003r monitor section

[themadarchitect]

[QUOTE=netnoggin;1593218]Might be a typo, so before it goes further - did you mean to say you imported a track at 24/48 instead of 24/44.1?
/QUOTE]

Sorry, forgot to answer this..No I meant 24/44.1

[Kempo]

Originally Posted by themadarchitect

I never made a comparison between PTHD and LE. Futher, SRC is only one sample inaccuracy inherently contained in fresh out the box Pro Tools LE systems. And one way or another in PT, you are always doing a sample rate conversion..whether it be importing or bouncing audio. And every plug in introduces a new phase relationship that is handled by converting samples to bend to the information being presented with the plug.

While the example I showed isn't the greatest test in the world (neither is that plug for monitoring) a drop in db's over an entire frequency range is NOT SAMPLE ACCURACY.. You can also run the same test with 2 identical sources at 16/44.1 Bounce a 16bit/44.1 file and then put an analyzer on both. You will see the same loss of fidelity in the frequency signals is similar.

The fact that it cannot represent the true levels (and this is only one short test, because sample rate accuracy measurement is a science onto itself way too big for a forum discussion IMHO) already shows inaccuracy in the way PTLE handles samples. If you remember the original question, it was why it didn't sound the same after it was printed. If he is hearing a relationship of frequencies at a certain db level, when he plays it back and that relationship is lower
, it will not sound as how it did when he was all ITB.

The next sample inaccuracy you get is during the D/A phase due to a entry level clock and converters in the unit (what he referred to as the monitor circuit). Once jitter is allowed into a system, the output experiences sample inaccuracy and is not the truest representation of what you SHOULD be hearing. See: Jitter Explained

There are then harddrive issues, memory handling issue and a ton of other issues in the chain while playback and conversion that introduce FURTHER SAMPLE ERRORS. And we have no need to discuss the A/D since its not relevant..but lets just say...further lack of sample accuracy there too.

So before it gets out of the studio, it has already experienced at LEAST TWO sample inaccuracies. (much much more in practice) I won't even go into the whole ADC issue thats been passed around like a hooker on the Jersey shore after a prison escape...but take in account its shifting SAMPLES to make them line up more accurately.........

Not trying to be a smart ass with the bold types either, cause I feel this is one of the few decent discussions that I've seen (except the "call out" thing) in a while. And I don't say this statement to be a jerk either:

but the definition of sample inaccuracy is sample inaccuracy if you catch my drift.

I'm not saying that LE is the only native DAW that experiences it..but right out the box, with the native program and interface dongle it is inaccurate..not retardedly so..but enough that I would avoid using LE for a PROFESSIONAL application without some upgrades to rid the system of these issues.

Ok, I understand PTLE has some flaw:

-Converters are mid to crap level (96 io is mid to crap level, 192io is mid level)
-Clock for Mbox, 002, 003... is not good (the same for 96io, a little bit better for 192io)
-Sample rate conversion is not top end (the same is for HD)
-Monitor controller is not good (my 003rack exibit 0.9 dB difference between L and R at max volume)

So for all for the above things I agree with madarchitect:

PTLE has some flaw

BUT

Jitter means nothing with bounced track or printed tracks.
The jitter is a clock incosistency and affect A/D and D/A, you can hear jitter but it absolutely don't modify the data inside a DAW.
So external summing is affected by jitter, monitoring is affected by jitter but not bouncing or printing tracks.

And again I don't understand what does "PTLE is not sample accurate" means.

PTHD suffer for the same problems of clocking and jitter, a little bit less with 192 io, but this doesn't influence the track printing or bouncing. You can always use a better clock, or converters as you can do for say 003 (as I do).
PTHD suffer the same phase distortion when inserting processing that alter signal above nyquist.

So I don't understand the differences between HD and LE in terms of "SAMPLE ACCURATENESS"

This is not a LE vs HD, I want only to understand why somebody state PTLE isn't sample accurate, and what this means.


P.S.
I mean printing track: recording audio tracks to another audio track through a bus.
I mean bouncing ...beh you know what I mean.

