NVIDIA is expected to launch its GeForce RTX 4060 Ti and RTX 4060 GPUs aiming at the mainstream segment in May of this year.
NVIDIA GeForce RTX 4060 Ti & RTX 4060 GPUs Reportedly Launching In May: AD106 160W For Ti & AD107 115W For Non-Ti
The NVIDIA GeForce RTX 4060 Ti & RTX 4060 graphics cards based on the Ada GPU architecture are expected to launch after the RTX 4070 which launches early next month on the 13th of April. This is going to be a full retail launch and our sources state that May is being targeted for both GPUs. We also managed to get some additional details on the PCBs being used by these cards so let's start at the Ti variant first.
NVIDIA GeForce RTX 4060 Ti Specifications "Rumored":
The NVIDIA GeForce RTX 4060 Ti is expected to utilize the AD106-350-A1 GPU core, a cut-down version of the full AD106 graphics chip, and based on previous rumors, it should pack 34 SMs or 4352 CUDA cores, an 8 GB GDDR6 memory running at 18 Gbps across a 128-bit bus interface, providing the card with 288 GB/s of bandwidth. There's also 32 MB of L2 cache on board the GPU which is an 8x increase over the GeForce RTX 3060 Ti.
The graphics card is expected to utilize the PG190 SKU 361 PCB. Leaked PCB shots revealed that while the reference Founders Edition makes use of the standard 16-Pin 12VHPWR connector, the custom models will be free to use either standard 8-pin connectors or the newer 16-pin plugs. We have been told that it is much cheaper to just go with 8-pin connectors so that should be the target as AIBs want to maximize the production value of their designs. As for the TDP, the card will end up between 150-160W which is around 25% lower than its predecessor, the RTX 3060 Ti.
NVIDIA GeForce RTX 4060 Ti Founders Edition Graphics Card (Image Credits: KittyYYuko):
NVIDIA GeForce RTX 4060 Specifications "Rumored":
For the GeForce RTX 4060, NVIDIA is expected to utilize the AD107-400-A1 GPU, the fully enabled AD107 Ada die. Do note that the same configuration is used by the GeForce RTX 4060 Laptop GPU. The chip is expected to feature 3072 CUDA cores and 8 GB of GDDR6 memory running at 18 Gbps memory speeds across a 128-bit bus interface. That gives the card 288 GB/s bandwidth and all of that will fall within a 110W - 115W reference TDP. The GPU is also said to feature 24 MB of cache (L2).
The NVIDIA GeForce RTX 4060 should end up with a 35% lower TGP than its predecessor, the RTX 3060 12 GB. While previous rumors pointed out the card using a PG190 PCB, our info suggests it would use the PG173 SKU 371 PCB design. It is likely that the PG190 variant is used by the FE and the PG173 is used by the reference AIC boards. Just like the RTX 4060 Ti, the Non-Ti variant will mostly rely on the standard 8-pin connectors and we will get some nice Mini-ITX variants of the card too, making for some truly efficient gaming PCs.
Using the rumored "reference" boost frequency of up to 2535 MHz as a reference, the NVIDIA GeForce RTX 4060 Ti should offer around 22 TFLOPs of compute performance with AIB models boosting it close to 24 TFLOPs.
FP32 Compute Horsepower Comparisons (Higher is Better)
Based on the numbers above, it looks like the NVIDIA GeForce RTX 4060 Ti graphics card is going to end up similar to the GeForce RTX 3070 Ti in performance. The 3070 Ti launched at $599 US so the 4060 Ti would have to be positioned close to $399 US to make it a worthy successor to the 3060 Ti which also had the same MSRP. The main issue is that while the card may do well at 1080p, higher resolutions would become a burden for it due to its lowly 128-bit bus interface and reliance on the slower G6 memory rather than G6X. The card would make for an extremely efficient design in the 150W TDP range but the rest remains to be seen when the card is made official and reviews come out.
NVIDIA GeForce RTX 40 Series Graphics Card Lineup (Rumored):
| Graphics Card | GPU | PCB Variant | SM Units / Cores | Memory / Bus | Memory Clock / Bandwidth | TBP | Power Connectors | Launch |
|---|---|---|---|---|---|---|---|---|
| NVIDIA Titan RTX ADA | AD102-400? | TBD | 144 / 18176? | 48 GB / 384-bit | 24 Gbps / 1.15 TB/s | ~800W | 2x 16-pin | TBD |
| NVIDIA GeForce RTX 4090 Ti | AD102-400? | TBD | 144 / 18176? | 24 GB / 384-bit | 24 Gbps / 1.15 TB/s | ~600W | 1x 16-pin | TBD |
| NVIDIA GeForce RTX 4090 | AD102-300 | PG136 | 128 / 16384 | 24 GB / 384-bit | 21 Gbps / 1.00 TB/s | 450W | 1x 16-pin | Q4 2022 |
| NVIDIA GeForce RTX 4080 | AD103-301/300 | PG139 SKU 360 | 76 / 9728 | 16 GB / 256-bit | 23 Gbps / 716.8 GB/s | 320W | 1x 16-pin | Q4 2022 |
| NVIDIA GeForce RTX 4070 Ti | AD104-400 | PG141 SKU 331 | 60 / 7680 | 12 GB / 192-bit | 21 Gbps / 504.0 GB/s | 285W | 1x 16-pin | Q1 2023 |
| NVIDIA GeForce RTX 4070 | AD104-251/250 | PG141-SKU 345/344/343 | 46 / 5888 | 12 GB / 192-bit | 21 Gbps / 504.0 GB/s | 200W | 1x 16-pin / 8-Pin | Q2 2023 |
| NVIDIA GeForce RTX 4060 Ti | AD106-350 | PG190 SKU 361 | 34 / 4352 | 8 GB / 128-bit | 18 Gbps / 288.0 GB/s | 160-150W | 1x 16-pin / 8-Pin | Q2 2023 |
| NVIDIA GeForce RTX 4060 | AD107-400 | PG173 SKU 371 | 24 / 3072 | 8 GB / 128-bit | 18 Gbps / 288.0 GB/s | 115-110W | 1x 16-pin / 8-Pin | Q2 2023 |
| NVIDIA GeForce RTX 4050 | AD107? | TBD | TBD | 6 GB / 96-bit? | TBD | 100W | 1x 16-pin / 8-Pin | Q3 2023? |
