Hardware Industry

NVIDIA’s next generation Maxwell core architecture based GeForce GTX 880, GeForce GTX 870 and GeForce GTX 860 will launch in October this year, as reported by GDM.OR.JP. The site which is part of one of the most major hardware retailers in Japan have been pretty accurate with their earlier leaks so we will assume that the October launch time-frame is quite accurate considering the launch was planned for Q4 2014 was narrowed down to October / November a few weeks ago and now we have a much more accurate schedule of NVIDIA’s Maxwell GPU launches.NVIDIA GeForce GTX 880

NVIDIA Maxwell Based GeForce GTX 880 and GeForce GTX 870 To Launch in Early October

The first two Maxwell cards to launch in the consumer market would be the high-performance GeForce GTX 880 and GeForce GTX 870. These cards will feature the GM204 core architecture which is the successor of the GK104/GK114 core from the Kepler generation. From little things we know thanks to the leaked PCB shot of NVIDIA’s alleged GeForce GTX 880 graphics card, we come to know that the core being used for the sample variant was GM204-INT41684-A1 however there has been a little variation in the codename and it is now called GM204-CS0-A1 (note that the CS stands for Consumer Sample) which means that final testing is under way and the chip will soon end up in the consumer market as the GeForce GTX 880. The chip is also the first of four variants which will make their way to the rest of the lineup including the GeForce GTX 870.

The GeForce GTX 870 on the other hand will be a cut down variant of the GK204 core featured on GeForce GTX 880 with some disabled SMM (Streaming Multiprocessor Maxwell) units. Its not know what specifications does the GM204 hold in general but it does feature a die size larger than its predecessor which is due to the existing 28nm process technology and the card will make use of high-frequency SKHynix H5GQ4H24MFR 0.5G clocked at 7 GHz. We will most probably be looking at a 256-bit since the memory arrangement as seen on the ES board is similar to the GeForce GTX 680 ES board and the final consumer variant. The card was shown with 8 GB GDDR5 memory which is a lot of VRAM for a reference card but we believe that the cards will ship in 4 GB variants. Aside from that, the total power consumption would be similar to Kepler around 195W but it will deliver performance on scale with some of the overclocked GeForce GTX 780 Ti’s we have seen. If the recent GeForce GTX 880 benchmark is taken for reference, we might be looking at a 20-30% performance uplift from Maxwell over GK110 which is insane and makes us wonder the performance uplift with the GM200 core which won’t ship to consumers till next year.


There have been countless leaks and rumors about Nvidia’s Second Generation Maxwell Architecture. Some of them have been pretty far fetched while others are almost confirmed. I have wanted to do an aggregate post summarizing all the authentic leaks so far for quite some time now but haven’t gotten the chance up-till now. I would also like to point out that the Majority of the Maxwell Mobility details have been spotted by the now well-known user Cloudfire (over at NotebookReview Forums) while the desktop variant leaks have multiple sources, including but limited to: Videocardz.com, SweClockers.com and Our Own.Maxwell-GTX-880-

Geforce Second Generation Maxwell Architecture – GM204, GM206 and GM200 Overview

Lets start with the desktop counterparts first. As you can probably guess, all Nvidia cores starting with the prefix GM2xx are Maxwell Second Generation cores. Now the first generation of the Maxwell Architecture was the GM107 which is present in the GTX 750 and GTX 750 Ti. Now the first generation of Maxwell introduced some very impressive performance gains on the same node. However, due to some drastic change of plans behind the curtains, Nvidia seems to have abandoned the GM1xx completely; at least as far as Desktop is concerned. Currently all reports have pointed to the GM2xx Dies being developed on the 28nm node and this begs the question what is the difference between GM1xx and GM2xx. There are two possibilities, one is what I want to hear: that the GM2xx might contain ARM cores and the other which I logically know is more probable: performance gains and a better architectural revision. The cores are as follows:

AMD Core Logo

Update: The teaser is actually done on an easy to spot ”cool truck” template found on the web (thanks to _John_ for that) but the image itself was posted by AMD PR. See my comment below for more information.

[Caution: Not Confirmed] I noticed something pretty interesting posted over at Planet3DNow. Basically the guys over there have managed to find an AMD PR image that was posted under the #AMDEvolved twitter handle (I couldn’t find the picture myself, but I probably didn’t look hard enough). The picture depicts 12 AMD Robots. Since the robots are an analogy for cores, it doesn’t take a genius to figure out that AMD is planning something with 12 Cores.

