Which host?

[hitman8081]

that, and that they work with alot of hardware and samples. it seems like gol expects people to buy a hot new pc for each version of fl studio.

That's strange reasoning. Gol/FL isn't responsible for your high CPU usage. Blame your [use of] plugins.

Freeze only helps so much. Even when implemented really elegantly it will still affect workflow somewhat. The best solution is simply trying to cut down on CPU use. Stick with plugs that are easy on the CPU.

i SERIOUSLY beg to differ dude. when i tryed ableton, i had a song at almost full cpu and lowered it to almost nothing after everytrack was frozen

I know there are some 3rd party freeze solutions. have you looked into those? They basically take the host out of the equation and are wrappers to the plugs you'd use. It might be worth trying?

i heard about these. suposidly there horrible. with 20 min freezes common as well as huge data storages.

[grymmjack]

Freeze is a solution that just isn't going to work the way most people want it in FL anytime soon.

"smart disable" on generators is an option in 6.4 beta and it works pretty good. smart disable on effects and generators really helps minimize CPU usage.

I think it's only fair at this point to get some specs from hitman about his hardware.

[grymmjack]

hitman, dude if you're using exclusively synth1 and maxing out your box, i gotta ask what kind of box you own.

While I can empathize with your plight, I cannot say I agree that it is fair to blame the programmer. His product excels at what it is designed for. Maybe because I program I can also empathize with gol/IL that coding a feature like freeze (which is a significant amount of work I bet given the way FL handles routing) is somewhat a waste of time because the user can already achieve the functionality and end-result of freeze with their own footwork. It's difficult to justify either side I think -- in some ways you could argue that gol does add workflow features that duplicate/enhance functionality that was already there before the workflow feature was implemented (a good example is the toolbar button for One-click audio recording in v6.4 beta -- this existed before on the Tools menu only I believe - the thing is that's much less difficult I'd bet since it's just pointing to the same functional code and the only modification was the creation of the UI parts to handle the button calling the existing code). Freeze is not so trivial.

I realize thinking from the perspective of the developer shouldn't be a users concern too, so don't go there

hi. i own a pentium 4 2.8 gig. before i was only using synth 1 and even though it was the only synth i was using, the maxing out the box came from, tons of effects, in a chain. from comp-eq-dimension-send ect. i thought to myself, out of cpu using synth 1??? but i use a comp-eq and slur of effects on every channel. then very recently i spent almost a week reviewing the quality/usabilty of every freeware synth and effect. anyway after, i added alot more synths that definitly arent low cpu. like hg fortune, angular monentum, bk synthslabs inro ect. thats where all these problems started.

when i compose in fl studio i tend to use a one instrument/one track kind of deal. the problem is, after exporting everything to wave. i need to 'hear' my track and how everything sounds together. i cant piece together solo tracks. im a good producer but im limited to the host i can use due to how i need to 'hear' my music in realtime. its also nay impossible to think pattern based. i do well with having one track set for each. intro, bridge, verse, chorus, breakdown, and extro.

so thats why im having alot of dificultys with fls manual freze.

Ok. Well do you create the songs piece by piece, you mention the intro, bridge, verse, chorus, breakdown and extra, then you say that it's hard to think pattern based, when essentially those are patterns. I'm thinking maybe a good solution would be to work iteratively if you don't already; create your intro, render it. Put it in a separate flp - call it songname_intro.flp, then file | save as | songname_verse, put rendered intro in as audio track, and use playlist to compose around that. after done with verse, render again. repeat the process for each 'segment'. After all is iteratively composed and rendered, create a new flp that's just used for arrangment of the rendered chunks...

You have more flexibility for arranging when you dont limit yourself. ie; it's easy to drag a 'chorus.wav' around on the playlist isn't it? it's easy to drag one object, so if you align your workflow WITH the way it was designed to work it wont seem so difficult.

If you need to make changes, go back to the segment (intro.flp, bridge.flp, verse.flp, choruse.flp, breakdown.flp, extra.flp) and make your changes and rerender and reimport into the arragment flp. If you have transitions that need to happen between segments (and we all do) you have to think creatively how to do it. You could isolate just the parts that transition from the segment renders ie; if you need a bridge to have some lead lines delay pulses fade out while the verse starts playing, then isolate that and render it separately and simply import the rendered stuff into the arrangement.

