Officially: Intel's Skylake promises a great improvement in graphics processing, modest improvement in general processing • HWzone

Officially: Intel's Skylake promises great improvement in graphic processing, modest improvement in general processing

Formal chip-based slides speak of an improvement of 11 in the overall processing capacity of its new processor generation, plus the cost-effectiveness and improvement of 40 to 50 percent of built-in graphics core capabilities

Get updates from everyone in TelgramGet updates from us all at TelgramJoin the channel now

The launch of the Skylake generation is around the corner, and according to all the reports and data we receive, it will not bring in truly breakthrough lines for the most performance enthusiasts. Now, a number of seemingly official slides of And offer a detailed overview of the performance improvement that awaits us in every sector and level.

For the general processing cores, Intel declares an improvement between 10 and 20 in single core performance and multi-core performance based on one of the two tests that make up the SPECint performance test. This figure sounds pretty good at all, but considering that Intel has been talking about a similar improvement in the generations of CPUs, It seems that we are looking at an improvement of less than 10 in general average performance, at least for the fixed and mobile models of the H and U series for which it is the sixth inertia since the revolution began..


Within the built-in graphical cores, the situation is much more optimistic. Intel continues to invest large resources and "floor space" in all of their chips to provide proper graphical processing performance that suppresses the discreet cards of and- Backward (toward the middle and high market), and in the second generation They expect an increase of 16 to 50 in performance, depending on the specific series under which the comparison is made.

The graphical performance of the built - in cores They are quite impressive today, but until the company does not start offering higher quality drivers with a wide and uniform support, it seems that everyone who is a high priority for gaming will still need a dedicated video card - so let's hope the chipset works to improve in this area as well.

As we already hypothesized, the Skylake generation will also focus, first and foremost, on improving efficiency and capabilities with minimal power consumption - making it the biggest leap forward in the Y series (the second generation of M) with up to 17 improvement in overall processing performance and up to 41 per cent of core graphics core performance. All this plus a predicted improvement of up to 30 (!) In battery life for corresponding specifications from the Current.


The modeled slides also confirm the interest of incorporating a lot of advanced capabilities and features right into the new generation chipset (rather than externally on the printed circuit board, as it has so far) - 2D camera hardware with resolution support (Plus 3D cameras with RealSense technology), A4WP wireless charging support, encoding and decoding of the latest HEVC standard, touch screen interface and sensors, upgraded digital voice signal processor for voice and speech, and eMMC 5.0, SDXC and USB interfaces Will be an integral part of the mobile Skylake processors, in a way that is expected to bring about savings in my life And cost savings for computer manufacturers and vendors.

The portable Skylake processors from the ultra low power series will be similar to chips And tablets (in the ARM architecture) than ever before

Now it seems that all the cards are already on the table - and we only have to wait patiently to find out if the rushing potential will be the big and surprising ace of the processors Stationary, or not really.

More links


9 תגובות

  1. Don't understand why Intel doesn't make a model or two without a graphics core and invest more in the processor itself? After all, most of us have an external card without the graphics core. What kind of logic do they have there to make such models ?? : s07:

  2. Don't understand why Intel doesn't make a model or two without a graphical kernel and invest more in processing power? After all, most of us have an external card without the graphics core. What kind of logic do they have there to make such models ?? : s07:

    I have a theory.
    Sounds reasonable.
    Don't know if it's true.

    You see it as a graphical processing unit, maybe they see it as a computational unit for additional uses that extend the processing capabilities, and thus an integral part of the processor.
    Just as efficient FLOATING POINT calculation capabilities are needed in addition to INTEGERS.

    So even if you don't use CPU for graphing, this component will allow for better software development standards and will always be there for that.

    If this component did not exist on every CPU, then you prevented the software industry from developing efficient code and uniform performance between computers.

  3. Don't think this theory is correct, the GPU if I'm not mistaken doesn't have to be fully replaced by an external GPU, the internal just isn't active. The CPU has no replacement and needs it to run everything on the computer, not just software development.
    It just seems easier for them to develop a uniform product. I'm not sure, but I think the cost of silicon per chip (operational cost without development costs, etc.) is also not that significant compared to everything else, so it won't save much to download the GPU.
    Exceptional Extreme Processors (significantly different, such as a different plug-in) do not have a built-in graphics card.

  4. So that if the screen card burns to you you can still use the computer with the built-in card

  5. A nice option if a burned card is not a reason to dedicate 60% of the processor space to a graphics core.
    Intel is preparing a mirror surface for much, much more powerful graphics cores. It seems.

  6. I suppose the real reason for the poor improvement in overall processing performance - is that they have no reason to introduce significant innovations in the processor segment. They are currently pretty monopoly, and if they have strong cards up their sleeve they keep them in case something arises in AMD, so the constant rejection of the announcements, so there is hardly anything new in any version of the processor, and investors where it is currently easy to show "improvement". As soon as the universe competes (if there is such a candidate at all) we will see tremendous leaps and new technologies that they will introduce to the processing world that will keep them ahead of all competitors. There is currently no reason.

  7. I thought about this option too, but I'm not sure it's true.
    Suppose that is true by the way, Intel actually stops the progress quite seriously. Our ability to move forward and develop new things depends first on the processing power available.
    Beyond that, I think that while AMD is shuffling, the other ARM-based chipset companies are actually showing impressive products.
    Currently, in the field of mobile Intel is not really a serious player, so here is an area where they do need to improve in processing power, at the moment their success is quite weak.
    The improvement of desktop processors exists and does not go anywhere, they simply created a situation where anyone who wants and needs this processing power simply has to pay more, this is of course XEON systems.
    This is probably more profitable for them, while the home user, for the most part, is having difficulty utilizing I5, and some of them can benefit from the graphical core of casual games. These are people who usually won't buy a separate video card.

    And in spite of all this, I too still want to see justification for these silly graphic cores.

  8. There is no need to improve the overall processing capability. It is also necessary to improve the graphical performance

Leave a Reply

Back to top button