Can anyone please tell me what is really the difference between "dedicated system memory", "dedicated graphics memory" and "shared system memory" when relating to graphics?
Thx in advance.
People are pretty sloppy about the use of the term memory for a computer.
but you only wanted to know about 3 general cases:
-Graphic chip external to the CPU will have a very small amount(if any) of dedicated memory that only it has access to. It will try to reserve a chunk of system motherboard RAM for its use (BIOS settings will determine the size)
- Graphics chip internal to the CPU, will have access to some of the CPU's very,very high speed memory, then get further memory from the system RAM on the motherboard.
- Graphics memory on a external card, is very high speed memory that only the graphics card gets to use. dedicated memory.
- Shared system memory: basically the operating system shares its slower RAM in the motherboard SIMM slots with the graphics chip.
- Dedicated graphics memory: the graphic hardware has memory only it can use
- System memory: areas of memory that only the Operation system can use.
Does reserving more system memory of rendering purposes improve performace, or would you only see a performance increase with more dedicated vram?
Both would increase performance for memory constrained graphics applications. System RAM is slower but cost a lot less that dedicated VRAM.
reserving more system RAM for a internal graphics processor helps up to a point where systems graphics performance is not constrained by the quanity of RAM. At that point the constraints become the speed of the RAM. For a game a cheap video card will have its own processor, memory access controllers to get at its own little stash of memory and will provide a lot more performance than the best current graphics processor inside of a desktop CPU.
For example: the Intel 4000 graphics processor built as a core within their new CPUs will be about equal in performance to a $30 to $40 external graphics card.
Rendering graphics or playing games will benefit more by reserving system memory for your other system's needs and using a discrete graphics with its own memory to separate the bulk of the rendering bandwidth making the system overall a lot more smooth and responsive.