, with 100 MHz steps as you go up the SKU ladder. Those are low frequencies for modern processors, no doubt, but Intel has always been very conservative when it comes to setting specs for base frequency. This is the speed that Intel guarantees the processors will run at when the CPU is fully loaded using a 15-watt TDP cooling design. Keeping in mind that we moved from dual-core to quad-core processors, it makes sense that these base frequencies would drop. Intel doesn’t expect users in thin and light machines to utilize all 8 threads for very long, or very often, and instead focuses on shorter use cases for multi-threaded workloads (photo manipulation) that might run at 3.x GHz. If this period of time is short enough, the cooling solution will be able to “catch up” and keep the core within a reasonable range.

Continue reading about the new 8th Generation Intel Core Processors!!

Author:
Manufacturer: AMD

A confusing market

I feel like I have been writing about AMD non-stop in 2017. Starting with the release of Ryzen 7 and following through last week’s review of the HEDT Threadripper processor, AMD has gone from a nearly-dormant state in 2015-2016 to a wildly active and successful organization with more than a dozen new product launches under its belt. Today we will reveal the company's first consumer products based on the new Vega GPU architecture, thrusting the Radeon brand back into the fight at the $400+ price segments.

At this point, with architecture teases, product unboxings, professional card reviews, and pricing and availability reveals, we almost know everything we need to know about the new Radeon RX Vega 64 and RX Vega 56 products. Almost. Today we can show you the performance.

I want to be honest with our readers: AMD gave me so little time with these cards that I am going to gloss over some of the more interesting technological and architectural changes that Vega brings to market. I will come back to them at a later time, but I feel it is most important for us to talk about the performance and power characteristics of these cards as consumers finally get the chance to spend their hard-earned money on them.

01.jpg

If you already know about the specifications and pricing peculiarities of Vega 64 and Vega 56 and instead want direct access to performance results, I encourage you to skip ahead. If you want a refresher those details, check out the summary below.

Interesting statistics from the creation of this review in a VERY short window:

  • 175 graphs 
  • 8 cards, 8 games, 2 resolutions, 3 runs = 384 test runs
  • >9.6 TB of raw captured video (average ~25 GB/min)

Radeon RX Vega 64 and Vega 56 Specifications

Much of the below is sourced from our Vega 64/56 announcement story last month.

Though the leaks have been frequent and getting closer to reality, as it turns out AMD was in fact holding back quite a bit of information about the positioning of RX Vega for today. Radeon will launch the Vega 64 and Vega 56 today, with three different versions of the Vega 64 on the docket. Vega 64 uses the full Vega 10 chip with 64 CUs and 4096 stream processors. Vega 56 will come with 56 CUs enabled (get it?) and 3584 stream processors.

Pictures of the various product designs have already made it out to the field including the Limited Edition with the brushed anodized aluminum shroud, the liquid cooled card with a similar industrial design, and the more standard black shroud version that looks very similar to the previous reference cards from AMD.

  RX Vega 64 Liquid RX Vega 64 Air RX Vega 56 Vega Frontier Edition GTX 1080 Ti GTX 1080 TITAN X GTX 980 R9 Fury X
GPU Vega 10 Vega 10 Vega 10 Vega 10 GP102 GP104 GM200 GM204 Fiji XT
GPU Cores 4096 4096 3584 4096 3584 2560 3072 2048 4096
Base Clock 1406 MHz 1247 MHz 1156 MHz 1382 MHz 1480 MHz 1607 MHz 1000 MHz 1126 MHz 1050 MHz
Boost Clock 1677 MHz 1546 MHz 1471 MHz 1600 MHz 1582 MHz 1733 MHz 1089 MHz 1216 MHz -
Texture Units 256 256 224 256 224 160 192 128 256
ROP Units 64 64 64 64 88 64 96 64 64
Memory 8GB 8GB 8GB 16GB 11GB 8GB 12GB 4GB 4GB
Memory Clock 1890 MHz 1890 MHz 1600 MHz 1890 MHz 11000 MHz 10000 MHz 7000 MHz 7000 MHz 1000 MHz
Memory Interface 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 352-bit G5X 256-bit G5X 384-bit 256-bit 4096-bit (HBM)
Memory Bandwidth 484 GB/s 484 GB/s 410 GB/s 484 GB/s 484 GB/s 320 GB/s 336 GB/s 224 GB/s 512 GB/s
TDP 345 watts 295 watts 210 watts 300 watts 250 watts 180 watts 250 watts 165 watts 275 watts
Peak Compute 13.7 TFLOPS 12.6 TFLOPS 10.5 TFLOPS 13.1 TFLOPS 10.6 TFLOPS 8.2 TFLOPS 6.14 TFLOPS 4.61 TFLOPS 8.60 TFLOPS
Transistor Count 12.5B 12.5B 12.5B 12.5B 12.0B 7.2B 8.0B 5.2B 8.9B
Process Tech 14nm 14nm 14nm 14nm 16nm 16nm 28nm 28nm 28nm
MSRP (current) $699 $499 $399 $999 $699 $599 $999 $499 $649

