NVIDIA Unleashes Maxwell GM204 Based GeForce GTX 980 and GeForce GTX 970 Graphics Cards

Today, NVIDIA finally unleashes their high-operation GeForce GTX 980 and GeForce GTX 970 graphics bill of fare. NVIDIA'due south latest graphics cards feature the 2nd generation Maxwell architecture which is the almost advanced GPU ever built by NVIDIA featuring keen functioning and delivering college efficiency in terms of ability input. The Maxwell architecture revolutionizes the graphics industry setting new standards for NVIDIA gamers and fans. NVIDIA GeForce GTX 980 Graphics Card

We accept waited for several years for Maxwell to hit the marketplace. While NVIDIA initially released the first cadre every bit the GM107 which was based on the showtime generation architecture blueprint, the GM204 is based on the latest and improved, second generation Maxwell core architecture which adopts some new technologies. Then earlier we go talk nearly the cards, let'due south take a epitomize of the architectural details we have come to know well-nigh GM204.

NVIDIA GM204 GPU Architecture

The GM204 is the heart of the next generation GeForce GTX 980 and GeForce GTX 970 graphics cards. The chip makes use of the second generation Maxwell core architecture that has faster per core functioning than outset generation Maxwell based fries (GM107) which were released with the GeForce GTX 750 and GeForce GTX 750 Ti graphics cards and has several new features which evangelize amend operation and great power efficiency making GeForce GTX 980 one of the most efficient flagship offering in history. NVIDIA has inverse since their Kepler generation of cards. Before Kepler, NVIDIA was known to release cards which ran hot and consumed a ton of power and the failure rates of the previous generation cards were pretty high. Though, NVIDIA did manage to release some cracking cards over the period, the G80 based GeForce 8000 GTX and the cost/performance rex, GTX 460 are yet considered i of the greatest NVIDIA cards that came to market.

Kepler changed sure things, NVIDIA moved away from their branding scheme where users were able to buy HPC chips rebranded for the GeForce audience. The starting time Kepler GPU, the GK104 was branded as the GeForce carte du jour and while information technology was fast, it wasn't the fastest compared to another chip which NVIDIA had in their hands for over a yr. I am talking about the GK110 which was geared towards the professional market place such every bit the Tesla super computer. The GK110 did launch a twelvemonth after but it since then, NVIDIA has configured their core lineup to span two generations, 1 with the gaming minded bit and the follow up would exist the full fledged HPC scrap. At that place'south a reason for NVIDIA to advertise the GeForce GTX Titan Z as a professional and gaming carte fifty-fifty though it clearly has a GeForce name in its branding. The Titan Z makes utilise of two GK110 fries which are the compute crunching beasts compared to the GK104 which focused on gaming features by excluding all the not essential features such as compute. So while the GM104 based GTX 980 will obviously replace GK110 based GTX 780 Ti in branding, the real comparing in branding should exist GM104 versus the GK104. Regardless of this, the GTX 980 is a superb card which beats GK110 on an existing process node.

NVIDIA Maxwell GM204 This is all achieved with the 28nm process node so one can imagine the numbers we can expect when NVIDIA hops to an even lower procedure in the future. Alright, so the GM204 has two variants, the GM204-400 which is fused on the GeForce GTX 980 and the GM204-200 which is fused on the GeForce GTX 970. The fully enabled GM204 chip features 4 GPC (Graphics Processing Clusters) which feature four SMM blocks each. These blocks include four logic units each which consist of 32 cores and then in total, a single SMM unit results in 128 Cores while the 16 blocks available on the GM204-400 chip equate to 2048 CUDA Cores. The GM204-200 has three less SMM units which result in a lower cadre count of 1664 thus making it around as fast every bit the GeForce GTX 780 while the GTX 980 volition tackle the GeForce GTX 980 with a practiced xv-20% performance atomic number 82.

The nigh critical details of the bit are the transistor number and we all remember that the GK110 chip was a performance and computing fauna at vii.08 Billion transistors while the GK104 included iii.54 Billion transistors. The GM204 includes v.2 Billion transistors crammed inside a die that measures around 398 mm2 just two mm2 shy of 400mm2. The GK104 and GK110 mensurate at around 294 mm2 and 581 mm2 respectively. The die size has been increased a lot compared to GK104 and that's the generational predecessor of the card. The GK110 will exist replaced by GM200 but that is far from launch at the moment just NVIDIA has managed to include more on the 28nm procedure yet keeping the power consumption at simply 165W on the GTX 980 and 148W for the GTX 970 which is just mind extraordinary.

The GM204 GPU features 128 texture mapping units which was the standard amount featured on the GK104 but the raster operation units accept been upped from 32 on GTX 680 and 48 on GTX 780 Ti to 64 on the GTX 980 graphics card. This is actually a larger update than GK110 but the GK110 does come with a very high TMU count of 240. NVIDIA compensates this past clocking the GM204 chip hence resulting in a higher per clock performance output when it comes to texture fill charge per unit. Maxwell was also meant to improve the mode GPU handles bandwidth and they are limiting the bandwidth dependancy of their cards by adding more enshroud of ii MB which is 512 KB L2 more than GK110. The GK104 had just 256 KB of L2 cache and so a major update there.

The theoretical compute of the scrap in single precision would be rated around four.vi TFLOPs which is really close to the GK110 which pumps out 5.1 TFLOps while the 1144 GT/s texture fill rate is a scrap depression but the pixel fillrate is considerably college at 72.1 GP/southward compared to 53.3 GP/s on GTX 780 Ti.

NVIDIA has some new software side enhancements through the hardware implemented in Maxwell which include Dynamic Super Resolution which is basically a second version of down sampling that functions to increment video quality at 1080P that matches 4K resolution. In that location'south also Delta Colour Compression which is similar to the colour pinch we saw on AMD'southward Tonga that compresses images to a lossless format so that overall quality is maintained allowing the GPU cadre to read and write the compressed data easily. A more than refined version which saves images in local memory to exist used subsequently to increase retentiveness efficiency is used in Maxwell.

Then there'due south Multi-Pixel Programming Sampling engineering which improves randomization of each sample and reduces quantification artifacts for meliorate geometry processing and anti aliasing filtering. An update on the brandish side is that GeForce GTX 980 adopts the HDMI 2.0 standard which goes in well with the new display standard of 3 Display Ports 1,2, ane DVI, 1 HDMI outputs set by NVIDIA for their flagship offer.

NVIDIA GM204 Block Diagram:

NVIDIA GM204 SMM Unit Block Diagram:

The NVIDIA SMM or SM (Streaming Multiprocessor Maxwell) units are a update over the Kepler SMX. Each logic unit of measurement is split into four parts consisting of 32 cores, each of the SM unit of measurement houses 128 cores. The 128 core count is lower than the 192 Cores featured on the SMX unit on Kepler but do note that the Maxwell second generation cores are a practiced forty+ faster than Kepler cores. The new design also simplifies the architecture and the overall scheduling resulting in a considerable drop in power consumption and delays.

NVIDIA Maxwell GM204 GPU Free energy and Retentiveness Bandwidth Efficiency

Probably 1 of the most major talks surrounding the Maxwell cards were their low retention bus compared to their GK110 based predecessors. The slide posted below conspicuously shows that due to a new and improved ram architecture, NVIDIA has enhanced the bandwidth efficiency where 7.0 Gbps DRAM can deliver an effective throughput of ix.iii Gbps in gaming. Hence even with lower bandwidth, the entire need of the available ring width has gone considerably down which results in better performance throughput and utilization. NVIDIA Maxwell GM204 Memory Bandwidth Reduction

On the other hand, the functioning numbers of Maxwell just continue on getting improve and amend with up to 3 times the energy efficiency of Kepler. Annotation that while people volition recall that NVIDIA should have compensated ability for more operation, the actual fact is that the carte du jour performs good and the lower TDP results in college stability and overclocking numbers from non-reference and custom variants. The Maxwell architecture volition also scale down from Tegra fries all the way to the acme terminate GM200 based HPC parts so free energy efficiency does affair. NVIDIA Maxwell GM204 Power Efficiency

NVIDIA Maxwell Technology and Features

NVIDIA is not merely introducing a new core compages simply along with it several new technologies. In that location are six primal updates to Maxwell that enable new algorithms and superior image quality compared to previous released cards.

