NVIDIA GeForce GTX 870 "Maxwell" Specifications Analysis - 13 SMM Units With 2nd Generation 1664 CUDA Cores, Performance Benchmarked

The specifications of NVIDIA's upcoming Maxwell based GeForce GTX 870 have been leaked Coolaler and we already covered the basic details of the carte in some other article which tin can exist seen hither. The NVIDIA Maxwell based GeForce GTX 870 graphics carte du jour is one of the three SKUs arriving to consumers in the upcoming months (September and October).

NVIDIA GeForce GTX 870 "Maxwell" Specifications Analysis

Last time, I did the analysis of the leaked GeForce GTX 880 PCB which was in engineering state and constitute out some intriguing details such equally the 256-Bit retention bus, the GM204 chip existence larger than GK104 while being based on the current 28nm procedure node. We didn't get to see any specifications of the card itself aside from a few details nosotros were able to gather from the pictures. Today, we accept a different sort of leak in the class of a screenshot which shows the GPU-z program running parallel to a 3DMark xi score of an unknown graphics menu.

The leaker from the main source has mentioned that the graphics card in question is NVIDIA'due south GeForce GTX 870 which is the skimmed downwards variant of the GeForce GTX 880 graphics which will be available to consumers in the upcoming months. The GeForce GTX 870 boasts the second generation Maxwell architecture and will feature the GM204 core architecture which will become the foundation of the flagship GeForce GTX 880 also. The GeForce GTX 860 is also expected to launch in Q4 2022 only it isn't expected whether the card volition characteristic GM204 core or the GM206 core but let'south forget that for a moment and focus on the details we accept in hands.

The GeForce GTX 870 has the Device Id "NVIDIA D17U-20" and the GPU Id for the graphics core is 13C2. Nosotros have our own speculation regarding that but before nosotros get into that, let's denote the unabridged specifications listed on the GPU-z panel. The GM204 SKU featured on the GeForce GTx 870 fuses 1664 CUDA Cores on the die along with 32 ROPs and 138 TMUs. From the kickoff generation Maxwell core architecture, we learned that a Maxwell SMM (Streaming Multiprocessor Maxwell) unit has 128 cores compared to 192 on the current generation Kepler SMX units. Along with that, nosotros have a 4 GB GDDR5 memory running across a 256-Chip 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 core based graphics cards. Lastly, we take the fill rate numbers which amount to 33.half-dozen GPixels/s Pixel and 145.0 GTexels/s Texture fill rates. The TMU count is a bit off since the maxwell cadre architecture holds eight TMUs per SMX and then we were supposed to see 104 TMUs in total instead of 138 and so information technology could be a GPU-Z fault which could mean that these specifications might non be the actual deal.

At present as yous may remember, the showtime generation Maxwell core architecture which is featured on the GM107 GPU has 2 times the operation per watt and the CUDA Cores, even though lower amount to more performance that is rated at 135% of the past generation. The key give-and-take here is start generation, since NVIDIA just previewed what their Maxwell compages is capable of and having ample time for optimizing and enhancing the architecture, I expect to run across better numbers beingness rated at 145-150% and we can see more than power efficiency on the new cards. The GeForce GTX 750 Ti which is NVIDIA's entry level Maxwell offering is rated at 64W and comes with splendid operation for its wattage. Because the updated design, I think at that place's no doubtfulness that NVIDIA GeForce GTX 870 will exist able to trump the GeForce GTX 780 in performance (although not by a large margin).

The GPU proper name is also interesting since 13C2 could probably mean 13 SMM units which are featured on the GPU which makes us think near the GeForce GTX 880. With a SMM or two in place, we can see the GeForce GTX 880 include 1792 or 1920 cores which although lower than the GK110 based cards is reasonably higher than what was featured on the GK104 chips and with the performance per cadre leverage, we tin can see the GeForce GTX 880 right next to the GeForce GTX 780 Ti in performance numbers. The GM200 is notwithstanding far abroad from launch and is expected next year but that just gives a glimpse of what to expect from the daddy core which can easily topple the flagship chips.