[themadarchitect]

Originally Posted by Kempo

Ok, I understand PTLE has some flaw:

-Converters are mid to crap level (96 io is mid to crap level, 192io is mid level)
-Clock for Mbox, 002, 003... is not good (the same for 96io, a little bit better for 192io)
-Sample rate conversion is not top end (the same is for HD)
-Monitor controller is not good (my 003rack exibit 0.9 dB difference between L and R at max volume)

So for all for the above things I agree with madarchitect:

PTLE has some flaw

BUT

Jitter means nothing with bounced track or printed tracks.
The jitter is a clock incosistency and affect A/D and D/A, you can hear jitter but it absolutely don't modify the data inside a DAW.
So external summing is affected by jitter, monitoring is affected by jitter but not bouncing or printing tracks.

And again I don't understand what does "PTLE is not sample accurate" means.

PTHD suffer for the same problems of clocking and jitter, a little bit less with 192 io, but this doesn't influence the track printing or bouncing. You can always use a better clock, or converters as you can do for say 003 (as I do).
PTHD suffer the same phase distortion when inserting processing that alter signal above nyquist.

So I don't understand the differences between HD and LE in terms of "SAMPLE ACCURATENESS"

This is not a LE vs HD, I want only to understand why somebody state PTLE isn't sample accurate, and what this means.


P.S.
I mean printing track: recording audio tracks to another audio track through a bus.
I mean bouncing ...beh you know what I mean.

I love the DUC..Where else can you debate such things..jitter is also introduced between master clocks and logical devices (read hardrives). Thus, realtime print, does cause the introduction of jitter if a master clock is flawed..not to mention a hard disks clock can be EXTREMELY jittery. But these are minute...just figure, nothing is perfect and there are tons of calculations happening as u listen to audio, and then as you print audio..there are errors )rightfully so at every chain in that experience.

But the proof is in what you hear..forget the technical jargon..if u can hear a difference, that means that there IS a difference..mixing skills, room acoustics and monitors aside...if a print doesn't sound good in the same room..u jsut eliminated the monitoring chain..guess what it has to be....

[Kempo]

Originally Posted by themadarchitect

I love the DUC..Where else can you debate such things..jitter is also introduced between master clocks and logical devices (read hardrives). Thus, realtime print, does cause the introduction of jitter if a master clock is flawed..not to mention a hard disks clock can be EXTREMELY jittery. But these are minute...just figure, nothing is perfect and there are tons of calculations happening as u listen to audio, and then as you print audio..there are errors )rightfully so at every chain in that experience.

But the proof is in what you hear..forget the technical jargon..if u can hear a difference, that means that there IS a difference..mixing skills, room acoustics and monitors aside...if a print doesn't sound good in the same room..u jsut eliminated the monitoring chain..guess what it has to be....

I just don't want to appear unfriendly, so I don't want to go against you.

I prefer to say thank you for your time, I really appreciated your explanation of what you thinks.

I just say for that I know about digital electronic architecture I don't think so

[netnoggin]

Originally Posted by themadarchitect

I never made a comparison between PTHD and LE.

You are correct. I didn't mean to indicate that by including my statement in the same post with the other stuff. It was actually Brad that brought up the difference between the two. But you did indicate the 003R was a "hobbyist" piece of hardware, which kind of caught my eye. While I'm well aware of the differences in hardware specifiations and even the specs of the chips contained in most of them, I don't know if we can really say what "professional-level" hardware is anymore. People are making a living doing amazing things with LE hardware, including music that is sold to end users for a profit. That does fit the textbook definition of professional. Not to mention many a recording has gone gold being born from much more humble beginnings than an 003R.

Futher, SRC is only one sample inaccuracy inherently contained in fresh out the box Pro Tools LE systems.

And Pro Tools HD too, according to the SRC comparison chart you pointed us to.

And one way or another in PT, you are always doing a sample rate conversion..whether it be importing or bouncing audio.

