'

Nvidia challenges AIB partners - Founders Edition GPUs will be factory overclocked

Nvidia wants to compete with AIBs like ASUS, MSI and Gigabyte

Nvidia challenges AIB partners - Founders Edition GPUs will be factory overclocked

Nvidia challenges AIB partners - Founders Edition GPUs will be factory overclocked

When people talk about reference graphics cards, people usually think about the GPU provided directly from manufacturers, after all, it is they who decide what their "reference" specs are. 

With their new RTX 20 series of graphics cards, Nvidia is redefining what most people would consider reference graphics cards, uplifting their new Founders Edition models to compete with their AIB partners. Now, in theory, Nvidia can provide higher performance levels than some AIB offerings, as Founders Edition graphics cards are no longer "reference". 

Nvidia has improved their Founders Edition graphics cards in two ways. First, they have upgraded their cooling solution, moving from a blower-style cooler to a new dual-fan solution, increasing airflow over the graphics card while also providing lower noise levels, both at stock and while overclocking. Nvidia states that this cooler will operate up to 10 degrees cooler than a reference blower and emit one fifth the noise when overclocked, a decrease of around 8 decibels.  

Turing Founders Edition Graphics cards will also feature higher clock speeds than "reference" RTX series graphics cards. Below is a table that showcases the clock speed differences between Founders Edition (FE) and Reference (Ref) RTX 2080 Ti, RTX 208 and RTX 2070 graphics cards. Every Founders Edition graphics card has a 90MHz increase in boost clock speed. 
  
 
 Geforce RTX 2080 Ti (FE)Geforce RTX 2080 Ti (Ref)Geforce RTX 2080 (FE)Geforce GTX 2080 (Ref) Geforce RTX 2070 (Fe)
Geforce RTX 2070 (Ref)
ArchitectureTuringTuringTuringTuringTuringTuring
CUDA Cores4,3524,3522,9442,944 2,3042,304
Ray Tracing Performance10 Gigarays10 Gigarays8 Gigarays8 Gigarays6 Gigarays6 Gigarays
Base Clock
1350MHz1350MHz1515MHz1515MHz1410MHz1410MHz
Boost Clock1635MHz1545MHz1800MHz1710MHz1710MHz1620MHz
Memory TypeGDDR6GDDR6GDDR6GDDR6GDDR6GDDR6
Memory Capacity11GB11GB8GB8GB8GB8GB
Memory Speed14Gbps14Gbps14Gbps14Gbps14Gbps14Gbps
Memory Bandwidth616GB/s616GB/s448GB/s448GB/s448GB/s448GB/S
Memory Bus Size352-bit352-bit256-bit256-bit
256-bit256-bit
SLIVia NVLinkVia NVLinkVia NVLinkVia NVLinkN/AN/A

Nvidia challenges AIB partners - Founders Edition GPUs will be factory overclocked  

Nvidia's new Founders Edition coolers feature "the world's first full-length graphics card vapour chamber," which allows the coolers to spread heat to the GPU's aluminium (painted black) fin stack for ultra-efficient cooling performance. 

Nvidia's new die-cast fan shrouds are diamond-cut and are anodised for scratch resistance, creating a durable, lightweight and "classy" cooler design. All of Nvidia's Founders Edition graphics cards include a backplate and NVLink compatible graphics cards will ship with a matching cover to deliver a consistent look for non-SLI users. 

Founders Edition graphics cards will also feature DisplayPort 1.4, HDMI 2.0b and VirtualLink. RTX 2080 and 2080 Ti Founders Edition models will feature three DisplayPort 1.4 connections, a single HDMI 2.0b connection and a single VirtualLink port. Nvidia's images for their Founders Edition RTX 2070 suggest that the graphics card will offer two DisplayPort 1.4 ports, a single HDMI 2.0b port, VirtualLink and a single DVI-D port. 

Nvidia challenges AIB partners - Founders Edition GPUs will be factory overclocked  

Nvidia is getting aggressive with their AIB partners, first by selling their Founders Edition graphics cards via a dedicated storefront and now by releasing Founders Edition graphics cards that can offer performance levels and cooling that could rival some of their custom designs. Affordable RTX 20 series graphics cards from AIBs may even provide less performance than Nvidia's Founders Edition models, which is new for the GPU market. 

