Author:
Manufacturer: AMD

BF4 Integrates FCAT Overlay Support

Back in September AMD publicly announced Mantle, a new lower level API meant to offer more performance for gamers and more control for developers fed up with the restrictions of DirectX. Without diving too much into the politics of the release, the fact that Battlefield 4 developer DICE was integrating Mantle into the Frostbite engine for Battlefield was a huge proof point for the technology. Even though the release was a bit later than AMD had promised us, coming at the end of January 2014, one of the biggest PC games on the market today had integrated a proprietary AMD API.

When I did my first performance preview of BF4 with Mantle on February 1st, the results were mixed but we had other issues to deal with. First and foremost, our primary graphics testing methodology, called Frame Rating, wasn't able to be integrated due to the change of API. Instead we were forced to use an in-game frame rate counter built by DICE which worked fine, but didn't give us the fine grain data we really wanted to put the platform to the test. It worked, but we wanted more. Today we are happy to announce we have full support for our Frame Rating and FCAT testing with BF4 running under Mantle.

A History of Frame Rating

In late 2012 and throughout 2013, testing graphics cards became a much more complicated beast. Terms like frame pacing, stutter, jitter and runts were not in the vocabulary of most enthusiasts but became an important part of the story just about one year ago. Though complicated to fully explain, the basics are pretty simple.

Rather than using software on the machine being tested to measure performance, our Frame Rating system uses a combination of local software and external capture hardware. On the local system with the hardware being evaluated we run a small piece of software called an overlay that draws small colored bars on the left hand side of the game screen that change successively with each frame rendered by the game. Using a secondary system, we capture the output from the graphics card directly, intercepting it from the display output, in real-time in an uncompressed form. With that video file captured, we then analyze it frame by frame, measuring the length of each of those colored bars, how long they are on the screen, how consistently they are displayed. This allows us to find the average frame rate but also to find how smoothly the frames are presented, if there are dropped frames and if there are jitter or stutter issues. 

screen1.jpg

Continue reading our first look at Frame Rating / FCAT Testing with Mantle in Battlefield 4!!

GTC 2014: NVIDIA Launches Iray VCA Networked Rendering Appliance

Subject: General Tech, Graphics Cards | April 1, 2014 - 04:42 PM |
Tagged: VCA, nvidia, GTC 2014

NVIDIA launched a new visual computing appliance called the Iray VCA at the GPU Technology Conference last week. This new piece of enterprise hardware uses full GK 110 graphics cards to accelerate the company’s Iray renderer which is used to create photo realistic models in various design programs.

NVIDIA IRAY VCA.jpg

The Iray VCA specifically is a licensed appliance (hardware + software) that combines NVIDIA hardware and software. On the hardware side of things, the Iray VCA is powered by eight graphics cards, dual processors (unspecified but likely Intel Xeons based on usage in last year’s GRID VCA), 256GB of system RAM, and a 2TB SSD. Networking hardware includes two 10GbE NICs, two 1GbE NICs, and one Infiniband connection. In total, the Iray VCA features 20 CPU cores and 23,040 CUDA cores. The GPUs used are based on the full GK110 die and are paired with 12GB of memory each.

Even better, it is a scalable solution such that companies can add additional Iray VCAs to the network. The appliances reportedly transparently accelerate the Iray accelerated renders done on designer’s workstations. NVIDIA reports that an Iray VCA is approximately 60-times faster than a Quadro K5000-powered workstation. Further, according to NVIDIA, 19 Iray VCAs working together amounts to 1 PetaFLOP of compute performance which is enough to render photo realistic simulations using 1 billion rays with up to hundreds of thousands of bounces.

DSC01431.JPG

The Iray VCA enables some rather impressive real time renders of 3D models with realistic physical properties and lighting. The models are light simulations that use ray tracing, global illumination and other techniques to show photo realistic models using up to billions of rays of light. NVIDIA is positioning the Iray VCA as an alternative to physical prototyping, allowing designers to put together virtual prototypes that can be iterated and changed at significantly less cost and time.

DSC01447.JPG

Iray itself is NVIDIA’s GPU-accelerated photo realistic renderer. The Iray technology is used in a number of design software packages. The Iray VCA is meant to further accelerate that Iray renderer by throwing massive amounts of parallel processing hardware at the resource intensive problem over the network (the Iray VCAs can be installed at a data center or kept on site). Initially the Iray VCA will support 3ds Max, Catia, Bunkspeed, and Maya, but NVIDIA is working on supporting all Iray accelerated software with the VCA hardware.

GTC 2014 IRAY VCA Renders Honda Car Interior In Real Time.jpg

The virtual prototypes can be sliced and examined and can even be placed in real world environments by importing HDR photos. Jen-Hsun Huang demonstrated this by placing Honda’s vehicle model on the GTC stage (virtually).

DSC01450.JPG

In fact, one of NVIDIA’s initial partners with the Iray VCA is Honda. Honda is currently beta testing a cluster of 25 Iray VCAs to refine styling designs for cars and their interiors based on initial artistic work. Honda Research and Development System Engineer Daisuke Ide was quoted by NVIDIA as stating that “Our TOPS tool, which uses NVIDIA Iray on our NVIDIA GPU cluster, enables us to evaluate our original design data as if it were real. This allows us to explore more designs so we can create better designs faster and more affordably.”

The Iray VCA (PDF) will be available this summer for $50,000. The sticker price includes the hardware, Iray license, and the first year of updates and maintenance. This is far from consumer technology, but it is interesting technology that may be used in the design process of your next car or other major purchase.

What do you think about the Iray VCA and NVIDIA's licensed hardware model?

