Looking for first to provide a MBP i7 Vega 20 CPU Thermal Test

Discussion in 'MacBook Pro' started by Trebuin, Nov 23, 2018.

  1. Trebuin macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #1
    Before I pull the trigger on purchasing a replacement for my 2011 MacBook Pro, I'm trying to determine which CPU will stay cooler when under load between the i7 & i9. I've already seen the thermals of the i9 which clocks down to between 1.7 & 1.8. That number tells me the i7 could have a chance to do better under load if it is more efficient with thermals. Could someone run a test to check that out?

    You'd need to install intel power gadget to monitor the heat while using Heaven benchmark to load the GPU while running enough of the following commands to get the CPU to ramp up. You should start with loading the CPU to 100% & verify using activity monitor as each null command will only load some of the CPU, not all. Run this in thermal:
    yes > /dev/null
    After you are done with the test, you'll need to use activity monitor to terminate each null command. This will not damage your MacBook as it will throttle like every other laptop. Also, you only need to run it until the CPU throttle stabilizes The result should look like this:
    upload_2018-11-22_10-34-29-png.805884
    (image provided by IdentityCrisis)

    Again, this is to evaluate the i7 Vega 20 MacBook Pro. Thanks to whoever benchmarks this!
     
  2. Padsterman macrumors member

    Joined:
    Nov 19, 2018
    Location:
    Boston
    #2
    I have the same concerns with the i9. From the YouTube reviews I have seen the Vega runs cooler than the 560x. Not sure if the i9 is worth the extra for me.
     
  3. Reppeg macrumors newbie

    Joined:
    Nov 22, 2018
    #3
    Hi, I ran this test today on a I9, 32GB, 1TB Vega 20. Ran it for many hours. Core clocked 2GHz most of the time actually Skärmavbild 2018-11-23 kl. 18.54.50.png
     
  4. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #4
    Thanks. Still looking for the i7 version of this test. Those numbers don't look bad, I just really want to know what the i7 will pull off.

    I actually really like it staying at that temp. Are the fans even running at full speed?
     
  5. Reppeg macrumors newbie

    Joined:
    Nov 22, 2018
    #5
    I forgot to mention a also ran Prime95 and cinebench at the same time.
     
  6. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #6
    The Dev null command is actually good enough. Just ran it as a refresher on my Mac & you actually stop it by quitting terminal...pretty easy. Still hoping that someone can provide an i7 check
     
  7. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #7
    I'm still really hoping that someone can help me out with this. I'm leaning i9 as a shot in the dark.
     
  8. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
  9. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #9
    Well, I'm about at trigger pull time. If someone can do this benchmark, I'll send $20 via PayPal or Apple Pay. Just let me know.
     
  10. Padsterman macrumors member

    Joined:
    Nov 19, 2018
    Location:
    Boston
    #10
    I could not wait so just pulled the trigger on an i7, 32gb ram, 1tb ssd, Vega 20. I am photographer and videographer using Adobe CC, photoshop, Premiere,and capture One etc . Will try to run tests when I get it. Play some games as well

    Bought from Adorama so no tax to my state
     
  11. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #11
    I'm pulling the trigger in about 6 hours...I'm serious about paying if someone can throw in the CPU thermal test before I pull that trigger.
     
  12. mashinhead macrumors 68030

    Joined:
    Oct 7, 2003
    #12
    I would like to know as well, to i7 or to i9?
     
  13. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #13
    Yea, as soon as I get an i7 benchmark, we can see what would be worth it for long life. That's really my considerations...you may have others.
     
