Just wondering where people set the master channel when mixing and recording stuff - I've heard that -18dbs is a good point.

Plus once you've finished the mix and are at the mastering stage do you push up the master levels as close to zero as possible (sorry I have no idea here ), or do you leave it where it is and let the final stage of processing deal with upping the loudness?

Quote Imran500:

---

Just wondering where people set the master channel when mixing and recording stuff - I've heard that -18dbs is a good point.

---

Digital right? And -18dB FS . Will also depend on your selected bit resolution.

Quote:

---

Plus once you've finished the mix and are at the mastering stage do you push up the master levels as close to zero as possible (sorry I have no idea here ), or do you leave it where it is and let the final stage of processing deal with upping the loudness?

---

Building a good gain structure will help.
I tend not to move the mix buss fader. Peak at around -6dBFS. Not much in terms of processing. I'll leave it to the ME who has the right room and right tools to achieve the right balance.

ken

--------------------
I'm All Ears.

Quote Ken Long:

---

Quote Imran500:

---

Just wondering where people set the master channel when mixing and recording stuff - I've heard that -18dbs is a good point.

---

Digital right? And -18dB FS . Will also depend on your selected bit resolution.

I'm recording in Reaper at 24bit-44.1. Er... what's FS?

Quote:

---

Plus once you've finished the mix and are at the mastering stage do you push up the master levels as close to zero as possible (sorry I have no idea here ), or do you leave it where it is and let the final stage of processing deal with upping the loudness?

---

Building a good gain structure will help.
I tend not to move the mix buss fader. Peak at around -6dBFS. Not much in terms of processing. I'll leave it to the ME who has the right room and right tools to achieve the right balance.

ken

---

Ok thanks for that, -6db sounds sensible

FS is Full Scale. The suffix helps identify what kind of system you are talking about.

24Bit will be more flexible than 16Bit as you have an extra 48dB of dynamic range to work with.

RE: -6dBFS : Of course, depending on the material, you may choose to add more/less processing, louder/softer etc...

ken

Ok thanks again - slowly - sometimes too slowly my mixing knowledge grows

You really shouldn't have any reason to touch the master fader. It should be at unity and left alone while mixing. The reference level of -18dbFS is the digital equivalent to the analogue reference of 0dbVU, which is the level your audio gear was designed to operate at. If you track each element of a mix at an average of -18dbFS then you will not likely run into the problem of clipping,(digital distortions that occur when the level exceeds full scale or 0dbFS). Clipping should be avoided at all cost because it ruins the quality of a mix. Also when mixing, it's a good idea to leave approximately 8 to 10db of headroom for the mastering engineer to work with. Not everybody does this and that's one of the reasons why some mixes come out squashed and lifeless after mastering. Headroom is the distance from 0dbFS. So ideally, you'd want a mix to peak not much more than -8dbFS, better still if it peaked around -10 or -12dbFS. You will have left plenty of headroom for the mastering engineer to work with and get your mixes to be loud, clean and dynamic.

Liz

--------------------
MIRRORMIX STUDIO
blog

Hi Liz!

Okay. I use all analog gear up until it goes into the hard disk recorder.
The hard disk recorder is calibrated at the factory to reflect 0vu like on an analog recorder.

How do I relate this to your -18dbFS?

Up till now I have just aligned the meters on my desk with those on the recorder and monitored my levels on the desk with occasional references to the recorder.
I have been getting clean unclipped recordings but surprisingly loud, comsidering I am keeping individual tracks at around -3 max and then only for peaks.
Loud enough uncompressed that I sound like I am in the loudness wars but with no compression!
Have I discovered pop nirvana?

Oh - desk and recorder are set up to +4



--------------------
Me? But I`m such a loveable old bugger!

I never touch the master fader in Logic, but I do change the outputs 1-2 level.

I listen to music at around 60dB SPL, or there abouts. I don't really listen at loud levels until I feel happy with a mix - then I crank the monitors and see what's really going on.

--------------------
Mac user; Logic and ProTools systems.