GDC 2014: Shader-limited Optimization for AMD's GCN

Subject: Editorial, General Tech, Graphics Cards, Processors, Shows and Expos | March 30, 2014 - 01:45 AM |
Tagged: gdc 14, GDC, GCN, amd

While Mantle and DirectX 12 are designed to reduce overhead and keep GPUs loaded, the conversation shifts when you are limited by shader throughput. Modern graphics processors are dominated by sometimes thousands of compute cores. Video drivers are complex packages of software. One of their many tasks is converting your scripts, known as shaders, into machine code for its hardware. If this machine code is efficient, it could mean drastically higher frame rates, especially at extreme resolutions and intense quality settings.

amd-gcn-unit.jpg

Emil Persson of Avalanche Studios, probably known best for the Just Cause franchise, published his slides and speech on optimizing shaders. His talk focuses on AMD's GCN architecture, due to its existence in both console and PC, while bringing up older GPUs for examples. Yes, he has many snippets of GPU assembly code.

AMD's GCN architecture is actually quite interesting, especially dissected as it was in the presentation. It is simpler than its ancestors and much more CPU-like, with resources mapped to memory (and caches of said memory) rather than "slots" (although drivers and APIs often pretend those relics still exist) and with how vectors are mostly treated as collections of scalars, and so forth. Tricks which attempt to combine instructions together into vectors, such as using dot products, can just put irrelevant restrictions on the compiler and optimizer... as it breaks down those vector operations into those very same component-by-component ops that you thought you were avoiding.

Basically, and it makes sense coming from GDC, this talk rarely glosses over points. It goes over execution speed of one individual op compared to another, at various precisions, and which to avoid (protip: integer divide). Also, fused multiply-add is awesome.

I know I learned.

As a final note, this returns to the discussions we had prior to the launch of the next generation consoles. Developers are learning how to make their shader code much more efficient on GCN and that could easily translate to leading PC titles. Especially with DirectX 12 and Mantle, which lightens the CPU-based bottlenecks, learning how to do more work per FLOP addresses the other side. Everyone was looking at Mantle as AMD's play for success through harnessing console mindshare (and in terms of Intel vs AMD, it might help). But honestly, I believe that it will be trends like this presentation which prove more significant... even if behind-the-scenes. Of course developers were always having these discussions, but now console developers will probably be talking about only one architecture - that is a lot of people talking about very few things.

This is not really reducing overhead; this is teaching people how to do more work with less, especially in situations (high resolutions with complex shaders) where the GPU is most relevant.

AMD FirePro W9100 Announced: Doing Work in Hawaii.

Subject: General Tech, Graphics Cards | March 26, 2014 - 05:43 PM |
Tagged: amd, firepro, W9100

The AMD FirePro W9100 has been announced, bringing the Hawaii architecture to non-gaming markets. First seen in the Radeon R9 series of graphics cards, it has the capacity for 5 TeraFLOPs of single-precision (32-bit) performance and 2 TeraFLOPs of double-precision (64-bit). The card also has 16GB of GDDR5 memory to support it. From the raw numbers, this is slightly more capacity than either the Titan Black or Quadro K6000 in all categories. It will also support six 4K monitors (or three at 60Hz), per card. AMD supports up to four W9100 cards in a single system.

amd-firepro-w9100.jpg

Professional users can be looking for several things in their graphics cards: compute performance (either directly or through licensed software such as Photoshop, Premiere, Blender, Maya, and so forth), several high-resolution monitors (or digital signage units), and/or a lot of graphics performance. The W9100 is basically the top of the stack which covers all three of these requirements.

amd-firepro-w9100-2.jpg

AMD also announced a system branding initiative called, "AMD FirePro Ultra Workstation". They currently have five launch partners, Supermicro, Boxx, Tarox, Silverdraft, and Versatile Distribution Services, which will have workstations available under this program. The list of components for a "Recommend" certification is: two eight-core 2.6 GHz CPUs, 32GB of RAM, four PCIe 3.0 x16 slots, a 1500W Platinum PSU, and a case with nine expansion slots (to allow four W9100 GPUs along with one SSD or SDI interface card).

amd-firepro-w9100-3.jpg

Also, while the company has heavily discussed OpenCL in their slide deck, they have not mentioned specific versions. As such, I will assume that the FirePro W9100 supports OpenCL 1.2, like the R9-series, and not OpenCL 2.0 which was ratified back in November. This is still a higher conformance level than NVIDIA, which is at OpenCL 1.1.

Currently no word about pricing or availability.

Source: AMD

NVIDIA SHIELD: New Features and Promotional Price Cut

Subject: General Tech, Graphics Cards, Mobile | March 25, 2014 - 03:01 PM |
Tagged: shield, nvidia

The SHIELD from NVIDIA is getting a software update which advances GameStream, TegraZone, and the Android OS, itself, to KitKat. Personally, the GameStream enhancements seem most notable as it now allows users to access their home PC's gaming content outside of the home, as if it were a cloud server (but some other parts were interesting, too). Also, from now until the end of April, NVIDIA has temporarily cut the price down to $199.

nvidia-shield-gamestream-01.jpg

Going into more detail: GameStream, now out of Beta, will stream games which are rendered on your gaming PC to your SHIELD. Typically, we have seen this through "cloud" services, such as OnLive and GaiKai, which allow access to a set of games that run on their servers (with varying license models). The fear with these services is the lack of ownership, but the advantage is that the slave device just needs enough power to decode an HD video stream.

nvidia-shield-gamestream-02.jpg

In NVIDIA's case, the user owns both server (their standard NVIDIA-powered gaming PC, which can now be a laptop) and target device (the SHIELD). This technology was once limited to your own network (which definitely has its uses, especially for the SHIELD as a home theater device) but now can also be exposed over the internet. For this technology, NVIDIA recommends 5 megabit upload and download speeds - which is still a lot of upload bandwidth, even for 2014. In terms of performance, NVIDIA believes that it should live up to expectations set by their GRID. I do not have any experience with this, but others on the conference call took it as good news.