I think part of the reason we get wrapped up in the workflow is because we are so f**king spoiled these days. I mean seriously we are spoiled, I come from tracking era where we were so limited and stuff it was a real chore to make music, to get a delays pulses to fade out over a verse we literally had to copy notes and pattern data and overlap tracks. There was no drag and drop visual alignment, no snap to grid, and the largest sample could be like 100k (and that was pushing it ALOT). Look again at what you have and realize it's potential by thinking outside of your own little box that you've created. WE are all guilty of this and I emphasize we because I fall into the lazy trap myself of wanting the tools to keep evolving and adjusting to MY own desires/needs/thoughts, but the real solution is to stop thinking like that and adjust the only thing I have total control of, myself and the way I work, to the tool.

Does it make sense? I know it's not normal, and we all like to compose with monolithic simple never ending timelines with all the tracks and shit stacked up but it's as you say causing a problem in CPU and stuff. So you adjust your mind to work in a way that will succeed and will make you happy. FL can do anything you want it to but you have to be prepared to reinvent the way you work according to your hardware limitations.

I'd say that if you do solve this problem in FL which you already have mastered for the most part, you will be happier and better off. Iterative render-as-you-go and gather-and-compile arranging is not a bad thing. In fact, I find it quite fun. Once you start learning the way to do it you can easily discern which sections of your music needs to be solo rendered, and it's a deviation from creative process because it's logical thinking and not really spontaneous 'play' thinking, your compositions will probably improve because you will have to put some work in ahead of time and because of this it will motivate you to think things through more and make wiser decisions. Atleast it does for me. I have a P4/1.9ghz, btw. Audio streams are the lowest CPU of anything, and as long as you render the effects you can seriously get crazy complex with audio clips and not strangle your CPU. So you can have your cake and eat it too, it just takes some time to bake the cake

[hitman8081]

dude, i love the new beta, its great, but gol says its juts that a beta. he says its for testing and not production work. is the program actauly operational? good tips btw.

[grymmjack]

hitman; So far so good with the beta for me. I'm not writing any music in it that i want to keep or anything to be sure, but I mean it's worth testing. The beta works great for me. As good or better than 6.1 IMO. I had a lockup just a few ago with Discovery pegging CPU at 100% but it happened in Cantabile as well so I don't think it's FL's fault. It seems to be happening when cutoff is tweaked rapidly and resonance is maxed.

[Barbarossa]

I think part of the reason we get wrapped up in the workflow is because we are so f**king spoiled these days. I mean seriously we are spoiled, I come from tracking era where we were so limited and stuff it was a real chore to make music, to get a delays pulses to fade out over a verse we literally had to copy notes and pattern data and overlap tracks. There was no drag and drop visual alignment, no snap to grid, and the largest sample could be like 100k (and that was pushing it ALOT). Look again at what you have and realize it's potential by thinking outside of your own little box that you've created. WE are all guilty of this and I emphasize we because I fall into the lazy trap myself of wanting the tools to keep evolving and adjusting to MY own desires/needs/thoughts, but the real solution is to stop thinking like that and adjust the only thing I have total control of, myself and the way I work, to the tool.

This is a good argument, but IMO only one side of the coin. The other side is, that with todays technology it would be possible to realize a good workflow.
This is the fundamental difference.
But what we see, is not always using the technology to get the best results, but to increase sells.
An example: Sonar6. A great sequencer, sure. But then try to edit notes in a MIDI-clip, that is lying on other clips... Or try to lasso notes in the piano roll. But the users rave about audio-snap...
Or think about good and important features that were thrown out Cubase, to be introduced later for payment again.

The difference to the earlier times is, that the machines were engineered to get the work done in an efficient way and due to technical restrictions, it was complicated to get more sophisticated things done. First came the engineers, then the managers to get the visions of the engineers done. Today this is inverted: the engineers are used to fulfill the plans of marketing.
New features are always implemented, but the not so good fundamentals are icecold ignored.
They calculate, Sonar6 will not sell any better, if MIDI editing would be adequate for such a beast. But if you can show the masses a video with SessionDrummer2 it will sell much better...

So the enhancements we see, are mostly not implemented because of a vision of the developers, but not to fall too far behind. This results in more and more half-baked implementations, being sold as the real thing (think about Steinbergs "Freeze" which is absolutely useless for most people) or old problems, that never get solved. The development time for 50 new designed plugin-UIs has a much higher ROI.

