Page 41 of 78 FirstFirst ...
31
39
40
41
42
43
51
... LastLast
  1. #801
    The Lightbringer Evildeffy's Avatar
    15+ Year Old Account
    Join Date
    Jan 2009
    Location
    Nieuwegein, Netherlands
    Posts
    3,772
    Quote Originally Posted by mrgreenthump View Post
    Sadly 4 GHz is around the limit of the manufacturing process they are currently using.
    Don't forget the fact that with current technology you actually hit a physics wall going beyond current speeds.
    I.E. require a nuclear reactor to power the CPU as well as a heatsink the size of your PC to cool just 1 core.

    Regardless ... I really would like if the i7-8700K got the leaked clockspeeds up to 4,7GHz ... but I still doubt it due to the power envelope and the requirements of the process going from Kaby -> Coffee, it's a pretty big leap if they want to remain inside of the envelope.

    Unless of course they do it much like they did with Skylake-X and say their TDP is value X but secretly go past it anyway.

  2. #802
    Quote Originally Posted by Linkedblade View Post
    It would be really impressive if intel and AMD came out with 1 core or two core skus with or without multithreading, then the clocks would be at crazy high amounts. But no, let's just keep increasing the core counts even though few applications the casual consumers use even take advantage of them.
    Ever heard of Pentiums and i3's? They exist. They are dual core. However, they clock no higher than the i5s and i7s because we have hit the limit. That's why we go more cores, because it is not currently possible to get clocks any higher. intel has been trying for the past 6 years, since Sandy Bridge, and they just get really really tiny gains each time. But sure, you are smarter than intel, so why don't YOU go and create a lightning fast single core CPU.

    Also consider that the casual consumer is not who they sell the most CPUs to. Businesses buy far more computers than individuals and they upgrade them more often. So THAT is who they design for because that is the larger market. They happen to make stuff for us too, but really that's secondary.

  3. #803
    Quote Originally Posted by Evildeffy View Post
    Regardless ... I really would like if the i7-8700K got the leaked clockspeeds up to 4,7GHz ... but I still doubt it due to the power envelope and the requirements of the process going from Kaby -> Coffee, it's a pretty big leap if they want to remain inside of the envelope.

    Unless of course they do it much like they did with Skylake-X and say their TDP is value X but secretly go past it anyway.
    4.7 GHz is the 1 core boost and 4.6 for 2 core, which should be pretty doable with 95W. Though remains to be seen where it goes when you 100% all cores.

  4. #804
    The Lightbringer Evildeffy's Avatar
    15+ Year Old Account
    Join Date
    Jan 2009
    Location
    Nieuwegein, Netherlands
    Posts
    3,772
    Quote Originally Posted by mrgreenthump View Post
    4.7 GHz is the 1 core boost and 4.6 for 2 core, which should be pretty doable with 95W. Though remains to be seen where it goes when you 100% all cores.
    Look at the power envelope of Skylake-X... Skylake was more power efficient than Kaby Lake.
    (HEDT or not is mostly no different, 7800X should be comparable due to uArch)

    Granted it is older but think of it in this way:
    By claiming 95W TDP they have improved their power efficiency by over 60% to get to 95W 6-core TDP vs. 91W 4-core TDP.
    This whilst remaining on 14nm and with the same uArch... see why I'm doubting this?

  5. #805
    The 6 core skylake x rund at base 3.5 and turbo's to 4.0 for an 140w package which I still find is lame as the 8 core variant has a base of 3.6 and turbo of 4.3 with turbo boost for 2 cores upto 4.5 for the same 140 watt package so there should at least be room for improvement on coffee lake 6 cores if you ask me.

    However I will believe it when I see it.

  6. #806
    the high stock clocks would be good for non-overclockers and reviews, but as far as OCing goes it should still be at least as good as a 7800X, hopefully better



  7. #807
    Dreadlord Mask's Avatar
    7+ Year Old Account
    Join Date
    Sep 2014
    Location
    Ironforge
    Posts
    772
    Doesn't seem like a big improvement. If the 7700k wasn't worth it for you to upgrade to, neither will this. Wait for the next generation. Hopefully Cannonlake CPUs will be impressive.

  8. #808
    Quote Originally Posted by Mask View Post
    Doesn't seem like a big improvement. If the 7700k wasn't worth it for you to upgrade to, neither will this. Wait for the next generation. Hopefully Cannonlake CPUs will be impressive.
    That's basically been the case since Sandy Bridge. The increases from gen to gen have been very small, so it's not worth upgrading unless you have nothing or what you have is dying.

  9. #809
    http://www.cpu-monkey.com/en/cpu_ben..._single_core-7

    this page added the 8700 single-core result

    previously there was only 8700K

  10. #810
    Quote Originally Posted by Life-Binder View Post
    http://www.cpu-monkey.com/en/cpu_ben..._single_core-7

    this page added the 8700 single-core result

    previously there was only 8700K
    I bet we can lock all cores to 4.7 and run it like that for an easy overclock.