I have no knowledge of this, so I won't speculate. But I didn't think PT did any SRC when importing files that matched the project settings. What would be the point?

And every plug in introduces a new phase relationship that is handled by converting samples to bend to the information being presented with the plug.

Agreed, of course. Insert a single plug and all bets are off regarding the output matching the input. But then, that's exactly what we want sometimes.

While the example I showed isn't the greatest test in the world (neither is that plug for monitoring) a drop in db's over an entire frequency range is NOT SAMPLE ACCURACY..

Technically I agree with you. But this is easily fixed by adjusting the level. It's not something inherently bad in the software that can't be overcome.

You will see the same loss of fidelity in the frequency signals is similar.

Again, I don't see this as a loss of fidelity, per se; only a level difference that can be compensated for.

The fact that it cannot represent the true levels (and this is only one short test, because sample rate accuracy measurement is a science onto itself way too big for a forum discussion IMHO) already shows inaccuracy in the way PTLE handles samples. If you remember the original question, it was why it didn't sound the same after it was printed. If he is hearing a relationship of frequencies at a certain db level, when he plays it back and that relationship is lower, it will not sound as how it did when he was all ITB.

I agree with you technically, especially regarding the original question. But I don't really think calling this a "sample inaccuracy" is the right way to look at it. Applying that logic, everything we do in any DAW produces these "inaccuracies". Every time we do a mix - any recording with more than one track for that matter produces these inaccuracies by definition because the samples coming out don't match what went in. It's what we do. It the whole reason for having a DAW - LE, HD, XZ, or whatever recording system you care to mention. Rarely is the requirement to just "capture and print".

The next sample inaccuracy you get is during the D/A phase due to a entry level clock and converters in the unit (what he referred to as the monitor circuit). Once jitter is allowed into a system, the output experiences sample inaccuracy and is not the truest representation of what you SHOULD be hearing. See: Jitter Explained

I completely agree. I know there are raging arguments across the internet as to how audible this is to humans using today's technology. I have no hard opinion except to say there is no doubt this is a measurable discrepancy in faithful reproduction of the original signal.

There are then harddrive issues, memory handling issue and a ton of other issues in the chain while playback and conversion that introduce FURTHER SAMPLE ERRORS.

This is where we really start to part company. You elaborate further in another post, saying "jitter is also introduced between master clocks and logical devices (read hardrives)". We can have a lengthy conversation sometime about hard drive timing signals, interface signaling, etc. (you mention you are "no big computer geek or engineer" - I am). But suffice it to say if hard drive timing variations were a factor in the audio signal chain (other than gross errors and delays that kill the transfer process completely), DAWs never would have made it off the ground at all. There's just too many ways to prove this doesn't have any effect - any. If your DAW software was dependent on hard drive timing it would be poor programming indeed.

Not trying to be a smart ass with the bold types either, cause I feel this is one of the few decent discussions that I've seen (except the "call out" thing) in a while. And I don't say this statement to be a jerk either

I don't think you are a smart ass or a jerk at all. I think you are speaking what you believe to be true, in a civil and intelligent way.

.but right out the box, with the native program and interface dongle it is inaccurate..not retardedly so..but enough that I would avoid using LE for a PROFESSIONAL application without some upgrades to rid the system of these issues.

My comments above still apply. I think that it's very hard to say what equipment is appropriate to create art. I do believe we have a responsibility to drive quality into the profession, and use the best tools we can. But when it's all over, the music is all that matters, not a picosecond of jitter that is orders of magnitude better than any tape machine flutter spec ever hoped to be. Lotta good music came from those tape machines...

[rqstudio]

While this may all be true, one of the most difficult things for any studio is to make mixes that sound AS good in the real world as they do in the studio. Before you start knocking digi, AGAIN, it sounds to me this is the studio's fault and they need to do some tweaking! Come on guys...

[netnoggin]

Originally Posted by rqstudio

While this may all be true, one of the most difficult things for any studio is to make mixes that sound AS good in the real world as they do in the studio. Before you start knocking digi, AGAIN, it sounds to me this is the studio's fault and they need to do some tweaking! Come on guys...