As for content, NVIDIA has expanded the number of supported titles to over a hundred, including new entries: Assassin's Creed IV, Batman: Arkham Origins, Battlefield 4, Call of Duty: Ghosts, Daylight, Titanfall, and Dark Souls II. They also claim that users can add other apps which are not officially supported, Halo 2: Vista was mentioned as an example, for streaming. FPS and Bitrate can now be set by the user. A bluetooth mouse and keyboard can also be paired to SHIELD for that input type through GameStream.

nvidia-shield-checkbox.jpg

Yeah, I don't like checkbox comparisons either. It's just a summary.

A new TegraZone was also briefly mentioned. Its main upgrade was apparently its library interface. There has also been a number of PC titles ported to Android recently, such as Mount and Blade: Warband.

The update is available now and the $199 promotion will last until the end of April.

Source: NVIDIA
Manufacturer: ASUS

Introduction and Technical Specifications

Introduction

02-card-profile.jpg

Courtesy of ASUS

The ASUS ROG Poseidon GTX 780 video card is the latest incarnation of the Republic of Gamer (ROG) Poseidon series. Like the previous Poseidon series products, the Poseidon GTX 780 features a hybrid cooler, capable of air and liquid-based cooling for the GPU and on board components. The AUS ROG Poseidon GTX 780 graphics card comes with an MSRP of $599, a premium price for a premium card .

03-fly-apart-image.jpg

Courtesy of ASUS

In designing the Poseidon GTX 780 graphics card, ASUS packed in many of premium components you would normally find as add-ons. Additionally, the card features motherboard quality power components, featuring a 10 phase digital power regulation system using ASUS DIGI+ VRM technology coupled with Japanese black metallic capacitors. The Poseidon GTX 780 has the following features integrated into its design: DisplayPort output port, HDMI output port, dual DVI ports (DVI-D and DVI-I type ports), aluminum backplate, integrated G 1/4" threaded liquid ports, dual 90mm cooling fans, 6-pin and 8-pin PCIe-style power connectors, and integrated power connector LEDs and ROG logo LED.

Continue reading our review of the ASUS ROG Poseidon GTX 780 graphics card!

Author:
Manufacturer: NVIDIA

DX11 could rival Mantle

The big story at GDC last week was Microsoft’s reveal of DirectX 12 and the future of the dominant API for PC gaming.  There was plenty of build up to the announcement with Microsoft’s DirectX team posting teasers and starting up a Twitter account of the occasion. I hosted a live blog from the event which included pictures of the slides. It was our most successful of these types of events with literally thousands of people joining in the conversation. Along with the debates over the similarities of AMD’s Mantle API and the timeline for DX12 release, there are plenty of stories to be told.

After the initial session, I wanted to setup meetings with both AMD and NVIDIA to discuss what had been shown and get some feedback on the planned direction for the GPU giants’ implementations.  NVIDIA presented us with a very interesting set of data that both focused on the future with DX12, but also on the now of DirectX 11.

15.jpg

The reason for the topic is easy to decipher – AMD has built up the image of Mantle as the future of PC gaming and, with a full 18 months before Microsoft’s DirectX 12 being released, how developers and gamers respond will make an important impact on the market. NVIDIA doesn’t like to talk about Mantle directly, but it’s obvious that it feels the need to address the questions in a roundabout fashion. During our time with NVIDIA’s Tony Tamasi at GDC, the discussion centered as much on OpenGL and DirectX 11 as anything else.

What are APIs and why do you care?

For those that might not really understand what DirectX and OpenGL are, a bit of background first. APIs (application programming interface) are responsible for providing an abstraction layer between hardware and software applications.  An API can deliver consistent programming models (though the language can vary) and do so across various hardware vendors products and even between hardware generations.  They can provide access to feature sets of hardware that have a wide range in complexity, but allow users access to hardware without necessarily knowing great detail about it.

Over the years, APIs have developed and evolved but still retain backwards compatibility.  Companies like NVIDIA and AMD can improve DirectX implementations to increase performance or efficiency without adversely (usually at least) affecting other games or applications.  And because the games use that same API for programming, changes to how NVIDIA/AMD handle the API integration don’t require game developer intervention.

With the release of AMD Mantle, the idea of a “low level” API has been placed in the minds of gamers and developers.  The term “low level” can mean many things, but in general it is associated with an API that is more direct, has a thinner set of abstraction layers, and uses less translation from code to hardware.  The goal is to reduce the amount of overhead (performance hit) that APIs naturally impair for these translations.  With additional performance available, the CPU cycles can be used by the program (game) or be slept to improve battery life. In certain cases, GPU throughput can increase where the API overhead is impeding the video card's progress.

Passing additional control to the game developers, away from the API or GPU driver developers, gives those coders additional power and improves the ability for some vendors to differentiate. Interestingly, not all developers want this kind of control as it requires more time, more development work, and small teams that depend on that abstraction to make coding easier will only see limited performance advantages.

The reasons for this transition to a lower level API is being driven the by widening gap of performance between CPU and GPUs.  NVIDIA provided the images below.

04.jpg

On the left we see performance scaling in terms of GFLOPS and on the right the metric is memory bandwidth. Clearly the performance of NVIDIA's graphics chips has far outpaced (as have AMD’s) what the best Intel desktop processor have been able and that gap means that the industry needs to innovate to find ways to close it.

Continue reading NVIDIA Talks DX12, DX11 Efficiency Improvements!!!

