Call for benchmarking your M1 Mac for photoshop application

[vett93]

If you use Adobe Photoshop and have a M1 Mac with 4K monitor, please benchmark your system with PugetBench for Photoshop: https://www.pugetsystems.com/labs/articles/PugetBench-for-Photoshop-1132/#Howtorunthebenchmark

My Mac Pro 6,1 is almost 7 year old and I have upgraded the CPU, RAM, and SSD. Its performance is shown below:

Most of the comparison videos on YouTube are for video processing. The few videos for photographers are focusing for importing and/or exporting, which is not very useful. So if you have a M1 Mac, please benchmark and share.

Please note that if you use the free version of the software, the results will be posted on their server. This is similar to Geekbench.

A score of 1,000 is based on the following system:

• Intel Core i9 9900K 8 Core
• NVIDIA GeForce RTX 2080 8GB
• 64GB of RAM
• Samsung 960 Pro 1TB
• Windows 10
• Adobe Photoshop CC 2019 (ver. 20.0.0)

 

[ArPe]

No point until the native version is out of beta and released. The current beta is buggy and incomplete.

 

[osplo]

No point until the native version is out of beta and released. The current beta is buggy and incomplete.

Sure but benchmarking now would give us a baseline which can only get better later. Any takers?

 

[vett93]

Sure but benchmarking now would give us a baseline which can only get better later. Any takers?

It can also give an indication of what performance level one can expect at this time. Would it be worth upgrading? Etc.

 

[Pressure]

This is what the test consists of:

General Tasks:

Open 18MP .CR2 RAW Image, Resize to 500MB*, Rotate, Magic Wand Select, Mask Refinement, Paint Bucket, Gradient, Content Aware Fill, Save .PSD File, Open .PSD File

Filter Tasks:

Camera Raw Filter, Lens Correction, Reduce Noise, Smart Sharpen*, Field Blur*, Tilt-Shift Blur*, Iris Blur*, Adaptive Wide Angle, Liquify

* denotes GPU-accelerated tasks

 

[vett93]

I have found the answer that I am looking for. The benchmark results are published on their website.

I searched for Apple computer results and you can see the results here: https://www.pugetsystems.com/benchm...toshop&application=&specs=apple#results-table

 

[pingchp]

How can I install plugin ? Open Adobe Creative Cloud app then nothing works.

 

[vett93]

How can I install plugin ? Open Adobe Creative Cloud app then nothing works.

I followed the instructions on their website:

In summary, you use Adobe Exchange to add the plug-in to your account. Then you go to Creative Cloud to install the plug-in to your Photoshop. Then you run the plug-in in Photoshop under the Window tab.

 

[BLUEDOG314]

While I'm not saying I read heavily into how this benchmark works, there seems to be no consistency in scores. Your Mac Pro is 33% faster than a 16 core 32 thread Zen 3 chip with an RTX 3080? Something is up.

 

[Gnattu]

The beta does not support camera RAW which is required for the benchmark.
You have to run via rosetta at the moment if you really want to know a score, which is not optimal.

 

[vett93]

While I'm not saying I read heavily into how this benchmark works, there seems to be no consistency in scores. Your Mac Pro is 33% faster than a 16 core 32 thread Zen 3 chip with an RTX 3080? Something is up.

It looks to me that Photoshop does not use much GPU. The GPU score of my MP is 95 in this benchmark which is less than the reference system of 100.

What is a Zen 3 chip?

BTW, I run the benchmark 6 times. The scores are less than 100 points apart.

 

[BLUEDOG314]

What is a Zen 3 chip?

The newest generation of AMD Ryzen 5000 series chips.

 

[vett93]

You are probably right. A score of 1,814 is probably too high, as compared to their reference system. I have left a comment/question on their website asking them to clarify. Thanks.

 

[ArPe]

Sure but benchmarking now would give us a baseline which can only get better later. Any takers?

Depends what you are benchmarking. Neural filters? They are being processed in the cloud. Traditional benchmarks such as those by Puget are heavily GPU reliant, but some features are buggy in the current native beta.

Camera RAW support isn’t ready so those benchmarks are out of the window. There are also some other file formats not ready yet.

Then there are newer Photoshop features such as the subject select tool. The native AS version is faster than the Rosetta version, but no faster than running it natively on Intel. What can improve with Machine Learning cores is the accuracy of the selection, after the app is continuously improved for ML.

In two months the native version should be complete. It’s 80% there.