@dendy Efficiency Post on Audiobus Forum

I very rarely visit the AudioBus forum any more but I was sitting here waiting for a video file to export and so I took a peek. LOL - the response to your post @dendy is the exact reason I don't go there much any more. It's just so toxic. If anyone dares to say anything good about something it seems the vultures are just waiting to swoop down and tear it apart! Of course I guess that's true of most message boards and forums these days. I stopped going to the forum of one particular app with a week or two because of the generally unwelcoming and passive-aggressive atmosphere. You're a brave man @dendy !

FWIW I rarely bother with any app other than NS2 these days. Sometimes I'll use certain apps create a loop or sample to use in Slate or Obsidian but that's about it. I just want my music experience (when I have the time) to be about making some music and not about having something go bonkers when I reach that point that it all falls apart. Without naming any app specifically, I have to say that most of them use a lot of CPU and don't honestly offer anything so unique that I can't get it within NS2. Plus NS2 is much more stable and efficient on my old Air 1 than any of them.

I like that this forum remains pretty civil and helpful. I hope it stays that way!

«1

Comments

  • Kudos to @dendy for trying to be helpful on that forum. And this one!!! 😊

    I agree with @anickt , I don’t have much time for music so I don’t want to waste any of it arguing with people who are more interested in arguing than creating. The NS2 forum fosters discussion of features & possible future features without the worry that you might be offending a ‘soldier’ of another DAW app.

  • i enjoyed that test i was curious, to be honest i didn't expected that difference will be so big... And Gadged can handle (in Vancouver) just SIX instances playing 16 notes poly (with filter on eqch channel) on iPad Air 1 without crackling !!!!

  • It's is totally out of any scale how NS is efficient, i can't stop wonderimg about that... i tried run 64 approx 10
    sec long reverbs on my Air 1 - on "high"
    buffer of course - and i ended with just 66% cpu load !!! That's true magic

  • T4HT4H
    edited March 3

    I’ve never seen efficiency so dissed! I don’t know what planet these other people are on, or what their backgrounds are...but having the ability to run tons tracks and vst instances with no glitching at low latencies has been the goal of digital producers in my neck of the woods for years, and many $1000’s have been spent chasing that goal. For instance my Mac Pro has 128gb of memory, and I store my vst libraries on 8 ssd’s. This is not a game for us! Having it for $30 is a dream come true and is opening doors that have been locked behind huge paywalls

  • Looks like the thread has been deleted on the AB forums.

  • edited March 3

    I also left the audiobus forum because of similar topics.
    NS2 really seems super optimized. But to be fair i can run f.e. 100s of stock synths and 1000s voices and 100s of FX inserts with Logic f.e. on my 5 years old macbook...even 100 instances of DRC for mac while maybe 3-4 on my iPhone.
    Even high end reverbs like Breeze 2 are so efficient that i can run 100 on this machine.
    But in the iOS world it is indeed something special maybe and Matt really took the time to get it to that great level. Stock synths and FX in DAWs are mostly very efficient and often really underrated.
    But of course even within a synth presets can be 100 or more time more cpu taxing.....sometimes worth it for specific sounds but sometimes really a waste of resources....like fighting in internet forums.
    Well, one forum less for me i have to deal with, means i am more efficient for the creative side of me :)

  • @richardyot said:
    Looks like the thread has been deleted on the AB forums.

    Ya I saw that. @dendy asked for it to be deleted and also said he's done with AB forums in the NS2 thread

  • @richardyot @T4H
    Looks like the thread has been deleted on the AB forums.
    Ya I saw that. @dendy asked for it to be deleted

    Yeah - when i saw that shit storm i realised that AB is not ready for objective exact comparisons of apps with "strong committed users base" :-) I asked for deleting that thread to avoid endless flamewar ..

    Plus i realised that AB forum is not good place for me. If i cannot be totally honest and if pure objective numbers works like trigger for shit storm of subjective negative opinions - that's not place for me. Even through there is also lot of nice people.

    @Cinebient
    But to be fair i can run f.e. 100s of stock synths and 1000s voices and 100s of FX inserts with Logic f.e. on my 5 years old macbook...even 100 instances of DRC for mac while maybe 3-4 on my iPhone.

    One of reasons is that DAW on desktop utilize full multicore power of CPU. DAW on iOS is runniong just on single core (host + all AU plugins are running in same audio thread).

  • I can get so many instances of Obsidian and Slate going with reverbs and delays on everything that it’s practically white noise and barely touch 50% CPU. I can’t do that with any other app or combination of apps. I find that amazing. Also, the people on this forum are helpful and enthusiastic. Don’t know why any of this would be a bad thing!