AMD 12 Core ProcessorA truck carrying 12 AMD ‘Cores’. The A Series Logo depicting APUs is also visible.

PR Material Hints at an A Series APU with 12 CPU Cores – Probably Steamroller

Before I let the wrong idea through let me explain why I have tagged it as ‘Not Confirmed’. See, AMD has a very funny way of naming cores. Kaveri has, according to AMD, 12 compute cores. 4 of these cores are CPU and 8 of these are CUs. Ofcourse this is pretty misleading since CPU and GPU cores are nothing alike, but the fact remains. It is possible that these 12 Robots depict 12 Compute Cores as opposed to just 12 Steamroller Cores and I wanted to make that clear. This is also the reason I have tagged the post as not confirmed. An interesting thing to note is that AMD introduced a new ‘robot’ in their “Core is Back” teaser video, which has a red color scheme. However these are the old robots with the standard white color scheme. This seems to suggest that all these robots are on the old steamroller architecture and not a new one. And since Kaveri already exists, boasting a 12 Compute Core A Series APU doesn’t make any sense and therein lies the logical conclusion: the truck must be boasting 12 Steamroller Cores.

Assassins Creed Unity Logo

Assassins Creed Unity is the next step in the famous game series that has quite the cult following. The game is powered by the Anvil Engine and Ubisoft recently did a tech demo for the upcoming iteration of AC which is Unity. The engine looks quite beautiful and boasts a plethora of interesting effects. CryEngine seems to be in trouble (if reports are to be believed) and the field is open for the next king of the hill when it comes to gaming engines.

Assassins_Creed_UnityImage courtesy of SweClockers.com

Ubisoft Anvil Engine Shown Off Using Assassins Creed Unity Tech Demo

Although Unreal Engine is currently the  favorite for taking over the vacuum that could be left behind CryEngine (if hypothetically speaking, its troubles continue) multiple tech demos have been emerging on the scene. Assassins Creed Unity’s Tech Demo shows of the Anvil Engine to great effect. The developers managed to recreate old Paris in an almost 1 to 1 scale. The demo showed off some very advanced lightning features, volumetric lightning and shadows. But perhaps the most promising thing (at least according to the demo) was the Advanced AI.


Some pretty interesting information has surfaced in a drivers update at Laptopvideo2go (via Videocardz). Apparently the .inf file to the driver update that contains support for two Maxwell powered Quadro Professional GPUs has been posted, along with a new nomenclature. This means that the video industry will soon be able to utilize the power increase and efficiency offered by the Maxwell architecture and get to enjoy an unfortunate nomenclature.NVIDIA Quadro K6000 GPU

Nvidia Driver Update Reveals Quadros with Maxwell Cores – Mobility Included

Now the biggest difference between Quadro and your mainstream Nvidia cards is the double precision floating point operation per cycle ratio. For Quadro variants this number is 1. Which means the card can process 1 double precision floating point operation per cycle per core (which is a 1/2 ratio). In sharp contrast a Fermi’s double precision floating point performance (aka FP64) is 1/8th of FP32 ( Single Precision). A Kepler 780’s FP64 performance is 1/16 of FP32 and finally a Maxwell 750’s FP64 is 1/32 of FP32. I am sure you can see a trend here. Basically if you want that sweet 1 to 2 ratio, you are going to have to pay the premium and quite a big one at that. As you can imagine, a Maxwell core inside a professional GPU will make quite a performance jump. VC managed to locate the IDs of the GPUs and alot them to the correct nomenclature:


[Rumor] Something very interesting cropped up on the forums of NoteBook Review (via User800). Basically somebody posted an alleged screen-shot of an Nvidia GPU with the name blurred out. The authenticity of this picture is not verifiable, but to the casual inspection it doesn’t appear doctored. The benchmark run appears to be of the Maxwell Flagship GTX 880. Maxwell-GTX-880-

Maxwell GTX 880 Gets Alleged 3dMark Fire Strike Benchmarks – Impressive Results

If you look very closely at the picture you can barely make out the “GT 8″. If you look even more closely, you can spot one half of the second 8 following the first. So the alleged benchmark run is clearly of a GTX 880. Now some users in the forum were commenting on whether this could be the mobile variant of the GTX 880. Well, I think, there is a reason the original user so conveniently put in the CPUZ shot, something that wasn’t really required. So that we can readily deduce it as a desktop variant. The i7 4770k is the Haswell LGA 1150 CPU and is most certainly not found in any notebook. There is a 0.1% chance that the mobile variant was being tested but I think we can safely ignore that possibility.

