i think it's on the high end of things to consider, i mean, how can you record, mix, and master what you cannot even hear if your room is poorly treated or not at all? putting egg cartons on the wall results on a dead high and midrange and an acoustically lifeless room, the exact opposite of what you want if you're doing any sort of recording

I think you're confusing how you'd treat a live room and how you'd treat the control room :/ (or the matter of taste in treating a live room)

In the control room you want the least refelections possible and diffusion helps by breaking up the high end reflections and reducing flutter echoes. Something you definitely want if you're mixing, and also if you want to record in a small room with loud things like drums.

But a lot of live rooms are intentionally dead for a reason. So you can add your own 'life' later. Some rooms are the opposite. Some of the best drums recordings are a good kit in a big wooden room.

So this:

[QUOTE]i think it's on the high end of things to consider, i mean, how can you record, mix, and master what you cannot even hear if your room is poorly treated or not at all? putting egg cartons on the wall results on a dead high and midrange and an acoustically lifeless room, the exact opposite of what you want if you're doing any sort of recording[/QUOTE]
is kinda wrong. It depends on what you want.

Egg cartons are just diffusers, splaying reflections to help parallell walls not have endless echoes back and forth. Bookshelfs with books do the same.

A control/mix room needs to be as dead as possible, it's really definition of treating the room, without getting into intense mathematics.

A dead control room or mix room is the wrong way to do things, you want the room to sound as natural as possible as in this is is an accurate representation of the sound wave and its frequencies. A dead room is an unnatural sound that would only be acceptable in an incredibly small studio or a vocal booth, where you would add reverb and ambiance later.

[quote=Convectuoso;18170492]In the control room you want the least refelections possible and diffusion helps by breaking up the high end reflections and reducing flutter echoes. Something you definitely want if you're mixing, and also if you want to record in a small room with loud things like drums.

...

A control/mix room needs to be as dead as possible, it's really definition of treating the room, without getting into intense mathematics.[/quote]


[quote=JoshIsNumber3;18170565]A dead control room or mix room is the wrong way to do things, you want the room to sound as natural as possible as in this is is an accurate representation of the sound wave and its frequencies. A dead room is an unnatural sound that would only be acceptable in an incredibly small studio or a vocal booth, where you would add reverb and ambiance later.[/quote]


These are both equally valid perspectives, and in general the contemporary idea of what is best tends to switch back and forth with the trends of the times. I believe the current trend is for a more natural sounding control room. I think there's something to be said about considering both in a home environment. Again, in general, it's easier to make a room dead-sounding than it is to make it live-sounding in a good way (especially if the room is small).


[quote=Convectuoso;18170492]Egg cartons are just diffusers, splaying reflections to help parallell walls not have endless echoes back and forth. Bookshelfs with books do the same.[/quote]


One other thing to keep in mind is how much better a bookshelf will perform. Egg Cartons tend not to have much effect because they have very little mass, which is one of the only things that can really alter how sound waves propogate (vacuums, resonators and wave guides are the only others I can think of, and they're all way harder to work with than pure mass).

Yeah on reflection and research, egg cartons are on the verge of pointless.


But lol idk how they even came up.


I dunno. With the amount of guesswork that goes with a live sounding room, I'd rather have a roomI only really have to worry about the response of the speakers.

IMO.

Room + monitor speakers (and all variables of these; position etc.) is the most important part of mixing. You could have Pro Tools HD and all the best gear, but if you can't monitor properly, it's all wasted.

I have learnt how to compensate when mixing with my AKG K701s, BUT i always check my mixes in the University studio (which is well-treated) and all other playback systems I can think of. I know of people who can mix just as well on headphones as they can on monitors, as they know they're headphones well enough to compensate for the generally poor low frequency response and generally poor accuracy of spatial qualities. I'm not at that point yet, but I'm getting there. Like I say, I always check back on good monitors in a good room before completing any projects.

[quote=Convectuoso;18171005]Yeah on reflection and research, egg cartons are on the verge of pointless.


But lol idk how they even came up.


I dunno. With the amount of guesswork that goes with a live sounding room, I'd rather have a roomI only really have to worry about the response of the speakers.

IMO.[/quote]

One thing I've always been curious about, but haven't ever tried (too lazy, lack of opportunity, etc) would be taking egg cartons and filling them with some kind of cheap putty or concrete or something similar. Because the shape does suggest a decent amount of diffusion/dispersion if there was any mass to fill it.