If you are a frequent reader of PC Perspective, you have already seen our reviews of the Vega Frontier Edition air cooled and liquid cards, so some of this is going to look very familiar. Looking at the Vega 64 first, we need to define the biggest change to the performance ratings of RX and FE versions of the Vega architecture. When we listed the “boost clock” of the Vega FE cards, and really any Radeon cards previous to RX Vega, we were referring the maximum clock speed of the card in its out of box state. This was counter to the method that NVIDIA used for its “boost clock” rating that pointed towards a “typical” clock speed that the card would run at in a gaming workload. Essentially, the NVIDIA method was giving consumers a more realistic look at how fast the card would be running while AMD was marketing the theoretical peak with perfect thermals, perfect workloads. This, to be clear, never happened.

Continue reading our review of the Radeon RX Vega 64, Vega 64 Liquid, and Vega 56!!

Author:
Manufacturer: AMD

RX Vega is here

Though we are still a couple of weeks from availability and benchmarks, today we finally have the details on the Radeon RX Vega product line. That includes specifications, details on the clock speed changes, pricing, some interesting bundle programs, and how AMD plans to attack NVIDIA through performance experience metrics.

There is a lot going on today and I continue to have less to tell you about more products, so I’m going to defer a story on the architectural revelations that AMD made to media this week and instead focus on what I think more of our readers will want to know. Let’s jump in.

Radeon RX Vega Specifications

Though the leaks have been frequent and getting closer to reality, as it turns out AMD was in fact holding back quite a bit of information about the positioning of RX Vega for today. Radeon will launch the Vega 64 and Vega 56 today, with three different versions of the Vega 64 on the docket. Vega 64 uses the full Vega 10 chip with 64 CUs and 4096 stream processors. Vega 56 will come with 56 CUs enabled (get it?) and 3584 stream processors.

Pictures of the various product designs have already made it out to the field including the Limited Edition with the brushed anodized aluminum shroud, the liquid cooled card with a similar industrial design, and the more standard black shroud version that looks very similar to the previous reference cards from AMD.

  RX Vega 64 Liquid RX Vega 64 Air RX Vega 56 Vega Frontier Edition GTX 1080 Ti GTX 1080 TITAN X GTX 980 R9 Fury X
GPU Vega 10 Vega 10 Vega 10 Vega 10 GP102 GP104 GM200 GM204 Fiji XT
GPU Cores 4096 4096 3584 4096 3584 2560 3072 2048 4096
Base Clock 1406 MHz 1247 MHz 1156 MHz 1382 MHz 1480 MHz 1607 MHz 1000 MHz 1126 MHz 1050 MHz
Boost Clock 1677 MHz 1546 MHz 1471 MHz 1600 MHz 1582 MHz 1733 MHz 1089 MHz 1216 MHz -
Texture Units 256 256 256 256 224 160 192 128 256
ROP Units 64 64 ? 64 88 64 96 64 64
Memory 8GB 8GB 8GB 16GB 11GB 8GB 12GB 4GB 4GB
Memory Clock 1890 MHz 1890 MHz 1600 MHz 1890 MHz 11000 MHz 10000 MHz 7000 MHz 7000 MHz 1000 MHz
Memory Interface 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 2048-bit HBM2 352-bit G5X 256-bit G5X 384-bit 256-bit 4096-bit (HBM)
Memory Bandwidth 484 GB/s 484 GB/s 484 GB/s 484 GB/s 484 GB/s 320 GB/s 336 GB/s 224 GB/s 512 GB/s
TDP 345 watts 295 watts 210 watts 300 watts 250 watts 180 watts 250 watts 165 watts 275 watts
Peak Compute 13.7 TFLOPS 12.6 TFLOPS 10.5 TFLOPS 13.1 TFLOPS 10.6 TFLOPS 8.2 TFLOPS 6.14 TFLOPS 4.61 TFLOPS 8.60 TFLOPS
Transistor Count 12.5B 12.5B 12.5B 12.5B 12.0B 7.2B 8.0B 5.2B 8.9B
Process Tech 14nm 14nm 14nm 14nm 16nm 16nm 28nm 28nm 28nm
MSRP (current) $699 $499 $399 $999 $699 $599 $999 $499 $649

