NVIDIA GeForce RTX 2060 12GB Has CUDA Core Count Rivaling RTX 2060 SUPER

[btarunr]

NVIDIA's surprise launch of the GeForce RTX 2060 12 GB graphics card could stir things up in the 1080p mainstream graphics segment. Apparently, there's more to this card than just a doubling in memory amount. Specifications put out by NVIDIA point to the card featuring 2,176 CUDA cores, compared to 1,920 on the original RTX 2060 (6 GB). 2,176 is the same number of CUDA cores that the RTX 2060 SUPER was endowed with. What sets the two cards apart is the memory configuration.

While the RTX 2060 maxed out the "TU106" silicon, the RTX 2060 12 GB is likely based on the larger "TU104," in order to achieve its CUDA core count. The RTX 2060 SUPER features 8 GB of memory across a 256-bit wide memory bus, however, the RTX 2060 12 GB uses a narrower 192-bit wide bus, disabling 1/4th of the bus width of the "TU104." The memory data-rate on both SKUs is the same—14 Gbps. The segmentation between the two in the area of GPU clock speeds appears negligible. The original RTX 2060 ticks at 1680 MHz boost, while the new RTX 2060 12 GB does 1650 MHz boost. The typical board power is increased to 185 W compared to 160 W of the original RTX 2060, and 175 W of the RTX 2060 SUPER.

Update 15:32 UTC: NVIDIA has updated their website to remove the "Founders Edition" part from their specs page (3rd screenshot below). We confirmed with NVIDIA that there will be no RTX 2060 12 GB Founders Edition, only custom designs by their various board partners.



NVIDIA is getting its add-in card partners to come up with several custom-design products based on the new SKU, which should occupy price-points below those of the RTX 3060 "Ampere." This could be an answer to AMD's Radeon RX 6600 (non-XT), which beats the RTX 2060 SUPER by 3% and the original RTX 2060 by 13%, at 1080p, in our testing. Technologically, the older "Turing" architecture won't find itself obsolete in the current market, as it maintains full DirectX 12 Ultimate compatibility.

View at TechPowerUp Main Site

 

[tomo82]

1. To the surprise of no one, it's been on the rumor mill for a while now.

2. It's still rubbish though! And why use fab capacity to build something old that nobody really wants? Are they using an old node that nobody is using?

 

[Prima.Vera]

 

[Selaya]

Yes, it's being made on TSMC 12nm. Recycling an obsolete node for a low-end ($600's low end these days haHAA) product isn't the worst idea, esp during a shortage.

 

[diopter]

So they increased the CUDA core count and the VRAM size in order to justify increasing the price of it for gamers and giving some gaming performance increase. Then they left the memory bandwidth the same as original 2060 in order to minimise the ethereum mining performance and keep it equivalent to the older lower spec card. It will appeal to gamers and not really to miners. This was the intention.

 

[simmomania]

'While the RTX 2060 maxed out the "TU106" silicon' - this isn't correct, 2070 (non-Super) is a full TU106

 

[bug]

I think I'll just wait till they re-release my 1060.

 

[Yraggul666]

@Selaya 600$ in our dreams, at this very moment two 2060 models are in stock at one of the major etailers in my piss poor excuse of a country;
one costs 912$ and the cheaper one is 820$......

 

[TheoneandonlyMrK]

Typically Nvidia, again with the massive amounts of SKUs with the same name but f all else in parity.
So confusing though I suppose I should be greatfull, my own 2060 (or weva a mobile 2060 is)won't go eol for a bit.

 

[deu]

rebinning of higher sku failed dies?

 

[tabascosauz]

Is this news story supposed to make me feel better about wanting to get a waterblock for my 2060 Super? Because by golly, it's working lol

Between where the 3070 and 6700XT currently stand here, all signs point to patiently waiting for the next gen...

Also, 185W is higher than either 2060S or 2070 - if the image is real about it being Founders and they really are putting the short FE cooler on this, it's pretty close to the limit of what that cooler can handle.

 

[Chomiq]

Is this news story supposed to make me feel better about wanting to get a waterblock for my 2060 Super? Because by golly, it's working lol

Between where the 3070 and 6700XT currently stand here, all signs point to patiently waiting for the next gen...