PlayStation 4 console

[Editorial] Before I begin a few words of caution. It goes without saying that the results of this benchmarking showdown will not be pleasant to some. Also, all the benchmarks were conducted virtually, which means that some slight variation may exist between real life results. The Equation System and the VGA Calculator used are the intellectual property of one Daeguen Lee (DG Lee) and all rights are reserved by him. The calculator can be accessed freely on his website iyd.kr (aka udteam.tistory.com).main title13

PS4’s GPU (Pitcairn) Benchmarked Against the Xbox One (Bonaire) with and without eSRAM

Wow, I don’t think there has ever been an instance of a universal benchmark, and for good reason too. Since an actual universal benchmark would instantly put the power hierarchy in harsh relief it has not been done. However what we are going to attempt today is the closest you will ever get to a universal real life benchmark and the results should be very accurate. Alright lets start. Since both the PlayStation 4 and Xbox One utilize AMD GPUs, that means that we are only going to use the AMD Equation System. A basic introduction of both the GPUs is in order. The PS4 GPU is codenamed Pitcairn and is basically a cut down variant of the Radeon 7870. It has a fairly high SP count and is clocked reasonably high too. The GPU has access to 8GB of Shared GDDR5 memory clocked a 1375 Mhz (5500 Mhz effective) and boasts throughput of upto  176 GB/s. Pitcairn has a 256 Bit bus which is quite reasonable for a GPU of this tier. The good thing about PS4’s GPU is that we have everything we need to calculate performance with almost zero real life variation.

AMD logo

During the Q2 financial conference call, AMD’s SVP and COO, Lisa Su, assured that their company will out 20nm products next year followed by 16nm FinFET later on. While there have been rumors that AMD might introduce 20nm products in late 2014, I believe that this statement is an end to all those reports since AMD’s entire lineup is to stick with the current 28nm process node this year.AMD 20nm

AMD Assures 20nm Products in 2015 – 16nm FinFET Later On

In Q1 2014, AMD forwarded the same statement that their graphics and semi-custom business products will both stick with 28nm. We have seen several 28nm products across AMD graphics and processors designs which include the Radeon R9 and Radeon R7 series for graphics while Steamroller, Jaguar powering AMD’s processors and accelerated processing units were also based on 28nm.

“We (AMD) will be shipping products in 20 nanometre next year and as we move forward obviously a FinFET is also important,” Lisa Su (Senior Vice President and Chief Operating Officer at AMD) via Fudzilla

This news is interesting for the moment since AMD is on the verge to introduce new graphics processing units for the consumer industry. These include the Tonga, Iceland and Maui which are reportedly launching in Q3 and Q4 2014. So we can more or less confirm that all three of these chips will feature the existing 28nm process design while their replacements which are now believed to arrive 2015 will be based on 20nm. The same goes for AMD’s upcoming semi-custom APUs which are part of AMD’s Skybridge initiative which allows pin-to-pin compatibility for x86 and ARM based processors. The Cambridge APU will feature this technology leveraging the efficiency while the K12 ARM core which is a 64-bit ARM custom designed core will use 16nm FinFET technology. It should be noted that AMD also has a true x86 architecture in development to feed high-performance needs which should not be expected prior to 2016-2017 so we may see it based on a smaller node.


After a long wait, we have finally received the first authentic confirmation of the Nvidia Flagship GPUs set to release in 2014. To be clear, we are definitely not looking at 20nm GPUs, instead its the same 28nm node with the advanced Second Generation Maxwell Architecture. The source of the leak is the insanely reliable Sweclockers.com.Maxwell-GTX-880-

Nvidia Maxwell GTX 880 and GTX 870 on the 28nm Node Landing in October/November

We had previously received estimation from Sweclockers of an expected release date of Oct/Nov but now they have confirmed it. This is big news people. This means that we are not going to be skipping the year 2014 entirely after all and will be receiving a new performance class form Nvidia based on the 28nm Maxwell. We have already seen the engineering sample of the Maxwell GTX 880 and this more or less confirms that we are going to be seeing leaked benchmarks very very soon. The core that will be housed is the GM204 circuit which will be used in the highest end GPU released (Geforce GTX 880). Now since reports indicate that Nvidia has completely scrapped the GM100 chips, we are looking at second generation Maxwell Architecture.

