Please can you post results of this. Thanks!
isn’t this from the Ballad of the Goodly Fere by Ezra Pound?
A serum poem
maybe i am im love again.
With the phase knob, you can ride the LFO path backwards, its quiety trippy to listen to it.
Little Rant on Music and Computer ressources
I don’t know for you guys but handle ressources with music and computers is a thing since I started to make music… For each new piece of software in time, you have to handle the results you get if you want to use it, especially if you don’t have much money to change computers every 4-5 years (I would say in the past I could use a computer for 8 years now I can’t.)
Computers change a lot in the last 20 years, 32 bits to 64 bits, Simple Core to Multi-core, ARM, … Speaking of the new Mac M1 and its 8gigs of RAM i get it a year ago change from an i7, 16 gigs of RAM with proper Video Card… to be honest I waited for a crazy bump in performance but I wasn’t impress at all !, even a bit disappointed. (Maybe the new M4 Mac mini 16gigs RAM would be a wow sensation)
The fact is if you want a beast computer it’s over 2000€ to really talk and not 2000€ spent 6 years ago. Computing has increased a lot (Not matter PC or MAC) back in the days you can say with 1500€ I can make myself a great computer I guess it’s no longer true.
If you use Kontakt with huge libraries you have to take 16gigs of RAM as the minimum. You have to think Apple change the way operating System is stored on the HardDrive (from Catalina) it takes more than 30gigs on the hard drive so the SSD minimum (at least for me) is 500go I can’t respect the 25% free disk space rule for the hard drive to run smoothly (and I don’t speak about system snapshots). Apple change also how the system operates with external devices and if in the past you can relocate a lot of stuff with symlinks it’s now a pain in the ass even if you lock your symlinks. It’s also more difficult to fake destinations with external hard drives naming…
SO optimizing ressources on a computer for music is becoming more and more a hassle. Because the tricks you find are also moving and become obsolete with hardware and software evolutions…
I would also say that software engineer seem to less optimise their codes, the amount of storage a software install takes, if you check Adobe Software, Affinity, Skylum, Ableton, Logic, iZotope (the worst I would say even if I like their products a MPS took over more than 10gigs on your storage for what 8 VSTs ???)
And also most of the installers (especially when it’s a bundle thing) won’t ask if you want a selective install, probably because there’s wrapper strategy and it’s depend on each and everyone software brand… And now with ara, clap, au, vst, vst3, aax and what not if you have 250gigs of ssd it can be really an issue, and forget about relocate your VST externally (except if the option is truly thinked by the software company Arturia did good on that) for instance Tone2, Native Instruments, XLN, Steinberg, Rob Papen are really sensitive, they don’t like relocate.
So today I think we have to take ressources handling as it is.
I’m not satisfied with my M1 mini, I will soon take the M4/16gigs/500 ssd maybe 1to if I can (I explain why in my little rant on Music and Computer ressources.
But SSD bump seriously the final price otherwise I would have it already.
But I didn’t notice CPU issue in a deal breaker manner not more than UHE DIVA in Divine mode… or Nowadays sessions (especially with a bunch of effect chain per tracks and busses). Also, don’t do stressful approach Tests > Do Real case scenario tests. I’m used to optimise my Ableton sessions. I go in a back and forth manner between a Midi session (backup) and an Audio session (in progress where I add up) every time I progress I bounce and put the midi in the Midi session. I can pickup any Midi track when I need to print automations in the arrangement step. I’m used to be anyway all print audio when mixing so.
i also create and arrange with Ableton + Reason Rack, but mixing under Logic Pro and so… I have extra steps to take care in the saving and organizing…
Not a big deal for me in the way I handle things.
NB: also I wouldn’t trust the CPU meter in Ableton. If you have audio artefacts then yeah it’s an issue. Another thing come to mind because none of my students go back and forth the buffer size they set it the minimum to avoid audio interruption and never go in preference again… That’s not what you should do. We record and play at the minimum buffer size and then we switch the buffer to help the CPU and continue our stuff… (I would love to set a shortcut or have a gui switch because yeah it’s a bit boring to go in the preferences so much…) and then to end up FREEZE and forget about it.
(seem to be less an issue on ARM but on INTEL still.)
From my Test on my M1 Mac mini 5 instances, no processing in Ableton Live, just 5 instances of plugins.
in Diva i have a Divine VA sound, but how many FX I have, can I use sample, multisample, granular, spectral ? No
in Serum2 did I have a Divine VA Sound, Can I do some crazy chords with the voicing, can I mix classic vintage components ? No
Do I need 5 instances of Diva in Divine or 5 instances of Serum 2 in High in Midi not frozen, at the same time. No
Does Pigments and Phase Plant perform any better ? no
And we talk only about 5 instances. If you expect recent computers to handle a mix and match of the Best VA emulation, the most Modern synthesis types, better Freq&rate without optimizing your ressources and make compromise. You will probably be disappointed…
Now if I handle my ressource better, i plug my soundcard, I set the freq&rate I need, I handle my buffer, I don’t adopt a stressful test but real case scenario, I have 1 instance of overlapping 5 notes chords for the pads (frozen), 1 instance of Diva playing the Bass (frozen) and 3 Serum 2 instances for the Lead the Pluck and the Arp :
I’m good to go. I will switch to Divine and High when I will bounce to audio. All the other stuff will be very fast in audio too. I can add my processing. I can even replace the synth effects if there is no effects automations involved into the sound design and busses those needs.
okay… so it’s not only me being annoyed by crackling sounds with all of those plugins… thanks for your post. Will read now ![]()
No. Just handle your ressources. ![]()
It’s worth reading the chapter “Computer Audio Resources and Strategies” in the Live manual.