NVIDIA GPUs pre-Fermi Are End of Life After 340.xx Drivers

Subject: General Tech, Graphics Cards | March 22, 2014 - 12:43 AM |
Tagged: nvidia

NVIDIA's Release 340.xx GPU drivers for Windows will be the last to contain "enhancements and optimizations" for users with video cards based on architectures before Fermi. While NVIDIA will provided some extended support for 340.xx (and earlier) drivers until April 1st, 2016, they will not be able to install Release 343.xx (or later) drivers. Release 343 will only support Fermi, Kepler, and Maxwell-based GPUs.

nvidia-geforce.png

The company has a large table on their CustHelp website filled with product models that are pining for the fjords. In short, if the model is 400-series or higher (except the GeForce 405) then it is still fully supported. If you do have the GeForce 405, or anything 300-series and prior, then GeForce Release 340.xx drivers will be the end of the line for you.

As for speculation, Fermi was the first modern GPU architecture for NVIDIA. It transitioned to standards-based (IEEE 754, etc.) data structures, introduced L1 and L2 cache, and so forth. From our DirectX 12 live blog, we also noticed that the new graphics API will, likewise, begin support at Fermi. It feels to me that NVIDIA, like Microsoft, wants to shed the transition period and work on developing a platform built around that baseline.

Source: NVIDIA

ASUS custom built ROG MARS GTX 760

Subject: Graphics Cards | March 21, 2014 - 02:33 PM |
Tagged: ROG MARS 760, nvidia, gtx 760, GK104, asus

If you can afford to spend $1000 or more on a GPU the ASUS ROG MARS GTX 760 is an interesting choice.  The two GTX 760 cores on this card are not modified as we have seen on some other two GPU cards, indeed ASUS even overclocked them to a base of 1006MHz and a boost clock of 1072MHz.  Ryan reviewed this card back in December, awarding it with a Gold and [H]ard|OCP is revisiting this card with a new driver and a different lineup of games.  They also pass this unique card from ASUS a Gold after it finished stomping on AMD and the GTX 780Ti.

1395035044ISFXADrH5Z_1_6_l.jpg

"The ASUS ROG MARS 760 is one of the most unique custom built video cards out on the market today. ASUS has designed a video card sporting dual NVIDIA GTX 760 GPUs on a single video card and given gamers something that didn't exist before in the market place. We will find out how it compares with the fastest video cards out there."

Here are some more Graphics Card articles from around the web:

Graphics Cards

 

Source: [H]ard|OCP

Microsoft DirectX 12 Live Blog Recap

Subject: Graphics Cards | March 21, 2014 - 02:25 PM |
Tagged: dx12, DirectX, DirectX 12, GDC, gdc 14, nvidia, Intel, amd, qualcomm, live, live blog

We had some requests for a permanent spot for the live blog images and text from this week's GDC 14 DirectX 12 reveal.  Here it is included below!!

  Microsoft DirectX 12 Announcement Live Blog (03/20/2014) 
9:53
Ryan Shrout: 

Hi everyone!

 
 
Thursday March 20, 2014 9:53 Ryan Shrout
9:53
Ryan Shrout: 

We are just about ready to get started - people are filing in now.

 
 
Thursday March 20, 2014 9:53 Ryan Shrout
9:53
[Comment From GuestGuest: ] 

?video

 
 
Thursday March 20, 2014 9:53 Guest
9:53
Ryan Shrout: 

Sorry, no video for this. They wouldn't allow us to record or stream.

 
 
Thursday March 20, 2014 9:53 Ryan Shrout
9:55
[Comment From GuestGuest: ] 

kk, no worries

 
 
Thursday March 20, 2014 9:55 Guest
9:55
[Comment From GuestGuest: ] 

Pictures?

 
 
Thursday March 20, 2014 9:55 Guest
9:55
Ryan Shrout: 

Yup!

 
 
Thursday March 20, 2014 9:55 Ryan Shrout
9:59
Ryan Shrout: 

Just testing out photos. I promise the others will be more clear.

 
 
Thursday March 20, 2014 9:59 Ryan Shrout
9:59
Ryan Shrout
 
 
 
Thursday March 20, 2014 9:59 
10:00
[Comment From SebastianSebastian: ] 

Looks like it's a very small event

 
 
Thursday March 20, 2014 10:00 Sebastian
10:00
Ryan Shrout: 

The room is much smaller than it should be. Line was way too long for a room like this.

 
 
Thursday March 20, 2014 10:00 Ryan Shrout
10:01
Josh Walrath: 

that is a super small room for such an event. Especially considering the online demand for details!

 
 
Thursday March 20, 2014 10:01 Josh Walrath
10:02
Ryan Shrout
 
Qualcomm's Eric Demers, AMD's Raja Koduri, NVIDIA's Tony Tamasi.
 
 
Thursday March 20, 2014 10:02 
10:03
Ryan Shrout: 

And we are starting!

 
 
Thursday March 20, 2014 10:03 Ryan Shrout
10:03
Josh Walrath: 

Have those boys gotten their knives out yet. Are the press circling them and snapping their fingers?

 
 
Thursday March 20, 2014 10:03 Josh Walrath
10:03
Ryan Shrout: 

Going over a history of DX.

 
 
Thursday March 20, 2014 10:03 Ryan Shrout
10:03
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:03 
10:04
Ryan Shrout: 

Talking about the development process.

 
 
Thursday March 20, 2014 10:04 Ryan Shrout
10:04
Ryan Shrout: 

All partner base.

 
 
Thursday March 20, 2014 10:04 Ryan Shrout
10:04
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:04 
10:05
[Comment From GuestGuest: ] 

why cant I comment ?

