Zenbook Pro Duo vs Surface Laptop 7 Snapdragon X Elite

Zenbook Pro Duo 10th Gen i9, 32GB RAM vs Surface Laptop 7 Snapdragon X Elite, 16GB RAM
SO I have had the ASUS for a few years now and nothing has ever marched it for performance with FL Studio but I was surprised how close the New Surface Laptop 7 came - they have clearly done a great job with the x86 emulation layer but notwithstanding, I look forward to the time when Image-Line release a Windows ARM based version of FL Studio!!
buymeacoffee.com/rd3d2t (Thanks)

[@vartannazarian3451]

This is the kind of test we needed

[@ianwalker2258]

Thanks, I am a user myself and wanted this and realised I might be able to do something (I know its not perfect but at least it is something :-) )

[@user-co3pj3yz6q]

Can you test a few third-party VSTi and VST plugins inside FL Studio? and audio interface, but guess the drivers won't emulate...

[@KevinShroff]

+1 please test stuff like serum, omnisphere, nexus, sylenth1

[@ianwalker2258]

Will see what I can do.

[@KevinShroff]

Which build of version are you on, on the Surface? I tried 24H2 with Parallels on my Macbook pro 14 but I couldnt even launch FL!

[@ianwalker2258]

Its 24H2 as well - obviously the ARM version emulating an x86. If you are using a Mac I would advise downloading the MAC O/S version of FLStudio from the Image-Line site

[@untml]

Hi, You have a fundamental flaw in your test. It's not possible to compare the performance of two devices by only examining the CPU graph of FL Studio. The relationship is not linear, as X% = Track/Process. This is a mistake Nick Batt from Sonic State also made with the first M1.
To correctly test performance and obtain meaningful data about each device, you need to follow the established method for testing DAW CPU performance. Create a project with the same track and commonly used processing, then multiply it until the sound crackles on each machine. Ensure the audio buffer is the same on all systems.
The issue with your current testing method is that (one example) doubling all the tracks in a project does not necessarily double the CPU resources needed to run it on all machines, so it's not an accurate indication of the true performance each machine is capable of. There are many reasons why this kind of test is flawed. I know it may not be easy to accept after all the hard work you put into it, but I would advise you to check out "Scan Audio's" testing methodology based on DAWbench (no FL studio but the The principal is the same) Again, I appreciate the work and effort you put into it, especially the testing of such unique devices.

[@ianwalker2258]

For sure, this is not meant to be a scientific test but just a "gauge" to aid comparison of devices performance when the primary use case is "FL Studio" .

[@untml]

@@ianwalker2258 Thank you for your reply. I understand your point, but I still think you should strive to create tests that are trully beneficial to others. The way you are conducting the test now is misleading, and I do not consider it a reliable "gauge."
Again, multitasking processing is not necessarily linear. Let's say the CPU graph of FL Studio shows 25% on one computer and also 25% on another. From your test, it would seem that both machines have equal processing power with FL Studio. Now, let's double the channel count in the project. It's possible that the FL Studio graph on computer 1 will show 50% and on computer 2 only 35%. However, if you push the channel count to the limit, you will see that computer 2 has more than double the processing power.
Maybe In addition to the tests you are already conducting, consider adding this kind of test, like DAWbench but for FL studio. It will not require much extra work and will provide results that truly reflect the capabilities of each machine. If you do this, I promise to subscribe to your channel and recommend it to others.