Jump to content
You must now use your email address to sign in [click for more info] ×

M1 Pro (200GB/s memory bandwidth) vs. M1 Max (400GB/s memory bandwidth)


Recommended Posts

  • Staff

Hi @Jan Tippman :)

1 hour ago, Jan Tippman said:

one of the M1 Pro and M1 Max differences is in memory bandwidth. Dou you think that twice memory bandwith is worth it? Did someone tested it already on pre-production machines? 

Unfortunately we're not able to release any benchmark information for these new devices, however I have spoken with our developers who confirmed that the increased memory bandwidth on the Max chip is used primarily for the GPU, rather than the CPU - and Affinity apps rely heavily on the GPU to render the canvas etc.

As for whether the cost is 'worth it' - this is a personal decision that will depend on your workflow and budget, and we're not able to recommend specific hardware as company policy, my apologies.

I hope this helps!

Please note -

I am currently out of the office for a short while whilst recovering from surgery (nothing serious!), therefore will not be available on the Forums during this time.

Should you require a response from the team in a thread I have previously replied in - please Create a New Thread and our team will be sure to reply as soon as possible.

Many thanks!

Link to comment
Share on other sites

Thank you for honest answer. :)

I cannot wait for some unofficial test results... 

If someone knows about some, please let me know. 

BTW, from the attached screenshots it looks, that memory bandwidth doesn't influence CPU performance. 

How is that possible? Is 200 MB/s so much, that higher bandwith is useless? (Just rhetorical questions.)  

Snímek obrazovky 2021-10-19 v 17.44.59.png

Snímek obrazovky 2021-10-19 v 17.48.15.png

Link to comment
Share on other sites

  • Staff
6 minutes ago, user_0815 said:

I assume that this also applies to Photo when working with several pixel layers and live filters. Is that correct?

Absolutely - with Metal Hardware Acceleration on Macs, Affinity uses the GPU to accelerate drawing the canvas in all apps - including Photo & our live filter layers :)

2 minutes ago, Jan Tippman said:

Thank you for honest answer. :)

I cannot wait for some unofficial test results... 

No problem at all, I too am looking forward to seeing what these new machines are capable of!

2 minutes ago, Jan Tippman said:

BTW, from the attached screenshots it looks, that memory bandwidth doesn't influence CPU performance. 

How is that possible? Is 200 MB/s so much, that higher bandwith is useless? (Just rhetorical questions.)  

As far as I understand, from a generic standpoint, the CPU does not need to load/access large 'chunks' of data through memory, unlike a GPU - so the capable 'bandwidth' is not a defining factor in CPU performance.

CPUs rely heavily on memory speed for better performance, as they need to access lots of small 'chunks' as quickly as possible. :)

Please note -

I am currently out of the office for a short while whilst recovering from surgery (nothing serious!), therefore will not be available on the Forums during this time.

Should you require a response from the team in a thread I have previously replied in - please Create a New Thread and our team will be sure to reply as soon as possible.

Many thanks!

Link to comment
Share on other sites

3 hours ago, Jan Tippman said:

I was probably wrong, I thought that memory speed and bandwidth goes hand in hand... 

The usual crude analogy is bandwidth is like the number of lanes on the highway & speed is like the speed limit of the highway. So fewer lanes with a higher speed limit might not perform as well as more lanes with a lower speed limit.

All 3 1.10.8, & all 3 V2.4.1 Mac apps; 2020 iMac 27"; 3.8GHz i7, Radeon Pro 5700, 32GB RAM; macOS 10.15.7
Affinity Photo 
1.10.8; Affinity Designer 1.108; & all 3 V2 apps for iPad; 6th Generation iPad 32 GB; Apple Pencil; iPadOS 15.7

Link to comment
Share on other sites

9 hours ago, R C-R said:

The usual crude analogy is bandwidth is like the number of lanes on the highway & speed is like the speed limit of the highway. So fewer lanes with a higher speed limit might not perform as well as more lanes with a lower speed limit.

It's nice comparison. It makes sense. Thank you. :)

Link to comment
Share on other sites

  • 2 months later...
On 10/20/2021 at 3:38 AM, Jan Tippman said:

It's nice comparison. It makes sense. Thank you. :)

I just put my order for the 10c cpu/ 24c core M1 Max with 32GB; the 400GB/sec speed and dual media encoders future proof the chip while 24 vs 32 gpu cores help stretch battery life. The 24 core gpus actually run slightly faster than the 32 core gpu core bc of total thermal limits, less cores can run faster and hotter… so that’s the sweet spot as far as performance / $ while maximizing battery life, specially an issue in the 14” chassis which has lower thermal envelope, although i ordered the 16” for screen size is #1 productivity driver IMO. Merry christmas!

2021 16” Macbook Pro w/ M1 Max 10c cpu /24c gpu, 32 GB RAM, 1TB SSD, Sonoma 14.4.1

2018 11" iPad Pro w/ A12X cpu/gpu, 256 GB, iPadOS 17

Link to comment
Share on other sites

☛ Affinity Designer 1.10.8 ◆ Affinity Photo 1.10.8 ◆ Affinity Publisher 1.10.8 ◆ OSX El Capitan
☛ Affinity V2.3 apps ◆ MacOS Sonoma 14.2 ◆ iPad OS 17.2

Link to comment
Share on other sites

On 10/19/2021 at 8:28 PM, Jan Tippman said:

I was probably wrong, I thought that memory speed and bandwidth goes hand in hand... 

the two relevant independent factors are latency and bandwidth. The main issue with cpu and ram is latency, not bandwidth . If there is a jump instruction, wrong prediction and cache miss (L1 to L3), waiting for ram takes ages, a nightmare for performance, potentially stalling a large pipeline of commands. Doubling bandwidth but same latency has benefits only in rare situations where the cpu is sequentially reading/writing huge chunks of ram. But such operations are often „outsourced“ to MMU, making virtual copies and postpone copy until a page gets written.

So ram bandwidth is important if cpu / gpu compete over ram BW, but does not help to boost cpu.

Mac mini M1 A2348 | Windows 10 - AMD Ryzen 9 5900x - 32 GB RAM - Nvidia GTX 1080

LG34WK950U-W, calibrated to DCI-P3 with LG Calibration Studio / Spider 5

iPad Air Gen 5 (2022) A2589

Special interest into procedural texture filter, edit alpha channel, RGB/16 and RGB/32 color formats, stacking, finding root causes for misbehaving files, finding creative solutions for unsolvable tasks, finding bugs in Apps.

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...

Important Information

Terms of Use | Privacy Policy | Guidelines | We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.