Quote EnlightenedHand:

---

You really shouldn't have any reason to touch the master fader. It should be at unity and left alone while mixing. The reference level of -18dbFS is the digital equivalent to the analogue reference of 0dbVU, which is the level your audio gear was designed to operate at. If you track each element of a mix at an average of -18dbFS then you will not likely run into the problem of clipping,(digital distortions that occur when the level exceeds full scale or 0dbFS). Clipping should be avoided at all cost because it ruins the quality of a mix. Also when mixing, it's a good idea to leave approximately 8 to 10db of headroom for the mastering engineer to work with. Not everybody does this and that's one of the reasons why some mixes come out squashed and lifeless after mastering. Headroom is the distance from 0dbFS. So ideally, you'd want a mix to peak not much more than -8dbFS, better still if it peaked around -10 or -12dbFS. You will have left plenty of headroom for the mastering engineer to work with and get your mixes to be loud, clean and dynamic.

Liz

---

As usual, Liz you ate talking a load of old .... codswallop!! NO, NO, sorry, I mean


......... common sense!!

P.S. Codswallop is an old English saying for ... political rubbish - or anything that is totally moronic!!

But of course I mean, you are absolutely right!!

Hi Ivan

They are reference levels. dBU refers to voltage (0.775V). You can go over this level when recording in analogue as you well know. dBFS refers to a digital system. 0dBFS is the maximum allowance for the amplitude of a signal in such a system before it clips so anything above that and you're distorting. And not in a good way.

ken

--------------------
I'm All Ears.

Can we clarify a few things as I'm not 100% on this yet - maybe I'm being thick, if so apologies this is just a hobby unfortunately!! Mixing at such a low volume is obviously a lot quieter than a commercial track - I guess we can turn the volume up on the monitors though. Using Cubase by the way.

I'm already using the SSL intersample peak meter accross the master bus to ensure this doesn't peak, but are you saying still keep the level way down - i.e. -18 on each track and peaking at a max of -8 on the master?

If we're not hitting the inter-sample peaks during mixing then why mix it lower than that? Is this about the conversion to 16-bit? Are you saying it's better to mix it lower and then limit it later to maximize the volume, rather than mix it louder and lower the volume if necessary later? If so why?

Quote HugoL:

---

Can we clarify a few things as I'm not 100% on this yet - maybe I'm being thick, if so apologies this is just a hobby unfortunately!! Mixing at such a low volume is obviously a lot quieter than a commercial track - I guess we can turn the volume up on the monitors though. Using Cubase by the way.

---

Yes, use the volume.

Quote:

---

I'm already using the SSL intersample peak meter accross the master bus to ensure this doesn't peak, but are you saying still keep the level way down - i.e. -18 on each track and peaking at a max of -8 on the master?

---

Track at -18dBFS. Recording 24Bit resolution. Mix so that you're sum arrives at -6dBFS (or, as Liz prefers, -8dBFS).

Quote:

---

If we're not hitting the inter-sample peaks during mixing then why mix it lower than that? Is this about the conversion to 16-bit? Are you saying it's better to mix it lower and then limit it later to maximize the volume, rather than mix it louder and lower the volume if necessary later? If so why?

---

No, the point that was being made was that you want to give the mastering engineer enough headroom to balance the mix and increase its loudness. If you are limiting the mix buss, or even mixing it hot, the ME won't have much to work with.

Mixing louder and lowering the mix buss fader later means you are processing the overall sound again (after having processed the separate tracks in the mix).

ken

--------------------
I'm All Ears.

Thanks Ken, but why is reducing the volume later if need be a problem? Surely it's just subtracting a constant number from everything? Not really heavy processing - and surely this is better than everything being much too low and having to boost it lots as you loose a lot of resolution and amplify the noise more?

But I'm gathering it's because when you increase the volume using limiting you're leveling everything out more, which is more effective when you have more dynamic range to play with - i.e. you can let it breath more?

Edited by HugoL (24/10/08 01:15 AM)

The other question would be for electronica it seems common to mix as you go - i.e. the seperation between composing, mixing and mastering can be much more vague. This is more inline with how I've been doing it so far, certainly.

We've already been advised to mix into a compressor, but if we're planning on just mastering stuff ourselves is there anything wrong with mixing with a set of mastering plugs accross the 2-bus as well to get things sounding much more finished as we go? Obviously if I was planning on releasing the stuff commercially I'd trust someone else and want to undo my probably far from ideal experiments, but if it's just for my own amusement?

