Over the past few hours, information has surfaced about graphics cards recently introduced by AMD, as RX 7900 XT(X). In which, according to the same due to design error coupled with a very tight release schedule, hand ended up with a design with a lower clock speed than expected. Is it true or, on the contrary, is it disinformation in order to boycott its launch? Let’s find out.
One of the big problems with information leaks is that they tend to add elements of their own harvest to the few numbers that companies give them to let them know about a future product. In the end, they finger themselves with their lies. However, they can be petty enough to say that one of their made-up rumors if it doesn’t come true is the fault of the chip designer or manufacturer.
The RX 7900 XT(X) 3 GHz controversy tweet
It all started in the last hours with a simple tweet, without the support of AMD itself, in which it is said that the Navi 31 chip must be redesigned due to the fact that the clock speed and the consumption curve are false and despite what the architecture could reach 3 GHz, it remains at 2.9 GHz under a consumption of 450 W.
Which is ironic considering that one of the things AMD has been pursuing for three generations now is to improve performance per watt. And one way to do that is to get higher speeds at the same voltage or existing ones at lower voltages. The point is, extraordinary claims require evidence of the same caliber and we cannot, at least until launch date, pick up an RX 7900 XT(X) and verify if the information is correct. In any case, a redesign would cost them between 9 and 12 months, enough for an RX 7950 XT(X) or who knows, an RX 8900 XT(X) based on RDNA 3+. So it wouldn’t be something they could solve in a few weeks or a few months.
Should they be faster than they are?
The larger a chip, the lower its clock speed, as the time it takes electrons to travel from one end of the chip to the other increases. Disaggregating a chip into multiple “chiplets” in an unaggregated fashion not only has the effect of reducing manufacturing costs, but also enables higher clock speeds. Officially, AMD only gave the Game Clock, 2.3 GHz, which is better than the 2 GHz of the RX 6900 XT, but we have to keep in mind that we have a more advanced crafting node. TSMC “5nm” vs “7nm” and a smaller chip.
However, this is a simplistic view and does not take into account an element that results from the disintegration of a chip and the increase in the communication distance between the parts of the GPU or the graphics chip.
What limits the clock speed of these graphics cards?
The explanation is in communication between GCD and DCM. And it is that when you move data out of a chip, the power consumption increases when the distance is greater. This is the price to pay for having chiplets. The solution in the future? The big three are studying optical interfaces for this. Logic tells us that’s why NVIDIA would have opted for a monolithic solution, the fact that they would be clock speed limited by data speed consumption
In other words, the Taiwanese founder had to create some kind of packaging so that at these data transfer rates the power consumption does not skyrocket and even then it takes some of the total power allocated to the GPU. . In any case, this does not prevent AMD from reviewing its RDNA 3 GPUs to make them more efficient throughout the commercial life of the architecture. Especially if we take into account the fact that we have already seen it with the RDNA 2 and the different versions of the RX 6000.