  • that it’s practically white noise

    Yeah, white noise is coool, it's basically all tracks ever made mixed together :))

    Like this:

    (put your volume down before you click it :)))

  • @dendy said:

    @Cinebient
    But to be fair i can run f.e. 100s of stock synths and 1000s voices and 100s of FX inserts with Logic f.e. on my 5 years old macbook...even 100 instances of DRC for mac while maybe 3-4 on my iPhone.

    One of reasons is that DAW on desktop utilize full multicore power of CPU. DAW on iOS is running just on single core (host + all AU plugins are running in same audio thread).

    Interesting. Does that mean that no iOS DAW can spread the load across different cores?
    Did Apple not say you can use all cores at the same time for applications?
    That would be a bummer for me. But if that means all the NS2 workload goes to one single core....then it is indeed even more impressive. That would also mean that single core performance is more important on iOS devices than the general multicore performance. So i guess my iPhone 6S plus is great for NS2 :)

  • @Cinebient
    Does that mean that no iOS DAW can spread the load across different cores?

    yeah exactly :) i was pretty much shocked about this fact .. simply host app in iOS is using realtime audio thread - and all AU plugins are running in same thread .. which basically means everything is running on single core. All other cores are sleeping :)

    So .. when choosing which iPad is good for audio apps - you should look just at single core performance :)

    From this point of view, 2018 PRO models are totally not worth the price, best iPad for music apps you can buy is 2018 normal iPad - basic model costs 340 EUR and single core geekbench score is 3500 ... 2018 PRO models have single core geekbench score around 4900 but lowest price for 11" model is 880 EUR ;-)

  • I like the AB forum very much, but some people will really take you to task for being less than scientific, which I think happened in this case. When I compare it to other music forums now and in the past, it’s actually pretty civilized and downright friendly (and this is is coming from someone who, on this very forum, failed to be civil). I’m sorry that @dendy is no longer posting there, as he is an excellent resource for all NS2 and other miscellaneous iOS-related things. I come here less often...

  • edited March 4

    Yeah I don’t see where he was being less than scientific. I think it was pretty apples to apples on compairing the basic things that a sampler would do and nothing very specific to what an individual sampler might do. I think people just don’t like to see their stuff get pointed out as less efficient. A guy pointed out that Obsidian wasn’t any more efficient, dendy gave more data, that wasn’t enough, so he gave more, then it was a ...Plenty of built in synths are extremely efficient, Obsidian is not outstanding in that sense... which is lame. Of course it’s true, but Obsidian is over the top more efficient. Hands down. PLUS it plays samples. So as soon as it got compared to BM3’s efficiency it was a wait a minute, we can’t quantitatively compare these apples to these oranges, even though people do it left and right with everything that comes out... how does x compare to y when plugins come out. I don’t see how this was any different...until it got to be a BM3/NS2 thing.

    Then it was there is very little to warrant claims of Obsidian being more efficient than any other built in synth in various DAWs, it is marketing speak, Modstep built ins = Efficient, Cubasis built ins = efficient, so on and so on, and non of those are AUv3 either, which is what the OP asked for. when dendy asked at the beginning what host he used and that NS2 had a built in that was great. The OP even asked about it: Obsidian? Is it really that efficient? so he was asking about it. Dendy’s First example was real world...talking about it as in he’s using it in his own music and that he could use 20 instances. That doesn’t HAVE to be scientific, it’s simply him using it IRL.

    Basically, it always comes down to somebody not liking something else because it isn’t theirs.

    If I ever get enough time to finish this tune I’m working on, i’m going to post the project file here as well. I’m going to strip out the 4 iSymphonic string channels and replace them with Obsidian instances because i’m tired of iSymphonic blowing up every 2 min (literally). I’m pushing +60 channels with one of the synths being non-Obsidian (A Ravenscroft piano) and the other a Synthmaster, but i’ll Probably just replace that one with an Obsidian patch. At one point at the biggest buildup, I’m at 50% sustained for 5 seconds and it bumps to 53% on my 2017 iPad Pro. There is literally zero chance I could do that in BM3. I’ve tried. I’ve got a tune on my soundcloud called A Day Off - WIP 1 I did in BM3 and I’m proud of it, but it was a beating to get those 20 tracks to not bury my CPU. I had to disable stuff to get through it and then just render it the best I could and that was my mix. I was done jackin with it. I like BM3 for what it is, but efficient...it is not. And for my normal style of music, I need track count. It’s dense...layers and layers and I don’t want to render to audio to get CPU when...I don’t have to.

    I think the whole reason dendy offered up the Obsidian thing in the first place was because, if the OP is asking for low CPU AU’s, maybe he could just use an efficient host with Obsidian and that would be helpful... get the most bang for the OP’s buck with the hardware he had. Then it spiraled into your data doesn’t mean anything because your real world example is bogus and he only wants AU’s anyway.