EDIT: Sorry ignore this post guys, day off / too much wine. I don't currently work this way, but it is tempting to make it sound finished as soon as possible.

Edited by HugoL (24/10/08 01:42 AM)

I don't understand why you'd want your master faders at anything but 0dB?!

Leave some headroom in your mix, but don't touch that master fader. If the listening level is too loud, turn your monitoring amp down!

Or am I missing something?

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote The Elf:

---

I don't understand why you'd want your master faders at anything but 0dB?!

Leave some headroom in your mix, but don't touch that master fader. If the listening level is too loud, turn your monitoring amp down!

Or am I missing something?

---

I turn the master fader down (in Reaper), when the clip ligts on the master buss start lighting up. Is there something wrong with this? What's the master fader for? Just 'fade-outs'?

--------------------
Battle flags are flown at the feet of a garden gnome.

If something is clipping then I'd trim the channel faders. That's simply about good gain management (channels < groups < masters...).

(as for fades, I'd always do those in an editor with dithering to smooth out the fade - not in my DAW).

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote The Elf:

---

I don't understand why you'd want your master faders at anything but 0dB?!

Leave some headroom in your mix, but don't touch that master fader. If the listening level is too loud, turn your monitoring amp down!

Or am I missing something?

---

Ok head (slightly) more attached today. Yeah I don't know about the original question - but I'm querying what level we should be peaking at on the master channel when mixing ITB. I always keep the master fader at unity.

I'm still not getting why - so long as we're not generating inter-sample peaks - it really matters if it's peaking at just under 0 rather than -10 ish. If a mastering engineer needs to reduce volume before processing to give more headroom then why is this a big deal? Surely he can turn it down slightly?

Or does reducing the volume (a divide not subtraction what was I thinking) introduce audible rounding errors? But if that were the case then surely digital volume changes would need dithering - which I don't think they do unless a word-length change is involved. Either way surely we aren't saying any digital volume attenuation is that bad for minor adjustments?!

Aaaah - please put me out of my misery!

44.1/24 bits, peaking anywhere between -18 dBFS and-12dBFS.

--------------------
No longer a forum member.

Quote The Elf:

---

If something is clipping then I'd trim the channel faders. That's simply about good gain management (channels < groups < masters...).

(as for fades, I'd always do those in an editor with dithering to smooth out the fade - not in my DAW).

---

Understood.

Apparently though, Reaper has a silly amount of headroom on individual channels. Rather than having to precisely adjust each fader to give the same gain reduction and thus the same mix balance, it makes more sense to me, to use the master fader unless there's some good reason not to.

What is the disadvantage of doing it this way?

--------------------
Battle flags are flown at the feet of a garden gnome.

So basically guys you are saying that the master fader should be left at 0db and all the other faders should be adjusted so the mix sounds roughly right peaking at -18dbs?

The only problem I get with this is that I have to turn loads of things way down - for example in the song I'm working on at the moment, I have routed guitars, bass and synths to their own FX tracks and these are all at like -20 to -25dbs (as are the kick and snare tracks).

It now all sounds fine, peaking at -18dbs so is there a problem with what I've done so far with regards to the low settings of the faders?

Quote HugoL:

---

Ok head (slightly) more attached today. Yeah I don't know about the original question - but I'm querying what level we should be peaking at on the master channel when mixing ITB. I always keep the master fader at unity.

I'm still not getting why - so long as we're not generating inter-sample peaks - it really matters if it's peaking at just under 0 rather than -10 ish. If a mastering engineer needs to reduce volume before processing to give more headroom then why is this a big deal? Surely he can turn it down slightly?

Or does reducing the volume (a divide not subtraction what was I thinking) introduce audible rounding errors? But if that were the case then surely digital volume changes would need dithering - which I don't think they do unless a word-length change is involved. Either way surely we aren't saying any digital volume attenuation is that bad for minor adjustments?!

Aaaah - please put me out of my misery!

---

Ok so I've been off trying to answer some of my own questions! Dither would be required for 16-bit digital volume attenuation - but not for 24-bit or 32-bit (as used by the Cubase mix buss).

I still can't see that peaking significantly lower than 0dBfs (-12 or whatever) is particularly necessary during mixing though providing it's not clipping. Even inter-sample peaks are about reconstruction in the D-A, right, which could still be fixed later - so long is nothing is actually clipping ever - though obviously you don't want distortion in your monitors whist mixing.

Any attenuation can be carried out prior to mastering processing as required, either in the analogue domain or digitally - especially if a 24 bit export is provided.

So am I missing something here? Is this nothing more than nice to have best practice just to eliminate any volume adjustment prior to mastering - which wouldn't be a big deal anyway?

Quote Imran500:

---

It now all sounds fine, peaking at -18dbs so is there a problem with what I've done so far with regards to the low settings of the faders?

---

I'd reduce gain higher up the chain to give me more room to work with the fader. Sometimes I'd use Cubase channel gain, but more often I'd use the gain trim in a (last in chain) channel insert compressor plug-in or such.

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote Imran500:

---

So basically guys you are saying that the master fader should be left at 0db and all the other faders should be adjusted so the mix sounds roughly right peaking at -18dbs?

---

Yep! Or actually, slightly *under* -18 dBFS.

Quote Imran500:

---

The only problem I get with this is that I have to turn loads of things way down - for example in the song I'm working on at the moment, I have routed guitars, bass and synths to their own FX tracks and these are all at like -20 to -25dbs (as are the kick and snare tracks).

---

Yes! You have to turn lots of stuff down! That is the whole point in mixing! I always track so that the signal *peaks* are hitting *each* channel at no more than -6dB (and sometimes at -10 dB). This is before hitting the red button.

Once I have those levels set properly, I find that I have less faders to pull way down to prevent peaking the master output.

I also sum all the channels, but use group channels rather than FX channels.

So with my gain staging, I watch the levels at:

1) mic -> mic pre
2) mic pre -> converters (sometimes there is a compressor or such in-between)
3) converters -> channel inputs on my DAW
4) channel outputs -> group inputs
5) group outputs -> main channel


By maintaining headroom throughout the chain, I am able to produce a clean signal all the way through the chain.

Quote Imran500:

---

It now all sounds fine, peaking at -18dbs so is there a problem with what I've done so far with regards to the low settings of the faders?

---

No, it sounds like you are doing OK, but if they are *too* low, you might have hit them too hard at the input stage, which is also bad.

--------------------
James
"Never interrupt your enemy when he is making a mistake" ~Napoleon Bonaparte~

Yeah its all been covered really and yeah the simple thing is

Leave your Master at 0 and don't bother touching it, the only thing I have on my Master is Compressor which helps glue the mix together and find that its pretty common practice to use a Compressor on the Master.

I then just go through tracks and try and get a balance of instruments that roughly peak at about -10 on the master fader. This leaves plenty of headroom if you want to get your track mastered or if you want to apply some form of Mastering plugin like T-racks, Izotope or PSP VW then you have a lot of space to work with.

Also remember to mix at a level which is comfortable for your ears and remember to test things by turning the volume down... I find it really helps when trying to balance out instruments and vocals.

--------------------
There is nothing Grim about this Reaper
We Fell From The Sky

Edited by Henry-S (24/10/08 11:27 AM)

Ok sounds like great sensible advice - but again if you do happen to have a hotter mix then that what is the real damage?

Liz and everyone is inferring the mastering engineer won't have headroom to work with and that will ruin everything. Surely a bit of attenuation on a 24-bit export prior to mastering whilst not best practice isn't actually going to be audible, right?

Quote Richard Graham:

---

Quote The Elf:

---

If something is clipping then I'd trim the channel faders. That's simply about good gain management (channels < groups < masters...).

(as for fades, I'd always do those in an editor with dithering to smooth out the fade - not in my DAW).

---

Understood.

Apparently though, Reaper has a silly amount of headroom on individual channels. Rather than having to precisely adjust each fader to give the same gain reduction and thus the same mix balance, it makes more sense to me, to use the master fader unless there's some good reason not to.

What is the disadvantage of doing it this way?

---

In theory, if the floating point processing is done correctly there should be no problems but (and it is a big "but") running your channels hotter than normal means your plugins run at a hotter input signal which can be detrimental as many plugs don't behave gracefully that way.

This has actually fundamentally changed the way I mix - it's all now suddenly more calm and relaxed.

