Thursday, January 19th 2023
NVIDIA GeForce RTX 4060 Ti Possible Specs Surface—160 W Power, Debuts AD106 Silicon
NVIDIA's next GeForce RTX 40-series "Ada" graphics card launch is widely expected to be the GeForce RTX 4070 (non-Ti), and as we approach Spring 2023, the company is expected to ramp up to the meat of its new generation, with xx60-segment, beginning with the GeForce RTX 4060 Ti. This new performance-segment SKU debuts the 4 nm "AD106" silicon. A set of leaks by kopite7kimi, a reliable source with NVIDIA leaks, shed light on possible specifications.
The RTX 4060 Ti is based on the AD106 silicon, which is expected to be much smaller than the AD104 powering the RTX 4070 series. The reference board developed at NVIDIA, codenamed PG190, is reportedly tiny, and yet it features the 16-pin ATX 12VHPWR connector. This is probably set for 300 W at its signal pins, and adapters included with graphics cards could convert two 8-pin PCIe into one 300 W 16-pin connector. The RTX 4060 Ti is expected to come with a typical graphics power value of 160 W.At this point we don't know whether the RTX 4060 Ti maxes out the AD106, but its rumored specs read as follows: 4,352 CUDA cores across 34 streaming multiprocessors (SM), 34 RT cores, 136 Tensor cores, 136 TMUs, and an unknown ROP count. The GPU is expected to feature a 128-bit wide GDDR6/X memory interface, and 8 GB could remain the standard memory size. NVIDIA is expected to use JEDEC-standard 18 Gbps GDDR6 memory, which should yield 288 GB/s of memory bandwidth. It will be very interesting to see how much faster the RTX 4060 Ti is over its predecessor, the RTX 3060 Ti, given that it has barely two-thirds the memory bandwidth. NVIDIA has made several architectural improvements to the memory sub-system with "Ada," and the AD106 is expected to get a large 32 MB L2 cache.
Sources:
kopite7kimi (Twitter), VideoCardz
The RTX 4060 Ti is based on the AD106 silicon, which is expected to be much smaller than the AD104 powering the RTX 4070 series. The reference board developed at NVIDIA, codenamed PG190, is reportedly tiny, and yet it features the 16-pin ATX 12VHPWR connector. This is probably set for 300 W at its signal pins, and adapters included with graphics cards could convert two 8-pin PCIe into one 300 W 16-pin connector. The RTX 4060 Ti is expected to come with a typical graphics power value of 160 W.At this point we don't know whether the RTX 4060 Ti maxes out the AD106, but its rumored specs read as follows: 4,352 CUDA cores across 34 streaming multiprocessors (SM), 34 RT cores, 136 Tensor cores, 136 TMUs, and an unknown ROP count. The GPU is expected to feature a 128-bit wide GDDR6/X memory interface, and 8 GB could remain the standard memory size. NVIDIA is expected to use JEDEC-standard 18 Gbps GDDR6 memory, which should yield 288 GB/s of memory bandwidth. It will be very interesting to see how much faster the RTX 4060 Ti is over its predecessor, the RTX 3060 Ti, given that it has barely two-thirds the memory bandwidth. NVIDIA has made several architectural improvements to the memory sub-system with "Ada," and the AD106 is expected to get a large 32 MB L2 cache.
164 Comments on NVIDIA GeForce RTX 4060 Ti Possible Specs Surface—160 W Power, Debuts AD106 Silicon
The only market that's shafted is the low end one where cards at 120 to 250 euros used to play games very decently.
Keep buying those high end and upper midrange Adas. Huang is laughing all the way to the bank with his billions of dollars in profits.
Revenue was up over 61% for 2022 over 2021.
If the 4070ti is "really" a 4060, then you are basically telling me a 4060 beats the snot out of amds best gpu in RT? Well no wonder huang is charging us that much, when his low end cards are competing with amds top range halo products, lol.
And as I see it, the only effective method to 'fight back' is to sit on your GPU for as long as you can.
Nothing else will be even remotely as efficient.
And note, this is not for NV GPU only, it's for AMD, Intel and so on.
They all deserve the same treatment.
The 3070 and 3060ti came out two years ago and are already getting long in the tooth for being supposed 1440p cards.
www.dsogaming.com/wp-content/uploads/2022/12/Warhammer-40000-Darktide-GPU-Benchmarks-2.png
The 2080ti hits 60 fps average at native 1440p. The 3060ti isn't that far off the 2080ti in performance, at least according to the tpup review, the difference is 10%. So a 3060ti with DLSS Q can absolutely get great performance even in this game at MAX settings. But even if it doesn't, I don't really know what point you are trying to make exactly? That there are games that run like dogshit? Sure, plenty, already gave you the example of ark. So?
