NVIDIA GeForce RTX 4060 Ti rumored to feature 4352 CUDAs, 8GB memory and 220W TDP

Published: Dec 13th 2022, 20:45 GMT   Comments

Please note that this post is tagged as a rumor.

NVIDIA RTX 4060 Ti with short PCB

Kopite7kimi, a hardware leaker that needs no introduction, has just revealed the preliminary specs of NVIDIA’s upcoming AD106 GPU. 

One should remind our readers that although accurate, early board information is not always final. NVIDIA is particularly known for sharing various specs with board partners, but oftentimes changing them at the last minute. In fact, what NVIDIA did with RTX 4080 12GB/4070 Ti shows exactly how everything can change, even after the product was announced.

According to the latest rumors, RTX 4060 Ti would get AD106-350 GPU with 4352 CUDA cores. This particular SKU would feature 8GB of GDDR6 memory clocked at 18 Gbps. This would be the first known model from RTX 40 series to feature GDDR6 memory instead of GDDR6X. Earlier leaks mentioned that AD106 features 128-bit memory bus, which means this card would have a bandwidth of 288 GB/s.

Furthermore, this model would launch with a TDP of 220W, which is 30W lower than the rumored RTX 4070. Interestingly, Kopite claims that the reference board for this card is very short, supposedly even smaller than RTX 4070 series. What may be important is that NVIDIA is reportedly still relying on the new 16-pin power connector. This could mean that all RTX 40 GPUs will get this connector, not just the high-power ones.

Just a few days ago, Kopite also shared the latest specs of RTX 4070 non-Ti. This SKU would feature 35% more cores, 4GB more memory capacity and 250W TDP. However, unlike the supposed RTX 4060 Ti, it would feature AD104 GPU instead.

NVIDIA is set to announce its RTX 40 Laptop GPU series as early as January next year. This means that AD104/AD106 GPUs should be ready for desktop cards as well. However, the release date of these cards is still unknown.

RUMORED NVIDIA GeForce RTX 40 Series Specs
VideoCardz.comRTX 4090RTX 4080RTX 4070 TiRTX 4070RTX 4060 Ti
Picture
Board-SKUPG136-SKU330PG136-SKU360PG141-SKU331PG141-SKU336/337PG190
ArchitectureAda (TSMC 4N)Ada (TSMC 4N)Ada (TSMC 4N)Ada (TSMC 4N)Ada (TSMC 4N)
GPUAD102-300AD103-300AD104-400AD104-250AD106-350
GPU Transistors
 
76.3B
 
45.9B
 
35.8B
 
35.8B
TBC
GPU Die Size
 
608 mm²
 
379 mm²
 
295 mm²
 
295 mm²
TBC
SMs
 
128
 
76
 
60
 
46
 
34
CUDA Cores
 
16384
 
9728
 
7680
 
5888
 
4352
Boost Clock
 
2.5 GHz
 
2.5 GHz
 
2.6 GHz
TBCTBC
Memory
 
24 GB G6X
 
16 GB G6X
 
12 GB G6X
 
12 GB G6X
 
8 GB G6
Memory Bus
 
384-bit
 
256-bit
 
192-bit
 
192-bit
 
128-bit
Memory Speed
 
21 Gbps
 
22.4 Gbps
 
21 Gbps
 
21 Gbps
 
18 Gbps
Bandwidth
 
1008 GB/s
 
736 GB/s
 
504 GB/s
 
504 GB/s
 
288 GB/s
Default TGP
 
450W
 
320W
 
285W
 
250W
 
220W
MSRP
 
1599 USD
 
1199 USD
TBCTBCTBC
Release DateOctober 12th, 2022November 16h, 2022January 5th, 2023TBCTBC

Source: Kopite7kimi




Comment Policy
  1. Comments must be written in English.
  2. Comments must not exceed 1000 characters. Comment splitting is not allowed.
  3. Comments deemed to be spam or solely promotional in nature will be deleted.
  4. Discussions about politics are not allowed on this website.
  5. Sharing relevant links is permitted; avoid bypassing the link and word filters. Our team will approve links.
  6. Comments complaining about the post subject or its source will be removed.
  7. Offensive language in comments or usernames result in a ban.
  8. Direct attacks/harassment result in immediate ban.
  9. VideoCardz isn’t and was never sponsored by AMD, Intel, or NVIDIA. Users claiming otherwise will be banned.
  10. Moderators may edit/delete comments without notice.
  11. If you have any questions about the commenting policy, please let us know through the Contact Page.
Hide Comment Policy
Comments