18

What are the main differences between each processor generation of Intel processors/chipsets? Is there a performance difference?

I've already seen processors from first to sixth generation, however I don't know/understand the exact differences.

2
  • Once upon a time, a lot; these days, barely anything. At one time, manufacturers and developers would hold off on new releases until there was a significant improvement. They would consolidate feature- and performance-enhancements, bug-fixes, etc. into a product that was truly an upgrade. Now, they throw out “new product” each month, with a new release every time they make even the slightest change. With software, it is annoying because you get a buggy, incomplete product that slowly gets better over time, but with hardware, it is damn annoying because it costs a lot to update every month.
    – Synetech
    Jul 22, 2012 at 6:47
  • 1
    It's called the "bleeding edge" for a reason, lol.
    – LawrenceC
    Apr 12, 2014 at 14:17

6 Answers 6

11

The term Generation is loosely applied to Intel processors to mean new and significant developments in processor architecture or functions.

  • A second generation Core processor is the family of processors known as Sandy Bridge, which among other things introduced shared cache and placed the memory controller, graphics and CPU on the same die. The first generation was composed the Core i3, i5 and i7 processors launched early last year.

  • The term can also be seen sometimes applied to processor families. The Core family of processors can sometimes be talked about as having had 3 generations (Lyndfield, Clarkdale and Sandy Bridge). But some processors like the i7, had more or different families (the Bloomfield and Gulftown). Example.

  • It can also be used to name different factory models within a similar architecture. Again, sticking to Core processors, The Core 2 Duo, Quad and Extreme being mentioned as one generation different than the i3, i5 and i7, while the sandy Bridge being the 3rd generation of Core processors. Example

All in all the term is not officially connoted to the media. Intel does seem to favor the term as meaning significant architecture and factory processing changes within the same family of processors. And they are ultimately the ones deciding what is named 2nd, 3rd, 4th and so on generation. They do it often. But the term has come to mean other things too, since Intel itself has never tried to enforce it. Hence being a term that can have both official and loose connotations, depending on the context.

5

This question is very broad, but the often the biggest difference is that they have been able to miniaturize the transistors, allowing for less energy usage, and less heat.

Of course, they can also change features, but this will vary with every comparison you could put forward. Here is a comparison of the i3:

Intel introduced the 1st generation Core i3 processors in 2010 and the 2nd generation Core i3 processors in 2011. The 2nd generation Core i3 processors are built on the Intel’s Sandy Bridge architecture, which is 32nm microarchitecture, while 1st generation Core i3 processors were built on Intel’s Nehalem architecture. Additionally, 2nd generation Core i3 processors include new features for improving the graphics performance of the processors such as Intel Quick Sync Video, Intel InTru 3D / Clear Video HD and WiDi 2.0 that were not available in 1st generation Core i3 processors.

http://rapidhow.com/2011/06/12/intel-core-i3-vs-2nd-generation-intel-core-i3-processors/

4
  • So actually 2nd genaration used to be better than 1st generation??? Can we assume this for all processor families?
    – Diogo
    Jul 25, 2011 at 12:43
  • 2
    Well, I would clearly assume that Intel would, or any manufacturer would not knowingly take steps backward. Does it guarantee it, no, but it is probably a safe assumption that 2nd generation is better than first barring something very freaky.
    – KCotreau
    Jul 25, 2011 at 12:44
  • 1
    I will say this: Given a choice, I would always take the newer one. I would probably be right 99% of the time.
    – KCotreau
    Jul 25, 2011 at 12:45
  • 1
    Better is relative: Ivy Bridge is smaller and reduces power consumption, but severely limits heat dissipation and conversely the ability to overclock. While I have my Sandy Bridge 2700k overclocked to 4.6GHz fairly easily and at a low temperature, the Ivy Bridge equivalent would run much hotter at the same frequency. Again, better is relative: is it more important to you that you can push your processor to higher GHz at the same temp or that your processor consumes as little power as possible? Dec 13, 2012 at 20:16
4

Wikipedia has a timeline of "generations", CPU models and new features introduced in them. But note that its generation count already disagrees with yours (it goes up to 9) and doesn't correlate with any clear technical property such as bus size or CMOS process size.

IMO the word "generation" as applied to CPUs is poorly defined and marketing driven - just a fancy way to say "this isn't just a new model, it's fundamentally better, so you must buy it!"

2

Suppose 2nd generation is first step of intel,then 3rd generation will be the improved (fixes all bugs,errors, graphics improvement,clock speed improvement,and the overall processing speed improvement.)

1
  • Can you provide details on how this is the case, including an example of the specific changes made in a particular generation?
    – bwDraco
    Oct 3, 2015 at 0:04
1

the progressions from generations by my thought are just the difference in architecture.. bigger the architecture, higher the battery consumption and vice a versa. this could be a factor that could influence your decision in buying a laptop/PC. as would the clock speed of the processor, size of cache and no of cores the processor has..

My question would be whether to go for a Core i5 2nd gen or a core i3 3rd gen ignoring personal utility and factoring in cost ( a core i5 is on an avg abt 80 -100$ more than an i3)??

1
  • This answer is a bit hard to understand. I suppose "the higher the architecture" means "the larger the semiconductor process", but each architectural update typically involves changes beyond simply shrinking to a smaller process. The details about how this is done are very, very complex; see [this answer] for an overview of the techniques used to improve processor performance.
    – bwDraco
    Oct 1, 2015 at 23:17
1

Intel's processors have a Tick-Tock cycle.

That is, processors released during a tock will have enough notably different design features from the previous tick to be considered a new microarchitecture. Those released during the tick are generally the same tock design but having undergone a die shrink, with some minor features added.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .