Serato Sample now with STEMS & VST3!!! (Maschine tutorial video)

Comments

  • Kymeia
    Kymeia Member Posts: 2,065 Expert

    Can it record samples itself?

  • AdelV
    AdelV Member Posts: 122 Advisor
  • Kymeia
    Kymeia Member Posts: 2,065 Expert
    edited August 24

    That is unfortunate - that was my only reason for not buying it last time it was on sale; I was hoping they would have added direct sampling by now.

  • D-One
    D-One Moderator Posts: 1,818 mod

    Been trying the beta for a couple of days and it's very cool. This will save me so much time as I used to do it separately in dedicated stem separation applications.

  • 6xes
    6xes Member Posts: 382 Pro
    edited August 24

    A question i would consider when using Serato 2.0 is its memory usage when in use as a VST3, as the problem i would anticipate, is that when the initial sample is dropped in the VST3, after the stems separation is completed..that all 4 stems remains resident in memory.... which in smaller memory configured setups this will be problematic, and needs to be seriously considered

    if however the sample dropped in the VST3 and the stems separation is saved to SSD for retrieval it may not be such a problem! this however is what needs to be checked and understood

    The comparison i make is within Virtual DJ, where the stems2.0 can be prepared vs on the fly stems2.0(unprepared) has a big increase in memory usage, and should be considered before purchasing.

    Prepared VDJstems2.0(prepared)being the better memory option

    if using a m1 8gb(ram) 256ssd configuration... you may notice that you will be pushed into swap memory quickly enough.. which is not a ideal position to be in

  • D-One
    D-One Moderator Posts: 1,818 mod
    edited August 24

    It seems to store things in memory, hard choice between that and disk storage, not sure I'd want all my MP3/Flac to take 4 times the space on my SSD, actually way more than that because I think it stores it as .wav / uncompressed... 10GB of samples/songs would become a problem, a 80GB problem, maybe more? Not sure... VDJ might store it in a compressed format maybe to avoid that, no ideia.

    The first instance seems to make Maschine or Ableton jump from 300/600MB of usage from an empty project to +-1.5GB, after that each new instance takes about 200/400MB. I tested with long songs (7min+) so if I do the math of 4 .wavs of that length it seems to add up.

    This is acceptable to me, as long as you don't need a crazy amount of instances with long songs loaded. With 3/4 long songs loaded, I am at less than 3GB of usage. I guess if you have a low-end computer you could crop your samples before importing.

  • 6xes
    6xes Member Posts: 382 Pro
    edited August 24

    thanks for testing @D-One

    when using a section of a stem albeit a part within the stem separation files and using serato sample...

    the simple reality is, we are only utilising small sections of the stem...(seems a costly exercise when VDJ is free)

    and although the luxury of putting cue points into place is there(also doable in VDJ).... when it comes to saving a project and reloading it... Serato2.0 inefficient use of memory, makes me wonder if sometimes its better to just sample out the sections you require and then take serato2.0 out of the plugin chain...

    perhaps it seems like its great for experimentation purposes...beyond that, i personally see some constraints being put on a systems memory when the integrated use of multiple audio-applications and such VST3 are involved(realtime context)....

    i will continue using VDJ-free for any and all stem separation purposes.. as prepared stems saved on the SSD will be less demanding of memory & then i can route audio directly into Maschine or what ever Daw.. and record the section i require..

    knowing these workarounds will save on system resources&money.. and have people understand the shiny new thing is not always the best thing moving forward!

  • 6xes
    6xes Member Posts: 382 Pro
    edited August 24

    @D-One

    another consideration...

    when you mention SSD storage... and comparing the price of Serat2.0 vs SSD space... which is the more cost effective method??, when you realise VDJ is free and does exactly the same operation as Serato... with better stem separation

    The pro's & cons i believe sit squarely with me... going on the cheap route(as i own serato sample1.x)... but hey... if people got money to throw around willy nilly.. i suppose its there perogative LOL

  • D-One
    D-One Moderator Posts: 1,818 mod
    edited August 24

    TBH I am just happy this exists and is successful, maybe other companies making samplers will open their eyes.


    @6xes :

    when using a section of a stem albeit a part within the stem separation files and using serato sample...

    the simple reality is, we are only utilising small sections of the stem...(seems a costly exercise when VDJ is free)

    Depends on what you use it for, Serato Sample for me is about convenience and speed while inside my production software, resampling a DJ software is neither (I assume that's what you mean when mentioning VDJ as an alternative), I also need to work non-destructively as much as possible.

    I am curious as I don't quite understand this workflow of yours, how can you modify ADSR, reversing, and all the tons of other things Samplers do but DJ software generally do not? (especially per chop/cue).