It'd still result in an uneven room. Egg cartons cannot 'control' low frequencies or even the low-mids, where most fundamentals lie

[QUOTE=Moseph;18168733]I was in a group that designed a dynamically-controlled EQ on a fixed-point DSP processor in school. In the research for that, we had to look into determining which features should be handled by the interrupt cycles on the chip. In researching for that, we came across the ASIO spec. I don't remember much about the details, but at a high-level of abstraction I did take away that this was one of the key considerations. We did a little bit of digging to see about CoreAudio (we were curious) but found nothing that gave hard technical data. However, based on speculation (not just ours, but we talked about it in Office Hours briefly as well) that most low-latency audio drivers would [I]need[/I] to do this to behave without glitchy behavior.


Anyway, that "most primitive" stuff, for the most part, is negligible considering the calculating speeds of modern processors. It's things like making sure the screen is refreshed and checking overflow status of buffers. Things that would generally result in a computer turning into an expensive paper-weight if they didn't happen. The vast majority of OS-based "behind the scenes" operations don't actively work in the interrupt cycles.




Raw pedagogy, basically. More pertinently, you're giving decent advice based on bad information, which in itself is bad. There's no gray area here to me: native processing does not suffer from quality issues (in fact, you might be able to argue the converse point based on fixed/floating-point accuracy).

The RAM issue is also, for the most part, pretty negligible for most pure effects, since they generally aren't holding onto a lot of past data. Obviously there will be exceptions to this (e.g., anything that uses convolution). This also can't be stated for synthesizers that use large wave-tables or lots of samples, since the RAM needs to hold onto a lot of raw data for quick access. The rudimentary algorithms themselves (even the big-name "emulations"), however, are generally on the scale of kilobytes.

I'll concede that there is a calculable difference in latency, but since modern processors run through cycles on the order of nanoseconds, you tend to have a lot of cycles of processing before you need to pass the next audio sample.

It's entirely possible that you think an Pro Tools|HD system is a requisite for good performance. But the stated reason (that OS processes happen before Audio processes) doesn't happen to be a good reason for that by virtue of the fact that for the most part it's incorrect.[/QUOTE]


which is why people still have latency issues with non TDM based computers? whatever reasearch you did in college you get an F. for both not citing your sources for such retarded claims and also for sucking dick at engineering. You're only further proving my point that you sound like you know what you're talking about, but don't know what the shit you're doing

[QUOTE=EmbraceRandom;18171187]Room + monitor speakers (and all variables of these; position etc.) is the most important part of mixing. You could have Pro Tools HD and all the best gear, but if you can't monitor properly, it's all wasted.

I have learnt how to compensate when mixing with my AKG K701s, BUT i always check my mixes in the University studio (which is well-treated) and all other playback systems I can think of. I know of people who can mix just as well on headphones as they can on monitors, as they know they're headphones well enough to compensate for the generally poor low frequency response and generally poor accuracy of spatial qualities. I'm not at that point yet, but I'm getting there. Like I say, I always check back on good monitors in a good room before completing any projects.[/QUOTE]


for low end you need a big room with some foam with a good NRC rating in the lower frequencies, baffles, ideally if you can get walls within walls with a little bit of space between them (both sound treated) would be awesome

[QUOTE=EmbraceRandom;18171187]Room + monitor speakers (and all variables of these; position etc.) is the most important part of mixing. You could have Pro Tools HD and all the best gear, but if you can't monitor properly, it's all wasted.

I have learnt how to compensate when mixing with my AKG K701s, BUT i always check my mixes in the University studio (which is well-treated) and all other playback systems I can think of. I know of people who can mix just as well on headphones as they can on monitors, as they know they're headphones well enough to compensate for the generally poor low frequency response and generally poor accuracy of spatial qualities. I'm not at that point yet, but I'm getting there. Like I say, I always check back on good monitors in a good room before completing any projects.[/QUOTE]

this is so right i don't even know what to say!

Xomblies

sup

wat is tdm

Time division multiplexing.

I guess I'll google it

Lol it's just kinda the architecture and the bus for audio in Pro Tools HD.

[quote=Xomblies;18174647]which is why people still have latency issues with non TDM based computers? whatever reasearch you did in college you get an F. for both not citing your sources for such retarded claims and also for sucking dick at engineering. You're only further proving my point that you sound like you know what you're talking about, but don't know what the shit you're doing.[/quote]

Latency exists in TDM systems as well. It's just generally very small. A/D/A conversion is a time-consuming thing (it's the inherent nature of needing to pace out your samples on the order of kHz, you can't just keep throwing processing at it because you still have to wait around for a temporal signal in the analog domain).

As for citing the exact sources, I mentioned the ASIO spec. That's a pretty well-known reference to consider. Admittedly the second-half was about conjecture, but it was the conjecture of somebody who's spent a few decades getting paid to educate for the EE department of an accredited university. So forgive me if that's not a good enough foundation to justify listening to his opinion.

In case you're not following, go here:

[url]http://www.google.com/url?sa=t&source=web&cd=1&ved=0CBIQFjAA&url=http%3A%2F%2Fstatic.helge.net%2F2010%2F06%2FASIO%2520SDK%25202.pdf&rct=j&q=ASIO%20steinberg%20spec&ei=fZuLTOS1M4SglAeJwplh&usg=AFQjCNH7C0L9kHFxA4kO58SDnjTgYp85Pg&sig2=te__hpv09-z24Rgl5-elNQ&cad=rja[/url]

Start reading at Page 42. Turns out that maybe I was wrong for Windows based systems (looks like it takes advantage of threading, which might actually make my assertion correct for the wrong reasons, but I don't know enough about Windows threading to speculate comfortably), but correct for OSX. Specifically, the buffers for I/O appear to accessed during the interrupt cycles on OSX.

Also to the point, there are people who can get negligible latency (on the order of 2 msec or less) out of native processing systems (I have been one of them). You just need to have sufficiently quick algorithms. Most "pure-effect" algorithms fall into this category (EQ, dynamics, your basic temporal effects).

Now, we're admittedly getting outside the scope of my focused studies (I was concerning myself with the operations of a TI CM55X chipset, and the inner-workings of a particular OS are more Computer Science than Signal Processing). But I'm still willing to wager that what's happening to move along the OS-sensitive operations in interrupt cycles is pretty minimal. That's kind of the point of the interrupt cycles: only the things that will break if they [I]don't[/I] happen get used there. But again, I'm still willing to wager that bussing a few bits here and there and clocking your hardware (the bare-bones stuff an OS is supposed to do) doesn't exactly tax a modern system into a catatonic state.

It's entirely possible I'm wrong, but you saying so means precisely dick to prove it (that's no better what you said I wasn't allowed to do myself). So either bring something to the table to show me otherwise, or let it go.

I'm going to repeat it again: it's entirely possible that you think a TDM system is a necessity. Saying it's necessary because the OS functions will always take precedent over the audio of a native processing system is not a valid reason because it's not particularly true.

i'll admit you can get low latencies with rtas vst etc, but unless you've got some retardedly fast processor, you'll overload your shit, tdm allows dedicated resources to JUST plugins/ ADA (basically most of pro tools' functions) at the lowest latency possible. YES, it's old tech, but for streaming audio it's still a lot better. That AISO is also used in pro tools bud, except with HD you've got an extra resource to handle the multi threading: TDM CARDS. As of late they've got 8 and 12 core macs that you can assign cores to PT and let the remaining handle system tasks, which are also more important than you think since you're going to be constantly accessing large amounts data off hard drives, building waveforms constantly and if you're editing; copying and pasting. Not to mention if you're using sample based instruments that's even MORE information to read off of other hard disks.

so yeah, if your computer is just chillin or you're using LFO's or something not sample based... sure, AISO and system ram are fine... but if you've got 10-16 drum tracks, 4-6 guitar tracks, 2 bass tracks and 8 vocal tracks, even at 44.1 and a 2 min song, are still going to be 30 megs a piece. i'll go with the standard 24 tracks that's 720 megs of audio to access just off of your session drive. if you're doing edits, or not working with consolidated tracks your computer NOW has to play a lot more files randomly. so moseph you can do the research if you want and waste your time or you can just TRUST someone who knows how to do audio like kfc does chicken... RIGHT. I've obviously got the chops to back what i'm sayin and i can help you out a lot if you would let me.

anyways you CAN get a decent sounding mix, recording off an LE rig or something like it, in fact i have heard tons of people do it... sorry about sounding like an asshole yesterday, lack of sleep and a passion for music don't mix when you're having one of these kinds of discussions

Let me back up for just a second here:

[quote=Xomblies;18167289]BTW, What are you trying to do here man? I'm trying to help people keep from making mistakes that add up to poor quality and you're trying to counter what i'm saying with some extremely subjective situation (more instances than this).[/quote]

What I think is getting lost in translation here is that about 85-95% of the stuff you say (from a technical standpoint) I agree with as being completely correct and proper. I generally only chime in or tack on when I think there's something to add (see comment about book cases above), or if see something that I believe is said in error. Having looked back on things I've chimed in regarding your particular advice posts, it looks like it's more frequently the latter case. That may be why you could get the idea that I'm beating up on you or something (you already know the difference in how I post when I'm actually beating up on you). That's definitely not the case. I know you're on the higher-end of the curve with regards to this stuff (your posts show that much), but I think we can both agree that neither one of us is on par with guys like Massenburg, Blumlein, Matthews, or even Frankle. At least not in their respective fields.