  14. mashinhead macrumors 68030

    Joined:
    Oct 7, 2003
    #14
    My concern is really throttling and if it's worth the $ for the i9. Based on reviews prior to the VEGA upgrade, I would have choosen the 2.2
     
  15. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #15
    True. The throttling may not be significant enough given a 1.7-1.9 on the i9 with GPU heating up the pipe too. What I'm looking to see is how the i7 handles on the heat pipe. If it is less than efficient, we could see lower than 1.7, possibly much lower. That would drive me to buy the i9. If it is more efficient (less power going to extra cache) then it could actually run higher than 1.9...in which case, I would save the money & go for the i7. I've been looking at the Geekbenchs of the graphics performance with the Vega 20s & I just don't see much i7s which is more of an indicator that the first possibility is likely. There are two (possibly 3) differences that I know of between this i7 & i9:
    1) different cache (requires a little more voltage?)
    2) different clock rates (if all other factors were equal then they down clock to the same rate.
    possible 3) i7 is less efficient at thermal control, the i7 would then would throttle lower than the i9
    --- Post Merged, Nov 24, 2018 ---
    I'm surprised no takers on the $20
     
  16. AppleHaterLover macrumors 68000

    Joined:
    Jun 15, 2018
    #16
    Isn't this like not good at all??

    1.9 Ghz sustained clock?
     
  17. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #17
    that' the i9. I'm looking for the i7 results. The results are actually fine for the i9.

    Edit: I woke up there & had my hopes up.
     
  18. Trebuin, Nov 24, 2018
    Last edited: Nov 24, 2018

    Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #18
    If no one is up for this, I'm going to make the call in about 30 mins to order...speak up before that or forever hold your peace.

    Well, no takers, I'm making the call going for the i9
     
  19. IdentityCrisis Suspended

    Joined:
    Sep 9, 2018
  20. Trebuin, Nov 24, 2018
    Last edited: Nov 24, 2018

    Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #20
    Dude, if that actually has the CPU temped, I'll pay you for finding that. I just got off the phone with apple, they're not able to ship direct to my deployed location & if they ship to my wife...I have to eat the taxes. Adorama is my next stop. I'll take a look at the video first.
    --- Post Merged, Nov 24, 2018 ---
    crap, that's the Vega 16. That'll make too much of a difference.

    Edit: jut finished the order, I pulled the trigger on the i9 because the thermals here are not that bad. 1.7-1.9 out of 2.4 base is not bad at all. The 4.8 GHz is designed for burst operations to make things feel snappier, not to be sustained. As others have pointed out, the GPU & CPU isn't going to run at max at the same time. I'll see that probably 6+ years down the road when the OS demands too much. Even then, I plan to disable hyper-threading which will drop the required voltage of the CPU when I don't need it. Hyper-threading is similar to how a fighter jet's afterburners work...you dump as much fuel as you can to hit a maximum thrust. In this case, you dump electricity (increasing heat) to add additional "virtual" cores that are built to utilize what the hard cores have not used. If a program is designed to efficiently use 100% of each core, then you should gain almost no advantage from hyper threading. You only gain an advantage when some of the core's capability is left unused. There are some operations that don't require the full capability of the core, especially when they use the extra features Intel & AMD build into these cores. Then you can essentially have more CPU power you can use. Finally, there are few i7s that show good Geekbench results compared to the i9s specifically addressing graphics performance. That tells me that the i7 is likely clocking lower than the i9s enough to drop a noticeable percentage.
     
  21. Thysanoptera macrumors 6502a

    Joined:
    Jun 12, 2018
    Location:
    Pittsburgh, PA
    #21
    So Vega 16 goes up to 45W? What happened to 35W limit? That's the first time I've seen someone motoring more than just fps or time.
     
  22. vaugha macrumors 6502a

    Joined:
    Nov 3, 2011
    #22
    I'm curious too. The video noted that vega 16 consumes more power hence worse battery life. He noted 30% better power efficiency which is offset by increased clock speeds as well. Can anyone confirm 45w on these vega units?
     
  23. IdentityCrisis Suspended

    Joined:
    Sep 9, 2018
    #23
    YEah, it was the only one I found for the i7. Mine still hasn't shipped that I ordered from Adorama. But I believe we are waiting on Apple now as it's a custom order.
     
  24. Trebuin thread starter macrumors 65816

    Trebuin

    Joined:
    Jun 3, 2008
    Location:
    Central Cali
    #24
    Luckily, I'm over seas so I'm not going to see it until next year maybe.
     
  25. Jerz macrumors regular

    Jerz

    Joined:
    Dec 21, 2013
    Location:
    Boston, MA
    #25
    I just ordered the i9 vega20 32gb 2tb last night as well
     

Share This Page

80 November 23, 2018