Hi, all... I'm interested in feedback regarding a recording I made recently of a high school wind ensemble. The sound file is: http://soundcloud.c…"]Ride (HHS-WE) by AdrMorgan on SoundCloud - Create, record and share your sounds for free[/]="http://soundcloud.c…"]Ride (HHS-WE) by AdrMorgan on SoundCloud - Create, record and share your sounds for free[/]. I've been recording high school concerts for about two years. I'm not completely settled on the best way to go about it; but I feel I have come along way.
The sound file was recorded with the following set up:
- Venue - An average high school auditorium. Hall size - 1500 seats. The HVAC fan is located in the stage-right wing; and it was not turned off for the recording.
- Ensemble Size - 60 musicians in front of sound shells.
- Mics - ORTF; AKG 414s. Their shapes make it a bit difficult to position precisely, so I created a template to get the mics angled at 110 degrees with the center of the mics spaced 17cm apart. The two mics are attached to a single mic bar.
- Mic-Stand Position - About 10 ft high, measured from the stage level; about 8 ft from the front of the ensemble. The mics are angled slightly downward.
- PreAmp - API 3124m+
- Recorder - Marantz CDR420
- Software and Post Production - Adobe SoundBooth. Very little post production work. I only changed the L/R stereo image slightly.
I have two questions: (1) Generally, how does it sound, in your opinion. (2) I have weak confidence in setting gain on the API. Over the past couple of years, I've vacillated between having the dB peaks hit between 0 and +3 on the API UV meter or having the loudest hit right at +18 on the UV meter The sound file above was the later (peaks hitting between 6 and 18). I set the output volume from the API preamp so that the Marantz UV meter hits between -18 and -12. Any gain-setting recommendations from your experience and from how the recording sounds?
I'm listening to this on pretty pedestrian headphones, so all comments should be viewed skeptically. Overall I thought it quite good. I felt there was a bit of a "hole in the center" feel (maybe because I listened on headphones.) So I might try moving back a bit (unless you tried to increase the spread in post - if so I'd reconsider that.)
-18 hitting the recorder sounds about right. I would think you are hitting the input stage of the API a bit hard for this music. About right for rock drums, but ,,,
Thanks, Bob. I didn't change the spread in post. I'm a little reluctant to do too much in post because I my reference speakers are just computer speakers. The recording equipment belongs to the school; and they were willing to spend some good money on quality recording equipment. I guess they just assumed that the person recording the concerts would have an equally good set set up for fine tuning the recording in post production. Thanks, again, for the feed back.
This is pretty funny. You are obviously a pissing cousin to me. I too am in the Herndon/Stirling area.
The school probably had this equipment because Bruce Kane & Sterling Productions & I, had most of the Northern Virginia schools. And in the schools, Bruce and I would always use API 3124m's and stuff like 414's, Sennheiser MKH 40's, Neumann TLM 170's/103's. But Bruce closed the company after more than 20 years this past May. Maybe that's why you got this gig?
While the 414's sound good, you'd probably be better served by running them as spaced omnis. You won't lose so much of the middle that way. It's a bigger broader wash of sound. That or configuring them right in the middle as an MS pair, patching middle to left and sides to right and then decoding in software. This method provides for a solid mono center image with complete mono compatibility by canceling the side microphone out. And then it's shooting right up the middle where it should, like they did in the old mono days gone by. You'll find that unique to be something of a wizard with microphone stand attachments to be able to configure the 414's as middle and side, one above the other, very close together. And about 4 to 8 rows behind the conductor. Otherwise it sounds like... well it sounds like a school band. Which is all you can really expect out of a bunch of kids.
So what kind of software are you using to post the project with? They probably didn't assume anybody would do anything to the recording? Just because they got it that way from Bruce or I, doesn't mean anybody else is expected to do the same kind of mastering jobs as we did. They just think you stick a couple of microphones up and that's it. And for most kids parents, that's all they need. We just went the extra mile because we are professionals. That's our edge over the competition. You're the competition and you're not the competition. Very infrequently did we do any actual multitrack productions for the schools. It wasn't something that was necessary or practical to do. Though near the end, we were using 8 & 24 track digital recorders, along with extra microphones, so Bruce could have a little extra pizzazz and flexibility. So most recordings were just two microphones and maybe one announcement microphone and/or solo microphone. And most of that we could do well, live. So then Bruce would only have to trim some things between numbers, etc..
Remember these recordings are not going to hit the Billboard top 100.
Mx. Remy Ann David
Thanks for the reply. Sterling Productions did the FCPS District Choir and Band concerts, I think. I've got some of those recordings since my older kids are former District Choir members. You've probably been to our school (Herndon HS), since the district often has their choir and band concerts there because our auditorium is one of the largest in the district.
As for the band on the recording above, it is the Herndon HS Wind Ensemble. The former director, who did a great job raising the musicianship of the kids and in building a top-notch program, bought the equipment in 2004. He bought 4 414s, but one "walked away". In the past, I think he just had a band booster member record the concerts; that's how I eventually got the responsibility (I'm now the HHS Band Parents Assoc. President). I think his motivation was to use the recordings as a learning tool for the kids and as a way for him to prepare for rehearsals.
When the former person recording the HHS concerts gave the equipment to me a couple of years ago, he also gave me a set of instructions that I knew when I saw them were suboptimal. I did some research and broke the mold by using three mics in a Decca-Tree-like configuration for my first concert and by changing the gain settings (lowering them...a lot). Initially, I got a lot of push back because that's not how it was done in the past (you know "tradition", and all). My first recording wasn't as good as I thought it could have been, but it was better than the earlier recordings -- and without all the clipping found on earlier recordings. Each recording has gotten successively better, though sometimes only in small ways. Now, the recordings are still being used in class rehearsals as learning tools; but, we also, sell the to families for basically the cost of the materials.
I've been interested in trying an MS pair configuration, but we don't have an attachment to properly position the mics. We use the Shure 15A with a mic bar (On Stage, I think is the brand). Do you have a recommendation for Mic attachment on a single stand for MS applications?
As for software, I use Adobe Soundbooth CS4. I see that Voxengo has a plug-in that is an MS encoder/decoder. Are you familiar with it?
As for the extent of post production, right now, I do very little. I'd like to do more; but I feel hamstrung with the stuff I use for post production (my laptop and headphones).
Space Omnis - I'll give that a try. One reason I went to ORTF was so that I could use just one stand. I'd also like to experiment with "flying" a Decca Tree rig, but it may too big of a project to work on. With a flying rig, there are no sight-line issues, so it is something interesting to me.
Thanks, again for the reply.
PS - I'm also helping out with sound for the HHS theater productions. This has been another evolving project...but fun.
SHURE, Atlas, Beyer and others make an assortment of microphone stand hardware items. I've used both Atlas and SHURE stereo microphone stand adapters for the most part. Sometimes a simple right angle elbow with a simple straight stereo microphone bar is all that you need to position your 414's into an MS configuration.
I have been using Adobe's Audition since it was released as Cool Edit, as shareware, back in 1996. Sound booth I'm not quite up to speed with? But as most software audio programs go, almost every one of them has some kind of MS decode and encode presets. Sometimes it's indicated as MS-LR and LR-MS. One you used for decoding and the other you use for encoding. With the microphones, there is no reason to encode just decode. We use the encoding feature for other purposes and reasons. Where we might want to add some dynamic compression to just the side channel? When combined back again, that technique provides for an enhanced and broader stereo image. Not something though you would want to use on a school band. Rock 'n roll? Absolutely.
Right. There really isn't anything you need to fool around with recording a school performance band. Thank goodness this is all very low-tech. Now adding some equalization can be beneficial. Beneficial equalization is frequently what's known as high pass filtering. Your microphones actually have those filters built into them. It's designed to compensate for an increase in the low-end response when used closely known as proximity effect. The other times it's used to eliminate low-frequency noise from outside traffic and HVAC systems. These extreme low frequencies whether the microphone can reproduce them or not are not really beneficial. In fact they're just the opposite.
There have been numerous versions of the 414 microphone throughout its nearly 50 year existence. Today's version, basically comes in two flavors. One flavor is that which comes closer to the older C-12 and includes a healthy rise in the presence. From 3 to 6 kHz. The other version is the more linear version with a much flatter response. People found that version to be better for overall recording purposes of a wider latitude. Where the older version sounds particularly good on vocals. However it can get a little harsh sounding on stuff like school bands. So depending on your version of 414, you might not need any postproduction equalization? Frequently not. The one helpful thing you might find is some dynamic range limiting? While we like a nice natural dynamic range... I can assure you... nobody really wants to hear a school band with its natural dynamic range. Most of the time it will be too soft with sudden loud explosive bursts. And that sucks. So, you want to employ some limiting. Too much and it will sound horrific, with the most ungodly noise trash and distortion you don't want. But just the right amount will make that school band sound more like the Boston Pops. And all of the parents will think their little darlings are wunderkind's LOL.