One of the goals of this forum (and most I would join), so far as I'm aware, is mutual [I]learning[/I]. This requires that not only questions get asked, but things get said in response. This isn't a forum that caters to experts, so a lot of stuff is bound to get misrepresented (case in point: my comment about ASIO as it applies to Windows). I'm more than willing to admit when I'm wrong, but I'm at least gonna want some evidence/rationale showing me why: giving me the finger and taking a dump on my lack of talent isn't gonna get anywhere.

Now, getting to the matter at hand...

Oddly enough, the most pertinent part is at the end:

[quote=Xomblies;18176160]so moseph you can do the research if you want and waste your time or you can just TRUST someone who knows how to do audio like kfc does chicken... RIGHT.[/quote]

I could, but that's the opposite of learning. Don't give us a fish, teach us how to fish, is what I'm getting at. You generally do alright in this respect, but there is that last 10% or so of stuff that needs some validation. That's not weird, and it's actually a good thing. If you're not stretching, you're not learning either.

Also, the KFC analogy might not be the best. It's kind of the Domino's of chicken...


[quote=Xomblies;18176160]I've obviously got the chops to back what i'm sayin and i can help you out a lot if you would let me.[/quote]


Here's a big pet peeve of mine, actually. Talent and good ears do not necessarily have any bearing on technical understanding. I'm not going to get into the specifics of how good I think I am, or how good you think I am, or any of that sort (clearly the consensus is known about all that).

Let me make an example. Once upon a time in the Jam Session you recommended the Heil PR-20 for snare drum over a Shure SM57. If I had made that same recommendation, but I said not to do it because the SM57 needs phantom power, whereas the PR-20 does not, what would your reaction be?

That's how I saw the OS-specific priorities thing. You contrasted audio streaming as a whole (rather than effects processing specifically) with OS-specific operations. That not only may not be pertinent to the issue at hand but it didn't agree with my particular understanding of reality. Hence the comment.

My old studio tech job was predominantly training people to use our facilities and acting as tech support when they hit problems (read into that what you want). When you decide you're gonna share knowledge like that, you sort of need to be able to have a basis to work off of, otherwise people aren't understanding what they did wrong in the first place.


[quote=Xomblies;18176160]i'll admit you can get low latencies with rtas vst etc, but unless you've got some retardedly fast processor, you'll overload your shit, tdm allows dedicated resources to JUST plugins/ ADA (basically most of pro tools' functions) at the lowest latency possible. YES, it's old tech, but for streaming audio it's still a lot better.[/quote]


You'll note that I never disputed it wasn't better. I said it wasn't better [I]because of OS priorities[/I] (the dispute was about prioritization). Yeah, the system is more efficient, it's throwing a lot more hardware at the problem. That's why they had PT systems up and running audio in like 1994, when the next best competitors were struggling to get decent MIDI synthesis.

Side-note: even though the definition of "retardedly fast" tends to change over time, I don't think you'd need one to do what you're talking about if your system is well put together. Modern systems are [I]fast[/I]. I've successfully done 18 channels (about 60% of what you're talking about) using a 1.8 GHz Pentium M, with only 1 GB of RAM. My current rig does 26 and is about 90% stable (I chalk that up to Presonus Drivers and Sonar not cooperating more than the system itself).


[quote=Xomblies;18176160]so yeah, if your computer is just chillin or you're using LFO's or something not sample based... [/quote]


You should also note that I conceded that point in a past post as well. But I'm also of the mentality that you shouldn't assume a user is gonna be using huge wave table-/sample-based synths unless that's mentioned.


[quote=Xomblies;18176160]you can assign cores to PT and let the remaining handle system tasks, which are also more important than you think since you're going to be constantly accessing large amounts data off hard drives, building waveforms constantly and if you're editing; copying and pasting. Not to mention if you're using sample based instruments that's even MORE information to read off of other hard disks.
...
i'll go with the standard 24 tracks that's 720 megs of audio to access just off of your session drive. if you're doing edits, or not working with consolidated tracks your computer NOW has to play a lot more files randomly.[/quote]


Except that's not exactly true is it? You're not streaming straight from the hard disk drives, you're streaming from buffers (the driver buffers) that are filled from data on the drive. I realize it seems like I'm nitpicking, but that's only because I am. Again, this is a learning thing here. So far as I know, the RAM doesn't need to be out the wazoo unless your edits and samples are off the charts, because the computer can generally jump to another page/cache faster than the buffers can unload at 441000 samples/sec.

