AMD Mantle Logo
Hardware

[Editorial] The Game Developers Conference is coming up and both DirectX and OpenGL have queued up some very very interesting reveals. These reveals, as you might have guessed are both dedicated at showing their respective solution to CPU Rendering Overhead and Low Level Access (Microsoft Only). If their claims have any weight what so ever, it wold effectively make Mantle Dead on Arrival. Mantle Dead on Arrival

Microsoft and Nvidia Queue Solutions to CPU Overhead and Low Level Access on GDC’14 – Mantle could be Dead on Arrival

As you probably already know, Mantle Low Level API improves the performance ingame by reducing the CPU Overhead and allowing more draw calls to be submitted to the GPU in a certain time frame. Basically it massively reduces load on the CPU, greatly helping CPU-Bound situations and Weak-CPU configurations. And of course as the name suggests, gives better access to Hardware. Now it seems that other developers have felt greatly threatened by the same because both DirectX and OpenGL are claiming to have solutions to CPU Overhead and improved Low Level Access. If this is true, then Mantle API would loose its edge entirely because Devs will not risk adapting a new API without major incentive. Here are the scheduled addresses for GDC’14:

Direct3D Futures (Presented by Microsoft)

“..discuss future improvements in Direct3D that will allow developers an unprecedented level of hardware control and reduced CPU rendering overhead across a broad ecosystem of hardware. “

AMD Catlayst Logo
Hardware

As you may know, the AMD Catalyst 14.2 V1.3 Beta Drivers are out and ready to roll. This particular beta driver is all about Thief. On a side note, Battlefield 4 and the Mantle Low Level API get updates too. We have also noticed improvement in Battlefield 4 multiplayer and single player after the 14.2 update. Without further ado the download link and instructions.

1Image Credits: Computerbase.de

AMD Catalyst Catalyst™ 14.2 Beta V1.3 Download and Install Instructions

Now the Drivers in beta, so that means that there are bound to be issues, so keep that in mind. Also most of the Toshiba, Sony Vaio and Panasonic Notebooks are NOT supported and the mobile Radeon series currently will not have mantle support (you can find a cracked version of 14.1 online though). This also enables frame pacing in NON-XDMA Crossfire configurations not to mention the small bug fixes and performance increases. The funnies bug is that in Thief when cross fire or 3D is enabled, the right eye will not render.  Anyways without further ado the download link and install instructions.

Download Link: AMD Catalyst™ 14.2 V1.3 Beta for Windows®

Installing The AMD Catalyst Software Driver

Current driver MUST be uninstalled before updating to AMD Catalyst™ 14.2 Beta V1.3 driver. For detailed instructions on how to correctly uninstall or install the AMD Catalyst Software Suite, please refer to the following support resources:

nvidia_gtx870m
Hardware

I have something very interesting today. The following pictures of the MXM GTX 870M and a GPUZ shot were sent to me by my source. You can see the core codename in the first picture, which, interestingly the does not match anything we know before. Also it is not present in the Techpowerup database. However the chip is undoubtedly an 870M. So what exactly are we looking at?

N15E-GT-A2 GTX 870M Maxwell Mobility

Geforce GTX 870M Die Shot and GPU-Z Leaked – 6GB VRam and Different Codename

Let me state for the record that I am not calling it yet. I will give the reasons later but for the moment I am leaving the decision whether or not we are looking at Maxwell up to the readers. Now let me provide you with what we already know. As you may recall, CLEVO slides were leaked two months back that showed the entire lineup as ‘Maxwell’ along with the codenames of the die. However just soon after another leak confirmed the GTX 880M as nothing more than a re branded Kepler GTX 780M. During this time details of the GTX 870M came to light. One of the important things was that the GTX 870M was going to have 3GB vRAM and not 6GB as the slides had originally said, something which the Techpowerup database confirmed. Also the codename mentioned for the GTX 870M was “N15P-GT-A2″. Remember this. Now observe the picture above and GPUZ screenshot below.

nvidia-geforce-gtx-logo
Hardware

Ok I just received an extremely strange piece of information from not one but two of my sources. The nature of said leak is so weird that i would urge you take it with a bag of salt. My sources (One and Two) tell me that though the first generation Maxwell GTX 750 and GTX 750 Ti do not currently support CUDA 6. They will be getting CUDA 6 Support. However, here comes the confusing part, they will not be getting UVM support. According to the same sources second generation Maxwell will have both CUDA 6 and UVM support.

maxwell03

Maxwell GTX 750 & GTX 750 Ti do not support Unified Virtual Memory – But will support CUDA 6?

From here on, and for this article only I will assume the leak is correct. Lets see, The first thing to note here is that though CUDA 6 has an essential component of UVM, UVM and CUDA 6 Support are not mutually inclusive. Which would help explain why the GTX 750 and GTX 750 Ti are going to have CUDA 6 Suport but no native UVM. Unified Virtual Memory is a great step towards a more unified architecture and memory scaling. Not to mention that it should simplify code by a lot too. Basically as you may already know, UVM is Nvidia’s take on unification architecture as opposed to HSA in AMD’s APUs. Where AMD’s APUS use a ‘truly’ unified architecture, Nvidia is attempting to bridge the gap by a more Soft approach.

nvidia-geforce-gtx-logo
Hardware

Something very interesting cropped up. Something that could change the GPU landscape very quickly and very drastically. If the heading hasn’t tipped you off, the thing is this: Nvidia’s Maxwell GPUs give more profitable hashrates than their AMD counterparts.

Maxwell GTX 750

Maxwell is perfectly poised to Steal AMD’s Mining Buzz – Has hell frozen over?

Ok I have a confession to make. I have speculated on the mining tables tilting in Nvidia’s favor before, but those usually involved Disruptive Business Changes such as CUDACoins. I never actually thought Nvidia would be able to faceoff AMD in a pure khash per watt and even khash per price basis. Well, hell apparently has frozen over because the Maxwell GTX 750 Ti has the honor of being the first card to actually be able to do that.

These benchmarks have been taken from TomsHardware :

AMD Kaveri APU
Hardware

An extremely important leak just arrived on our doorstep and (for the first time I think) the leak is provided by Extremetech.com which noticed a very important piece of evidence posted on SA forums. The leak more or less confirms that Excavator cores and the Carrizo APUs will be on the 28nm Process but as always I would recommend that pinch of salt. More interestingly however it also hints that part of the production will shift back to Global Foundries and away from TSMC.

Carrizo APUs with Excavator Cores 28nm Glofo

‘Excavator Cores’ will be on 28nm with ‘Carrizo APU’ – Production Shift to Global Foundries

There has been some serious speculation whether AMD would shift to 20nm for the Carrizo APU and now it seems to have finally been answered. It appears that AMD will not be shifting to 20nm design and instead stick with a 28nm design. Also part of the production (and I don’t know how big that part is) will shift back to Global Foundries.

Excavator Cores 28nm Carrizo APU

nvidia-geforce-gtx-logo
Hardware

Ok the NDA has finally lifted and the official architecture documentation is up. Its finally time to take a look at Maxwell in depth.  It manages to Double the Performance per Watt while staying on the same 28nm Process.  Which, there is no other way to put it, is nothing short of a miracle. Lets see just how exactly it manages that.

Nvidia Coporation

Maxwell 28nm Miracle  – How the Architecture makes Doubling Performance Per Watt Possible

Firstly, as I am sure most of you are aware, Maxwell does not work on SMX units. It works on SMMs, short for Streaming Maxwell Multiprocessors. Each SMM houses 128 CUDA Cores as opposed to the 192 housed by SMXs. Now unlike in the Kepler architecture where the CUDA Cores are housed in a single core fashion, Maxwell houses Cuda Cores in 4 subsets of each SMM. Almost 4 Separate “Cores” within the SMM. Lets call these “Major Cores” (as opposed to CUDA Cores) to avoid confusion. Do realize that this only refers to the 1st Generation of Maxwell and the division by four could change in the next generation. The Major Cores tactic allows Nvidia to achieve much higher efficiency rates and increase performance by 135% Per Core. Take a look at this diagram of Maxwell SMMs.

GeForce GTX 750 Ti
Hardware

Today, NVIDIA officially announces their GeForce GTX 750 Ti and GeForce GTX 750 graphics cards which are based on the first generation Maxwell core architecture. With almost twice the performance per watt of its predecessors, the NVIDIA Maxwell core architecture is going to become one of the most innovative GPU core architecture after Kepler in terms of efficiency.NVIDIA GeForce GTX 750 Ti Front

NVIDIA Launches GeForce GTX 750 Ti and GeForce GTX 750

The first generation Maxwell core would be integrated inside two graphics cards known as the GeForce GTX 750 Ti and the GeForce GTX 750. These are the first mid range cards NVIDIA has launched in their GeForce 700 series lineup since the rest of the graphics cards were based off the GK110 or the GK104 Kepler cores. The Maxwell GPU for both cards is codenamed GM107 so we will talk about it first before moving in the specifications of the cards.

The NVIDIA Maxwell GM107 core is the first of two Maxwell based chips which will be based on the 28nm process . The GM107 which is already available on the GeForce GTX 750 Ti and the GeForce GTX 750 while the other chip is codenamed GM108 and is the base Maxwell model which would be available as a part of NVIDIA’s GeForce 800M mobility lineup. The GM108 would be fused in the entry level mobility chips which include the GeForce GT 840M and possibly a few more models.

GeForce-GTX-Titan_5
Hardware

NVIDIA will also unleash their GeForce GTX Titan Black graphics card today featuring the full compute power of the GK110 core. The GeForce GTX Titan Black is the second generation Titan graphics card featuring the GK110 core with double precision cores unlocked and faster clock speeds compared to the GeForce GTX 780 Ti.GeForce GTX Titan Black Official

NVIDIA GeForce GTX Titan Black Delivers Unprecedented Amounts of Power to Developers and Enthusiasts

NVIDIA had initially launched their GeForce GTX Titan back in February 2013 which was the world’s first graphics card to feature the GK110 core which was at the time being mass produced specifically for server needs. Super-computers such as the Titan which the card was named after was the first to utilize the latest core with its high performance CUDA cores, 2668 in count at the time. The GeForce GTX Titan was a great card but came at a big price so it was mostly a enthusiast only product but still outsold the GeForce GTX 690 which was NVIDIA’s Dual GK104 based card available at the same $999 US price range.

Moving forward, NVIDIA introduced several GK110 based cards, the GeForce GTX 780 and GeForce GTX 780 Ti. The Ti model was the first to introduce the full GK110 core to the masses (but still had gimped double precision performance). The card came at a lower $699 US price range and bested the Titan in performance. AMD on the other hand had some interesting Hawaii models in the pipeline with highly competitive pricing models which was a reason for their popularity among the buyers. The fight between the GeForce GTX 780 Ti and Radeon R9 290X still remains unsettled since both cards are good in their own terms. The GTX 780 Ti is faster in NVIDIA optimized games, runs cooler with a low TDP but costs higher. On the AMD side, the R9 290X performs fantastic for its price and even manages to outperform the 780 Ti in AMD optimized titles but runs really hot and loud plus you can hardly find one in the market due to the crypto mining craze.

Battlefield 4 logo
Hardware

You might recall I have been ranting on and on about how Mantle will work wonders on Multiplayer gaming because the situation is so CPU bound. Well, I have finally gotten my hands on standardized Battlefield 4 Multiplayer Benchmarks with Mantle. Since Multiplayer is an extremely tricky thing to benchmark because the conditions change with every run, these benches are quite unique.

Battlefield 4 Mantle Update

Battlefield 4 Multiplayer Benchmarks with AMD Mantle API Vs Direct X

The benches were given to us by Hardwareluxx.  As always the preliminaries first. The specs of the test rig are as follows:

  • Intel Core i7-3960X 3.3 GHz overclocked to 3.9 GHz
  • ASUS P9X79 Deluxe
  • ADATA XPG Gaming Series Low Voltage 4x 2 GB PC3-12800U CL 9-9-9-24
  • ADATA S510 SSD 60GB
  • Windows 8 Pro 64 bit

A complicated method was used to standardize the multiplayer benchmark. The map was Conquest Large and 64 players connected to it. The test rig was then manipulated by commandline after every 10 minutes going through every combination given below. The result was then extrapolated into an average using the frame response time to give the final output of FPS. Now without ado, the benchmarks, i have added the Single player benches as comparison to emphasize my earlier rants.

nvidia-geforce-gtx-logo
Hardware

The GTX Titan Black’s final specifications have been confirmed as well as its launch date, thanks to Videocardz. The Titan Black, a 6GB and DP unlocked version of the GK110 Core will be launching next week and will make its way into stores quietly.

Nvidia GTX Titan Black

Final Specs of the GTX Titan Black Confirmed – Will Be Launching Silently Next Week

The GTX Titan Black will be launching very silently with no fuss. It will be priced at the $999 mark and will have fully unlocked DP performance of the complete GK110 Core.  Without further ado the final specs of the Nvidia Titan Black:

  • Base Clock: 889 Mhz
  • Boost Clock: 980 Mhz
  • Memory Clock: 7008 Mhz
  • Memory: 6Gb GDDR5
  • 380 Bits
  • 2880 CUDA Cores
  • 240 TMUs
  • 48 ROPs
  • 250W TDP
  • 6pin + 8 Pin Power Connector
AMD_APU_Kaveri_min
Hardware

Well here i have the final Kaveri benchmarks and these are the much demanded HSA benchmarks (kinda late, but better late than never right?). The post is dvided into two sections. Section One will contain Arithmetic and Multimedia benchmarks ie Non-HSA. Then, Let. There. Be. HSA.

AMD Kaveri HSA Features

Non-HSA Kaveri Vs i5-4670K Benchmarks – Arithmetic Functions and Multimedia Crunching

Ok here are the preliminaries:

  • SAMSUNG DDR3 PC3-10600 4GB x2 @ 2133Mhz
  • Catalyst 13.35 Beta (HSA Enabled)
  • Samsung 840 Series SSD 500GB
  • A10-7850k @ 4Ghz
  • A10-6800k @ 4.4Ghz
  • A8-3870k @ 3Ghz
  • Core i5 – 4670k @3.8Ghz
  • Core i3 – 4330 @ 3.5Ghz

As you can probably see the clocks are at stock boosts more or less. These benches are courtesy of the reliable DG’s Nerdy Story.