With their Founders Edition graphics cards, Nvidia is taking the margins that usually typically would go to retailers and AIB partners like ASUS, EVGA, MSI or Gigabyte. Now, Nvidia has challenged their custom overclocked models, basically saying "I don't need you" to their clients, no doubt annoying their partnered AIBs and retailers. 

You can join the discussion on Nvidia's challenge to their AIB partners and their factory overclocked Founders Edition graphics cards on the OC3D Forums.  

«Prev 1 Next»

Most Recent Comments

21-08-2018, 07:11:01

MiNo
Hang on.



Overclocking is something that is done to a product after it leaves the manufacturers hands. You make it go faster than intended; from this comes the term "Over".


Aftermarket cards can be Overclocked by EVGA/MSI/ASUS since they are "Over" the specs from the manufacturer (Nvidia). And any card can be Overclocked by the end user.


But the manufacturer CANNOT overclock their own creation. It is simply against the definition of the word. A founders Edition just runs at the speed is does and is built for. In this case, perhaps by using binned chips. But they are not overclocked. Rather the chips given to others are then simply of lower quality - or "underclocked".Quote

21-08-2018, 07:20:18

WYP
Quote:
Originally Posted by MiNo View Post
Hang on.



Overclocking is something that is done to a product after it leaves the manufacturers hands. You make it go faster than intended; from this comes the term "Over".


Aftermarket cards can be Overclocked by EVGA/MSI/ASUS since they are "Over" the specs from the manufacturer (Nvidia). And any card can be Overclocked by the end user.


But the manufacturer CANNOT overclock their own creation. It is simply against the definition of the word. A founders Edition just runs at the speed is does and is built for. In this case, perhaps by using binned chips. But they are not overclocked. Rather the chips given to others are then simply of lower quality - or "underclocked".
Well, Nvidia's Founders Editions are not "Reference" designs anymore.

Nvidia now makes a "Reference" design for others to follow and surpasses it with their new models. I see your point, but TBH it is arguing semantics. AMD did similar things with their air-cooled and water-cooled Vega 64 GPUs, both were AMD designs but one performs better. Both could be considered reference models.

With Nvidia's new Founders Edition cards we now have a situation where some AIB models could offer lower performance levels than Nvidia Founders model.Quote

21-08-2018, 07:54:25

MiNo
Quote:
Originally Posted by WYP View Post
TBH it is arguing semantics

I agree, but feel it relevant still. Else almost any improvement to any product can be called overclocking / tuning or something like that. The term then looses its meaning.Quote

21-08-2018, 09:47:02

AlienALX
Quote:
Originally Posted by MiNo View Post
I agree, but feel it relevant still. Else almost any improvement to any product can be called overclocking / tuning or something like that. The term then looses its meaning.
It has been going on for years. Gigabyte's SOC "Super over clocked". EVGA do it too. Superclock and etc were names they used for years.

In its rawest definition no it isn't overclocking, but technically it is. And so is Turbo boost on CPUs and etc. They all raise the clocks and lower them accordingly.

The new Strix has OC on the box large in white. So don't blame the end user for getting confused Quote

21-08-2018, 10:34:20

MiNo
Quote:
Originally Posted by AlienALX View Post
It has been going on for years. Gigabyte's SOC "Super over clocked". EVGA do it too. Superclock and etc were names they used for years.

In its rawest definition no it isn't overclocking, but technically it is. And so is Turbo boost on CPUs and etc. They all raise the clocks and lower them accordingly.

The new Strix has OC on the box large in white. So don't blame the end user for getting confused

That is pretty much the opposite of the point I was making. Please read the post again:-)


Nvidia : Cannot call their own stuff OC'd. They produce a card with chip that can run a given speed. No matter what, that is the 'product speed'. Nvidia cannot say a chip runs at "500" speed, but they make it run at "600" speed. That makes the "500" bogus, and "600" become the norm. By running it at "600", that then becomes the 'spec' of that chip. Argh. Please understand :-)



EVGA/ASUS and the rest CAN. They get a chip, and make it run faster than Nvidia says it can. That is OC.Quote
Reply
x

Register for the OC3D Newsletter

Subscribing to the OC3D newsletter will keep you up-to-date on the latest technology reviews, competitions and goings-on at Overclock3D. We won't share your email address with ANYONE, and we will only email you with updates on site news, reviews, and competitions and you can unsubscribe easily at any time.

Simply enter your name and email address into the box below and be sure to click on the links in the confirmation emails that will arrive in your e-mail shortly after to complete the registration.

If you run into any problems, just drop us a message on the forums.