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.
164 Comments on NVIDIA GeForce RTX 4060 Ti Possible Specs Surface—160 W Power, Debuts AD106 Silicon
RTX 6080 XT GB $650
Which means the 4070 Ti is 22% faster at a 21% higher price. Performance per dollar is basically the same, on a card with a die that is 47% the size of the one used in the 3080, with a tiny little memory bus, much simpler PCB design and VRAM capacity larger by just 20%.
And this is almost 2.5 years after the release of the 3080. The manufacturing cost has not increased by a factor of 2.13 to justify the same performance per dollar!
Asus TUF GAMING GeForce RTX 4070 Ti 12GB Video Card $799.99
GTX 1080 Ti: $699
GTX 980 Ti: $649
GTX 780 Ti: $699
GTX 680: $499
GTX 580: $499
GTX 480: $499
GTX 280: $649
9800 GTX: $299
8800 GTX: $599
7800 GTX: $599
Funny how you managed to avoid showing that $300 jump from the GTX 1080 Ti to the RTX 2080 Ti, eh? Well, hiding the truth doesn't change reality, it just makes you dishonest and then nobody takes your words seriously. I know that I won't because what you just did is a marketing trick which means there's a strong chance that you're a paid shill for nVidia. Even if you're not, I just turned any credibility you ever had to dust.Not THIS gamer because I've been around for decades as well. It's the young noobs that don't know better.
4070 TI - $799 MSRP
3070 TI - $599 MSRP
2070 Super - $499 MSRP
1070 TI - $399 MSRP
The 780 vs the 980 was a 38% increase in performance for a $100 price CUT. The 1080 was a $50 price increase for a 50% increase. The 2080 was a 39% increase for a $100 premium, and the 3080 was the same price for a 63% jump in performance. Now the 4080 is 42% more powerful for five hundred more dollars. Where is the technological progression? Tech is supposed to come down in price/performance over time.
Value is the thing that is important, except to halo product buyers. The trouble is, now the XX70 TI level has become a halo product.
pcpartpicker.com/search/?q=rtx+3070+ti
RTX 3070 Ti 8GB $669.99
The 8800 GTS 512 and 9800 GTX were on par with the 8800GTX. The top end 9 series from Nvidia was just basically a rebrand top end 8800 series. I think it was the most stagnant of all generational changes in my GPU history that I can remember.
RTX 4070 Ti 12GB Video Card MSRP $799.99
Some people believe that it's fair to compare the 4070Ti to something 22% slower (because similar price/perf) others think it's worth comparing to the 3090 and see better performance for less money in the 4070Ti. Both are technically correct.
At the end of the day, we're far from getting the BEST perf/$ uplift we had in the past but there it is some uplift.
RX 6900 XT 16GB Video Card $809.00
pcpartpicker.com/product/kVqPxr/asus-tuf-gaming-geforce-rtx-4070-ti-12-gb-video-card-tuf-rtx4070ti-12g-gaming
Asus TUF GAMING GeForce RTX 4070 Ti 12GB Video Card $799.99
IIRC, there actually was a small difference between the 8800 GTX and 9800 GTX but I think that it was mostly the fact that the 9000-series was PCI-Express v2.0 while the 8000-series was PCI-Express v1.x. The stagnation at the time was because the 8800 GTX just annihilated the HD 2900 XT and was even faster than the HD 3870. There wouldn't be serious competition from ATi until they fixed the TeraScale architecture with HD 4000-series. That's when the HD 4870x2 knocked the GTX 280 off of its perch as the fastest card in the world. Jensen had just sat back and let his R&D department work its butt off to have their next product ready and waiting for ATi's response while just refreshing the G80 to the G92 with the new interface.
That's why nVidia was able to react so quickly to the HD 4870x2 with the GTX 295. I think that nVidia expected to keep one step ahead of ATi in this way. However, just like ATi flubbed early TeraScale at the worst possible time, when nVidia had created the G80, nVidia flubbed the GF100 at the worst possible time. This is because ATi released their Evergreen HD 5000-series and it just obliterated everything that nVidia had at the time.
Guru3D's Hilbert Hagedoorn's review of the Radeon HD 5970 (code name Hemlock) had the funniest paragraph that I've ever read in a tech review. I actually laughed my butt off reading his article because I'd never seen someone say something like this before (or since for that matter) in a tech review:
"(Tom Clancy's) HAWX is very lenient towards ATI cards thanks to DX10.1, but with so much brute power, NVIDIA has been working nervously to optimize their drivers. And as such the more recent drivers make the GeForce cards much more competitive. As such the GTX 295 pushes 42 FPS at 2560x1600 with 4xAA (on average), but the Radeon HD 5970 brutally sodomizes the GTX 295 here with nearly doubled up performance."
Let me help you with the imagery...
EVGA GeForce GTX 295 before the HD 5970:
HD 5970:
GTX 295 after the HD 5970 had done its "brutal giving of the sword":
"Yes, it's childish and stupid, but then, so is high school!"
- Ferris Bueller, 1986 :roll:
So to me: