The cheaper GeForce RTX 4000 graphics models that are coming to the market now could bring a reduction in consumption against their predecessors after a long time, instead of increasing it.
Even before the end of the year, information was leaked about how the upcoming cheaper Nvidia graphics with Ada Lovelace’s new architecture should look like: GeForce RTX 4070 and GeForce RTX 4060 Ti. However, it looks like they may be launched in a slightly different form than the company intended at the time. According to proven leaker Kopite7kimi, these graphics will have a significantly lower TDP than previous leaks and could be quite economical and efficient GPUs.
When Kopite7kimi communicated information about the GeForce RTX 4060 Ti for the first time (a month or so ago), we learned that the card has an AD106-350 chip, which is said to contain 4352 shaders (34 SM blocks) and 32MB of L2 cache. It’s too early to tell about clocks, but the card is said to use 8GB of GDDR6 memory, which will run at 18.0GHz (effectively). This will give it a throughput of 288 GB/s, since the AD106 GPU is only supposed to have a 128-bit bus.
It was also said then that this card, which in reference form should have a very short PCB (with a 16-pin power connector), will have a TDP of 220 W. That would be the same consumption as the previous generation GeForce RTX 3070 model. And higher, than the RTX 3060 Ti (200 W) had, not to mention the older RTX 2070 and GTX 1070 models.
But this is said to have changed. Kopite7kimi now reports that Nvidia has reconsidered the policy and the GeForce RTX 4060 Ti is said to drop down to a TDP of just 160W, which is quite a drastic reduction. In this case, the card would even have a lower TDP than GeForce RTX 3060 (170 W). So it could probably be a relatively energy-efficient graphics card. It is true, however, that with this reduction, the performance that the graphics are aimed at will probably also shift down, compared to the previously intended 220W configuration.
After correction, 160W. https://t.co/thAbqh0Ndq
– kopite7kimi (@kopite7kimi) January 18, 2023
RTX 4070 also more economical?
According to Kopite7kimi, more models have undergone this change of targets, and it is said that a reduction in TDP can also be expected for the GeForce RTX 4070. Its parameters were also revealed last month – it should have an AD104 chip with a total of 5888 shaders (46 active SMs), 32MB L2 cache and 192-bit GDDR6X memory with a capacity of 12 GB and a throughput of 504 GB/s (21 GHz effective clock).
According to the first version of the specifications, this card was supposed to have a TDP of 250 W. In its case, however, the new value is not yet certain. Perhaps it could be somewhere around 200 W. Again, this would lead to a reduction in consumption in a specific class of cards, which we would quite welcome, because graphics manufacturers have been pushing consumption higher and higher for a long time (especially Nvidia in recent years).
Kopite7kimi reports that the card is not internally labeled PG141-SKU336/337 (as originally stated) but PG141-SKU343/344/345. This could indicate some other changes. As with the GeForce RTX 4060 Ti, it is probably possible that with the reduction in consumption, the card will shift slightly lower in performance, as it will show lower clocks in games.
Why does Nvidia reduce consumption? It is possible that the chips produced by the 4nm process behave better than the company originally calculated, and thus it is possible to achieve better performance and consumption ratios with them. But it is also possible that it is an adaptation to the situation in a world where the price of electricity is rising sharply and there is suddenly greater pressure to save. But it could also be a reassessment of the competitive situation, where Nvidia thinks it won’t need as much performance against rival Radeon graphics…