Yup, Bruce of Sterling Productions sent me to your school plenty of times over the past 20 years. I'm sure you have the equipment you have because that's what Bruce and I used. So naturally, after dozens of times of us showing up with API 3124's & 414's, they knew that was what to get. Too bad somebody walked away with one of those 414's. Actually I can't quite believe the school would have shelled out that kind of money? I actually think that's relatively irresponsible of them. Public high schools that are paid for by the taxpayers don't need to have the most top-of-the-line audio equipment by any manufacturer. The 3124's costs in excess of $2500 each where a simple $350 Mackie would work as fine and last as long. We're a couple of $250 each condenser microphones would have made more sense than 4 $1000 microphones. Sure that's what we had but we're professionals. You wouldn't give a 16-year-old a Jaguar, Maserati, Lamborghini, Cadillac would you? No. So that was a waste of the taxpayers money. That's a $200 hammer or $100 toilet seat. And that action gets me angry that was passed through that way. So, just make sure, when your kid graduates, don't give back the equipment. LOL. Because that's the stuff you want. That's the stuff everybody wants. That's the stuff few people can afford.
I traded one of my two 414's for an old 67. So I'm down to only one of those dang things.
Mx. Remy Ann David
oops... Forgot to talk about creating an MS pair with the 414's.
Yes because you want to use a single stand, MS is certainly the way to go. In fact it's actually my favorite stereo microphone technique. Decca tree? It never terribly impressed me, either. Works out well for a lot though that like that. Bruce at Sterling Productions preferred spaced Omni's. So when I worked for Bruce, I would give him what he wanted unless he would defer to my choice. And I don't ever remember Bruce ever doing MS? He just wasn't an MS kind of guy. He was always kind of spaced out with Omni's LOL. But I am an MS nut. MS is unique and ultra cool. With MS, you can vary the stereo width and never have any problems with monaural compatibility. Which makes it quite unique for musical recording purposes. In a sense it gives you the ability to electronically position your microphones in the stereo imaging. Can't do that after you've set them up with other types of stereo microphone techniques. And having a single stand makes this extremely practical to do. Your only real criteria is how close you can get the capsules together of the microphones. Now the cool thing is, any method of mounting the microphones will work. So you can have one on top of the other. You can have one behind the other. You can have one sitting straight up facing the front of the stage and one lying on its side facing the side walls. So unlike XY and ORTF, the two microphones do not need to be set up identically and symmetrically. Sometimes I've found the need of a 3 inch gooseneck piece. Other times an elbow. Sometimes just the stereo microphone bar. With the microphones mounted not as you would do for the other two techniques.
Now when you record MS, the middle/center/Mono microphone, should always be facing the stage and in cardioid mode. Now that 414 has more than one cardioid type. The standard cardioid pattern is the optimal one to use. And of course, the side or what will become the left/right, it is generally directly below and perpendicular to the Mono Center Middle microphone. And it is in figure of 8 position. So the capsules are facing the left and right walls. You will always notice that your record levels between left and right will be significantly different from one another. In fact it is normal for the side channel (recorded as the right channel) to appear to be significantly lower in average level. So those side facing capsules are simply recording the ambience of the room. But when you combine both together and utilize the MS decode preset, it will sound like you had a pair of cardioid microphones right on the band, exactly where you wanted them. Now in the MS decode preset, varying the side channel level, will widen or narrow your stereo image. And if you listen to the single center mono channel, you're going to hear how bands were recorded years ago when only a single microphone and monaural recordings were made. And because of the phase trick played with that side channel microphone, if the stereo recording is collapsed to monaural, say on a simple monaural table radio, the side channel will cancel completely and you will be listening to only that single center mono cardioid microphone.
This same MS technique is used to bring you FM stereo and formally stereo analog television. That's the way the left and right channel is encoded, to be broadcast. Your FM stereo is not transmitted left and right but rather mono and side/sides. This also allows for different kinds of competitive dynamics processing to be used to get ratings with. And it can be used in different recording techniques for different purposes.
One of the important things of MS to remember is that if you engage the low-frequency cut switch on one microphone, you should do it on the other. If you engage the pad switch on one microphone, you should do it to the other. This provides for the best MS performance since MS relies upon the differences in the phasing. Any settings that are not identical to both microphones will present phase aberrations. And with phase aberrations, consistency is kind of down the drain. I know what I'm doing when I want to manipulate it differently. MS is usually done with matching microphones where the most predictable results can be had. But you can use all sorts of different kinds of microphones and polar patterns with MS for all sorts of wild different ways to record things in stereo. A lot of purists don't like MS because they think that phase cancellation is bad. And where in MS, it relies upon phase cancellation to work. And phase cancellation to make it all monaural compatible. So for folks who don't quite understand phasing and timing problems along with acoustic gobbledygook, most are too timid to even try MS. So I think you have the right and bold idea. I mean we are like on the same frequency... man. Wow so cool. Hope ya had a great Thanksgiving? Hope ya have a good holiday coming up in a month.