AMD R9 Series Logo

Videocardz.com has just released some new information about AMD GPUs. The GPUs in question are ofcourse Tonga, Iceland and Hawaii. As for the Maxwell Mobility news, we have already covered the 870MX a while back. Now we had already told you that the speculated Hawaii XTX with 3072 SP doesn’t exist and now it seems that VC is stating the same, citing a Hawaii core with higher clocks. AMD Tonga GPU

AMD Iceland, Tonga and Hawaii Revision Detailed – R9 390X Allegedly Leaked

Now before we begin, the leak from our source also mentions the fabled R9 390X citing some specs. However the leak of the R9 390X is slightly shady. Here are the reasons why:

weekly shorts logo

We are back with the second installment of WCCF’s Weekly Shorts. The rules are simple, of the many articles we post, we shortlist not more than 7 entries and summarize them in not more than 5 lines. The previous episode of the WCCF’s Weekly Shorts can be found here.weekly shorts main image

Weekly Shorts – Peeping EA, AMD Carrizo, PS4’s 60FPS Optimization, Benefits of Sand, Anti-Theft Protection Debunked and the GM200 Core

1. EA is Spying on You by File Snooping with the Origin Client – Official Investigation Begins

Origin is one of the biggest gaming clients out there and it turns out that it does more than just let you play EA Games. In this report we take a look at what some dedicated gamers have uncovered. Basically the Origin Client is also an accomplished file snooper and is actively spying o your usage habits. Origin Reps have stated that they have begun an official investigation. Read More about Origin’s File Snooping.

Tegra K1 Logo

The folks over at Digital Storm have revealed an in-depth benchmark analysis of the Tegra K1 SoC and the results are quite impressive, atleast as far as the graphics side is concerned. The compute side on the other hand is fairly mediocre and is just ahead of the competition. However since the up side of the Tegra K1 is the gaming capability and 64Bit support, the benchmarks can be taken very positively overall.NVIDIA Tegra K1

Tegra K1’s Gaming Benchmarks Blow Away Competition – CPU Performance, Not So Much

Alright so the Tegra K1 tested is housed inside the Xiaomi Mi Pad so we are looking at a real world scenario with constrained TDP and clocks. This means that the results we see here are very much indicative of the final product to be used in tablets and we can expect a slight increase in performance for devices like the Shield 2 and the Tegra K1 VCM. The gaming result is very impressive and if you couple this with the Unreal Engine 4 Compatibility demoed earlier you have a very nice combo. The 64Bit K1 is the only chip that currently supports the Unreal Engine 4 Full Desktop Render in conjunction with AEP (Android Extension Pack). This makes the Tegra K1 a very special chip indeed. Although it is only a matter of time before the compatibility is added to rival counterparts as well, until that happens, the soc has an advantage. Anyways, enough of the preliminaries, here are the benchmark themselves:

AMD Kaveri APU

[RUMOR] Notice the shiny tag at the start. Continue. We have received some information regarding AMD’s upcoming APU Carrizo, this information is unverified so I will be treating it as a rumor. The report comes from the Italian site bitsandchips.it and states that AMD’s upcoming flagship APUs will have Stacked DRAM while maintaining the 28nm Node.

Carrizo APU RoadmapThe Quite Old ‘Leaked’ Roadmap

Carrizo APU will be based on the 28nm Node and have HBM (High Bandwidth Memory)

Now it goes without saying, that you need to keep that pinch of salt handy throughout this post. However this news, if true, is very interesting. We know for a fact that APUS benefit a lot from good memory and if these APUs will truly support HBM then we can expect to see some very substantial performance per clock gains while jumping from Kaveri to Carrizo even while staying on the same node. Another important point to note is that with the Carrizo APU the implementation of HSA will be perfected resulting in probably significant in compute as well as gaming.

