The last few days I've been reading about GPU problems in the MBPs. I'm not too familiar with what sort of GPUs exist for notebooks and what not, and after reading peoples' suggestions for this card or that card, I was wondering if someone could shed some light on the differences between ATI vs NVIDIA... what makes one better than the other? Is there any real significance to the nomenclature of the different models? Or is that just how they ID the cards. (Like how an IS250 refers to the engine size, does the 8600 in our MBP cards signify something?) Yeah, just wanted to get a better grasp of the whole GPU world. Thanks!
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
GPU Heirarchy
- Thread starter nanvinnie
- Start date
- Sort by reaction score
wrldwzrd89
macrumors G5
All modern GPUs from ATI and nVidia use hierarchical model numbers as a way of distinguishing their approximate capabilities and generation. For example, ATI's X200 series of cards represent their 11th generation - the specific model numbers denote power. The higher the number, the more capable and powerful the GPU. nVidia's card numbers work in much the same way.
The hardware that powers each GPU is what determines how capable it is. There are 4 factors that influence GPU power: Number of shaders, size of onboard RAM (if any), speed of CPU, and size of cache. As these things increase, so does the power of the GPU.
The hardware that powers each GPU is what determines how capable it is. There are 4 factors that influence GPU power: Number of shaders, size of onboard RAM (if any), speed of CPU, and size of cache. As these things increase, so does the power of the GPU.
Although Nvidia just changed their name scheme to GTX xyz, the same basic principle applies. The first number is the architecture generation, the second number is the performance bracket, and the rest of the numbers are any revisions or differentiations of the cards within that bracket.
ATI has the simpler nomenclature now:
For example, the 4850 and 4870. The 4 represents the generation of their architecture (in this case, the R7xx cards), the 8 represents the performance bracket (the higher the better in general, in this case 8 is their high-end performance cards) while the 50 and 70 differentiate the cards (the 50 means the 4850 is lower in performance than the 4870) etc.
NVidia has it ismilar with their new name scheme such as GTX 260 and GTX 280 where the 2 is the generation (for their GT200 chip), and the 60 and 80 show that the 280 is the higher performing than the 280.
It used to be far more confusing when both sides used letters such as GTX, GTS, GT, GS, GSO and XTX, XT, PRO, GT, LE, etc.
If you're looking at older cards, go for the numbers first then know that for Nvidia it is: GTX > GTS > GT > GS in general and ATI is XTX > XT > PRO > GT
ATI has the simpler nomenclature now:
For example, the 4850 and 4870. The 4 represents the generation of their architecture (in this case, the R7xx cards), the 8 represents the performance bracket (the higher the better in general, in this case 8 is their high-end performance cards) while the 50 and 70 differentiate the cards (the 50 means the 4850 is lower in performance than the 4870) etc.
NVidia has it ismilar with their new name scheme such as GTX 260 and GTX 280 where the 2 is the generation (for their GT200 chip), and the 60 and 80 show that the 280 is the higher performing than the 280.
It used to be far more confusing when both sides used letters such as GTX, GTS, GT, GS, GSO and XTX, XT, PRO, GT, LE, etc.
If you're looking at older cards, go for the numbers first then know that for Nvidia it is: GTX > GTS > GT > GS in general and ATI is XTX > XT > PRO > GT