There's no more struggling with the levels when things get too hot - mainly because things don't get hot anymore!

The mix sounds right at -18dbs and I just turn the amp up!! It's so obvious but until someone in the know tells you, you basically haven't got a clue.

This is why I love this forum - straightforward and knowledgeable people who have no problems taking a bit of time to help others!

Quote archdake mkII:

---

Quote Richard Graham:

---

Quote The Elf:

---

If something is clipping then I'd trim the channel faders. That's simply about good gain management (channels < groups < masters...).

(as for fades, I'd always do those in an editor with dithering to smooth out the fade - not in my DAW).

---

Understood.

Apparently though, Reaper has a silly amount of headroom on individual channels. Rather than having to precisely adjust each fader to give the same gain reduction and thus the same mix balance, it makes more sense to me, to use the master fader unless there's some good reason not to.

What is the disadvantage of doing it this way?

---

In theory, if the floating point processing is done correctly there should be no problems but (and it is a big "but") running your channels hotter than normal means your plugins run at a hotter input signal which can be detrimental as many plugs don't behave gracefully that way.

---

Hang on though, don't my channel plug-ins appear pre-fader? Certainly my compressor and gate plug-ins don't need to be changed when I change the channel fader level!

--------------------
Battle flags are flown at the feet of a garden gnome.

Quote Richard Graham:

---

Hang on though, don't my channel plug-ins appear pre-fader? Certainly my compressor and gate plug-ins don't need to be changed when I change the channel fader level!

---

Yes, and it is here that you could make some attenutaion to give you more room on the fader - compressor output gain, for instance. A gain control is not always about turning a signal up.

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

I'm a bit surprised that there is such confusion about this essential element of mixing. It boils down to this.

When you're mixing in the box, your master fader should not be adjusted at all in any way volume wise. Some people like to use mix buss compression, where they put a compressor on the master fader to achieve a certain type of sound. That's fine so long as you leave the master fader set to unity. Don't move it off of unity.

As far as the whole -18dbFS thing is concerned. That's something that you want to pay attention to while tracking. You want to make sure that each element of the song is tracked at an average of -18dbFS which would be 0dbVU in the analogue domain. After you finish tracking each element your master levels could easily come close to 0dbFS upon simple playback with all of the track's faders set to unity, that's fine because it's just playback. When it comes time for you to mix you'll have to pull all of channel faders down, (not the master. Don't touch the master), and then push them up as you go along with the process of mixing and automating and panning and all of that. All of the while you should be keeping an eye on the master to see that the total track's summed level doesn't peak much more than anywhere from -12 to -6dbFS. That leaves headroom for the mastering engineer to process your stereo mix and get it louder while remaining clean and dynamic. Headroom is good room all of the time. If your master buss is reading close to 0dbFS then you are mixing too hot,(and probably tracking too hot also), and you are painting the mastering engineer into a corner with respect to how much louder they can get your music. An unmastered track should be VERY quiet compared to a mastered commercial release. That means that you've preserved the track's headroom, which is again the distance down from full scale or 0dbFS. Headroom is good room all the time.

If you've tracked and/or mixed too hot and the master fader is clipping and you simply pull it down you're still distorting your D-A converter and you just can't see it because the clip light has gone off. It's the equivalent to burning a steak and then while it's hot dousing it with ice water and saying it isn't burned anymore. It's still a burnt steak. You can always finish cooking something that's undercooked,(that would be the equivalent of leaving plenty of headroom for the mastering), but you can't un-burn something, just as you can't un-clip digital audio. You have to make sure to preserve your headroom at all times during mixing by not letting the master levels peak any higher than anywhere from -12dbFS to -6dbFS, the greater the headroom the better the mix will lend itself to mastering and to being louder in the long run.

In the words of Bob Katz, "It's not how loud you get it. It's how you get it loud."

Liz

--------------------
MIRRORMIX STUDIO
blog

What about out the box? I like to push my master fader quite hard - sometimes around +15dBU.

I mix at this level...
Just kidding. I can't even speak at that level never mind listen.

--------------------
http://soundcloud.com/anthony-wall/sets/audio-reel
http://songsforvoiceandpiano.com/

Quote snipsnip:

---

What about out the box? I like to push my master fader quite hard - sometimes around +15dBU.

