Available only on the Mac App Store or Windows Store. RIP OpenGL, no more GPU test!
Please post your scores below.
Please post your scores below.
Last edited:
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Long awaited Cinebench R20 released
- Thread starter SoyCapitanSoyCapitan
- Start date
- Sort by reaction score
3350
CPU only test now. Although I use cinema 4d R20 on my other 2 MacPros both fitted with dual 1080ti's using redshift renderer and they fly!
I'm glad they removed their GPU test because it was early-2000s-looking silliness that wasn't representative of anything.
Almost nine years later, Mac Pro with Westmere Xeons still has decent scores. Hold better than I thought - but that is pure Intel fault.
MP5,1 is the modern SE/30.
MP5,1 is the modern SE/30.
Last edited:
Amazing isn’t it!
thats the reason all 3D rendering is heading towards being GPU powered, at least with GPU’s there still seems to be a decent speed bump with each new generation.
Or, is it Maxon's fault for still compiling their code for a nine-year old processor?
When code is recompiled to use the new instruction set extensions, it can be significantly faster.
Compiling for MMX and running on a system with AVX2 should be a crime.
If AVX and AVX2 where so good and could be applicable to everything as Intel wanted to sell us, every Pro app should be using by now, no?Or, is it Maxon's fault for still compiling their code for a nine-year old processor?
When code is recompiled to use the new instruction set extensions, it can be significantly faster.
Compiling for MMX and running on a system with AVX2 should be a crime.
2010-2020 will be know as the decade that ended Moore law.
Actually, it's proof that Moore's law is still going strong.
Moore's law was never about *faster* CPUs, it was about more transistors per mm² on the die.
For several generations, more transistors did turn out faster CPUs. Bigger caches, better branch prediction, µOp architectures - and by coincidence faster clocks.
The "faster clock" thing is hitting a ceiling, so new CPUs work to increase the instructions-per-cycle. One big way this is done is to extend the MMX/SSE SIMD instructions to wider and wider registers, with more complete instruction coverage.
AVX uses 256-bit wide registers, with good FP coverage and basic integer instructions. This is basically "loop unrolling" in hardware. A single instruction can do 8 FP32 or "int" operations in parallel.
AVX2 extends AVX so that the 256-bit registers can do a fairly full set of both integer and floating instructions. AVX512 doubles the size of AVX2 registers to 512-bits - so each instruction can do 16 FP32 or "int" operations.
There are also new instructions to accelerate encryption/decryption and other tasks.
But, this all goes to waste if your programs are still compiled for Woodcrest 65nm processors.
Last edited:
Actually, I disagree. The SE/30 was an all in one machine and had limited expandability. I had one. Upgraded the RAM, and ran a second Black & White (not monochrome) monitor. Couldn't do much else.
IMHO, the cMP 5,1 is the modern Macintosh II (the first really expandable Mac). I had a IIci, Processor upgradable, NuBus slots and etc.
My Score - 2272
Lou
Last edited:
My sister bought a 4 core mac mini 2018. She uses Tesseract for OCR. Her new machine is quite bit faster at that task than my 3.5 Ghz 2014 iMac. I suspect that AVX-512 is a big part of that.
On the 5 machines I use the most:
2014 Mac Mini: 765
2010 17" MBP: 762
2009 (?) MacPro: 3338
2018 MSI WS63: 1829
Unknown Year HP Z800: 3072
2014 Mac Mini: 765
2010 17" MBP: 762
2009 (?) MacPro: 3338
2018 MSI WS63: 1829
Unknown Year HP Z800: 3072
What is the configuration of the Z800? As an FYI the Zx00 series uses similar levels of technology as the 4,1 and 5,1 Mac Pros.
[doublepost=1552052806][/doublepost]6,1 Mac Pro: 6-core, 3.5GHz, 16GB RAM, D300:
Single: 293
Multi: 2109
Multi: 2109
@iGobbleoff: Interesting to note it appears CB20 is not indicating threads in its processor reporting.
[doublepost=1552090786][/doublepost]
[doublepost=1552090786][/doublepost]
Any chance you could run a single thread test on the Mini?