AMD R9 Series Logo
Hardware

Litecoin, what are we going to do with you. Just recently today the Retail price of the R9 290X hit an all time high of 899.99 Dollars at Newegg. Perhaps even more surprisingly the $900 models soon were ‘Sold Out’ at the same $899.99 Mark. It goes to show how quickly miners will snatch up any high end radeons floating around. Ofcourse the Component Shortage we talked about earlier doesn’t help either.

R9 290X 900 bucksImage Credits Only: The Tech Report, LLC

R9 290X Gets Sold Out on $900 Retail – ASUS Listing Pulled, and Price Bumped Back Down to $700

However, not only have the $900 SKUs all sold out, but it appears that something in the back end took place because a couple of changes have happened.

  • The ASUS Listing is no longer on display.
  • The Gigabyte Listing has been reduced to $699 after sell out.
  • Only two SKUs, one from Visiontek and Sapphire remain, both listed at $699.99
  • All “Sold Out” SKUs are no higher than $799.99 and no lower than $699.99, compared to a few hours earlier when the maxima was $899.99 and minima was $849.99.
nvidia-geforce-gtx-logo
Hardware

NVIDIA’s Maxwell based GeForce GTX 750 Ti and GeForce GTX 750 graphics cards have been detailed in official slides which were leaked by Vitamenese forums. NVIDIA Maxwell is the latest generation of graphics core being developed by the green team which brings enhanced performance and higher power efficiency as detailed in the architecture preview of the Maxwell core.NVIDIA GeForce GTX 750 TiImages are courtesy of Tinhte.vn!

NVIDIA Maxwell GeForce GTX 750 Ti and GTX 750 Official Specs Confirmed

The slides were displayed by NVIDIA APAC at a conference in Ho Chi Minh City, Vietnam. They are quite interesting and shows NVIDIA’s plans for their 2014 GPU lineup which will introduce the latest Maxwell core architecture. It can be seen from the slides that NVIDIA’s inital Maxwell lineup would be based on the first generation of the GPU core which features the SMM (Streaming Multiprocessor Maxwell), The SMM or Streaming Multiprocessor of Maxwell will replace the SMX of Kepler and each of the SMMs are assembled into four blocks yet are defined as part of a single SMM which means the core architecture has got quite dense with Maxwell. Each of these blocks hold 32 CUDA cores so a single SMM with four of these operation units results in 128 CUDA cores. The SMM has 128 CUDA cores compared to 192 CUDA cores on the SMX.

The GM107 GPC ‘Graphics Processing Cluster’ consists of five of these streaming multiprocessors which are connected to a Raster Engine. Each SMM consists of Polymorph Engine 2.0 which includes the Vertex Fetch, Tessellator, Viewport transform, attribute setup and stream output. Each SMM has 8 texture mapping units which equates to 40 on the whole chip and 16 ROPs while connected to two 64-bit memory controllers. More details on the GM107 Maxwell core are available here.

Nvidia
Hardware

Disclaimer: The Conclusion given at the end has been given in a completely non-professional capacity. Though the Guidance has been given to the best of my abilities we cannot be held liable in any way whatsoever.

Nvidia Corporation (NASDAQ: NVDA) just recently outed its Q4 Results for 2014. The reported non-GAAP diluted EPS of $0.32 is an almost full double of the analyst’s estimate of $0.18 (+ $0.14). Jen-Hsun Huang, CEO of Green attributed the results to the PC Gaming Sector which has been responsible for an ‘outstanding year’ for GPU Sales. The Gaming PC Market is shifting to the Mid – High end spectrum while PC sales in general have seen a Year By Year Decline.

Nvidia Corporation (NVDA) Earnings Per Share 14 Cents more than Analysts Expectation – GPU Sales, GRID Computing and Tegra K1 Attributed with the Success

Nvidia Corporation (NVDA)Ok lets start with a little overview. Geforce GTX GPU sale increased by nearly 50% this year. The Tesla K40 the highest end niche product was also launched. Denver was demonstrated and Tegra K1 targeted the auto industry. However operating costs increased and with the Tegra K1 no longer an ace up its sleeve, it remains to be seen how Green will fare.



Creative Commons License This work by WCCF (Pvt) Ltd. is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Permissions beyond the scope of this license may be available on the Terms and Conditions page. WCCF 2011 is designed by UzEE, Inc.
Ran 29 queries in 1.318  seconds.
Find us on Google+