Latencymon - I don't get it

Hi everyone,

Latencymon shows way better result for my old Surface Pro than it does to my newly purchased gaming laptop. Can someone please explain to me why this is the case and what it really means? I’m really not that good at computer technology.

The Surface Pro specs:
i5-4300 1.90GHz - 2.5GHz
4GB RAM
Windows 10 Home x64
No audio interface
I didn’t tweak anything in this machine for performance.

My new Dell 7577
i7-7700HQ 2.80GHz - 2.81 GHz ( I disabled Turbo Boost)
16GB RAM
Windows 10 Home x 64
Audiobox 22VSL
I did quite a few performance tweaks for this machine, most of which are listed on Steinberg website. I also disabled some unnecessary drivers, such as the NVIDIA graphic driver and Realtek Audio driver.

How does the high latency reported by Latencymon actually affect music production work on the Dell? Most of the time it hovers around 35-125 while occasionally peaking at 700-800, and as high as 2000. Does it mean I’ll be able to load less VSTs, or will experience pops and crackles, or something else? And is it low enough that I can stop worrying about it, or is there something wrong that I should investigate further?

Thanks for reading :slight_smile:


FWIW, here’s a thread with questions I had about Latency Mon answered by some pretty smart dudes: https://www.soundonsound.com/forum/viewtopic.php?f=20&t=2204&start=400&_=1500768000000&qt-forum_new=0

Thanks Alexis, I learn a lot from reading that thread, and also after watching this video from Native Instruments: https://www.youtube.com/watch?v=BnbfVFqkKOA

I found out that my problem was caused by storport.sys. My guess is the Toshiba XG-5 drive I’m having is currently OEM only and doesn’t have any dedicated proprietary driver for it to run more effectively.

From actual working experience, I can say that Cubase runs pretty reliably for me at buffer size 512. So what exactly does the bad result from Latencymon mean? Does it only mean that if I run my audio interface at lower buffer size (64, 128), I’ll experience dropouts?

Amazing vid in your link - thanks!

(Should be a sticky here!)

Ideally, you want to run your latency no greater than a few milliseconds, as any more and you will start to notice the delay between when you generate a note and when you hear it, so you will want to run with the lowest buffer size that you can.

With a properly set up system (makes it sound like it is straightforward), you should be able to select the smallest buffer size.