Blender 4.0.1 BMW27 demo test

[SchumannFrequency]

Render time: 119.2 seconds

This test is famous in part because Intel or AMD have frequently used the test to demonstrate the advances of their latest-generation CPUs.

Hardware: Intel 12600KF (stock) -- Kingston 6200 MHz CL36 -- Sapphire RX 7600 -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB
Software: Alpine Linux, river wm, Blender 4.0.1

You can find the test on this page: https://download.blender.org/demo/test/BMW27_2.blend.zip
Please include your specific hardware and software when you post a result.

These are some old results for this BMW27 test.

 

[Marcus Vinicus]

Hardware as per signature below
Windows 11 Pro 23H2 22635.3139
Render time: 63.68 seconds

 

[mclaren85]

Reserved.
I will post the result.

 

[Hardi]

82.32

5950X PPT/TDC/EDC - 190/135/160
32GB 3800MHz

 

[Gica]

Blender 4.0
14700KF @ 7900X mode (max 105W power consumption)
RTX 3070 Ti
DDR4 32GB@4000MHz CL18

CPU: 90.7 sec
OptiX: 11.46 sec

 

[SchumannFrequency]

Render time: 115.74 seconds

Hardware: Intel 12600KF (stock) -- Kingston 6200 MHz CL36 -- Sapphire RX 7600 -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB
Software: Alpine Linux, river wm, Blender 4.0.2 (Flatpak version)

This is 4 seconds faster than my previous result. There are two possible explanations why it is faster now.
- Blender 4.0.2 vs. 4.0.1
- I am now using the Flatpak version instead of the package in the standard repos.

I am going to be under 108 seconds render time if I were to select the Linux distro that has the highest performance in Blender.

 

[dom1234]

Render time: 149.14 seconds
Update: GPU takes 33 seconds

Apple M2 Pro 10 Core (6P + 4E) @24W
Apple M2 Pro 16 Core GPU @15W
Blender v4.1, macOS Sonoma 14.5

 

[SchumannFrequency]

Render time: 101,89 seconds

Hardware: Intel 12700KF (stock) -- G.SKILL RIPJAWS @3600 CL18 (stock) -- Sapphire RX 7600 -- ASRock B760M-ITX/D4 WiFi -- fractal design DEFINE NANO S -- bequiet! SYSTEM POWER 10 550W -- DeepCool AG500BK ARGB -- EVO 850 500GB
Software: mageia, LXQt, Mesa open-source driver, XFS file system, Blender 3.3.16

 

[A Computer Guy]

Blender 4.2, 5950x stock everything w/ DDR4-3200 ECC (seems I didn't get anywhere close to 93 seconds or I'm not looking in the right spot for render time)

Render time: 119.2 seconds

How are you getting render time? I don't see an output for that.

 

[SchumannFrequency]

How are you getting render time? I don't see an output for that.

I had simply taken the time reported by the app and converted it to seconds.
In your case, it reports 1:41.36. Converted, this is 101.36 seconds.

 

[HAL9090]

6.18 Seconds (Once with RTX4090 SUPRIM X 24GB RAM) (Linux-x86_64) NVIDIA Driver Version: 535.183.01 Ubuntu 24.04 LTS (128 GB RAM)

65.51 seconds (CPU: Core™ i9-14900K × 32)

infernal machine...

 

[Solaris17]

Blender 4.2.1

 

[AVATARAT]

9700x@PBO
2x32GB@6200MT/s
Time: 01:39.01

 

[SchumannFrequency]

9700x@PBO
2x32GB@6200MT/s
Time: 01:39.01

Thanks!

 

[DR4G00N]

Blender 3.6.15

5600 @ 4.65GHz (BMW CPU): 212.76sec

RTX 3060 Laptop 80W OptiX (BMW GPU): 13.37sec

 

[stahlhart]

CPU render:

GPU render, CUDA, ECC off (MSI RTX 4090 Suprim X Liquid):

Blender 4.2, remainder of system specs to the left...

 

[penumbra]

Cpu: 71.21
Gpu: 05.33 (optix) 10.76 (cuda)

 

[tps3443]

This is my "OLD-SCHOOL" Xeon w3175x (28/56) Running at a mild 4.5Ghz all-core overclock. System ram is 192GB DDR4 3800CL14-14-14-32 (G.Skill Samsung B-Die)

CPU Gets it done in 51 seconds. Definitely a very fast chip in 2025, it has no trouble pushing the water-cooled and power modded 1200W RTX5090FE, that little sucker takes down RTX Pro 6000's for fun on the weekends...

 

[Paramount]

7950x3d

63.73 sec

 

[Det0x]

Baby 16 core EPYC reporting in
4585PX is basically the server version/clone of the desktop 9950X3D with same boost limits and FMAX etc

Blender version = 4.5
For pure CPU render it took 45.27 seconds

While my 5090 only used 2.69 seconds for the same job

Oh and i also ran the normal blender benchmark

CPU score = 767.63

Blender - Open Data Benchmark

Blender Open Data is a platform to collect, display and query the results of hardware and software performance tests - provided by the public.

opendata.blender.org

GPU score = 17151.17

Blender - Open Data Benchmark

Blender Open Data is a platform to collect, display and query the results of hardware and software performance tests - provided by the public.

opendata.blender.org

 

[A Computer Guy]

Baby 16 core EPYC reporting in
4585PX is basically the server version/clone of the desktop 9950X3D with same boost limits and FMAX etc

Blender version = 4.5
For pure CPU render it took 45.27 seconds
View attachment 410028

While my 5090 only used 2.69 seconds for the same job
View attachment 410029

Oh and i also ran the normal blender benchmark

CPU score = 767.63

Blender - Open Data Benchmark

Blender Open Data is a platform to collect, display and query the results of hardware and software performance tests - provided by the public.

opendata.blender.org

View attachment 410030

GPU score = 17151.17

Blender - Open Data Benchmark

Blender Open Data is a platform to collect, display and query the results of hardware and software performance tests - provided by the public.

opendata.blender.org

View attachment 410031

That's interesting. Someone around here had a post a while ago asking if anyone knew if those little Epycs could run in consumer motherboards. I suppose your screenshot answers the question. Is there any advantage to the baby Epyc over the 9950X3D?

 

[Det0x]

That's interesting. Someone around here had a post a while ago asking if anyone knew if those little Epycs could run in consumer motherboards. I suppose your screenshot answers the question. Is there any advantage to the baby Epyc over the 9950X3D?

No, in my testing they are pretty much exactly the same CPUs with different names