The NVIDIA Maxwell core architecture adds the new tiled resources and multi-projection technology for voxel grids (hereafter VXGI) which enhances global illumination. The DirectX 11.2 API makes use of 3D Tiled Resources that allows hardware managed virtual memory for the graphics processing unit and has several Tier-2 features supported such equally Shader LOD clamp and mapped status feedback, mini/max reduction filtering and reads from not-mapped title returns 0. NVIDIA Maxwell GM204 New Technology

Conservative Raster Technology:

First upwardly, we have conservative raster technology which improves voxeliazation, improving the accuracy of voxel coverage calculation. A mapped path of pixels will be covered if they are already covered by a triangle which is the bourgeois raster enabler which notices both orange and majestic colors and covers them conserving the time information technology requires for calculation. This enables new rendering algorithms and the upshot of this voxelization tech improves functioning by 3 times with the new hardware enabled acceleration available on Maxwell. NVIDIA Maxwell Conservative Raster

MFAA or Multi-Framed Sampled Anti-Aliasing Algorithm:

NVIDIA has been ahead in the anti-aliasing game for some time releasing new algorithms each passing generation. Their recent updates include MLAA, FXAA, TXAA and at present, NVIDIA introduces the latest MFAA (Multi-Framed Sample Anti-Aliasing) engineering which is an ultra efficient anti aliasing software design that delivers 30% more functioning and the same quality as 4xMSAA. NVIDIA Maxwell GeForce GTX 980 Performance

NVIDIA Dynamic Super Resolution - 4K Quality on a 1080P Brandish:

One of the new features Maxwell supports in DSR or Dynamic Super Resolution. You tin can call information technology a new version of downward sampling which has become a trend in PC gaming. The applied science is enabled on GeForce 900 series cards only and tin be enabled through GeForce Experience (set to enabled by default). The main purpose of down sampling is to deliver higher resolution quality down scaled to a smaller resolution monitor. So regardless of your monitor size, it tin can display superior prototype quality than what information technology'south built to show as standard. NVIDIA Maxwell Dynamic Super Resolution DSR

NVIDIA Flex, Gameworks, FlameWorks, HairWorks, GodRays Technologies:

NVIDIA's Flex is the latest unified GPU PhysX system which allows developers to utilize a combination of rigid body and fluid simulations. In past game development processes, information technology was hard to permit the two simulations work bated each other due to their complex nature merely NVIDIA's Flex with the correct tools would unify this process allowing the apply of both rigid body and fluid simulations.

Next up is the new GI Works SDK which is the brusk term for Global Illumination Works which allows real-time global illumination in any scene required. Currently, developers use pre-backed global illumination effects in their scenes placing several low-cal sources in a particular place which is a brunt for developers and at the same time, it gives off a non-dynamic presentation. This is solved with the use of existent-time global illumination which is more realistic and offers a more dynamic experience to gamers.

Last upwards is the Flame Works SDK which includes a moving picture-quality volumetric effect solution to render flame and smoke. NVIDIA is calculation these features along with various other effects in alot of upcoming titles such as Batman: Arkham Origins, Witcher 3: The Wild Hunt, Assassins Creed Iv: Blackness Flag, Lookout Dogs. Some of the new titles such as Project Cars and the multi-million dollar funded Star Citizen are also offering rich NVIDIA Turbulence and NVIDIA PhysX and PhysX Particles support plus HBAO+, TXAA, Cloth Simulation and many more to name.

NVIDIA too showcased several slides during the result at their conference at GDC 2022 which confirm that their next generation FleX Unified PhysX and Turbulence particle effects are officially headed for PC and would be inte grated in Unreal Engine iv and CryEngine. The Turbulence particles will exist added to Unreal Engine iii, 4 and Cry Engine via a patch while FleX would be headed to Unreal Engine 4. Only PC is the supported platform for these new features and so titles developed exclusively for PC or multi-plat titles which are optimized for PC will adopt the new features. NVIDIA Maxwell GM204 GPU Features

NVIDIA GeForce GTX 980

The Flagship GeForce 900 Maxwell

The NVIDIA GeForce GTX 980 is the flagship GeForce 900 serial offering and the fastest Maxwell card to launch in the market. From tiptop to lesser, the GeForce GTX 980 is a well built carte du jour featuring better performance, low power consumption and several new gaming and architecture side enhancements. The NVIDIA GeForce GTX 980 include 2048 CUDA Cores, 128 TMUs, 64 ROPs. The core clock is maintained at 1126 MHz core and 1216 MHz boost while the memory is clocked in at 7 GHz effective clock which results in 224 GB/s bandwidth. The TDP of the card is fix at 165W while the power is fed through dual 6-Pin power connectors.