    6 real Intel cores is a lot of power.

    9% ipc increase, nice. That, plus 2 more cores = finally upgrading my 4790k. 169 to a 218 single core performance vs the two.
    Last edited by Zenfoldor; 2017-08-09 at 10:05 PM.

  11. #811
    Quote Originally Posted by Mask View Post
    Doesn't seem like a big improvement. If the 7700k wasn't worth it for you to upgrade to, neither will this. Wait for the next generation. Hopefully Cannonlake CPUs will be impressive.
    Considering they are running up against the laws of physics, that isn't likely.

    All of these designs are just further refinements of an architecture that started after the Core 2 lineup. Die shrinks and small improvements, but the underlying architecture is largely unchanged.

    There's going to have be an architecture breakthrough to see very serious gains at this point. which isn't impossible, but as Ryzen showed, isn't that easy - (and this is not a dig at AMDs ability, but more about showing how hard it really is) - "All" AMD could do is bring their IPC close to Intel's with a MASSIVE architecture shift which took 5-6 years to develop.

    We may be looking at the end of real huge IPC gains just because of physics.

  12. #812
    I feel like people are starting to use IPC and single thread performance interchangeably.
    The whole problem with the world is that fools and fanatics are always so certain of themselves, but wiser people so full of doubts.

  13. #813
    IPC determines single thread performance.

  14. #814
    Quote Originally Posted by Kagthul View Post
    IPC determines single thread performance.
    IPC x Clock Speed determines single thread performance. IPC is only half the equation. Just look at Ryzens. Their clock speed is behind, but the single threaded performance is similar.

    As here in the Cinebench Single threaded benchmark:
    http://www.guru3d.com/articles_pages...review,10.html

    The 1600 at stock is right in between the 5960X and the 6950X. The 1600 has a slightly higher clock speed than the 5960X and performs slightly better, so the IPC is likely similar. The 6950X is also clocked just slightly lower, yet performs better, so it's clear it has better IPC.

  15. #815
    Moderator Cilraaz's Avatar
    15+ Year Old Account
    Join Date
    Feb 2009
    Location
    PA, USA
    Posts
    10,139
    It's starting to look like the 8700k will actually be worth the cost of upgrading from my 2500k.

  16. #816
    Warchief
    7+ Year Old Account
    Join Date
    Nov 2016
    Location
    The pit of misery, Dilly Dilly!
    Posts
    2,061
    Quote Originally Posted by Life-Binder View Post
    maybe to further segment it from the more expensive more premium models

    or maybe its lower binned


    I dont think its manual OC potential will be any lower though
    Its probably lower binned, they usually do this chips from the same batch that are lower quality

    - - - Updated - - -

    Quote Originally Posted by Life-Binder View Post
    http://www.cpu-monkey.com/en/cpu_ben..._single_core-7

    this page added the 8700 single-core result

    previously there was only 8700K
    Wow, my 5820 is starting to look like a potato.

    Possible upgrade to i9? doubtful, maybe in a few years, the 8700 looks interesting however, that single core performance is insane.

  17. #817
    http://ranker.sisoftware.net/show_ru...482f1ccf4&l=en

    SiSoftware Official Live Ranker
    Details for Result ID Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz (6C 12T 4.3GHz/4.7GHz, 4GHz IMC/4.4GHz, 6x 256kB L2, 12MB L3)

  18. #818
    Quote Originally Posted by Lazyyrogue View Post
    Wow, my 5820 is starting to look like a potato.

    Possible upgrade to i9? doubtful, maybe in a few years, the 8700 looks interesting however, that single core performance is insane.
    Of course, it has a freaking 4.7 GHz single core boost clock, apparently.
    R5 5600X | Thermalright Silver Arrow IB-E Extreme | MSI MAG B550 Tomahawk | 16GB Crucial Ballistix DDR4-3600/CL16 | MSI GTX 1070 Gaming X | Corsair RM650x | Cooler Master HAF X | Logitech G400s | DREVO Excalibur 84 | Kingston HyperX Cloud II | BenQ XL2411T + LG 24MK430H-B

  19. #819
    Quote Originally Posted by Zenfoldor View Post
    I bet we can lock all cores to 4.7 and run it like that for an easy overclock.

    6 real Intel cores is a lot of power.

    9% ipc increase, nice. That, plus 2 more cores = finally upgrading my 4790k. 169 to a 218 single core performance vs the two.
    It's more like a 4.5% ipc increase, since it was clocked about 4.5% faster than the 7700k in that chart. So half the improvement was from ipc, half from clockspeed. Also do we have any idea how these will actually clock? 4.7 looks good, but without temps, voltage, or what kind of cooling was used we can't really use this information to compare with previous stuff. It could have been a golden sample for all we know.

  20. #820
    so far 4.7 looks to be the default single core boost on this CPU

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •