Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Geekbench 4 is out. CPU and GPU..

Last edited:
AndreeOnline said:
Mac Pro 4.1-->5.1 12c 3.46GHz
R9 280X

Single: 2870
Multi: 20655

Compute: 107519

https://browser.geekbench.com/v4/cpu/7204
https://browser.geekbench.com/v4/compute/2938
Click to expand...

GB4 is a more accurate idea of what these old 12 core Macs can do. For a while people were going around with an overinflated idea of what their systems were capable of compared to a new system because GB3 wasn't equivalent enough to a real world workload.

My Win10 system for new hardware reference:

6700K
GTX 1070

Single core 5320
Multi core 17499
OpenCL Compute 155688
 
I kind of like the fact that the GB4 is free for 64 bit benchmarks, paid for 32bit-- the opposite of GB3.

32bit code doesn't reflect what's actually run on most machines.
 
  • Like
Reactions: keysofanxiety and frou
Hey SoyCap,

Always nice to get updated benchmarks. I'd much prefer accurate results that translates into real world performance than puffy scores. But I always find it hard to say how trust worthy these things are. Time will tell.

Look how my Mid 2012 MacBook Pro has a 30% advantage over my 12c Mac Pro in 'single' =)

Still, the purpose of my Mac Pro build was render speed in FCPX, out of Resolve and Maxwell in C4D. I can honestly say I've been happy so far in those scenarios.

I'm waiting to see what Apple does for the Mac Pro—with a personal deadline this fall/winter. If nothing appears I'll get back to building PCs (which I do miss a bit). It's just so EXTREMELY annoying that everything is centered around single socket. Might look into Supermicro or similar. Time to read up on home built dual CPU systems.
[doublepost=1472625481][/doublepost]
jerwin said:
I kind of like the fact that the GB4 is free for 64 bit benchmarks, paid for 32bit-- the opposite of GB3.

32bit code doesn't reflect what's actually run on most machines.
Click to expand...

It was nice to run the vanilla bench for free. But CUDA requires purchase which only cost me like $7 for single user OS X.
 
  • Like
Reactions: SoyCapitanSoyCapitan
It hurts that the fastest possible cMP (3.46 GHz) gets spanked in Single Core by anything remotely recent - even lowish-end mobile chips :eek:
 
frou said:
It hurts that the fastest possible cMP (3.46 GHz) gets spanked in Single Core by anything remotely recent - even lowish-end mobile chips :eek:
Click to expand...

Yes, this is true. The delta between Single and Multi on the X5690 is especially large with Single only being like 13% of potential power.

But I think in situations where performance relies on Single thread power, the X5690 can sometimes hide behind a well balanced system with fast PCIe SSD and so on, whereas when true grunt is needed it's still great bang for the buck.

It's about time to turn this machine into a render station. Here it would shine since it's quiet—and filled with modern HDDs you can have all the space you need for non speed critical storage of large media files and image sequences. Oh, well.

Anyway: bring on the RX 470/480 results!
 
My Skylake hackintosh:

Single core: 6283
Multi core: 20702
https://browser.geekbench.com/v4/cpu/11650

The GPU compute test quit before finishing.

Screen Shot 2016-08-31 at 4.21.07 AM.png
 
Last edited:
  • Like
Reactions: Synchro3
Not working on MacOS 10.10.5? o_O

Code: 
Process:               Geekbench 4 [4615]
Path:                  /Applications/Geekbench 4.app/Contents/MacOS/Geekbench 4
Identifier:            com.primatelabs.Geekbench4
Version:               ???
Code Type:             X86-64 (Native)
Parent Process:        ??? [1]
Responsible:           Geekbench 4 [4615]
User ID:               501

Date/Time:             2016-08-31 13:15:03.315 +0200
OS Version:            Mac OS X 10.10.5 (14F1909)
Report Version:        11
Anonymous UUID:        250C95C6-CA0B-F80C-AACC-2AFC90E655E6


Time Awake Since Boot: 1200 seconds

Crashed Thread:        0

Exception Type:        EXC_BREAKPOINT (SIGTRAP)
Exception Codes:       0x0000000000000002, 0x0000000000000000

Application Specific Information:
dyld: launch, loading dependent libraries

Dyld Error Message:
  Library not loaded: /System/Library/Frameworks/Metal.framework/Versions/A/Metal
  Referenced from: /Applications/Geekbench 4.app/Contents/MacOS/Geekbench 4
  Reason: image not found

Binary Images:
    0x7fff60c87000 -     0x7fff60cbd92f  dyld (353.2.4) <708E385E-50C6-32A3-983C-42E97BEE4346> /usr/lib/dyld
    0x7fff94000000 -     0x7fff941b0ff3  com.apple.QuartzCore (1.10 - 361.19) <E59000AA-DC30-3F73-A103-0240576C9556> /System/Library/Frameworks/QuartzCore.framework/Versions/A/QuartzCore