The GeForce GTX 980 is making employ of an update revision of the NVTTM libation introduced on the GeForce GTX Titan Black with a all black naming logo etched on the shroud well-nigh the I/O plate and a all black heatsink which can be spotted from the mirror cutting out in the centre of the shroud. The carte obviously makes of vapor sleeping room which is cooler off by a blower fan. We were unable to find the Dual Axial fan pattern which NVIDIA had patented dorsum a few months and was rumored to be a function of the new graphics carte series but I expect the card even as it is will do a groovy job cooling the menu considering it can dissipate oestrus of up to 275W while GeForce GTX 980 will have a maximum thermal dissipation ability of just nether 170W. So that'due south a ton of cooling being supplied to the core and we can expect massive overclocking headroom for a bill of fare which is already clocked past the 1216 MHz barrier.

Back to the cooler pattern, the NVTTM does include some minor changes along the display ports isolating information technology inside the shroud entirely. One of the changes I like the well-nigh is the add-on of the backplate which is carried over from the GeForce GTX Titan Z. The card features two SLI Gold fingers which volition allow iv-Way SLI Multi GPU functionality. The GeForce GTX 980 is fed power through dual six-Pin connectors and while there is space for an eight-Pin connector, NVIDIA will just characteristic ii 6-Pivot as a reference blueprint leaving its AIB partners to exercise the rest in the form of custom designs. Display outputs include DVI, HDMI and iii display ports which is i reason for the unusually big size of the display connector. The bracket is besides updated with a new layout since the cut outs for exhaust look similar to the ones featured on the GeForce GTX Titan Z.

The PCB has been modified to a more brute design, NVIDIA can be seen using viii Samsung K4G41325FC-HC28 128M x 32. A full of viii of these modules have been featured which equate to 4 GB GDDR5 VRAM across a 256-chip bus. The voltage controller has been moved below the power connectors and the power delivery includes 5 Phases compared to 6 on the GeForce GTX 780 Ti. At the aforementioned time, nosotros tin come across a large assortment of VRMs bated the chokes which will evangelize unprecedented corporeality of overclocking performance fifty-fifty on the reference designs. The NVIDIA GeForce GTX 980 will retail at $549 The states while non-reference models will retail at around $599 United states pricing.

NVIDIA GeForce GTX 970

The $329 US and Sub-150W Maxwell

The NVIDIA GeForce GTX 970 is the about surprising role in the Maxwell lineup coming in at a price of simply $329 United states of america. NVIDIA's GeForce GX 970 features 13 SMM units placed in 4 GPC (Graphics Processing Clusters). Since each SMM unit has 128 CUDA cores, 32 in each logic unit (32 x iv), the full number of CUDA cores equates to 1664 on the dice. From the first generation Maxwell core architecture, nosotros learned that a Maxwell SMM (Streaming Multiprocessor Maxwell) unit has 128 cores compared to 192 on the electric current generation Kepler SMX units. The specifications equate to a full of 1664 CUDA Cores, 104 TMUs and 64 ROPs.

Forth with that, we have a 4 GB GDDR5 memory running across a 256-Bit memory interface clocked at 1753 MHz (7.00 GHz Effective) which pumps out 224.4 GB/s bandwidth. The core clock is maintained at 1051 MHz and 1178 MHz heave clock something which I was expecting if the cards were to be able to have on the GK110 cadre based graphics cards. Lastly, we have the fill charge per unit numbers which amount to 33.vi GPixels/due south Pixel and 145.0 GTexels/s Texture fill rates. The GeForce GTX 970 will exist available in both reference and non-reference variants at launch which volition retail at a range of $329 to $349 US. Display ports on the reference models will stick with the DVI, HDMI and 3 display ports. The card uses HDMI 2.0 engineering science and will be powered by dual 6-Pin connectors. AIB partners may offer unlike brandish output configurations but the cards would be fully compatible with Chiliad-Sync monitors.

