Why the GeForce RTX 4060 Ti could be a killer GPU for compact gaming PCs

pcb geforce rtx 3080

Small PCs like Intel’s Raptor Canyon NUC 13 Extreme are totally cool, but you’re not going to shove a GeForce RTX 4080 into one of those things. You could get away with a Radeon RX 7900 XTX, but it’s still going to be a snug fit and a pretty good sized chassis, as small form factor machines go. If you want to build something even smaller, you’ll need a much smaller GPU. Apparently, the GeForce RTX 4060 Ti might fit the bill.

Over the past few days, regular leakster @kopite7kimi has released new specs for both the GeForce RTX 4060 Ti and the standard GeForce RTX 4070. As an update, the card formerly known as GeForce RTX 4080 12GB, which has now been “not launched,” is expected to reappear under the name GeForce RTX 4070 Ti. These two cards would be the next two steps down from that model.
rtx4060ti kopite7kimi tweet

Looking at the specs, we see that the GeForce RTX 4070 is cut from the same AD104 die used for the RTX 4070 Ti. If the rumor is accurate, it would take big cuts to its computing capacity and probably its raster graphics capability as well. Those are on the order of 30%. However, the rumor seems to indicate that it will keep the same memory bus width and memory transfer rate, giving it the same memory bandwidth.

The coolest part is the GeForce RTX 4060 Ti, at least if kopite7kimi’s predictions are accurate. Presumably it will be based on a mostly intact AD106 die. It puts it down to 4,352 shaders, which if we assume a boost clock of 2.6GHz, gives it a peak single-precision computation speed of 23 TFLOPs. Not too shabby for a x60 part. However, the AD106 is known to only have a 128-bit memory interface, giving NVIDIA the option of 8 or 16GB.

ada vs ampere dec2022 chart

Worse, the RTX 4060 Ti will apparently feature lower clocked GDDR6 memory instead of the GDDR6X that the GeForce RTX 4070 will apparently use. The combination of narrow bus and middling transfer speeds gives the hypothetical GeForce RTX 4060 Ti just 288GB/ sec of memory bandwidth, less than the RTX 3060, RTX 2060, GTX 1660 Super, and even the GeForce GTX 1080. Ada’s drastically increased L2 cache seems to help with that, though.

The leaker notes that the reference card is “very short” and still uses the CEM5 connector. This is another name for the 12+4 pin 12VHPWR connector. It’s interesting to see it here; kopite7kimi seems to expect the RTX 4060 Ti to launch with a total board power of 220 watts. While by no means a fuel-efficient part, compared to the extremely thirsty models at the top of the product pile, it’s quite streamlined. It will likely consume less power while gaming too, just as we’ve seen with the RTX 4090 and its smaller sibling.

Leave a Reply

Your email address will not be published. Required fields are marked *