I wasn't knocking - I guess in a way I was defending. Good grief.

You're right, we're off the OP's beaten path considerably. As I said, the music is what matters.

The monitor section of the 003R does have it's limits, but I've never used it much. I always come out of the direct outputs. You lose a little functionality that you paid for (mute, level control, etc.), but if you want a cleaner signal, less things in the chain is always a better bet. If you really need the functionality, getting a Presonus unit or one of the Dangerous Music units will restore it and then some.

As for the other problem:

my mixes have been coming out to bright and the low mids a bit muddy when listening at home on the stereo or pc, at the studio my mixes sound great, Im aware that the acoustics of my room influence my mixes

As rqstudio said, it might be the studio's fault, but your hard pressed to find a studio that magically turns out mixes that translate well, no matter what the equipment or acoustics. Good translation is directly in the hands of the good engineer, and starts with good source material recorded with good mic positioning and other techniques. But in order for the engineer to do this, he or she must trust what they are hearing. This is why engineers (especially those that freelance) generally have many examples of reference music with them. These are music and mixes that the person knows well and trusts to sound good and translate well to many listening scenarios. I have read more than once where prominent engineers are quoted as saying they will even break out these reference mixes in the middle of a session, just to re-baseline themselves.

So I would start out by picking your personal reference mixes. It doesn't matter what they are; it's whatever works for you. But they should be recordings that you know translate well to many types of end devices, from studio monitors to iPods to PC speakers to your car to the neighbor's cheap stereo - you get the idea. I have a handful that span different genres. One example - Lately I kinda like Garth Brooks' "Long Neck Bottle" for C&W. To each his own.

Once you have your trusted references, use your ears (and eyes) to compare your material against the reference and try to deduce what's different. Listen closely. Listen through the mix and pick out individual instruments and try to determine why the reference material works so well. This is part of that critical listening ability that you really must have to do decent mixes. And if you struggle with it at first, don't fret. Critical listening is just like a muscle - if you exercise it, you will find it develops over time. But you must accept one fact of life - if you don't have any success at developing this skill, you are better served by putting your gear on Ebay.

The key point to reference mixes is no monitoring system and acoustic environment is perfect. More important than attaining perfection is understanding the anomalies and limitations of your recording environment so you can adapt to them. If you don't understand how those things affect your listening, you will invariably make counterproductive adjustments to your material to compensate.

A classic example is turning up the bottom end on mixes to unwittingly compensate for either lack of frequency response of the monitors or maybe the low end of your studio is being sucked up by room modes or wall-bounce phase cancellation at the mixing position. But the result is the same - things may sound good in the studio, but muddy or boomy when played on anything else, because you overcompensated. This does not mean you can't make good mixes in this room with this equipment. It just means you have to understand it happens, and mix accordingly. Leave the bottom a little lower in the studio, and it will come out right on the other gear. Again, the key is knowing your environment. Back to the reference material - if you have the above situation, your reference stuff probably sounds a little lacking in the bottom end when played in this studio environment.

Part of my recording and mixing regimen is to regularly listen to test mixes on other speakers and in different environments (and different volume levels - extremely quiet, extremely loud, etc.) as I go. I even have a small FM transmitter wired into my signal chain, so I can step out to the car and listen with ease. It never ceases to amaze me the needed adjustments I pick up on while revving the engine. Things have a different perspective with a S/N ratio of something less than 5 because of wind and engine noise.

I hope this give you some ideas. What it really comes down to is the gear is important, but not the magic bullet. Your ears are the magic bullet. I would venture to say while it may technically be deficient in some ways, your 003R monitoring section is not what will make or break your recordings.

A parting thought:
"George Martin, producer for The Beatles, actually believed that the limitations of the four-track stimulated creativity and that the 1967 album Sgt Pepper’s Lonely Hearts Club band would not have been as good if recorded on a 24-track."
http://blog.amatistudios.co.uk/2010/...mid-1960s.html