 
 
Thursday March 20, 2014 10:05 Guest
10:05
Ryan Shrout: 

GPU performance is "embarrassing parallel" statement here.

 
 
Thursday March 20, 2014 10:05 Ryan Shrout
10:05
Scott Michaud: 

You can, we just need to publish them. And there's *a lot* of comments.

 
 
Thursday March 20, 2014 10:05 Scott Michaud
10:05
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:05 
10:05
Josh Walrath: 

We see everything, Peter.

 
 
Thursday March 20, 2014 10:05 Josh Walrath
10:05
Ryan Shrout: 

CPU performance has not improved at the same rate. This difference rate of increase is a big challenge for DX.

 
 
Thursday March 20, 2014 10:05 Ryan Shrout
10:06
Ryan Shrout: 

Third point has been a challenge, until now.

 
 
Thursday March 20, 2014 10:06 Ryan Shrout
10:07
Ryan Shrout: 

What do developers want? List similar to what AMD presented with Mantle.

 
 
Thursday March 20, 2014 10:07 Ryan Shrout
10:07
Ryan Shrout: 

DX12 "is no dot release"

 
 
Thursday March 20, 2014 10:07 Ryan Shrout
10:08
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:08 
10:08
Ryan Shrout: 

It faster, more direct. ha ha.

 
 
Thursday March 20, 2014 10:08 Ryan Shrout
10:08
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:08 
10:08
Ryan Shrout: 

Xbox One games will see improved performance. Coming to all MS platforms. PC, mobile too.

 
 
Thursday March 20, 2014 10:08 Ryan Shrout
10:08
Josh Walrath: 

Oh look, mobile!

 
 
Thursday March 20, 2014 10:08 Josh Walrath
10:09
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:09 
10:09
Ryan Shrout: 

New tools are a requirement.

 
 
Thursday March 20, 2014 10:09 Ryan Shrout
10:09
Josh Walrath: 

We finally have a MS answer to OpenGL ES.

 
 
Thursday March 20, 2014 10:09 Josh Walrath
10:09
Scott Michaud: 

Hmm, none of the four pictures in the bottom is a desktop or Laptop.

 
 
Thursday March 20, 2014 10:09 Scott Michaud
10:09
Ryan Shrout: 

D3D 12 is the first version to go much lower level.

 
 
Thursday March 20, 2014 10:09 Ryan Shrout
10:09
[Comment From GuestGuest: ] 

The last one is a desktop...

 
 
Thursday March 20, 2014 10:09 Guest
10:10
Scott Michaud: 

Huh, thought it was TV. My mistake.

 
 
Thursday March 20, 2014 10:10 Scott Michaud
10:10
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:10 
10:10
Ryan Shrout: 

Yeah, desktop PC is definitely on the list here guys.

 
 
Thursday March 20, 2014 10:10 Ryan Shrout
10:11
Ryan Shrout: 

Going to show us some prototypes.

 
 
Thursday March 20, 2014 10:11 Ryan Shrout
10:11
Ryan Shrout: 

Ported latest 3DMark.

 
 
Thursday March 20, 2014 10:11 Ryan Shrout
10:12
Ryan Shrout: 

In DX11, one core is doing most of the work.

 
 
Thursday March 20, 2014 10:12 Ryan Shrout
10:12
Ryan Shrout: 

on d3d12, overall CPU utilization is down 50%

 
 
Thursday March 20, 2014 10:12 Ryan Shrout
10:13
Ryan Shrout: 

Also, the workload is more spread out.

 
 
Thursday March 20, 2014 10:13 Ryan Shrout
10:13
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:13 
10:13
Ryan Shrout: 

Interesting data for you all!!

 
 
Thursday March 20, 2014 10:13 Ryan Shrout
10:13
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:13 
10:14
Ryan Shrout: 

Grouping entire pipeline state into state objects. These can be mapped very efficiently to GPU hardware.

 
 
Thursday March 20, 2014 10:14 Ryan Shrout
10:14
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:14 
10:15
Ryan Shrout: 

"Solved" multi-threaded scalability.

 
 
Thursday March 20, 2014 10:15 Ryan Shrout
10:15
Scott Michaud: 

Hmm, from ~8ms to ~4. That's an extra 4ms for the GPU to work. 20 GFLOPs for a GeForce Titan.

 
 
Thursday March 20, 2014 10:15 Scott Michaud
10:15
[Comment From JayJay: ] 

Multicore Scalability.... Seems like a big deal when you have 6-8 cores!

 
 
Thursday March 20, 2014 10:15 Jay
10:16
Josh Walrath: 

It is a big deal for the CPU guys.

 
 
Thursday March 20, 2014 10:16 Josh Walrath
10:16
Ryan Shrout: 

D3D12 allows apps to control graphics memory better.

 
 
Thursday March 20, 2014 10:16 Ryan Shrout
10:16
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:16 
10:17
Ryan Shrout: 

API is now much lower level. Application tracks pipeline status, not the API.

 
 
Thursday March 20, 2014 10:17 Ryan Shrout
10:17
[Comment From JimJim: ] 

20 GFlops from a Titan? Stock Titan gets around 5 ATM.

 
 
Thursday March 20, 2014 10:17 Jim
10:17
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:17 
10:18
Ryan Shrout: 

Less API and driver tracking universally. More more predictability.

 
 
Thursday March 20, 2014 10:18 Ryan Shrout
10:18
Ryan Shrout: 

This is targeted at the smartest developers, but gives you unprecedented performance.

 
 
Thursday March 20, 2014 10:18 Ryan Shrout
10:18
Ryan Shrout: 

Also planning to advance state of rendering features. Feature level 12.