---

Depends on your recording medium and its limitations.

ken

--------------------
I'm All Ears.

true,

I actually know the answer anyway, Im just SO bored at work i was trying make converstion.

Sorry, I still don't get it.

Example scenario: the recording isn't clipping (because I've left 12dbs headroom). The channels aren't clipping (because in my DAW, Reaper, they allegedly have massive headroom). The plug-ins aren't clipping (because they are pre-fader, and also have massive headroom).

The only thing clipping is the master output, because the faders on the channels are well up and the Master fader is set to unity.

Solution: turn the master fader down. Why not?

Why bother to have a master fader at all if it is never supposed to move off unity? It doesn't make sense!

--------------------
Battle flags are flown at the feet of a garden gnome.

Quote Richard Graham:

---

Why bother to have a master fader at all if it is never supposed to move off unity? It doesn't make sense!

---

Couldn't agree more! I would be happy to see it removed altogether to be honest!

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote snipsnip:

---

true,

I actually know the answer anyway, Im just SO bored at work i was trying make converstion.

---

Oh. Well in that case:

Michelle Obama or Cindy McCain?

ken

--------------------
I'm All Ears.

If your master output buss is clipping then the recording IS clipping. It doesn't matter that no individual channels are clipping. The sum of all the channels added together is greater than any individual channel's output. Therefore your master output can clip without any individual channels clipping at all. It's that simple. You're adding all of the output of all of the channels together at the output buss.

You don't adjust the master fader during mixing because that's not it's purpose. It's purpose is simply to adjust playback volume. That is all. On an actual console that's one way you'd adjust the playback volume. When you're mixing you don't want to toggle the playback volume at the master fader because it will not allow you to see if you are clipping the converters or not. That's why you leave it alone. Like I wrote earlier, you can't un-clip a signal. Turning down the master fader doesn't un-clip anything. It just hides the fact that you are clipping from you. The D-A converters have already clipped and the distortions that come along with it are already there. It might not sound that bad to you because you've got the playback volume down but when a mastering engineer gets it they will know that it sounds like sh*t when they try to process it. And they will likely conclude that you've either tracked too hot, mixed too hot or both.

Think of it like this. If 0dbFS is the ceiling of a room and you need more space to fit things in that room from floor to ceiling,(headroom), how would you get more floor to ceiling space by lowering the ceiling? That doesn't make sense at all. Yet that's exactly what you're doing when you turn down the master fader as you notice clipping going on. The solution is to not take up as much floor to ceiling space in the room to begin with. If you need to take up more later,(during mastering), then you can. But you can't get headroom back once it's gone. You can't un-bake a cake and you can't un-burn a steak. The damage is done even if you hide it and it only gets worse as you further process the audio during mastering. You'll end up with an album that sounds as absurdly mixed as Californication or something worse. No dynamics, no clarity, no depth, no good.

Liz

--------------------
MIRRORMIX STUDIO
blog

Hey Richard! I'd like to discuss a few points.

Quote Richard Graham:

---

Sorry, I still don't get it.

Example scenario: the recording isn't clipping (because I've left 12dbs headroom).

---

You mean on each track?

Quote:

---

The channels aren't clipping (because in my DAW, Reaper, they allegedly have massive headroom).

---

An analog channel can clip at +27 or +30dBu and, yes, that's massive compared to an average clipping point of +20dBu.
Virtual [software] channels can't exhibit "massive" headroom because all that matters regarding headroom is if the floating point processing is correctly implemented when the channel level exceeds 0dBFS.

Quote:

---

The plug-ins aren't clipping (because they are pre-fader, and also have massive headroom).

---

If the pre-fader level is relatively high, even though they can process near -or above- 0dBFS levels is irrelevant. The only thing that matters is how gracefully they do it.

Quote:

---

The only thing clipping is the master output, because the faders on the channels are well up and the Master fader is set to unity.

---

If you've tracked everything with a nominal of -20 or -18dBFS you should only get the occasional near 0dBFS peak unless you're heavy handed with EQ boosts and stems.

Quote:

---

Solution: turn the master fader down. Why not?

Why bother to have a master fader at all if it is never supposed to move off unity? It doesn't make sense!

---

With digital, clipping the master fader is less critical than analog but keep in mind my comments.