But if you add a Ti suffix and shave off 50$, then they'll have my attention (also considering the 160W).
60 fps avg minus 10% is 54 avg, and lows will be 40 or sub 40. Thats... pretty much the gist of what im saying a page long now. Im glad we reached consensus and you confirmed all those goalposts were moved for nothing ;)
TL DR; mid range gaming didnt improve at all. You get less gpu for your money throughout the stack since Turing. Midrange gaming was great about 5-6 years back, between Maxwell and Pascal.
I mean your whole premise is fundamentally flawed. You mentioned Pascal. Do you consider the 1080ti good enough for 1080p at the time? Cause even back when it was released, there were games OLDER than the actual card that it would struggle to run at 1080p!!! So I guess - using your logic - the 1080ti wasnt good enough for 1080p even back in 2017. Does that make sense to you? Cause that's what you are saying about the 3060ti. "oh look, there is a game it can't run at steady 60 (with DLSS OFF) at max settings, therefore it's not a good 1440p card". Okay :D
So.....
RTX 2060 Super was faster than the gtx 1080 for much less $$
RTX 3060 Ti was tied with the rtx 2080 super for much less $$
RTX 4060 Ti maybe tied with rtx 3070 for the same price?
I swear this freaking gpu prices are upside down.
What's next? rtx 5060 ti tied with rtx 4060 ti for more money?
lol....
NVIDIA believes DLSS 3 and frame generation will make up for it's cut downs to previous generations.
They may well be right, but this 4060Ti will be at least 500 bucks and here in EU lots more.
People like me who always buy budget to mid-range,
we're getting screwed over.
Here's hoping AMD remedies the situation!
At this point, upgrading every generation is just a waste of money because you'll just be paying infinitely more money for an infinitely smaller positive impact on your gaming experience. Truth be told, I'd probably be just fine today if I were still using my RX 5700 XT. If the vast majority of gamers are perfectly fine with GPUs as weak as the GTX 1650, then even the RX 580, a card that is over 30% faster, wouldn't be something that most gamers would feel the need to upgrade. In fact, most gamers would be upgrading TO the RX 580 if they needed to upgrade at all. That sounds really weird to say. :laugh: I'm sorry, are you referring to last-gen or current-gen because that description hilariously fits BOTH! :laugh: I don't know if that's what he's doing because the RX 6800 XT wasn't the fastest card of its generation to begin with. There were 4 or 5 cards in that gen that were already faster. It didn't bother me that my RX 6800 XT wasn't the fastest card because it was the best value high-end card by a landslide. Now I know that the bar was set so low as to be underground but the one thing that I was proud of was that I paid $500 less for my RX 6800 XT than was the going rate at the time. Having said that, I still paid $500 too much for it and I still swear at myself in the mirror over it. I agree with you that there's no reason to get annoyed that your card is slower than the new gen because it if wasn't, the new gen wouldn't exist. Anyone who has ever bought a video card before knows that while its viability will last for many years, it won't be on the top charts for more than three at the most. Yeah, but it doesn't beat the RX 6800 XT in any meaningful way because a lot of its excess performance is more applicable to it remaining at the high-end than anything else. I personally won't see a slide in performance from my RX 6800 XT for many years because I tend to game at 1440p60Hz.
I doubt it will cost more than $500, though (MSRP, not real prices). There still has to be room for the 4070. But even $500 is a ridiculous price for a card with a microscopic die of 190 mm2, with laughable 8 GB of VRAM and a bus width cut in half.
www.newegg.com/p/pl?d=6800xt
shop.asus.com/us/90yv0ij1-m0aa00-tuf-rtx4070ti-12g-gaming.html
Asus TUF GAMING RTX 4070 Ti 12GB Video Card $799.99
If you use DLSS/FSR most higher end cards will do 4K just fine.
If not, 4080, 7900XTX and especially 4090 is the only good options, unless you don't have high demands for image quality and framerate.
Everything in this generation is out of proportion. And yes, if you are upgrading from and old system, these cards are better than anything before, but it does not change the fact that performance per dollar improvement is almost non-existant. The manufacturing cost argument would only be valid if these GPUs had the same die sizes as the previous generation. But they do not, not even close.
shop.asus.com/us/90yv0ij1-m0aa00-tuf-rtx4070ti-12g-gaming.html
Asus TUF GAMING RTX 4070 Ti 12GB Video Card $799.99
Not sure what you are defending here. Do you really believe the 40 series is appropriately priced in terms of performance per dollar, ignoring the naming of the cards?
RTX 3080 Ti 12GB MSRP: $1200
RTX 4070 Ti 12GB MSRP: $800