[bullshark]

Reaper: Like the audio side of things but I have Vegas for that and I hate the way MIDI is handled; furthermore it doesn't look like MIDI will get looked at in the forseeable future.

I just stumbled in on the Cockos forum tonight, well, it looks like I might have to eat those words.

No problems, really, I can face it; anybody care to pass the salt?

[koolkeys]

Bullshark,
i feel with you. I'm in quite the same position (for many months already).
I'd suggest to ask dedicated questions about the workflow, instead using names for it ("freeze").
I.e. don't ask how to record in eXT, ask them in their forum, HOW they record. Ask about every mouse-click that is necessary.
Otherwise you will always receive most stupid answers like "use easy-on-the-CPU PlugIns".
You know where your problems are. So just ask how this certain task can be done and all the fanboys soon will shut up, when it comes to explain the workflow and it turns out to be a long list, while in other programs this can be done with 3 clicks.

I will say one thing. If you think freeze is there to help your HOST save on CPU, you are completely and utterly wrong. Freeze will NOT make your host faster.

The problem is the plugin developers. They are making more CPU intensive plugs, and those that use so much more RAM then they used to. THEY are the reason that freeze is needed. If you blame it on your host, you are placing blame in the wrong place, period. If you use the same plugs you've always used, but use a new host version, your CPU shouldn't rise much at all. If it does, it's most likely something you changed, unless a considerable amount has been added to the host.

Again, free does nothing for your host at all. It just helps you run larger plugins that are totally independent of the host. This is why people suggest running less CPU intense plugs. Because that is THE RIGHT WAY TO SAVE!!!! Any other blame is probably just finding a reason to gripe about the host.

That being said, if you are really that unhappy with your host, move on. It's not hard. And with more updates coming soon, such as EXT2 and Tracktion3, and the slew that have already been released, you don't have to sit and complain, you can just move on. Something will meet your needs. But if your current setup is bothering you to the point where you just aren't making music the way you once did, then maybe it's time to move on, even if it's only temporary. Find something that helps you create.

For me, Tracktion and EXT do everything I could ever want. I hope you all find the setup that inspires you.

Brent

[Sascha Franck]

Freeze only helps so much. Even when implemented really elegantly it will still affect workflow somewhat. The best solution is simply trying to cut down on CPU use. Stick with plugs that are easy on the CPU.

Well, unfortunately you're sort of right, with one exception: Samplitude. Someone showed me the freeze functionality briefly a while ago and my jaws are still dropped.
Basically it seem as if you'd just freeze and then move on arranging, cutting, copying and doing whetever to the frozen tracks. No other hosts allows for such a thing, so, let's say in Cubase or Logic, whenever you want to edit *anything* on a frozen track (and be it only moving things for a few bars), you need to un-freeze things and then re-freeze them. This is just as lousy as it could get. Heck, in Cubase (SX at least, no idea what they did in Cubase 4) you can't even mute a frozen track in your project page but need to open the mixer first.

Anyways, as far as the OPs host journey goes, I kinda feel with it. When Emagic shafted their Windows users, I started looking around for a proper alternative sequencer for Windows.
And while I love eXT to death as a modular sub-host (it beats anything, really, especially with those additional MIDI and audio routing plugins from Nicfit and the likes), I really didn't like the sequencing workflow. Starts with no proper dual monitor support (which I just need), non-configurable keycommands (it's driving me mad on the laptop) and continues basically the same way as bullshark put it. Yeah, you *can* (and *should*) adopt to the eXT way of doing things, but sometimes things just don't "click" with ones brain. Which, in my case, is really sort of weird, because the modular stuff in eXT almost instantly "clicked" with me.

I will obviously watch eXTs development, but in the meantime I got so sick of the hosts I tried (I even own Cubase...) that I took the plunge and bought a Macbook. Defenitely not an alternative for most people. But I just had to do so. And I'm rather happy about the decision.

[billybk1]

Basically it seem as if you'd just freeze and then move on arranging, cutting, copying and doing whetever to the frozen tracks. No other hosts allows for such a thing, so, let's say in Cubase or Logic, whenever you want to edit *anything* on a frozen track (and be it only moving things for a few bars), you need to un-freeze things and then re-freeze them. This is just as lousy as it could get. Heck, in Cubase (SX at least, no idea what they did in Cubase 4) you can't even mute a frozen track in your project page but need to open the mixer first.

