Benchmarking the eight-core 2019 MacBook Pro with Vega 20 graphics

AppleInsiderAppleInsider
Posted:
in Current Mac Hardware edited June 2020
Apple latest and greatest 2019 MacBook Pro with 8-core i9 processor and Radeon Pro Vega 20 graphics is on our test bench, so let's see how fast it really is.

2019 MacBook Pro
2019 MacBook Pro


We've already benchmarked the new 2019 13-inch MacBook Pro with a 2.4GHz quad-core 8th-generation i5 Processor. That 13-inch MacBook Pro retails for $1,799 before taxes.

However, the substantially more powerful 15-inch MacBook Pro is also available for users looking for the ultimate portable Mac. We're testing the 8-core 2.4GHz i9 processor that can Turbo Boost up to 5GHz, 32GB of RAM, 1TB SSD, and VEGA 20 graphics.




Benchmarking

For our benchmarks, we run a set of standardized tests on all new machines that come into our labs, all evaluating different aspects of the machine.

2019 8-Core i9 Geekbench 4 results
2019 8-Core i9 Geekbench 4 results


Starting off, we threw the Geekbench 4 benchmark suite at the new MacBook Pro. While we acknowledge that it isn't the be-all and end-all of tests, it is a very good measure of burst processing, similar to the vast majority of workloads.

We reached a multi-core score of 29976 and a single core score of 5879. That is roughly a 20% gain over the 2018 2.9GHz i9 MacBook Pro which earned 24226 for the multi-core.

Following Geekbench, we next fired up Cinebench R20. In 2018, when we reviewed the previous generation MacBook Pros, we used Cinebench R15. R20 is a newer release and the results are incompatible making it not a direct comparison between the two figures. That said, going forward we will exclusively be relying on R20.

Our machine earned an impressive 3096 score as the average of ten back-to-back tests. Through the ten tests, our highest score we reached was 3160 and the lowest was still 3012.

Cinebench R20 and Intel Power Gadget the tools we use when testing the thermal performance of the new Pro. We won't go into all the details here -- you can see those in our full thermal analysis -- but the short version is the thermal condition in the 2019 model is greatly improved versus any 2018 model. The 2019 version is able to maintain clock speeds under load of 3.0GHz rather than the 2.4GHz Apple advertises, with only a slight decrease to just over 2.9GHz when the GPU is under load as well.

Unigine Heaven benchmark results
Unigine Heaven benchmark results


To test the graphics aptitude, we use the Unigine Heaven benchmark. This is more of a gaming benchmark and general performance indicator in a mixed-CPU and GPU task, and is still a good way to push the graphics to the limits.

The 2019 i9 MacBook Pro with Vega 20 gathered 82.3 average FPS and a score of 2072. The max FPS was 147.4. This was all with the quality set to medium, tessellation and stereo 3D disabled, anti-aliasing off, and running at full screen.

When we compare to the 2018 i9 MacBook Pro with Radeon Pro 560X graphics, we see huge gains. The previous generation garnered 62.5 average FPS and a score of 1575, a roughly 25% lower score. This same 2018 MacBook Pro with a 16GB Radeon VII Thunderbolt 3 eGPU and external display set to the same resolution as the MacBook Pro scored 92.0 frames per second, and a 2318 score -- with notably lower CPU temperatures.

Blackmagic Disk Speed Test
Blackmagic Disk Speed Test


We use the Blackmagic Disk Speed Test to measure the flash storage. Our model pulled in 2627.4 MB/s for read speeds and 2626.3 MB/s for write speeds. Nothing much has changed in this regard from the last model. This will be lower with smaller drives, because of lesser parallelization in writes, so opt for the 1TB drive at the least for best performance.

Overall these machines are delivering impressive performance increase over the previous generation for Apple's first 8-core portable machines. Stay tuned to AppleInsider for our full review in the coming days.

Save up to $230 on 2019 MacBook Pros

If you want to grab a new 2019 MacBook Pro, Apple Authorized Reseller Adorama is knocking $100 to $230 off select models with coupon code APINSIDER, plus no interest financing when paid in full within 12 months with the Adorama Credit Card. For a full rundown of the latest MacBook Pro deals and product availability, be sure to check out our 2019 MacBook Pro Price Guide.

$200 to $230 off 2019 15" MacBook Pros $150 to $200 off 2019 13" MacBook Pros

Comments

  • Reply 1 of 16
    entropysentropys Posts: 4,163member
    It is great these regular updates. Keep it up Apple!
    caladanianwatto_cobra
  • Reply 2 of 16
    KITAKITA Posts: 392member
    It definitely depends on the application, but it looks like the new 8 core i9 is giving a reasonable boost in performance over last year's 6 core i9 in real world applications:

    Dave2

    Of course, the lack of NVIDIA graphics is unfortunate. It's a shame that Apple and NVIDIA couldn't sort out their differences.