    Nothing beats free but Serato Sample cost me 50$ on sale on pluginboutique.com, I'd never pay 150 tho..


    when it comes to saving a project and reloading it... Serato2.0 inefficient use of memory, makes me wonder if sometimes its better to just sample out the sections you require and then take serato2.0 out of the plugin chain...

    Well, how would you reload your DAW/Maschine project when the Sample slices/cues are in another software?

    If you have no issues committing you can just resample and get rid of the plugin in Maschine, or bounce/freeze in a DAW, same stuff you do with big 5/10GB Kontakt instruments. Personally, I never needed such, not even when I had 16GB of RAM... But I have trouble committing... If a beat gets picked up and will be released I go back and search for more samples in the same song, adjust slices, create new sections with the same or more chops, etc..

    So far I did my stem separation separately from the production stage, lately I have been using Ultimate Vocal Remover which is free and unlike what the name implies it also does 4 stems too, the best of all is you can install a bunch of fancy modern algorithms at the press of a button and pick the ones you like best - These algorithms should be way better than things designed for DJ'ing, sometimes I do it to full tracks sometimes just small sections I want to sample then bring it to Maschine/DAW, this is my preferred way of working, I'll use Serato Sample mainly for speed, just to get things going and with serious beats that will get released I'll just re-do the sampling for best quality. This workflow might suit you I guess as an alternative to VDJ.

    How much RAM do you have to be so concerned about it? 4GB or something?

  • Schmapps1
    Schmapps1 Member Posts: 107 Advisor

    I think having Serato Sample available as a VST to use stem separation is huge.

    it’s not like it’s all that hard to just do a couple different passes of various stems, resample what you want, and then disable the VST if memory use is a big concern right

  • D-One
    D-One Moderator Posts: 1,818 mod
    edited August 25

    Humm but it doesn't do the same exact operation at all... triggering cue points does not equal an actual Sampler. Serato Sampler is really simple as far as samplers go, sure, but it does more than just trigger cues. Unless you talking about some sort of Sampler built into VDJ that offers functionality above the usual DJing ones?

    Well... SDD's are arguably cheaper if that's what you're getting at but I never reached a RAM limit making music but I keep buying SSD's every few years. I bought a 2TB NVME SSD recently for 130$ (lower quality controller) it's the same price that my 32GB RAM kit costs now, although I paid a lot more at the time.

  • 6xes
    6xes Member Posts: 382 Pro
    edited August 25

    @D-One

    as you so allude to, ....alot of it depends on the practical application of your workflow & how much internal memory you have when your project is in the near complete stage...

    this is going to bring you to a point where you are being forced to freeze tracks... becos of where Serato will store its stems(in memory).

    This memory constraint coupled with the added CPU usuage of the VST will affect more parts of the Daw then people will realise... at which point, your hand will be forced into freezing or sampling the small portions within Serato2.0 that you actually use....

    This is something people wont anticipate when using it in the experimental stage of being wow'ed by the new shiny toy... but the simple realities of system-resources will come to the fore when used in the creation phase&moving forward into the completion stage, more noticeable on smaller memory configurations

  • 6xes
    6xes Member Posts: 382 Pro
    edited August 25

    @D-One

    does the Serato sample2.0 allow you to utilise separate midi channel to trigger the cue-points...

    for instance if you put 2 instances of serato on separate tracks... can you trigger via midi independant cue-points on the separated tracks/instances??

    becos last i used serato sample 1.x

    each cue would be triggered by the same midi channel... thus independant cue triggering on midi channels cannot be utilised.. *shrugs (unless you share the sample slots)

  • 6xes
    6xes Member Posts: 382 Pro

    Dont get me wrong...

    the memory considerations is valid if for instance you have a 4gb or even a 8gb PC setup, and although my somewhat critical view of memory usage is plain to see...

    im not at a disadvantage, as i have Serato2.0 access as a previous purchaser of Seratosample 1.x

    it is something people just need to understand when using such plugins!

    cheeRs

  • Percivale
    Percivale Member Posts: 166 Advisor
    edited August 25

    Imagine creating live remixing templates (think age-old MIDI packs being sold) which would (based on settings selectable from hardware), apply intro, verse, bridge and chorus with decent AI control on voice/melody/bass/drum. Assisted performance making if you would like it, i.e. to free up the performer to curate what's incoming/next, respond to the need of the played-to crowd at spur of moment and do a live play over impromptu etc.

    Add random, intensity (think Maschine variation control), bpm range, key sync/range and genre feel all selectable from hardware. You could almost programme out a 1 hour performance and do the DJ thingy.

Back To Top