Affinity Photo 2 performance testing

Messages
2,704
Name
Steve, Coventry, England
Edit My Images
Yes
Anybody with Affinity Photo 2 willing to run the test and post the result, with details of their CPU and RAM?

Takes less than 5 minutes to run.



You will get this (hopefully much better numbers, this is an old slow laptop)

bmk.jpg

Ryzen 5 2500U @ 2GHz 16GB RAM
 
I didn't even know it had a benchmark test. But this really confirms that the new iPad Pro M4 is really overpowered for what it's used for. I'll run it on my other machines later today for comparison.

IMG_0110.jpeg

Apple M4 @ (ARM) @4,4GHz, 8 GB RAM
 
Anybody with Affinity Photo 2 willing to run the test and post the result, with details of their CPU and RAM?

Takes less than 5 minutes to run.



You will get this (hopefully much better numbers, this is an old slow laptop)

View attachment 439467

Ryzen 5 2500U @ 2GHz 16GB RAM

So what do you do with this information once you have run the test?
 
Anybody with Affinity Photo 2 willing to run the test and post the result, with details of their CPU and RAM?

Takes less than 5 minutes to run.



You will get this (hopefully much better numbers, this is an old slow laptop)

View attachment 439467

Ryzen 5 2500U @ 2GHz 16GB RAM
Likewise I too had no idea that v2 had its own benchmark built in


Of note in the instructions it advises temporarily disconnecting from the internet....to ensure no internet related background activity affecting the test....

To perform the benchmark:
  1. Ensure no other apps are running in the background. This includes services that sync cloud storage, so disconnect from the Internet to ensure this does not happen while performing the benchmark.
  2. In Affinity Photo 2, close all document windows.
  3. Select Help>Benchmark.
  4. Select Run Benchmark.


I'm just off out, but is there a link to this test?

I can run it on a M4 Pro MacBook 16 with 24 GB RAM and an ASUS Zenbook duo UX582 with 32GB RAM and the 3070 TI (8GB) and the i9 chip
 
Last edited:
MacOS Monterey
3.3Ghz i5 Intel Quad core
40GB 1867 DDR3
Benchmark.jpg
 
Especially on Intel, it would be interesting to know the processor model, as some i5s are faster than some i9s
 
So what do you do with this information once you have run the test?
If you think it should be better you can go to EDIT>SETTINGS and try enabling / disabling hardware acceleration, and or check your graphics drivers are up to date etc.
 
Not on my PC now but will run the test later on today :fingerscrossed:
 
Okay back home from Wex (must not buy the 40-150 F2.8).

No tabs open or anything, just affinity 2

Mac and Asus. It won't let me upload the images

Vector (Single Core)
461 Asus . Macbook: 818

Vector (Multi Core)
3256 Asus. Macbook: 6469

Raster (multi core)
470 Asus. Macbook: 1852

Raster (Single core)
12394 - RTX 3070 TI. macbook: 34674
12393 - intel Iris XE,

Combined (multi)
535 Asus. 1793 MacBook

Combined (multi)
5307 - RTX 3070 TI. Macbook: 30953
6243 - intel Iris Xe


I have the result images in the phone. Laptop was plugged in and MacBook on battery
 
Last edited:
Where can we find an explanation of the numbers?
FWIW & IMO
Benchmarks are primarily aimed at comparison differences.

The higher each of the benchmark figures the better, speedier, the PC or Mac performance.
 
Ok just did the test....


PC Specs
SYST INFO .jpg

Initial Benchmark

FIRST ONE 2024-11-24 161032.jpg

A second run a few moments after the above initial one

SECOND ONE 2024-11-24 161241.jpg

So for whatever factors influence the 'runs' there can be differences.

Of note, unless doing some heavy processing that is dependent on the GPU, the GPU fans are silent but when stressed the fans speed up audibly......and I heard them during the benchmark testing runs.
 
Ok just did the test....


PC Specs
View attachment 439542

Initial Benchmark

View attachment 439543

A second run a few moments after the above initial one

View attachment 439544

So for whatever factors influence the 'runs' there can be differences.

Of note, unless doing some heavy processing that is dependent on the GPU, the GPU fans are silent but when stressed the fans speed up audibly......and I heard them during the benchmark testing runs.


Very interesting, your i5 is one of the best buys, and you see it is faster that the i9 above :) and you results show how much difference the video card makes.

Lots of things can make minor differences on each run
 
Very interesting, your i5 is one of the best buys, and you see it is faster that the i9 above :) and you results show how much difference the video card makes.

Lots of things can make minor differences on each run
Yes, I was very carefull about selecting both the i5 version/generation and the GPU........I was trying to keep my self-build budget under control :LOL: The only reason IIRC to get the CPU with the onboard GPU was a a backstop in case I had any GPU issues ;) thinking about the build, I think I started the build testing with the onboard GPU and installed the RTX the next day???

NB I already had the nVidias RTX3060 bought in 2022 for my old build to replace the GTX 1050Ti. The primary reason for the new build was because the old PC had a 1st Gen i5 Quad core (IIRC the only Quad Core of that generation and did not support AVX2 which the newer software required ! :(
 
Last edited:
I've now run it on my Mac Mini M2 Pro with 16GB RAM, and my iPad Mini 7 with 8 GB of RAM. It's interesting that my GPU results are higher for the Mac Mini M2 Pro than for the iPad Pro M4.

Mac Mini M2 Pro
Affinity Photo M2 Pro.jpg

iPad Mini 7
Affinity Photo iPad Mini 7.PNG
 
Screen Shot 2024-11-24 at 19.28.24.png
i9 Macbook, coming up for 5 years old.
 
Thanks for all the results, interesting, and confirms thoughts that most people had.

What I see from it


Apple (with and Apple chip) is generally faster than anything Intel (including Apples with Intel chips)

Graphics cards are a major contributing factor

Combined single GPU figures are the best result to judge/compare editing speeds

Single core speeds are the most significant processor figure, multi core speeds less so.

An i9 is not faster than an i5 just because it is an i9




On my old laptop, the main time I felt the lack of speed was when merging/stacking, it was taking 96 seconds for a five image merge!
I would have been happy if the new machine came in under 30 seconds, it actually does it in 19.5 seconds

The total file size of the five images is 26MB, and I time from clicking OK on this screen
1734080281334.png



To the time the options (natural/detailed) on the left here appear
1734080346511.png




My “guesstimate” of how long other machines (the ones people posted results for) would take on the same testing
1734080400707.png
Of course those could be way out ;)




Which machine is best?

If you want the very best editing experience, then it’s Mac
If you want six times the storage, more memory and still a very good editing experience at half the price, then it’s PC (with careful choice of components)

I think the main point would be that most fairly recent machines with a half decent graphics card will be comfortable for the average non-professional user, and it goes up from there
 
I just ran my macbook again and the scores have altered. Some slightly higher and some slightly lower. Nothing running on the laptop bar Affinity. They are thereabouts that same, but it is interesting how the same machine, same set up, alters when you run it a few times. This was the same on the Asus laptop also.

It would be interesting to see a base M4 mac mini vs your pro model above Chris.
 
MacBook Pro M1 13 with 16Gb RAM.

Initial run

First run.jpg

Second run

Second run.jpg
 
HP ENVY x360 Convertible
Processor 12th Gen Intel(R) Core(TM) i7-1260P 2.10 GHz
Installed RAM 16.0 GB (15.7 GB usable)
System type 64-bit operating system, x64-based processor
Pen and touch Pen and touch support with 10 touch points

Photo_BFXlLvJJCt.png
This with my M4 Pro Mac mini with 24Gb memory.

View attachment 440884
 
Thanks, have added that one.

This is what it looks like so far

pre.jpg

The numbers on the right are how long it takes Affinity to process a 5 image HDR merge. (this was the only thing that took too long for me on my old machine)
The red figures are actual results, the rest are guestimated.
The green figures are likely to be around 2.5 time greater going by some rough tests people have done.
The black figures are likely to be fairly close.

What is quite obvious is that the high benchmark results of the Apples do not translate directly to faster processing times on Affinity

There could be a number of reasons

How the benchmark is run
How well the machines process the type calculations used in this particular operation
How well the Affinity code is optimised for Intel or Apple

What is also clear is that any Apple from an M1 Pro up is going to be very comfortable using Affinity, but the situation is not so clear with Intel, the CPU and GPU on an intel system have to be chosen carefully.

And to try to answer the frequent question "what computer is best / do I need?" was one of the aims, and to me it has made that a lot clearer.

Usinf PS would probably give similar results, but I have not tried it.


Best performance for money? i5 13600 plus 8GB RTX 3069ti / 4060ti
 
Back
Top