What is the flow processor of the graphics card, and what exactly does the number of stream processo

Well,,The stream processor of the graphics card is its CPU, which is called the GPU upstairs...

It's like the core of the CPU of the host is the same as the processor of the graphics card.

The stream processor directly maps the multi-graph data stream to the stream processor for processing, and there are two types: programmable and non-programmable. The stream processor can optimize the shader engine more efficiently, it can process the stream data, and also output a stream data, which can be applied to other superscalar stream processors

Processors (SPS), stream processors can be run in groups or in large numbers, which greatly improves the parallel processing capabilities.

The number of stream processors has a decisive role in the performance of the graphics card, and it can be said that the main difference between high, medium and low-end graphics cards in addition to the different cores is the number of stream processors.

Hehe. You haven't looked at it closely.

The R&D team of A card is countless times behind that of N card.

Someone has counted that the capacity of the stream processor of 6 A cards can be close to that of one N card.

Therefore, a card of the same level needs many stream processors to reach the capacity of an N card.

Of course, the power consumption and heat generation are also large.

The classification of the N card is clear.

GS is the mobile version of GT.

That is, the shrunken version.

It is used on mobile platforms such as laptops.

This is because the computing requirements of this type of platform are lower than those of equivalent desktops.

So it's understandable to shrink a bit.

Of course. It will also be reduced.

The number of stream processors determines the computing speed of the graphics card, but does not determine the computing power, the bit width determines the output speed, and the video memory determines the upper limit of computing.

The GPU is responsible for the calculations and coordination.

The drive is responsible for connecting to the system.

Driver updates are also important... Wait, the graphics card is the same as the black computer, and it's quite complicated.

That's the relationship.

Before the DX10 graphics card came out, there was no such thing as a "stream processor". The GPU is made up of "pipelines", which are divided into pixel pipelines and vertex pipelines, and their number is fixed. In simple terms, the Vertex Pipeline is primarily responsible for 3D modeling, and the Pixel Pipeline is responsible for 3D rendering.

Since their number is fixed, this is a problem, when a game scene requires a lot of 3D modeling without too many pixels, it will cause the vertex pipeline to be tight and the pixel pipeline to be idle a lot, and of course the opposite is true. This results in a shortage of some resources and a waste of others.

In this case, people put forward the "unified rendering architecture" for the first time in the DX10 era, and the graphics card canceled the traditional "pixel pipeline" and "vertex pipeline", and changed it to a stream processor unit, which can perform both vertex operations and pixel operations, so that in different scenes, the graphics card can dynamically allocate the number of stream processors for vertex operations and pixel operations, so as to make full use of resources.

The number of stream processors has become a very important indicator to determine the performance of the graphics card, NVIDIA and AMD-ATI are also constantly increasing the number of stream processors of the graphics card to make the performance of the graphics card grow by leaps and bounds, for example, AMD-ATI graphics card HD3870 has 320 stream processors, HD4870 reaches 800, HD5870 reaches 1600!

It is worth mentioning that the GPU architecture of N-card [3] and A-card [3] is not the same, and the allocation of the number of stream processors is also different. There is no comparison between the two sides. Each stream processor unit of N card only contains 1 stream processor, and A card is equivalent to 5 stream processors in each stream processor unit, (A card stream processor 5) For example, although HD4850 is 800 stream processors, it is actually only equivalent to 160 stream processor units, and the frequency of a card stream processor is the same as the core frequency, which is why 9800GTX+ only has 128 stream processors, but the performance is comparable to HD4850 (the frequency of N card stream processors is about twice the core frequency).