NVIDIA Maxwell GeForce GTX 980 and GeForce GTX 970 Reviews:

NVIDIA GeForce GTX 980 Review @ Anandtech
NVIDIA GeForce GTX 980 Review @ Techpowerup
NVIDIA GeForce GTX 980 Review @ Hardwarecanucks
NVIDIA GeForce GTX 980 Review @ Guru3D
NVIDIA GeForce GTX 980 Review @ HardOCP
NVIDIA GeForce GTX 980 Review @ PCPer
NVIDIA GeForce GTX 980 Review @ Chip-Tech
NVIDIA GeForce GTX 980 Review @ Overclock3d
NVIDIA GeForce GTX 980 Review @ TechReport
NVIDIA GeForce GTX 980 Review @ Hexus
NVIDIA GeForce GTX 980 Review @ MaximumPC
NVIDIA GeForce GTX 980 Review @ Techspot
NVIDIA GeForce GTX 980 Review @ Tweaktown
NVIDIA GeForce GTX 980 and GeForce GTX 970 Review @ PCPOP
NVIDIA GeForce GTX 970 Review @ Wccftech

NVIDIA GM204 Die Shot:

NVIDIA GeForce GTX 970 and GTX 980 Specifications:

	GeForce GTX 570	GeForce GTX 580	GeForce GTX 670	GeForce GTX 680	GeForce GTX 770	GeForce GTX 780	GeForce GTX 780 Ti	GeForce GTX 970	GeForce GTX 980
Codename	GF110	GF110	GK104	GK104	GK104	GK110	GK110	GM204	GM204
Process	40nm	40nm	28nm	28nm	28nm	28nm	28nm	28nm	28nm
GPU Cadre	Fermi	Fermi	Kepler	Kepler	Kepler	Kepler	Kepler	Maxwell	Maxwell
SM Units	15 x 32	sixteen 10 32	7 x 192	8 x 192	viii x 192	12 10 192	fifteen x 192	xiii x 128	16 ten 128
CUDA Cores	480	512	1344	1536	1536	2304	2880	1664	2048
ROPS	40	48	32	32	32	48	48	64	64
TMUs	60	64	112	128	128	192	240	104	128
Cadre Clock	732 MHz	772 MHz	915 MHz	1006 MHz	1046 MHz	863 MHz	875 MHz	1051 MHz	1126 MHz
Boost Clock	1464 MHz	1544 MHz (Shader Clock)	980 MHz	1058 MHz	1085 MHz	900 MHz	928 MHz	1178 MHz	1216 MHz
Retentiveness	one.ii GB GDDR5	1.5 GB GDDR5	two GB GDDR5	2 GB GDDR5	2 GB GDDR5	3 GB GDDR5	iii GB GDDR5	four GB GDDR5	four GB GDDR5
Memory Bus	320-Fleck	384-Bit	256-bit	256-bit	256-bit	384-Bit	384-Bit	256-bit	256-scrap
Memory Clock	3.eighty GB/s	4.0 GB/due south	half-dozen.0 GHz	half-dozen.0 GHz	7.0 GHz	6.0 GHz	seven.0 GHz	7.0 GHz	vii.0 GHz
Memory Bandwidth	152.00 GB/due south	192.4 GB/s	192.0 GB/southward	192.0 GB/southward	224.5 GB/s	288.six GB/s	336.0 GB/s	224.5 GB/s	224.five GB/southward
Texture Make full Rate GT/s	43.92	49.41	102.v	128.viii	134	166	210	145.0	TBC
TDP	219W	244W	170W	192W	220W	250W	250W	148W	165W
Power Connectors	six+6 Pin	8+6 Pin	vi+6 Pin	6+vi Pin	8+6 Pin	8+6 Pin	8+vi Pin	six+vi Pin	six+6 Pin
DirectX 12 Back up	Yes	Yep	Yep	Yeah	Yeah	Yes	Aye	Aye	Yes
Launch	December seventh 2022	November 09 2022	May tenth 2022	March 22nd 2022	May 30th 2022	May 23rd 2022	December 2022	18th September 2022	18th September 2022
Toll	$349 US	$499 Us	$349 U.s.	$499 Usa	$349 The states	$499 Usa	$699 US	$329 Reference $329+ Custom	$549 Reference $549+ Custom