    Meanwhile, NVIDIA is pushing Quadro RTX 5000 laptops (falling under RTX Studio branding) to compete directly against the new MBP 15 for creators.

    These things are in a completely different league than AMD's current mobile offerings in the MBP.

    Also, the new 15" and 17" Studio models from Razer look rather familiar...




    williamlondon
  • Reply 3 of 16
    thx1965thx1965 Posts: 2member
    KITA said:
    It definitely depends on the application, but it looks like the new 8 core i9 is giving a reasonable boost in performance over last year's 6 core i9 in real world applications:

    Dave2
    Hi @KITA - where did you get those Premiere Pro testing results from?

    I always prefer real-world vs. abstract benchmark tests. That's the only thing that tells us if a certain processor or CPU is worth the extra money.
    I understand that the Razer Blade i7 was able to take advantage of NVIDIA acceleration inside Premiere Pro. Do you know how the MBP render test was run? Was the Premiere project set to OpenGL or Metal?
  • Reply 4 of 16
    KITAKITA Posts: 392member
    thx1965 said:
    KITA said:
    It definitely depends on the application, but it looks like the new 8 core i9 is giving a reasonable boost in performance over last year's 6 core i9 in real world applications:


    Hi @KITA - where did you get those Premiere Pro testing results from?

    I always prefer real-world vs. abstract benchmark tests. That's the only thing that tells us if a certain processor or CPU is worth the extra money.
    I understand that the Razer Blade i7 was able to take advantage of NVIDIA acceleration inside Premiere Pro. Do you know how the MBP render test was run? Was the Premiere project set to OpenGL or Metal?
    The results came from Dave Lee's video review:



    I'm not too sure of the settings on the Mac.
    edited May 2019
  • Reply 5 of 16
    davendaven Posts: 696member
    KITA said:
    thx1965 said:
    KITA said:
    It definitely depends on the application, but it looks like the new 8 core i9 is giving a reasonable boost in performance over last year's 6 core i9 in real world applications:


    Hi @KITA - where did you get those Premiere Pro testing results from?

    I always prefer real-world vs. abstract benchmark tests. That's the only thing that tells us if a certain processor or CPU is worth the extra money.
    I understand that the Razer Blade i7 was able to take advantage of NVIDIA acceleration inside Premiere Pro. Do you know how the MBP render test was run? Was the Premiere project set to OpenGL or Metal?
    The results came from Dave Lee's video review:



    I'm not too sure of the settings on the Mac.
    No. I think the reason they didn't change the chassis is that they market it as thin not because of tooling. IMHO - Apple isn't going to tell us why. I suspect that they studied the market and that most buyers don't have long cpu demanding runs. However, is it for people who do have many long cpu intensive tasks? Probably best to do those on a desktop Mac or switch to a bulkier Windows laptop that has better thermal management.
  • Reply 6 of 16
    amar99amar99 Posts: 181member
    I couldn't help but notice you were running the MBP on battery power. Its peak performance is only reached when the power adapter is connected, from what I understand.
    watto_cobra
  • Reply 7 of 16
    Mike WuertheleMike Wuerthele Posts: 6,861administrator
    amar99 said:
    I couldn't help but notice you were running the MBP on battery power. Its peak performance is only reached when the power adapter is connected, from what I understand.
    The tests were performed plugged into power. They are not substantively different with a near-full battery, though.
    edited May 2019 caladanianchiawatto_cobra
  • Reply 8 of 16
    Mike WuertheleMike Wuerthele Posts: 6,861administrator

    thx1965 said:
    KITA said:
    It definitely depends on the application, but it looks like the new 8 core i9 is giving a reasonable boost in performance over last year's 6 core i9 in real world applications:

    Dave2
    Hi @KITA - where did you get those Premiere Pro testing results from?

    I always prefer real-world vs. abstract benchmark tests. That's the only thing that tells us if a certain processor or CPU is worth the extra money.
    I understand that the Razer Blade i7 was able to take advantage of NVIDIA acceleration inside Premiere Pro. Do you know how the MBP render test was run? Was the Premiere project set to OpenGL or Metal?
    This test is accurate, but Premiere Pro is highly optimized for Nvidia cards with no real regard to the fact that they aren't available in the Mac. As such, this test highly favors the Razer -- which is fine.

    Final Cut Pro (obviously) is optimized for AMD.
    edited May 2019 chiawatto_cobra
  • Reply 9 of 16
    amar99amar99 Posts: 181member
    amar99 said:
    I couldn't help but notice you were running the MBP on battery power. Its peak performance is only reached when the power adapter is connected, from what I understand.
    The tests were performed plugged into power. They are not substantively different with a near-full battery, though.
    Sorry, I'm just going by the actual video, in which there is very clearly no power cord attached to the MBP, and in which the battery level is low and is not in a "charging" state.

    I mean, at 1:03 in the video you can't tell me it's plugged into power.