Now we already know that AMD is working with Hynix to create Stacked DRAM.  We also know that this memory will come in two types, namely 3DS and HBM (Don’t be fooled by the lack of 3D in this name, both are stacked). The memory that is in question here is the HBM variant type which will feature the highest bandwidth and I know for a fact that there are two types already in production. Namely the 2-Hi and 4-Hi variants. You can find the detailed analysis of the same in my Pascal Architecture Analysis. Now the max bandwidth of a single HBM is 128-256 GBps (compare this to the 28GBps of GDDR5), so we are looking at an insane growth in bandwidth, albeit at reduced clocks (most probably around 1000Mhz).


It looks like Zauba.com has become a pretty good source to spot graphics processing units as the shipping site has revealed several NVIDIA GPUs over the last few months. This has happened once again as 3DCenter was able to find a shipping manifest of NVIDIA’s upcoming flagship GeForce Maxwell GM200 GPU which will hit the consumer market next year.GeForce Maxwell GM200 GPU

NVIDIA’s Flagship GeForce Maxwell GM200 GPU Spotted

So what makes this spotting so special? There are no images, no specifications, no assurance or whatsoever that this might be true but recent leaks such as the GM204 PCB reveal and the codenames matched earlier listings on the site which makes these listings somewhat believable if not entirely true. The GM200 chip mentioned in the shipping list originated from Taiwan where it was manufactured and will be discharged in Banglore, India. The sample was dispatched on 8th July 2014 which means that the GM200 GPU is already in-works but will take some time to reach mass production before it begins shipment to market. The GM200 GPU is codenamed as GM200-INT0-A1 which means this is the first revision of the core and we may see newer revisions (B1) in the future but for now the A1 revisions is what will be featured on graphics cards for servers and consumers.

While the GM204 chip which was revealed a while back is bigger than its GK104 predecessor, it’s still a bit small compared to GK110. Its hard to tell if the GM200 GPU will be larger or equivalent in size as compared to GK110, its last generation Kepler equivalent but from what we have seen, it will be a huge die returning to the legacy of the big die sizes which NVIDIA has been known for but will be extremely power efficient and house several more cores. The chip is planned for launch in 2015 which would give ample time for 20nm production through TSMC but previous reports show that NVIDIA might ditch 20nm in favor of 16nm FinFET so that could be a possibility but will result in a longer delay period. From what little details we can gather, TSMC will commence volume production of 16nm FinFET+ nodes in Q1 2015 which would mean a Q2/Q3 2015 time for GM200 GPUs to appear on the new process or they could move the launch a bit earlier by sticking with 20nm which is already under production since Q1 2014 but prioritized for Apple till Q2 2014 which is over now.

Colorful iGame GTX 780 _2

Colorful, the Chinese graphics card manufacturer widely known for their iGame series are readying new modular cooling designs for the NVIDIA GeForce 800 series. Announced during a conference in mainland China, Colorful also showed off their new iGame logos which will be introduced when the new Maxwell generation of graphics cards hit the market in Q3 2014.Colorful GeForce GTX 780 Ti Concept Cooling

Colorful Aiming Modular Cooling For Upcoming GeForce 800 Series Graphics Cards

While Colorful as a brand is limited to APAC region, they are still widely renowned for their products in the entire market with creation such as the iGame GTX 780 Ti Kudan, iGame GTX 780 Fire Warrior, iGame GTX Titan Liquid Cooled, iGame GTX 680 Kudan and iGame GTX 660 Ti Ice Knight. All these graphics card are known for the badass and amazing design schemes on the cooling and PCB which allow extreme levels of overclocking and modding to enthusiasts and overclockers. Colorful has an entire portfolio of graphics card series for NVIDIA while their sister brand known as Colorfire make customized AMD Radeon graphics cards covering the entire graphics market in the region.

In June 2014, Colorful did establish an overseas partnership with Newegg which has gone under effect and the US based online retailer widely selling PC hardware have now started to offer Colorful manufactured graphics cards. The Colorful GeForce series has three tiers under the iGame brand, Kudan (Enthusiast), Ice Knight (Overclocking) and Fire Warrior (High-End). All these graphics cards offer a shot of their own with beastly designs, customized boards and additional PWM add-in cards on some products.

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 23 queries in 1.156  seconds.