 
 
Thursday March 20, 2014 10:18 Ryan Shrout
10:19
Scott Michaud: 

Titan gets around ~5 Teraflops, actually... if it is fully utilized. I'm saying that an extra 4ms is an extra 20 GFlops per frame.

 
 
Thursday March 20, 2014 10:19 Scott Michaud
10:19
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:19 
10:19
Josh Walrath: 

Titan is around 5 TFlops total, that 20 GFLOPS is potential performance in the time gained by optimizations.

 
 
Thursday March 20, 2014 10:19 Josh Walrath
10:19
Ryan Shrout: 

Better collision and culling

 
 
Thursday March 20, 2014 10:19 Ryan Shrout
10:19
Ryan Shrout: 

Constantly working with GPU vendors to find new ways to render.

 
 
Thursday March 20, 2014 10:19 Ryan Shrout
10:20
Ryan Shrout: 

Forza 5 on stage now. Strictly console developer.

 
 
Thursday March 20, 2014 10:20 Ryan Shrout
10:20
[Comment From Lewap PawelLewap Pawel: ] 

So 20GFLOPS per frame is 20x60 = 1200GFLOPS/sec? 20% improvement?

 
 
Thursday March 20, 2014 10:20 Lewap Pawel
10:21
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:21 
10:21
Scott Michaud: 

Not quite, because we don't know how many FPS we had originally.

 
 
Thursday March 20, 2014 10:21 Scott Michaud
10:21
Ryan Shrout: 

Talking about porting the game to D3D12

 
 
Thursday March 20, 2014 10:21 Ryan Shrout
10:22
Ryan Shrout: 

4 man-months effort to port core rendering engine.

 
 
Thursday March 20, 2014 10:22 Ryan Shrout
10:22
Ryan Shrout: 

Demo time!

 
 
Thursday March 20, 2014 10:22 Ryan Shrout
10:22
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:22 
10:22
Ryan Shrout: 

Rendering at static 60 FPS.

 
 
Thursday March 20, 2014 10:22 Ryan Shrout
10:23
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:23 
10:23
Ryan Shrout: 

Bundles allows for instancing but with variance.

 
 
Thursday March 20, 2014 10:23 Ryan Shrout
10:24
Ryan Shrout: 

Resource lifetime, track memory directly. No longer have D3D tracking that lifetime, much cheaper on resources.

 
 
Thursday March 20, 2014 10:24 Ryan Shrout
10:24
Ryan Shrout: 

"It's all up to us, and that's how we like it."

 
 
Thursday March 20, 2014 10:24 Ryan Shrout
10:24
Ryan Shrout: 

Does anyone else here worry that DX12 might leave out some smaller devs that can't go so low level?

 
 
Thursday March 20, 2014 10:24 Ryan Shrout
10:25
Josh Walrath: 

I would say that depends on the quality of tools that MS provides, as well as IHV support.

 
 
Thursday March 20, 2014 10:25 Josh Walrath
10:25
Scott Michaud: 

Not really, for me. The reason why they can go so much lower these days is because what is lower is more consistent.

 
 
Thursday March 20, 2014 10:25 Scott Michaud
10:26
Ryan Shrout: 

And now back to info. Will you have to buy new hardware? I would say no since they just showed Xbox One... lol

 
 
Thursday March 20, 2014 10:26 Ryan Shrout
10:26
[Comment From killeakkilleak: ] 

Small devs will use an Engine, not make their own.

 
 
Thursday March 20, 2014 10:26 killeak
10:26
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:26 
10:26
Ryan Shrout: 

On stage now is Raja Koduri from AMD.

 
 
Thursday March 20, 2014 10:26 Ryan Shrout
10:27
Scott Michaud: 

Not true at all, actually. Just look at Frictional (Amnesia). They made their own engine tailored for what their game needed.

 
 
Thursday March 20, 2014 10:27 Scott Michaud
10:27
Ryan Shrout: 

AMD has been working very closely with DX12. Heh.

 
 
Thursday March 20, 2014 10:27 Ryan Shrout
10:27
Josh Walrath: 

Shocking!

 
 
Thursday March 20, 2014 10:27 Josh Walrath
10:28
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:28 
10:28
Josh Walrath: 

Strike a pose!

 
 
Thursday March 20, 2014 10:28 Josh Walrath
10:28
Ryan Shrout: 

There is tension: AMD is trying to push hw forward, MS is trying to push their platform forward.

 
 
Thursday March 20, 2014 10:28 Ryan Shrout
10:28
Ryan Shrout: 

Very honest assessment of the current setup between AMD, NVIDIA, MS.

 
 
Thursday March 20, 2014 10:28 Ryan Shrout
10:28
[Comment From GuestGuest: ] 

Scott, with the recent changes with CryEngine, UE4 going subscription based more Indies might just go that route.

 
 
Thursday March 20, 2014 10:28 Guest
10:28
Ryan Shrout: 

DX12 is an area where they had the least tension in Raja's history in this field.

 
 
Thursday March 20, 2014 10:28 Ryan Shrout
10:29
Scott Michaud: 

Definitely. But that is not the same thing as saying that indies will not make their own engine.

 
 
Thursday March 20, 2014 10:29 Scott Michaud
10:29
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:29 
10:29
Ryan Shrout: 

Key is that current users get benefit with this API on day 1.

 
 
Thursday March 20, 2014 10:29 Ryan Shrout
10:29
Ryan Shrout: 

"Like getting 4 generations of hardware ahead."

 
 
Thursday March 20, 2014 10:29 Ryan Shrout
10:29
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:29 
10:31
Josh Walrath: 

That answers a few of the burning questions!

 
 
Thursday March 20, 2014 10:31 Josh Walrath
10:31
Ryan Shrout: 

Up now is Eric Mentzer from Intel.