    In the video they even read out the exact value displayed on the screen as shown in the video. I'm sorry, I just don't understand how you can say it was plugged into power lol.

    watto_cobra
  • Reply 10 of 16
    OutdoorAppDeveloperOutdoorAppDeveloper Posts: 1,293member
    Compared to a typical RTX 2080 laptop at about the same price, the new MacBook Pro graphics benchmark is dreadful: Heaven 4.0 Mac: 2072 Windows Laptop: 5417 https://laptopmedia.com/video-card/nvidia-geforce-rtx-2080-laptop/
    CheeseFreeze
  • Reply 11 of 16
    Mike WuertheleMike Wuerthele Posts: 6,861administrator
    amar99 said:
    amar99 said:
    I couldn't help but notice you were running the MBP on battery power. Its peak performance is only reached when the power adapter is connected, from what I understand.
    The tests were performed plugged into power. They are not substantively different with a near-full battery, though.
    Sorry, I'm just going by the actual video, in which there is very clearly no power cord attached to the MBP, and in which the battery level is low and is not in a "charging" state.

    I mean, at 1:03 in the video you can't tell me it's plugged into power.

    In the video they even read out the exact value displayed on the screen as shown in the video. I'm sorry, I just don't understand how you can say it was plugged into power lol.

    If you don't understand the realities of a clean shot in video production, without the cabling, I'm not sure what to tell you.
    edited May 2019 chiafastasleepwatto_cobra
  • Reply 12 of 16
    amar99amar99 Posts: 181member
    Mike Wuerthele said:
    amar99 said:
    amar99 said:
    I couldn't help but notice you were running the MBP on battery power. Its peak performance is only reached when the power adapter is connected, from what I understand.
    The tests were performed plugged into power. They are not substantively different with a near-full battery, though.
    Sorry, I'm just going by the actual video, in which there is very clearly no power cord attached to the MBP, and in which the battery level is low and is not in a "charging" state.

    I mean, at 1:03 in the video you can't tell me it's plugged into power.

    In the video they even read out the exact value displayed on the screen as shown in the video. I'm sorry, I just don't understand how you can say it was plugged into power lol.

    If you don't understand the realities of a clean shot in video production, without the cabling, I'm not sure what to tell you.
    LMAO no need to patronize. Just say you did the benchmarks off cam and came up with the same results.
  • Reply 13 of 16
    amar99amar99 Posts: 181member
    And btw, @Mike Wuerthele, I've been a reader of AppleInsider for probably 15 years. So this wasn't a criticism. I just work with Macs every day, so I'm interested in the new MBPs!
    watto_cobra
  • Reply 14 of 16
    Mike WuertheleMike Wuerthele Posts: 6,861administrator
    amar99 said:
    amar99 said:
    amar99 said:
    I couldn't help but notice you were running the MBP on battery power. Its peak performance is only reached when the power adapter is connected, from what I understand.
    The tests were performed plugged into power. They are not substantively different with a near-full battery, though.
    Sorry, I'm just going by the actual video, in which there is very clearly no power cord attached to the MBP, and in which the battery level is low and is not in a "charging" state.

    I mean, at 1:03 in the video you can't tell me it's plugged into power.

    In the video they even read out the exact value displayed on the screen as shown in the video. I'm sorry, I just don't understand how you can say it was plugged into power lol.

    If you don't understand the realities of a clean shot in video production, without the cabling, I'm not sure what to tell you.
    LMAO no need to patronize. Just say you did the benchmarks off cam and came up with the same results.
    Not patronizing. However....


  • Reply 15 of 16
    amar99amar99 Posts: 181member
    Mike Wuerthele said:
    amar99 said:
    amar99 said:
    amar99 said:
    I couldn't help but notice you were running the MBP on battery power. Its peak performance is only reached when the power adapter is connected, from what I understand.
    The tests were performed plugged into power. They are not substantively different with a near-full battery, though.
    Sorry, I'm just going by the actual video, in which there is very clearly no power cord attached to the MBP, and in which the battery level is low and is not in a "charging" state.

    I mean, at 1:03 in the video you can't tell me it's plugged into power.

    In the video they even read out the exact value displayed on the screen as shown in the video. I'm sorry, I just don't understand how you can say it was plugged into power lol.

    If you don't understand the realities of a clean shot in video production, without the cabling, I'm not sure what to tell you.
    LMAO no need to patronize. Just say you did the benchmarks off cam and came up with the same results.
    Not patronizing. However....




    Alright, I genuinely wasn't trying to pick a fight here. It was confusing seeing an unplugged computer, and I thought maybe the benchmarks had been run without being plugged in which would have made their results not quite as high as they could be.My bad
    watto_cobra
  • Reply 16 of 16
    brycesteinerbrycesteiner Posts: 1unconfirmed, member
    The new MBP with Vega 20 looks great! I want to see everything the same, using the Vega 20 option but the comparison being the i9 2.3 vs 2.4 processor. I can't imagine there being much difference there but have you seen any?
Sign In or Register to comment.