Latest Nvidia Blackwell Specs Rumor Emerges

The remarkable Nvidia Blackwell specifications leak reveals the exact number of TPCs and GPCs for each new Nvidia gaming GPU, as well as the memory bus and GDDR configuration attached to each model. While this leak doesn’t provide the precise specifications of future RTX 50 graphics cards, it offers the closest insight yet into the kind of power Nvidia’s upcoming graphics chips should be capable of generating.

The author fully expects Nvidia’s upcoming 5000 series graphics cards to dominate the high-end segment of the best graphics card guide, as AMD is not expected to compete with an equivalent RDNA 4 graphics card. However, there is still much to learn about Nvidia’s next offerings before such recommendations can be made.

This new Nvidia Blackwell spec leak comes from regular leaker on X/Twitter @kopite7kimi, who simply shared the following list of Nvidia GPU specs:GB202 12*8 512-bit GDDR7

GB203 7*6 256-bit GDDR7
GB205 5*5 192-bit GDDR7
GB206 3*6 128-bit GDDR7
GB207 2*5 128-bit GDDR6

These GB numbers refer to the code names for Nvidia’s GPUs, with the B in the code names referring to the Blackwell architecture of the GPUs. Current-generation GPUs based on the Ada Lovelace architecture have code names ranging from AD102 to AD107. The list provided here seemingly corresponds to the full range of chips, from the flagship GB202 GPU that is expected to be used in RTX 5090-class cards, to the GB207 that is expected to be used in RTX 5060-class cards.

The next two numbers in this list are the GPC and TPC configurations of the GPU. GPCs, or graphics processing clusters, are the largest sections Nvidia defines in its GPUs. Within each GPC are many repeated graphics processing parts.

One of those parts is called a TPC or texture processing cluster, which again defines a smaller section of the GPU containing several smaller processing components. The GPC/TPC relationship for the current AD102 GPU can be seen in the image below.

From the list above, we can see that GB202 will have 12 GPCs, each containing eight TPCs. Now, at this point, we have to make assumptions about what is contained in those TPCs. If we assume that Blackwell will have the same configuration of components inside each TPC as is used in Ada Lovelace, then that means GB202 will contain a total of 192 streaming multiprocessors (SMs). By comparison, AD102 has 144 SMs.

Nvidia Blackwell is indeed awesome

From the information provided, we can see that, as expected, Blackwell will offer an increase in the sheer number of pure graphics processing components for some GPUs, but not for others. This means that the leap from lower-tier cards to high-tier cards might be even bigger with Blackwell than with Ada.

It is also worth noting that this leak suggests that Nvidia will use an exact number of TPCs per GPC for each GPU, whereas this sometimes varies with Ada Lovelace GPUs. For instance, for AD103, one GPC contains only four TPCs, not six like the rest. Therefore, we must emphasize that the SM numbers above are estimates, so only limited conclusions can be drawn at this time.

It is also important to remember that these GPU numbers often do not correspond with the configurations used in graphics cards. This is because GPUs are defined based on the maximum potential features they are designed to include.

However, when it comes to manufacturing, errors in the process mean that parts of the GPU often have to be deactivated to make the chip work. That’s why the RTX 4090, for example, has only 128 SMs when its AD102 GPU can theoretically have up to 144.

So, it is clearly too early to say exactly what the specs of a future RTX 5080 might be, but we’re at least one step closer if we can believe @kopite7kimi.

For more of our thoughts on what we definitely know about the latest graphics cards, see why we think the RTX 4070 Super and AMD Radeon RX 7900 GRE are the best graphics cards around right now.

More info about Nvidia Click Here

Don’t miss out on staying updated at Technobrads.com.