 
 
Thursday March 20, 2014 10:31 Ryan Shrout
10:31
[Comment From KevKev: ] 

Thank you! Great news guys!

 
 
Thursday March 20, 2014 10:31 Kev
10:31
Scott Michaud: 

You're welcome! : D

 
 
Thursday March 20, 2014 10:31 Scott Michaud
10:32
[Comment From JimJim: ] 

OH, intel and AMD in the same room....

 
 
Thursday March 20, 2014 10:32 Jim
10:32
Scott Michaud: 

Intel, AMD, NVIDIA, and Qualcomm in the same room...

 
 
Thursday March 20, 2014 10:32 Scott Michaud
10:32
Ryan Shrout: 

Intel has made big change in graphics; put a lot more focus on it with tech and process tech.

 
 
Thursday March 20, 2014 10:32 Ryan Shrout
10:32
Josh Walrath: 

DX12 will enhance any modern graphics chip. Driver support from IHVs will be key to enable those features. This is a massive change in how DX addresses the GPU, rather than (so far) the GPU adding features.

 
 
Thursday March 20, 2014 10:32 Josh Walrath
10:32
[Comment From GuestGuest: ] 

so this means xbox one will get a performance boost?

 
 
Thursday March 20, 2014 10:32 Guest
10:32
Scott Michaud: 

Yes

 
 
Thursday March 20, 2014 10:32 Scott Michaud
10:33
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:33 
10:33
Scott Michaud: 

According to "Benefits of Direct3D 12 will extend to Xbox One", at least.

 
 
Thursday March 20, 2014 10:33 Scott Michaud
10:33
Ryan Shrout: 

Intel commits to having Haswell support DX12 at launch.

 
 
Thursday March 20, 2014 10:33 Ryan Shrout
10:34
Ryan Shrout: 

BTW - thanks to everyone for stopping by the live blog!! :)

 
 
Thursday March 20, 2014 10:34 Ryan Shrout
10:34
Josh Walrath: 

Just to reiterate... PS4 utilizes OpenGL, not DX. This change will not affect PS4. Changes to OpenGL will only improve PS4 performance.

 
 
Thursday March 20, 2014 10:34 Josh Walrath
10:34
Ryan Shrout: 

If you like this kind of stuff, check out our weekly podcast! http://pcper.com/podcast

 
 
Thursday March 20, 2014 10:34 Ryan Shrout
10:34
Ryan Shrout: 

No mention of actual DX12 launch time quite yet...

 
 
Thursday March 20, 2014 10:34 Ryan Shrout
10:34
[Comment From MagnarockMagnarock: ] 

Finish?

 
 
Thursday March 20, 2014 10:34 Magnarock
10:34
Ryan Shrout: 

And Intel is gone. Short and sweet.

 
 
Thursday March 20, 2014 10:34 Ryan Shrout
10:34
Scott Michaud: 

Still have NVIDIA and Qualcomm, at least.

 
 
Thursday March 20, 2014 10:34 Scott Michaud
10:35
Scott Michaud: 

So -- not finished.

 
 
Thursday March 20, 2014 10:35 Scott Michaud
10:35
Ryan Shrout: 

Up next is Tony Tamasi from NVIDIA.

 
 
Thursday March 20, 2014 10:35 Ryan Shrout
10:35
Ryan Shrout: 

NVIDIA has been working with MS since the inception of DX12. Still don't know when that is...

 
 
Thursday March 20, 2014 10:35 Ryan Shrout
10:35
[Comment From AlexAlex: ] 

PS4 doesn't use OpenGL, but custom APIs instead...

 
 
Thursday March 20, 2014 10:35 Alex
10:35
Scott Michaud: 

True, it's not actually OpenGL... but is heavily heavily based on OpenGL.

 
 
Thursday March 20, 2014 10:35 Scott Michaud
10:36
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:36 
10:36
Ryan Shrout: 

They think it should be done with standards so there is no fragmentation.

 
 
Thursday March 20, 2014 10:36 Ryan Shrout
10:36
Ryan Shrout: 

lulz.

 
 
Thursday March 20, 2014 10:36 Ryan Shrout
10:37
Scott Michaud: 

Because everything that ends in "x" is all about no fragmentation :p

 
 
Thursday March 20, 2014 10:37 Scott Michaud
10:37
Ryan Shrout: 

NVIDIA will support DX12 on Fermi, Kepler, Maxwell and forward!

 
 
Thursday March 20, 2014 10:37 Ryan Shrout
10:37
Ryan Shrout: 

For developers that want to get down deep and manage all of this, DX12 is going to be really exciting.

 
 
Thursday March 20, 2014 10:37 Ryan Shrout
10:38
Ryan Shrout: 

NVIDIA represents about 55% of the install base.

 
 
Thursday March 20, 2014 10:38 Ryan Shrout
10:38
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:38 
10:38
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:38 
10:39
Ryan Shrout: 

Developers already have DX12 drivers. The Forza demo was running on NVIDIA!!!

 
 
Thursday March 20, 2014 10:39 Ryan Shrout
10:39
Ryan Shrout: 

Holy crap, that wasn't on an Xbox One!!

 
 
Thursday March 20, 2014 10:39 Ryan Shrout
10:39
Scott Michaud: 

Fermi and forward... aligning well with the start of their compute-based architectures... using IEEE standards (etc). Makes perfect sense. Also might help explain why pre-Fermi is deprecated after GeForce 340 drivers...

 
 
Thursday March 20, 2014 10:39 Scott Michaud
10:40
Ryan Shrout: 

Support quote from Tim Sweeney.