If you are a frequent reader of PC Perspective, you have already seen our reviews of the Vega Frontier Edition air cooled and liquid cards, so some of this is going to look very familiar. Looking at the Vega 64 first, we need to define the biggest change to the performance ratings of RX and FE versions of the Vega architecture. When we listed the “boost clock” of the Vega FE cards, and really any Radeon cards previous to RX Vega, we were referring the maximum clock speed of the card in its out of box state. This was counter to the method that NVIDIA used for its “boost clock” rating that pointed towards a “typical” clock speed that the card would run at in a gaming workload. Essentially, the NVIDIA method was giving consumers a more realistic look at how fast the card would be running while AMD was marketing the theoretical peak with perfect thermals, perfect workloads. This, to be clear, never happened.

vega-44.jpg

With the RX Vega cards and their specifications, the “boost clock” is now a typical clock rate. AMD has told me that this is what they estimate the average clock speed of the card will be during a typical gaming workload with a typical thermal and system design. This is great news! It means that gamers will have a more realistic indication of performance, both theoretical and expected, and the listings on the retailers and partner sites will be accurate. It also means that just looking at the spec table above will give you an impression that the performance gap between Vega FE and RX Vega is smaller than it will be in testing. (This is, of course, if AMD’s claims are true; I haven’t tested it myself yet.)

Continue reading our preview of the Radeon RX Vega 64 and Vega 56!

Author:
Manufacturer: AMD

Software Iteration

The software team at AMD and the Radeon Technologies Group is releasing Radeon Crimson ReLive Edition 17.7.2 this evening and it includes a host of new features, improved performance capabilities, and stability improvements to boot. This isn’t the major reboot of the software that we have come to expect on an annual basis, but rather an attempt to get the software team’s work out in front of media and gamers before the onslaught of RX Vega and Threadripper steal the attention.

radeonsw-4.jpg

AMD’s software team is big on its user satisfaction ratings, which it should be after the many years of falling behind NVIDIA in this department. With 16 individual driver releases in 2017 (so far) and 20 new games optimized and supported with day one releases, the 90% rating seems to be about right. Much of the work that could be done to improve multi-GPU and other critical problems are more than a calendar year behind us, so it seems reasonable the Radeon gamers would be in a good place in terms of software support.

radeonsw-7.jpg

One big change for Crimson ReLive today is that all of those lingering settings that remained in the old Catalyst Control Panel will now reside in the proper Radeon Settings. This means matching UI and streamlined interface.

radeonsw-14.jpg

The ReLive capture and streaming capability sees a handful of upgrades today including a bump from 50mbps to 100mbps maximum bit rate, transparency support for webcams, improved optimization to lower the memory usage (and thus the overhead of running ReLive), notifications of replays and record timers, and audio controls for microphone volume and push-to-talk.

Continue reading about the latest Crimson ReLive driver updates!

Author:
Manufacturer: AMD

Specifications and Design

Just a couple of short weeks ago we looked at the Radeon Vega Frontier Edition 16GB graphics card in its air-cooled variety. The results were interesting – gaming performance proved to fall somewhere between the GTX 1070 and the GTX 1080 from NVIDIA’s current generation of GeForce products. That is under many of the estimates from players in the market, including media, fans, and enthusiasts.  But before we get to the RX Vega product family that is targeted at gamers, AMD has another data point for us to look at with a water-cooled version of Vega Frontier Edition. At a $1500 MSRP, which we shelled out ourselves, we are very interested to see how it changes the face of performance for the Vega GPU and architecture.

