crazy cpu usage

[calimerox]

hi there, 

i noticed in the last versions of mixbus 32c the cpu usage increased a lot lot on my machine. 

running mixbus32c now with one stereo track on 96khz and with 3 plugins (ava mastering eq, presswerk u-he and overtone fc70) i already reach 100% cpu on a pretty fast and new laptop. 
i havent had that before and even on my old thinkpad from 2011 i had way better performance.. is there something i can try or check? if not mixbus32c becomes completely unusable for me. 

also as a side note: it is weird to install mixbus32c new, and there are no mixbusses    and you have to first find the shortcut etc to enable them, etc.. it is a minor detail but everytime i set up a new machine it is weird. especially when the software has the busses in the name  

thanks!

Code:

cat /proc/cpuinfo
processor    : 0
vendor_id    : GenuineIntel
cpu family    : 6
model        : 165
model name    : Intel(R) Core(TM) i7-10750H CPU @ 2.60GHz
stepping    : 2
microcode    : 0xea
cpu MHz        : 899.999
cache size    : 12288 KB
physical id    : 0
siblings    : 12
core id        : 0
cpu cores    : 6
apicid        : 0
initial apicid    : 0
fpu        : yes
fpu_exception    : yes
cpuid level    : 22
wp        : yes
flags        : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep bmi2 erms invpcid mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp pku ospke sgx_lc md_clear flush_l1d arch_capabilities
vmx flags    : vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid ple pml ept_mode_based_exec
bugs        : spectre_v1 spectre_v2 spec_store_bypass swapgs itlb_multihit
bogomips    : 5199.98
clflush size    : 64
cache_alignment    : 64
address sizes    : 39 bits physical, 48 bits virtual

[Sojuzstudio]

What Linux distribution do you use? Have you optimized it for audio use?

[calimerox]

i have optimized it a little bit just on my laptop, like setting ulimit, etc.. but not extensively. using also pulseaudio and alsa in contrast to jack which i use in the studio...  not sure if this plays into the cpu usage?

i m on ubuntu

[calimerox]

this is for example just opening mixbus32c. no plugins, no nothing. and the cpu load is at 46%, fluctuating between 25 and 60%

as soon as i import a stereo track and use 2 or 3 plugins i m at the limit.  and this with a computer I run davinci resolve projects in 4k pro res and with 25 tracks of audio and effects in fairlight with no problem.

[madmaxmiller]

this is for example just opening mixbus32c. no plugins, no nothing. and the cpu load is at 46%, fluctuating between 25 and 60%

as soon as i import a stereo track and use 2 or 3 plugins i m at the limit.  and this with a computer I run davinci resolve projects in 4k pro res and with 25 tracks of audio and effects in fairlight with no problem.

96kHz - how many buffers? Your screenshot shows DSP load, not CPU load... - unless you are recording and monitoring through Mixbus you don't need ultra-low latency.
MMM

[calimerox]

this is for example just opening mixbus32c. no plugins, no nothing. and the cpu load is at 46%, fluctuating between 25 and 60%

as soon as i import a stereo track and use 2 or 3 plugins i m at the limit.  and this with a computer I run davinci resolve projects in 4k pro res and with 25 tracks of audio and effects in fairlight with no problem.

96kHz - how many buffers? Your screenshot shows DSP load, not CPU load... - unless you are recording and monitoring through Mixbus you don't need ultra-low latency.
MMM

buffer is at 10.7ms (1024 samples) maybe that is a bit low? I will try increasing the buffer size..

[calimerox]

I set the buffer to 4096 samples (42ms)and it significantly dropped dsp load and made the performance more stable. thanks for that hint! I worked in the past mostly with 48khz projects so this might have made the difference. for recording this might be already a too high latency? wouldnt be a problem for me atm as I am working mostly in the box..

[Dingo]

If you want to compare Resolve with Mixbus test at 44.1 or 48 kHz which are the only two settings in Resolve.

[madmaxmiller]

I set the buffer to 4096 samples  (42ms)and it significantly dropped dsp load and made the performance more stable. thanks for that hint! I worked in the past mostly with 48khz projects so this might have made the difference. for recording this might be already a too high latency? wouldnt be a problem for me atm as I am working mostly in the box..

Yes, double frequency at same buffers = half latency, but also half the time to deliver every packet to the sound device.
MMM

P.S. I'm always monitoring in the sound device (Motu Ultralite AVB atm) with virtually zero latency and therefore i can have high buffers all the time, it just lands on the hard drive a few ms later.

[Sojuzstudio]

Latency can sometimes be an issue. Does Motu Mixer work on Linux? In fact, I look forward to hoping that RME Fireface will come with Linux drivers and Totalmix. I could finally leave Windows out altogether.

Now that I've been testing the latest Mixbus 32 C Avlinux MXE for a week, it may run at a higher dsp load than before. Or I just imagine.