    You have to take the good and the bad with every forum but the defensive nature of the BM3 crowd over there is tiring. I could see why dendy walked off.

  • Well, ok. It did get out of hand over there. Your remarks are well-taken. I thought Dendy’s post was more of an apples to oranges comparison, and some took exception to that. I can’t go back to that thread now, so I’ll take your word for it.

    Anyway, NS2 may be the most efficient host for AUV instances (non-Obsidian) - at least that is what I am finding in my own non-scientific experience - Cubasis and BM3 may be comparable. Maybe that would be a test worth trying for someone so inclined. You appear to be saying that Obsidian can pretty much cover everything (making this point moot), but there are AUVs that give me a particular sound that I do not find easily achievable with Obsidian. That’s not a knock at all, btw. I like what Obsidian does, and I use it plenty.

  • IMHO dendy did nothing wrong on that AB thread - he was actually quite helpful. dendy’s comparisons were spot on accurate as real world examples of what musicians do: make music with lots of tracks. Sampler tracks compared to sampler tracks sounds pretty common sense to me. Dendy also didn’t instigate anything in that AB forum thread. Just the facts. Dendy wrote nothing that suggested anything derogatory about BM3. There are just too many people on the internet who get super ‘tribal’ about the group they identify with so that if anything has even the smell of potentially pointing out anything that might be negative about thier favorite DAW, they turn super argumentative.

    AB Forum can be pretty nice, but I rarely go there due to some people trying to turn it into A(rguing) B(ullies) Forum. Sad really. We should be discussing how to use the strengths of various apps to work-around the weakness of others. You know. Productive conversation. Not petty school yard virtual shoving matches to beat down others who don’t agree which DAW is best. Which is best is subjective and we can afford more than one iOS DAW. Anyone who bought an iPhone or iPad can afford that. We should remain thankful we aren’t paying $300+ for desktop DAW.

  • Fun-fact: haven't even opened an au in it yet ;)

  • edited March 5

    @ALB
    I can’t go back to that thread now, so I’ll take your word for it.

    it was basically very simple.. same sample loop (10 sec long) loaded into sampler, with basic lowpass filter applied, playing 16 notes simultaneously. Then duplicating instances until i reached 80% cpu load. Very simple. Basically sampler performance test, which speaks about how efficient is sample playback and filtering in that given app. It tells you how far you can go during building song in that app, until you start reaching CPU limit (if you use just built in stuff).

    You can then extrapolate this performance test on all other parts of apps, in most cases it is very precise. Extrapolation is standard scientific method ;-)

  • edited March 5

    @SlapHappy said:
    ..... negative about thier favorite DAW, they turn super argumentative.

    !!! I didn’t know such a word existed!

    Actually, top to bottom describes my wife!
    I would even go a step further and call her “Argumentalist”.
    I seriously starting to wonder if she visit AB forums too...

    On topic, I believe there is not much to prove to anyone. NS is and will forever be the most efficient iOS daw ever. And no. Ain’t going at AB forum to state that....

    ... because my wife may lurk there. No, really.

  • 😁😁 It describes all wives. ...Eventually.

  • Wifes are alway right !!! It's seriously dangerous, sometimes even life threatening, to argue !!!

  • @dendy said:

    @ALB
    I can’t go back to that thread now, so I’ll take your word for it.

    it was basically very simple.. same sample loop (10 sec long) loaded into sampler, with basic lowpass filter applied, playing 16 notes simultaneously. Then duplicating instances until i reached 80% cpu load. Very simple. Basically sampler performance test, which speaks about how efficient is sample playback and filtering in that given app. It tells you how far you can go during building song in that app, until you start reaching CPU limit (if you use just built in stuff).

    You can then extrapolate this performance test on all other parts of apps, in most cases it is very precise. Extrapolation is standard scientific method ;-)

    Ok - understood. Sorry that you and others have had bad experiences with the AB forum. I think many valued your input there all the same.

  • T4HT4H
    edited March 5

    I have a copy of the whole thread captured before it was deleted that I could post so everyone here could have the proper context of the experiment as well as the tone of the posts. He specifically prefaced his post by saying it wasn’t an attack on anyone’s DAW etc. I also have the posted experiment files.

  • edited March 6

    So now we’re starting a foum war eh? Matt, you need better pr.

  • @jenkweb said:
    So now we’re starting a foum war eh? Matt, you need better pr.

    You discuss news on more than one forum, right? Should we not discuss NS2 on the audiob.us forum at all because we should only discuss it here?

    Welcome to the internet. We will discuss what we want.