I believe SONAR 6 allows you edit & re-arrange frozen tracks.

[Barbarossa]

I just stumbled in on the Cockos forum tonight, well, it looks like I might have to eat those words.

No problems, really, I can face it; anybody care to pass the salt?

Maybe it is of interest for you: Programming god Justin already released a new version he did during his holidays (!) where MIDI-data are integrated into the project. He also states, that he has lot more in the pipeline for MIDI.


Sascha,
you mentioned Samplitude as a good example for freeze and that it were the only host allowing for that. Well, i'm on the search of a host and yesterday i did my first closer look at Live6.
You think the freeze of Sam is the best? Then you don't know Live6's not yet.

You know, Live offers besides the traditional view of a sequencer, a so called SessionView. There audio-tracks and MIDI-tracks are represented vertically with free slots, capable to hold the clips. There's no timeline in the session view. The clips you click, are playing back as loops. Different clips can be loaded in the slots of such vertical tracks, but only one of them is always played back. Therefore it can easily be tested, how the different clips/variations work with the rest. The performances in the SessionView can be recorded into the normal arrange-view. So far so great for building arrangements that made Live quite famous.
But a MIDI-track holds not only the MIDI-clips but can also hold the VSTi- with all FXs in the case of native sound generation.

And now comes Live6's incredible Freeze into play, that lifts it even above Samplitude: you can freeze the MIDI-clip (which means the whole unlimited audio-generation chain) and in the SessionView then you can drag and drop it simply from the slot in the MIDI-track on a slot in an audio-track. And know what? This clip behaves like a normal audio clip!
So in Live6 you not only can work with the freezed material like with any audio-clip, like in Samplitude, you even can use it in the SessionView to test and build arrangements with the ease Live offers! Incredible.

[bullshark]

I just stumbled in on the Cockos forum tonight, well, it looks like I might have to eat those words.

No problems, really, I can face it; anybody care to pass the salt?

Maybe it is of interest for you: Programming god Justin already released a new version he did during his holidays (!) where MIDI-data are integrated into the project. He also states, that he has lot more in the pipeline for MIDI.

Yes, that's the post I stumbled upon. Looking good for the future.

You know, Live offers besides the traditional view of a sequencer, a so called SessionView. There audio-tracks and MIDI-tracks are represented vertically with free slots, capable to hold the clips. There's no timeline in the session view. The clips you click, are playing back as loops. Different clips can be loaded in the slots of such vertical tracks, but only one of them is always played back. Therefore it can easily be tested, how the different clips/variations work with the rest. The performances in the SessionView can be recorded into the normal arrange-view. So far so great for building arrangements that made Live quite famous.

That's the pattern paradigm which I find impossible to work with. I always end up trying to do impossible math equation when using those, like, for example, trying to have a 2 bars drum sequence with a 3 bars bass pattern finish at the same time as, say, an 11 bars melody. Then I try to make a one bar variation to the drums while the melody keeps playing...I usually end up banging my head against the wall; don't mind the bruises so much but the plaster repair cost are killing me

[Barbarossa]

Hell, what kind of ugly music are you doing?
Again i have to admit to you, that also this second suggestion from me was nothing good for you.

[bullshark]

Hell, what kind of ugly music are you doing?

You don't want to know, I even scare myself with it sometime

More seriously thought, I made an extremely simple song for this month KVR contest; VERY simple: a drumkit, a bass, a guitar soundfont and an e piano. I couldn't even fit that into a patern paradigm (I have Live Lite, remember?). I mean, the melody extend for about the lenght of the song but the bass is a two bars patern BUT there are 2 bars in the song that are not the same; same with the drums, 4 bars patern BUT with slight variations...and I won't even get into the chord progression (e piano), I get headache just thinking about trying to fit that into a patern.

But kudos to those who can.

[Sascha Franck]

Yeah, for me pattern based hosts usually don't cut it either (even if I find that very Live featureset you mentioned, Barbarossa, interesting).
Another huge problem for most of them being simple things such as pickups. They simply don't fit into patterns usually dealing with full stop bars (or any number of them). Maybe that's why we don't hear pickups (or anticipated kicks and stuff) often in dance music?