 
 
Thursday March 20, 2014 10:40 Ryan Shrout
10:41
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:41 
10:41
[Comment From CrackolaCrackola: ] 

Any current NVIDIA cards DX12 ready? Titan, etc?

 
 
Thursday March 20, 2014 10:41 Crackola
10:41
Ryan Shrout: 

Up now is Eric Demers from Qualcomm.

 
 
Thursday March 20, 2014 10:41 Ryan Shrout
10:42
Scott Michaud: 

NVIDIA said Fermi, Kepler, and Maxwell will be DX12-ready. So like... almost everything since GeForce 400... almost.

 
 
Thursday March 20, 2014 10:42 Scott Michaud
10:42
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:42 
10:42
Ryan Shrout: 

Qualcomm has been working with MS on mobile graphics since there WAS mobile graphics.

 
 
Thursday March 20, 2014 10:42 Ryan Shrout
10:42
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:42 
10:42
Ryan Shrout: 

Most windows phones are powered by Snapdragon.

 
 
Thursday March 20, 2014 10:42 Ryan Shrout
10:42
Josh Walrath: 

We currently don't know what changes in Direct3D will be brought to the table, all we are seeing here is how they are changing the software stack to more efficiently use modern GPUs. This does not mean that all current DX11 hardware will fully support the DX12 specification when it comes to D3D, Direct Compute, etc.

 
 
Thursday March 20, 2014 10:42 Josh Walrath
10:43
Ryan Shrout: 

DX12 will improve power efficiency by reducing overhead.

 
 
Thursday March 20, 2014 10:43 Ryan Shrout
10:43
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:43 
10:44
Ryan Shrout: 

Perf will improve on mobile device as well, of course. But gaming for longer periods on battery life is biggest draw.

 
 
Thursday March 20, 2014 10:44 Ryan Shrout
10:45
Ryan Shrout: 

Portability - bringing titles from the PC to Xbox to mobile platform will be much easier.

 
 
Thursday March 20, 2014 10:45 Ryan Shrout
10:45
[Comment From David UyDavid Uy: ] 

I think all Geforce 400 series is Fermi. so - Geforce 400 and above.

 
 
Thursday March 20, 2014 10:45 David Uy
10:45
Scott Michaud: 

I think the GeForce 405 is the only exception...

 
 
Thursday March 20, 2014 10:45 Scott Michaud
10:45
Ryan Shrout: 

Off goes Eric.

 
 
Thursday March 20, 2014 10:45 Ryan Shrout
10:45
Ryan Shrout: 

MS back on stage.

 
 
Thursday March 20, 2014 10:45 Ryan Shrout
10:46
Ryan Shrout: 

And now a group picture lol.

 
 
Thursday March 20, 2014 10:46 Ryan Shrout
10:46
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:46 
10:47
Ryan Shrout: 

By the time they ship, 50% of all PC gamers will be DX12 capable.

 
 
Thursday March 20, 2014 10:47 Ryan Shrout
10:47
Ryan Shrout: 

Ouch, targeting Holiday 2015 games.

 
 
Thursday March 20, 2014 10:47 Ryan Shrout
10:48
Ryan Shrout: 

Early access coming later this year.

 
 
Thursday March 20, 2014 10:48 Ryan Shrout
10:48
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:48 
10:48
Josh Walrath: 

Yeah, this is a pretty big sea change.

 
 
Thursday March 20, 2014 10:48 Josh Walrath
10:48
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:48 
10:49
Ryan Shrout
 
 
 
Thursday March 20, 2014 10:49 
10:49
Scott Michaud: 

50% of PC Gamers sounds like they're projecting NOT Windows 7.

 
 
Thursday March 20, 2014 10:49 Scott Michaud
10:49
Ryan Shrout: 

They are up for Q&A not sure how informative they will be...

 
 
Thursday March 20, 2014 10:49 Ryan Shrout
10:50
Josh Walrath: 

OS support? Extension changes to D3D/Direct Compute?

 
 
Thursday March 20, 2014 10:50 Josh Walrath
10:50
Ryan Shrout: 

Windows 7 support? Won't be announcing anything today but they understand the request.

 
 
Thursday March 20, 2014 10:50 Ryan Shrout
10:51
Ryan Shrout: 

Q: What about support for multi-GPU? They will have a way to target specific GPUs in a system.

 
 
Thursday March 20, 2014 10:51 Ryan Shrout
10:51
Ryan Shrout: 

This session is wrapping up for now!

 
 
Thursday March 20, 2014 10:51 Ryan Shrout
10:51
Ryan Shrout: 

Looks like we are light on details but we'll be catching more sessions today so check back on http://www.pcper.com/

 
 
Thursday March 20, 2014 10:51 Ryan Shrout
10:52
Scott Michaud: 

"a way to target specific GPUs in a system" this sounds like developers can program their own Crossfire/SLi methods, like OpenCL and Mantle.

 
 
Thursday March 20, 2014 10:52 Scott Michaud
10:52
Ryan Shrout: 

Also, again, if you want more commentary on DX12 and PC hardware, check out our weekly podcast! http://www.pcper.com/podcast!

 
 
Thursday March 20, 2014 10:52 Ryan Shrout
10:52
Ryan Shrout: 

Thanks everyone for joining us! We MIGHT live blog the other sessions today, so you can sign up for our mailing list to find out when we go live. http://www.pcper.com/subscribe

 
 
Thursday March 20, 2014 10:52 Ryan Shrout
10:57
Scott Michaud: 

Apparently NVIDIA's blog says DX12 discussion begun more than four years ago "with discussions about reducing resource overhead". They worked for a year to deliver "a working design and implementation of DX12 at GDC".

 
 
Thursday March 20, 2014 10:57 Scott Michaud
11:03