Are you buying a new computer and do not know if it is better to choose an integrated graphic card, a dedicated graphic card or a shared graphic card? Here we will explain what is the difference between an integrated, dedicated and shared graphic card.
WHAT IS THE DIFFERENCE BETWEEN AN INTEGRATED, DEDICATED AND SHARED GRAPHIC CARD?
Regardless of whether it is a desktop computer or a laptop, when you are about to buy a new computer you are often faced with the usual dilemma: is it better to have an integrated graphic card, a dedicated graphic card or a shared graphic card? To try to clarify any doubt about it I will therefore explain what is the difference between each of these types of GPU.
INTEGRATED GRAPHIC CARD
The integrated video card, also sometimes called integrated graphics, shared graphics solutions, integrated graphics processors (IGP) or unified memory architecture (UMA), is a particular type of video adapter that does not have its own video memory and then uses a part of the computer RAM memory as its video memory. An integrated video card has an amount of graphic memory in the order of the MB and can be integrated precisely or in the motherboard as part of the chipset, or in the same chip together with the CPU (just like, for example, in solutions APU of AMD or Intel’s HD Graphics products ).
The integrated video cards are less expensive to implement than dedicated video cards and almost always have very limited performance. Their use is therefore recommended to those who must use the computer purely for non-binding purposes, such as surfing the Internet, consulting the e-mail or maybe watching some videos online.
DEDICATED VIDEO CARD
Among the various types of existing graphic cards, the dedicated graphic card, sometimes called a discrete graphic card, is the best solution from a performance point of view. The dedicated term refers to the fact that, unlike the integrated graphic card, the dedicated graphic card has a video memory of its own, of various GB, of the type GDDR3, GDDR5, GDDR5X, or even HBM, HBM 2 or HBM 3. Generally the dedicated graphic cards interface with the PC motherboard through an expansion slot such as PCI Express, AGP, or in some cases even through the obsolete PCI.
Motherboard permitting, the dedicated graphic cards have the advantage of being able to be replaced or updated very easily, even without asking anyone for help. The dedicated video cards can also be used together with other integrated or dedicated graphic cards (in the latter case through technologies such as NVIDIA SLI or AMD CrossFire), thus allowing to exploit the potential of one or the other type according to their needs. The dedicated graphic cards generally have prices that can range from hundreds to thousands of dollars and are therefore recommended for those who use computers mainly or to play, or for those who must work with very heavy software.
SHARED GRAPHIC CARD
The shared graphic card is simply another way to call the integrated video card. The shared term refers, in fact, to the characteristic of integrated video cards that, not having a graphic memory of their own, share a certain amount of RAM memory of the computer as their graphic memory. For example, if you use a shared graphic card or an integrated 256MB graphic card on a computer with 8 GB of RAM (or 8,192 MB ), the total memory available on that computer will drop to 7,936 MB.
It should be noted, however, that the amount of shared video memory usually never exceeds the single gigabyte and that, if desired, it is also possible to increase or decrease it as desired through the BIOS/UEFI settings of the motherboard.
HYBRID GRAPHIC CARDS AND EXTERNAL GRAPHIC CARDS
In addition to the video cards already mentioned, on the market it is sometimes possible to find two other types of video cards: hybrid graphic cards and external graphic cards:
- By sharing part of the computer’s RAM memory and using a small dedicated cache memory, hybrid graphic cards look like a cross between integrated graphic cards and dedicated graphic cards, making them a little more expensive than the first but with a price definitely lower than the latter. The most common implementations of this particular type of graphic cards are through the HyperMemory technologies of AMD and the NVIDIA TurboCache;
- external graphic cards, also called eGPU, are mostly used on laptops to compensate for the poor performance offered by the integrated or dedicated graphic card that is already on board. In order to use an external video card, however, it is necessary to use an appropriate Thunderbolt port (version 2 or 3), usually relegated to Apple’s MacBook only.
So, at this point, you should finally have realized that there is a difference between an integrated, dedicated and shared graphic card.