  • edited March 6

    @drez @T4H

    Better not provoke certain people :-) It's nice place here. Full of constructive debates, lot of great music and other resources sharing, so keep it this way .. Do not give them reason to start flamewar again, and especially not here ...

    Btw. because Gadget users aren't so sensitive, i can afford to share other interesting test .. i tried this "sample playback with basic filtering" in Gadget and i was pretty much shocked - just SIX instances of Vancouver, each playing 16 notes of that 10 sec long sample !!! Just compare it to 44 Obsidian instances on same device .. unbelievable :-)

    Wanted to try it also in Cubasis but it behaves very strange - it looks like it starts drop voices to hold on CPU load under 50%. Even through i set "128 voices" poly in settings, it was like it is not definitely playung 128 voices, instead of it is dropping voices to hold on CPU load low .. Cubasis is strange in some details.

  • Hi - I've just joined this forum as I'm really excited to check out how NS2 develops. I might as well jump in on this thread first seeing as I was part of this discussion at ABF. A couple of things to say here. Firstly, I will reflect on how I interact online as I thought the criticism of that particular test was fair, especially given the context. However, the comments in this thread have given me pause for thought. If you know me from ABF then you'll know that my posts are pretty good natured and I will do my best to keep it that way going forward. Community is very important to me even in this warped online world so thank you for keeping me in check.

    @dendy this might sound harsh but I'm being honest here for the record, I did find your constant promoting of NS2 on ABF a little unbalanced and stifling. However, I also appreciated that you consistently tried to help everyone with NS2 questions which was good of you.

    The Audiobus forum is a brilliant place to stay a part of iOS music development and it is really buzzing right now. NS2 should be a big part of that discussion so I just want to say that Michael runs that forum with a view that everyone is welcome there and to my knowledge no-one gets asked to leave unless banned for moderation reasons. Sure, you need a thick skin as there can be strong opinions but any criticism should be constructive and if it turns personal or offensive then it generally gets moderated very quickly. iOS is already a small niche of the bigger music production scene so a respectful community open to all developers, apps and musicians will only help move things forward. Seeing the different developers collaborate on ABF is amazing and one of the joys of that forum.

    I'll probably lay low here as I'm really keen to see what comes next for NS2 but I was waiting for audio tracks to arrive before jumping in. @richardyot's drum packs are making me think I shouldn't wait any longer though...

    Anyway, have a blast making music and peace to you all.

    Gus

  • Welcome!
    Hopefully you’ll enjoy your stay and Dendy’s genuine, unprovocative, sincere, enthusiastic altruism.

    Took me half an hour to look at the dictionary and find the most elegant words to describe his character, but he deserves it. ;)

  • Thanks @Zupi - that's a nice character reference.

  • Welcome Gus!

    You wrote,
    “I did find your constant promoting of NS2 on ABF a little unbalanced and stifling.”

    I couldn’t disagree more. The NS2 thread on AB forum was all about NS2, so “unbalanced”...?
    I don’t think so. It should be ALL about NS2, and not about any other app. A lot of us are enthusiastic about NS2 because it works really well for our particular workflow. Sharing that enthusiasm in helpful posts explaining what NS2 can do did not come across as “stifling” to me. Perhaps that’s just because I am a NS2 enthusiast. I think that dendy’s effort in AB forum was however futile. Some people just don’t want to hear it. He tried to help and had to put up with a lot of crap for his effort.

    There is a lot of ‘destructive’ criticism on that forum. That’s fine. Some people like to act a little more harsh online than they would in real life to blow a little steam off. There are certainly far worse forums online. There are a lot of good people on the AB forum, but there are enough negative posts that I am always leary that I’ll step in some shit with anything I might post, no matter how unintentional. The tone in a lot of threads (not all for sure) prevented me from getting very involved. When I did it was only ever relating to NS1 or NS2. Thinking I was being helpful, then acussed of the opposite - well that’s not fun. Why bother? I don’t care what random people on the internet feel is the best DAW app and I am not getting into arguments over it. It is just easier not to visit AB forum, unless there is some new app I might want to read about, maybe, but no posting. It’s not worth it.

    Since this forum is really only about one app, more of the people in this forum are here to learn about the app and share their positive experiences. Much less combative. That might lead some on the other forums to accuse us of being ‘NS2 fanboys’, but I think that is a bit childish. Hopefully the spirit here won’t put you off. That is, you generally don’t need thick skin here. I feel like you would be just the sort of civil chap that is affable enough to blend right in. Constructive criticism is welcome, but hopefully is balanced. There were some from the other forums that came here initially after launch with nothing but negative posts, and that is hardly constructive. There will probably always be a few bad apples, but most of them seem to be bored with this forum now and don’t come around much any more.

Sign In or Register to comment.