Also, 185W is higher than either 2060S or 2070 - if the image is real about it being Founders and they really are putting the short FE cooler on this, it's pretty close to the limit of what that cooler can handle.

I doubt this will get any better with next gen. Fingers crossed my 1060 doesn't expire before that.

 

[bug]

I doubt this will get any better with next gen. Fingers crossed my 1060 doesn't expire before that.

I'm in the same boat. I don't really game anymore, but I do have a backlog my 1060 can help me with. Having just bought a 12600k, if my video card gives up the ghost, I can make do with the IGP. No way I'm paying the prices we're seeing today for a video card.

 

[tomo82]

I'm in the same boat. I don't really game anymore, but I do have a backlog my 1060 can help me with. Having just bought a 12600k, if my video card gives up the ghost, I can make do with the IGP. No way I'm paying the prices we're seeing today for a video card.

Same here with 980ti... except no IGP to fall back on (hang in there buddy! ... *gulp*)

 

[bug]

Same here with 980ti... except no IGP to fall back on (hang in there buddy! ... *gulp*)

Technically, I'm still on my 6600k, so no IGP either. the 12600k is on its way, the (free!) adapter kit for my Freezer 34 is already here and Gigabyte still has to actually make some Aorus Pro DDR4 boards... But I'm getting there.

 

[DeeJay1001]

1. To the surprise of no one, it's been on the rumor mill for a while now.

2. It's still rubbish though! And why use fab capacity to build something old that nobody really wants? Are they using an old node that nobody is using?

2000 series cards are based on TSMC 12nm, to my knowledge no current or next generation CPU, GPU, or other primary computer components use TSMCs 12nm.

My guess is one of two things are happening.
1. Nvidia has these dies already produced and are simply finishing off the stock.
2. TSMC had production contracts available for 12nm at an attractive price and Nvidia took them up on their offer.

 

[bug]

2000 series cards are based on TSMC 12nm, to my knowledge no current or next generation CPU, GPU, or other primary computer components use TSMCs 12nm.

My guess is one of two things are happening.
1. Nvidia has these dies already produced and are simply finishing off the stock.
2. TSMC had production contracts available for 12nm at an attractive price and Nvidia took them up on their offer.

Probably the latter, since the chips themselves appear to be a new design.

 

[r9]

2060 on d1ck pills.

 

[Vya Domus]

Yes, it's being made on TSMC 12nm. Recycling an obsolete node for a low-end ($600's low end these days haHAA) product isn't the worst idea, esp during a shortage.

They didn't recycle anything, it's literally the same chip on the same node.

 

[RandallFlagg]

So 2060 Super with the memory bandwidth of a 2060 and double the memory.

Not cost effective for crypto mining.

All depends on how many of these they can make.

 

[zlobby]

When is the rehash of 980 expected?

 

[Selaya]

They didn't recycle anything, it's literally the same chip on the same node.

That's what I meant. Maybe reposting'd have been more apt? The point is just that they either keep making those or are making them again, to improve availability on the lowend

 

[bug]

When is the rehash of 980 expected?

After the "launch" of 4080, I guess.

 

[Muaadib]

Is this news story supposed to make me feel better about wanting to get a waterblock for my 2060 Super? Because by golly, it's working lol

Between where the 3070 and 6700XT currently stand here, all signs point to patiently waiting for the next gen...

Also, 185W is higher than either 2060S or 2070 - if the image is real about it being Founders and they really are putting the short FE cooler on this, it's pretty close to the limit of what that cooler can handle.

Same boat here, was contemplating how and why to upgrade then saw prices and now I'm looking at aftermarket coolers for my gpu to prolong its life.

Technically, I'm still on my 6600k, so no IGP either. the 12600k is on its way, the (free!) adapter kit for my Freezer 34 is already here and Gigabyte still has to actually make some Aorus Pro DDR4 boards... But I'm getting there.

6600K has an iGPU, HD530 iirc.

 

[xorbe]

This card WON'T be supported on Win7, right? Or are they going to release a 472 driver update?