Let’s start with a look at the specifications of this version of the Vega Frontier Edition, which will be…familiar.

  Vega Frontier Edition (Liquid) Vega Frontier Edition Titan Xp GTX 1080 Ti Titan X (Pascal) GTX 1080 TITAN X GTX 980 R9 Fury X
GPU Vega Vega GP102 GP102 GP102 GP104 GM200 GM204 Fiji XT
GPU Cores 4096 4096 3840 3584 3584 2560 3072 2048 4096
Base Clock 1382 MHz 1382 MHz 1480 MHz 1480 MHz 1417 MHz 1607 MHz 1000 MHz 1126 MHz 1050 MHz
Boost Clock 1600 MHz 1600 MHz 1582 MHz 1582 MHz 1480 MHz 1733 MHz 1089 MHz 1216 MHz -
Texture Units ? ? 224 224 224 160 192 128 256
ROP Units 64 64 96 88 96 64 96 64 64
Memory 16GB 16GB 12GB 11GB 12GB 8GB 12GB 4GB 4GB
Memory Clock 1890 MHz 1890 MHz 11400 MHz 11000 MHz 10000 MHz 10000 MHz 7000 MHz 7000 MHz 1000 MHz
Memory Interface 2048-bit HBM2 2048-bit HBM2 384-bit G5X 352-bit 384-bit G5X 256-bit G5X 384-bit 256-bit 4096-bit (HBM)
Memory Bandwidth 483 GB/s 483 GB/s 547.7 GB/s 484 GB/s 480 GB/s 320 GB/s 336 GB/s 224 GB/s 512 GB/s
TDP 300 watts
~350 watts
300 watts 250 watts 250 watts 250 watts 180 watts 250 watts 165 watts 275 watts
Peak Compute 13.1 TFLOPS 13.1 TFLOPS 12.0 TFLOPS 10.6 TFLOPS 10.1 TFLOPS 8.2 TFLOPS 6.14 TFLOPS 4.61 TFLOPS 8.60 TFLOPS
Transistor Count ? ? 12.0B 12.0B 12.0B 7.2B 8.0B 5.2B 8.9B
Process Tech 14nm 14nm 16nm 16nm 16nm 16nm 28nm 28nm 28nm
MSRP (current) $1499 $999 $1200 $699 $1,200 $599 $999 $499 $649

The base specs remain unchanged and AMD lists the same memory frequency and even GPU clock rates across both models. In practice though, the liquid cooled version runs at higher sustained clocks and can overclock a bit easier as well (more details later). What does change with the liquid cooled version is a usable BIOS switch on top of the card that allows you to move between two distinct power draw states: 300 watts and 350 watts.

IMG_4728.JPG

First, it’s worth noting this is a change from the “375 watt” TDP that this card was listed at during the launch and announcement. AMD was touting a 300-watt and 375-watt version of Frontier Edition, but it appears the company backed off a bit on that, erring on the side of caution to avoid breaking any of the specifcations of PCI Express (board slot or auxiliary connectors). Even more concerning is that AMD chose to have the default state of the switch on the Vega FE Liquid card at 300 watts rather than the more aggressive 350 watts. AMD claims this to avoid any problems with lower quality power supplies that may struggle to hit slightly over 150 watts of power draw (and resulting current) from the 8-pin power connections. I would argue that any system that is going to install a $1500 graphics card can and should be prepared to provide the necessary power, but for the professional market, AMD leans towards caution. (It’s worth pointing out the RX 480 power issues that may have prompted this internal decision making were more problematic because they impacted the power delivery through the motherboard, while the 6- and 8-pin connectors are generally much safer to exceed the ratings.)

Even without clock speed changes, the move to water cooling should result in better and more consistent performance by removing the overheating concerns that surrounded our first Radeon Vega Frontier Edition review. But let’s dive into the card itself and see how the design process created a unique liquid cooled solution.

Continue reading our review of the Radeon Vega Frontier Edition Liquid-Cooled card!!