The core count and configuration for the GeForce GTX 880 looks around right since looking at the trend in the tabular array provided below, you can annotation that the skimmed down *70 graphics card has generally ane or two SM units disabled. The GK110 is the simply chip in the line which has more than chips disabled simply that is due to the extended lineup due to which we had several cards in this line aka GeForce GTX 780, GeForce GTX Titan, GeForce GTX 780 Ti, GeForce GTX Titan Blackness (all of which are configured around the same core). In case there volition be a college SM units SKU, then NVIDIA will most definitely launch it equally a Ti part every bit was rumored past a few sources a while back. If there are more than disabled SMMs, then NVIDIA will launch them as a dissimilar card rather than releasing a 4 SMM disabled function

The iv GB GDDR5 road seems to be the reference route for NVIDIA this fourth dimension around moving onwards from a ii GB GDDR5 VRAM configuration on several Kepler cards. The chips will nonetheless be configured effectually a 256-Bit bus just this just seems to be the route for graphics makers these days because AMD is going this road with their adjacent operation scrap, codenamed Tonga (Radeon R9 285) which is considered faster than the Tahiti but has a 256-Chip retention bus compared to 384-Scrap on Tahiti based Radeon R9 280 series.

 NVIDIA GeForce GTX 870 Leaked Performance Numbers:

The performance numbers of the GeForce GTX 870 are interesting considering they are really shut to the GeForce GTX 780. The card was tested with a Intel Cadre i7-4820K CPU and is currently a test sample so final performance updates and drivers are not in place considering the GPU-z couldn't even recognize the card properly until the database is updated. In the 3DMark 11 Performance and Extreme modes, the GeForce GTX 870 was able to score P11919 points and X4625 marks respectively. A GeForce GTX 780 on the other mitt on average scores around X4500 points and P12000 points. With the GeForce GTX 870 all the same optimized and without any sort of official drivers, I say these numbers are skillful considering this menu volition retail at $349 U.s.a. and become you performance on par or better than a carte du jour based on a flagship chip from current generation.

NVIDIA Maxwell GM107 Architecture

The NVIDIA Maxwell GM107 architecture is nigh clearly built from scratch withal looks like an hybrid of both Kepler and Fermi. The SMM or Streaming Multiprocessor of Maxwell volition supplant the SMX of Kepler and each of the smm are assembled into 4 blocks notwithstanding are divers as role of a single SMM which means the core architecture has got quite dense with Maxwell. Each of these blocks agree 32 CUDA cores and then a unmarried SMM with four of these functioning 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 ii.0 which includes the Vertex Fetch, Tessellator, Viewport transform, attribute setup and stream output. Each SMM has 8 texture mapping units which equates to forty on the whole chip and 16 ROPs while connected to ii 64-flake memory controllers.

In that location's also a handful of cache on Maxwell which is 2 MB in full compared to 256 KB on Kepler GK107 which reduces GPU queuing. IPCC has been increased on the GPU core and balancing improvement to workload has been done. NVIDIA has improved their H.264 encoding and decoding with NVENC on Maxwell so that would result in better operation in ShadowPlay technology and improved sleep states have been implemented to reduce power input while the GPU is running idle. All of this is packed in a die which measures 148mm²  which ways transistor density has been upped by 15% on the aforementioned 28nm process pattern.

• NVIDIA GM200.GM210 (Maxwell Architecture, High-Performance for Telsa/Quadro Arrives later for Cosnumers, Successor of GK110)
• NVIDIA GM204 (Maxwell Architecture, High-End Consumer, Successor of GK104, Get-go GeForce 800 Series Products likely to characteristic)
• NVIDIA GM206 (Maxwell Architecture, Operation Minded, Successor of GK206, Mid-Range GeForce 800 Series products to characteristic)
• NVIDIA GM107/207 (Maxwell Architecture, Entry Level, Successor of GK107, Entry Level GeForce 800/700 Serial To feature, Already introduced on GTX 750 Ti / GTX 750)

Practice tell us if we missed out annihilation in the Disqus comments section beneath!

NVIDIA GeForce GTX 870 and GTX 880 Specifications (What We Know So Far):

	GeForce GTX 470	GeForce GTX 480	GeForce GTX 570	GeForce GTX 580	GeForce GTX 670	GeForce GTX 680	GeForce GTX 770	GeForce GTX 780	GeForce GTX 780 Ti	GeForce GTX 870	GeForce GTX 880
Codename	GF100	GF100	GF110	GF110	GK104	GK104	GF114	GK110	GK110	GM204	GM204
Procedure	40nm	40nm	40nm	40nm	28nm	28nm	28nm	28nm	28nm	28nm	28nm
GPU Core	Fermi	Fermi	Fermi	Fermi	Kepler	Kepler	Kepler	Kepler	Kepler	Maxwell	Maxwell
SM Units	14 x 32	15 x 32	15 x 32	16 ten 32	7 ten 192	8 x 192	eight x 192	12 10 192	14 x 192	xiii x 128	16 x 128
CUDA Cores	448	480	480	512	1344	1536	1536	2304	2880	1664	2048
ROPS	forty	48	40	48	32	32	32	48	48	TBC	64
TMUs	56	sixty	threescore	64	112	128	128	192	240	TBC	128
Core Clock	607 MHz	700 MHz	732 MHz	772 MHz	915 MHz	1006 MHz	1046 MHz	863 MHz	875 MHz	1051 MHz	1126 MHz
Boost Clock	1215 MHz	1401 MHz (Shader Clock)	1464 MHz	1544 MHz (Shader Clock)	980 MHz	1058 MHz	1085 MHz	900 MHz	928 MHz	1178 MHz	1216 MHz
Memory	i.2 GB GDDR5	1.5 GB GDDR5	ane.2 GB GDDR5	ane.5 GB GDDR5	2 GB GDDR5	2 GB GDDR5	two GB GDDR5	3 GB GDDR5	3 GB GDDR5	iv GB GDDR5	four GB GDDR5
Memory Passenger vehicle	320-Chip	384-Bit	320-Bit	384-Bit	256-flake	256-flake	256-fleck	384-Bit	384-Bit	256-bit	256-bit
Retentiveness Clock	3.34 GB/south	3.69 GB/s	3.80 GB/south	4.0 GB/s	6.0 GHz	6.0 GHz	7.0 GHz	6.0 GHz	vii.0 GHz	7.0 GHz	7.0 GHz
Memory Bandwidth	133.34 GB/s	177.4 GB/s	152.00 GB/south	192.four GB/due south	192.0 GB/s	192.0 GB/s	224.v GB/s	288.6 GB/s	336.0 GB/s	224.five GB/s	224.v GB/s
Texture Fill Charge per unit GT/due south	34	42	43.92	49.41	102.5	128.8	134	166	210	145.0	TBC
TDP	215W	250W	219W	244W	170W	192W	220W	250W	250W	148W	165W
Ability Connectors	8+6 Pin	8+6 Pin	6+6 Pivot	8+half-dozen Pivot	six+6 Pin	6+6 Pin	8+half dozen Pin	eight+6 Pin	8+6 Pin	vi+half-dozen Pin	half dozen+6 Pivot
DirectX 12 Support	Yep	Yep	Yes	Yep	Yes	Yes	Yes	Aye	Yes	Yes	Yeah
Launch	March 26th 2022	March 26th 2022	December 7th 2022	November 09 2022	May 10th 2022	March 22nd 2022	May 30th 2022	May 23rd 2022	December 2022	18th September 2022	18th September 2022
Price	$349 Usa	$499 Usa	$349 U.s.a.	$499 U.s.a.	$349 Us	$499 US	$349 US	$499 US	$699 United states	$299 Reference $329+ Custom	$549 Reference $549+ Custom

Source: https://wccftech.com/nvidia-geforce-gtx-870-maxwell-specifications-analysis-13-smm-units-2nd-generation-1644-cuda-cores-performance-benchmarked/

Posted by: mcgrawvithembity.blogspot.com

Share this post