Model: MacPro5,1, BootROM MP51.007F.B03, 12 processors, 6-Core Intel Xeon, 2.66 GHz, 24 GB, SMC 1.39f11
Graphics: NVIDIA GeForce GTX 670, NVIDIA GeForce GTX 670, PCIe, 4095 MB
Graphics: ATI Radeon HD 2600 XT, ATI Radeon HD 2600, PCIe, 256 MB
Memory Module: DIMM 1, 4 GB, DDR3 ECC, 1333 MHz, 0x830B, 0x4E54344743373242385042304E462D434720
Memory Module: DIMM 2, 4 GB, DDR3 ECC, 1333 MHz, 0x830B, 0x4E54344743373242385042304E462D434720
Memory Module: DIMM 3, 4 GB, DDR3 ECC, 1333 MHz, 0x830B, 0x4E54344743373242385042304E462D434720
Memory Module: DIMM 5, 4 GB, DDR3 ECC, 1333 MHz, 0x830B, 0x4E54344743373242385042304E462D434720
Memory Module: DIMM 6, 4 GB, DDR3 ECC, 1333 MHz, 0x830B, 0x4E54344743373242385042304E462D434720
Memory Module: DIMM 7, 4 GB, DDR3 ECC, 1333 MHz, 0x830B, 0x4E54344743373242385042304E462D434720
AirPort: spairport_wireless_card_type_airport_extreme (0x14E4, 0x8E), Broadcom BCM43xx 1.0 (5.106.98.100.24)
Bluetooth: Version 4.3.6f3 16238, 3 services, 19 devices, 1 incoming serial ports
Network Service: Ethernet Bond (bond0), Ethernet, bond0
Serial ATA Device: Samsung SSD 840 EVO 1TB, 1 TB
Serial ATA Device: Samsung SSD 840 EVO 500GB, 500,11 GB
Serial ATA Device: HL-DT-ST DVD-RW GH80N
Serial ATA Device: SAMSUNG HD501LJ, 500,11 GB
Serial ATA Device: SAMSUNG HD501LJ, 500,11 GB
Serial ATA Device: WDC WD20EFRX-68AX9N0, 2 TB
Serial ATA Device: WDC WD20EFRX-68AX9N0, 2 TB
Serial ATA Device: MARVELL VIRTUALL
Serial ATA Device: Samsung SSD 840 EVO 500GB, 500,11 GB
Serial ATA Device: Samsung SSD 840 EVO 500GB, 500,11 GB
Serial ATA Device: MARVELL VIRTUALL
USB Device: Logitech Camera
USB Device: iPhone
USB Device: Keyboard Hub
USB Device: Apple Optical USB Mouse
USB Device: Apple Keyboard
USB Device: Hub
USB Device: DYMO LabelWriter 400
USB Device: Throttle - HOTAS Warthog
USB Device: Joystick - HOTAS Warthog
USB Device: BRCM2046 Hub
USB Device: Bluetooth USB Host Controller
FireWire Device: built-in_hub, Up to 800 Mb/sec
Thunderbolt Bus:

When checking the app:

/Applications/Geekbench 4.app/Contents/MacOS/Geekbench 4

Strange! Or do I need to update MacOS to 10.11 first?

Cheers
 
m4v3r1ck said:
Not working on MacOS 10.10.5? o_O
Click to expand...

Argggggg! :rolleyes:

Thank you for your interest in Geekbench 4. We are sorry to hear that you are running into issues. Geekbench 4 requires OS X 10.11.4 or later.
Click to expand...

Why not just put this requirement on the FRONTPAGE:

ScreenCap%202016-08-31%20at%2014.12.32.jpg


or the "Buy" page:

ScreenCap%202016-08-31%20at%2013.54.26.jpg


Using "Mac OS X (<10.10) and MacOS (>10.11)" together for the app is really quite confusing!

Cheers
 
Last edited:
  • Like
Reactions: zoomfinder
frou said:
It hurts that the fastest possible cMP (3.46 GHz) gets spanked in Single Core by anything remotely recent - even lowish-end mobile chips :eek:
Click to expand...
With GB3 it didn't account for heavy load and therefore did not accurately reflect real life. What I mean by this is my 5,1 Mac Pro, despite having lower GB3 scores than a 2012 rMBP, was able to transcode video faster. This was because the rMBP was not able to sustain 100% utilization across all threads during the operation whereas the Mac Pro could. I suspect this was a thermal limitation.
 
frou said:
It hurts that the fastest possible cMP (3.46 GHz) gets spanked in Single Core by anything remotely recent - even lowish-end mobile chips :eek:
Click to expand...

Wow im a little suprised how well my laptop did! Beats out single core on my mac pro's as well, but looses in multicore performance but for a laptop this is pretty good. I love my laptop and I was originally planning to get a 15 mac pro but they never updated them so i got this one instead, i'm definitively pleased.

https://browser.geekbench.com/v4/compute/8333

1.jpg


https://browser.geekbench.com/v4/cpu/29290

2.jpg
 
Too bad. I am still running 10.10. Can't run Geekbench 4.

Hope this questions is not too stupid: for Photoshop, should I pay attention to single-core performance or multi-core performance? Based on the Geekbench 4 Mac performance list, my machine (mid 2010 MP5,1, 2.93 GHz, 12 cores) is still quite good in multi-core performance, but quite lousy in single-core. I wonder if I should expect a major boost in Photoshop performance if I switch to MP6,1 or MP7,1.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back