CPU meter in Live takes into account only the load from processing audio, not other tasks the computer performs (e.g., managing Live’s user interface). So yes, in a sense you shouldn’t trust it as a means to determine the overall cpu load, but the way it works is pretty straight-forward and should be accurate.
mac mini m2 pro here, will report my findings soon when i get my hands on serum 2
the sfz import is irresistible to me lol
Replying to myself here figured out my woes related to the arp patches, clearly my ancient rig is underpowered for the amount of voices in use on some of them, cut the polyphony and they run much better… ![]()
wait a minute! that LFO is impossible! what the… care to share an audio example?
Right now, I have four instances of Serum 2 running in Live on my M4 Pro Mac mini (24 GB RAM). In each instance I have at least the three main oscillators running, and sub and noise in two of them. They all have anywhere between 3-8 effects running. I latched an arp in each with at least 3 notes in each instance. I am also using an effects Rack in Live for extra eq and limiters. According to Live, I am using 52-54% of my resources, but when I look at Activity Monitor, Live is using between 6.1% and 6.5%, so I am not sure what Live is registering–maybe some other kind of limit on resources. Anyway, the M4 Pro Mac mini could seemingly handle another 20 instances of Serum running simultaneously and barely break a sweat.
Activity monitor does show that Live is using all of the Performance Cores (I only have Live, Activity Monitor and Safari running right now).
On 626 presets I suggest to delete those 11 presets and we’re good to go. ![]()
INST - Acoustic Modeled Waterphorium
INST - Doors To The Secret Garden
KY - Delicacy
LOOP - Balafon Dreams
LOOP - Downtempo Cave
ORCH - Choir And Strings
PD - Early Morning Days
PD - Ether
PD - Granular Ambiance
PD -Replicant
STR - Pedal Chords
More seriously that’s the most CPU intensive presets
(at least for me)
the software I talk about earlier in this thread… with the video on how to do it with bitwig but the idea is similar for any other conversions
https://www.mossgrabers.de/Software/ConvertWithMoss/ConvertWithMoss.html
He had some really useful tipps - i.e. using the note number to modulate the release, to have a longer release on higher note on his cello/pad patch. I wish we had a similar modulation source on the Elektrons. (Ok A4 might have it, with keymap -but i think digitone could profit a lot from this.)
Just been playing with the spectral sampler/oscillator. Need to watch your ears a bit as some of the warp modes resulted in some very sharp frequency spikes (I already have tinnitus, need to watch myself) but it’s capable of some really mad atmospheres and pads.
Turned the scan rate down really low on one of the factory drone samples (like 2%) and then added a load of reverb and ended up heading into some really unusual sci-fi soundtrack territory.
Watch your volume, some of this is a bit “sharp”
What an update! For my ITB production, I use Ableton Suite (and its devices) with a healthy dose of Serum. Sure, sometimes I pull in plugs like Little Alter Boy and such, but I can pretty much guarantee that anything I’m producing these days is going to use Serum quite a bit.
So this update makes it even MORE of a “go to” for me. I’m already finding more “out there” moving sounds with Serum 2. The inclusion of the new oscillators, filters, effects routing, clips, arpeggiator and more make it much more of a self-contained “bread and butter” synth for all KINDS of purposes so far beyond wavetable stuff.
Plus a multi-sampling oscillator that loads standard .sfz files? Are you kidding me?
![]()
Plus a multi-sampling oscillator that loads standard .sfz files?
Yeah I was also like what? when I saw that, because the single sampler is cool but for hybrid electronic/acoustic sounds this is so cool to be able to get an acoustic layers that sounds nice across the keyboards and behave also nice between timbre and velocity.
Like a Afro House Pluck with real muted guitars or Tonal Percussions ![]()
This is super dope !!! especially if you are more into the hybrid acoustic vibe of Afro House sounds mixture than integrate Melodic House/Techno vibes of Electronic sounds… it’s like on Sound selection there’s those two very distinctive side (and personal taste is what it is but the melodic h/t is not what I like the most I’m more towards carl Craig/armonica/fnx omar vibes than Anima/Bodzin stuff if that make sense for those who know a bit Afro House not the popular tunes but the more underground club side of it… like Hyenah) it’s amazing when you dig into a modern genre very deep how there’s excellent stuff in every of those genre. for some tracks it’s remembering some tribal/ritual African vibes into the house music of the sound factory NYC… so I’m totally digging it… anyway I digress ![]()
on M4 max here, picked random multisample presets and high cpu pad presets. I was still getting okay performance until 80 and took 85 to crash ableton, and my buffer size was 64 samples lmao. I definitely paid to much for this laptop, but its kinda worth it with how much fun im having with serum 2
Lol, I have an M4 Max as well but never really stress tested it in that way. Good to know that 85 Serums is the limit ![]()
lol thats nuts, i need a new m4 ![]()
appreciate you pushing it to the limit