[quote=Moseph;18176213]So far as I know, the RAM doesn't need to be out the wazoo unless your edits and samples are off the charts, because the computer can generally jump to another page/cache faster than the buffers can unload at 441000 samples/sec.[/quote]
Lol full drum kit Beat Detective at 96khz with plug ins on everything, man I need to try that.

Convection beats

[quote=Convectuoso;18176303]Lol full drum kit Beat Detective at 96khz with plug ins on everything, man I need to try that.[/quote]


Wait a second, I think I see my problem. The RAM pages will have the entire data files available because there's no smart way to arbitrarily select which samples stay on the disk and aren't copied. The CPU will know which of the samples to grab out of RAM because the DAW controls that, so the driver buffers get filled (by the results of an ADD operation). But the RAM itself will likely have immediate access to all points in the larger data files.

If that's the case though, then it should suggest multiple edits (and thus, lots of small files) would be less of a problem, which I know isn't the case in practice. Unless there's a fragmentation issue? Or maybe it's just done with pointers, rather than creating a new "base" track of the smaller file. I guess that will be something to look into.

[quote=viciouscycle;18176307]Convection [B]h[/B]eats[/quote]
See what I did there?


@Moseph:


In Pro Tools it never makes new files when you make edits, it just makes regions, which are just snap shots of the waveform. I think they're saved within the Pro Tools session file.

Dunno if that helps in anyway or if I'm just stating the obvious.

Reaper is none destructive and I'm assuming also doesn't create anything new

[quote=Convectuoso;18176482]See what I did there?


@Moseph:


In Pro Tools it never makes new files when you make edits, it just makes regions, which are just snap shots of the waveform. I think they're saved within the Pro Tools session file.

Dunno if that helps in anyway or if I'm just stating the obvious.[/quote]


No, wasn't certain of the particulars for Pro Tools (though I probably knew that at some point). Sonar does something similar unless you bounce the file or set it up to save as it's own file. I guess that does raise up the question about the beat detective/editing then: do the edits actually use up any more RAM?

Once it does the edits and cross fades everything (which apparently on an old G3 can take 2 hours lol) I see no change in performance.

Modern computers throw most preconceptions of DAWs out the windows I tell you.

Hi gice.
Im still unsure about whether to buy DFH or SD2.0.

My laptop specs are Dual Core AMD 2.5ghz, 4 GB RAM. Will either of these run ok with my system, bearing in mind that ill be using Pod Farm, ReCabinet(KeFir) and some other plug ins?

Im also pretty unsure about how these programs work. If i map out the midi using Cubase and then send it to an instrument track do i use DFH as a Vst to track the sound into the instrument track, or does it work the same as Pod Farm - as soon as you remove the Vst it leaves a dry signal?

Could do with someones expert knowledge!

Cheers then.

[quote=benfan;18178864]Hi gice.
Im still unsure about whether to buy DFH or SD2.0.

My laptop specs are Dual Core AMD 2.5ghz, 4 GB RAM. Will either of these run ok with my system, bearing in mind that ill be using Pod Farm, ReCabinet(KeFir) and some other plug ins?

Im also pretty unsure about how these programs work. If i map out the midi using Cubase and then send it to an instrument track do i use DFH as a Vst to track the sound into the instrument track, or does it work the same as Pod Farm - as soon as you remove the Vst it leaves a dry signal?

Could do with someones expert knowledge!

Cheers then.[/quote]

I'll let somebody who uses DFH confirm, but I know you're on the right track for at least SD 2.0. I'm pretty sure they operate the same way though.

The key difference between what I think you're saying and what actually happens is that MIDI, by itself, won't generate audio. It's instruction data: you need some kind of synthesizer to respond to it. So you're "dry signal" is actually inaudible digital instructions. You'll see MIDI meters moving/peaking around but conceptually it's like plugging a cable into a guitar and then not plugging the other end into an amp. The signal is being generated, it's just not being delivered anywhere.

Otherwise, if you're saying what I think you're saying, you've got the right idea.

SD2.0. More options. DFH only sounds good for a few genres and isn't as customizable.

Addictive drums is really snazzy too.

Also. You can just make an instrument track in Cubase. No need to make a MIDI track and route it to an instrument track.

Get Metal Foundry if you want metal drums