Let me know how it goes? Since you are so close to me, feel free to stop by my website and drop me a line or call. 202-239-7412 will get you my answering service and I'll get the message later and can return your call. I'm here off of 606 and 50. On the far side of Dulles Airport these days. Just a couple miles away from Sterling Productions which is now closed as of the end of May 2012.
I'm almost positive, while I don't know Adobe Sound Booth, I could probably show you how to do MS decode without any preset. But I really think it must have a preset? I've seen entry-level software that has it. You'll find it under something called " channel mixer ", in Adobe's Audition. Because it's really a way of mixing up a pair of channels into a completely different pair of channels, or so to speak. Basically, this is what people hate about software. In the control room as we just grab it patch cords and press buttons that are fairly obvious. Otherwise ya have to know your software about as well as your wife's private parts. Or at least should know.
So here's what you do, you sell the one 414 that has no partner. With that money, you purchase a small Mackie and a couple of bargain small diaphragm condenser microphones and give that back to the school. They'll never know the difference and neither will the parents. And that's your fee for recording. Just don't tell them you charged them. You could even be super gracious and get them 4 cheap small diaphragm condenser microphones. Really, honestly. Why? Because small diaphragm condenser microphones will sound much better when people don't understand how bad large diaphragm condenser microphones like 414's can actually sound. So you'll be doing them a beneficial favor. And ya get a refund on your taxes that way called 3124 and 414. Hey I'll give you two grand for all of that and the replacement equipment! And only because it will be BETTER. Better for them. Better for you. And better for me. After all, there is nothing state-of-the-art about that API 3124. That's old technology from the 1960s! And who wants that crap? Something new like a transformer less Mackie has better specifications. And it looks cooler! We could even get a Behringer mixer that has built-in reverb effects that are all digital! So much better than that old junky API stuff. I'll be happy to dispose of it for you. And the simply awful 414's that sound terrible with off axis sounds. They shouldn't have those either. I can dispose of those also when I throw my other 414 out. Yeah... I can be a big help.
MS on a single stand is good to take a stand on.
Mx. Remy Ann David
Wow... This is great information. Thanks.
Regarding the recording equipment, our band booster account paid for it. The county, like other school districts I presume, funds performing arts so poorly that getting any equipment or instruments is very difficult. The booster organizations oftentimes fund many large purchases. I get your point regarding the equipment being overkill; and that the money could have been spent getting more equipment with the same amount of money spent on the 414s and the 3124+. With more equipment we could be more versatile.
Thanks, again. Your information is very helpful.
Researching things seems to imply that I can make a Y cable for the side mice to decode the side channels by reversing the "+" and "-" path on one of the male connectors. On the preamp, I'd bring the front mic to channel one; and in channels two and three, I'd terminate the male connectors from the side mic and pan channel two to left and channel three to right. This should work, right? I'd set the output volume of channels 2 and 3 to suit the desired stereo width. The downside I see is not being flexible in being able to set the stereo width as I could be if I did the decoding in post-production software.
No, you don't need to use a Y cable. The side microphone is handled as a single channel. It is recorded as a single channel. The Y table would only be needed, if you were doing your MS decode, on an analog console. The patching for an analog console would be your left (mono center cardioid microphone) will normally be pan to Center on your pan pot. The side microphone would then go into say, inputs number two and number three. On channel 3, you would invert the phase and pan that to the right. Channel 2, would stay in phase and would be panned to the left. The two levels between those two inputs, have to be matched exactly. So if one is moved up the other also must be moved up etc.. And by raising and lowering those both together, in relationship to the fixed gain of the mono center channel allows you to bring in the stereo and by slightly lowering your center mono microphone, can widen the stereo image. All the way up to the point where the side channels are completely out of phase in the left and right channel, which will cause it to cancel it collapsed to mono. And so the need to keep the center channel mono microphone up at all times. If this manipulation of phase that creates this precise stereo imaging from these two channels with that technique on an analog console.
In digital software, the preset essentially does the same thing to the digital stream. Some software displays this with faders, four faders, in the window. Which is why in software inextricably comes under a drop-down menu indicated as something like " channel mixer ", as in, Adobe Audition. Where then you'll find the MS drop-down within the channel mixer. So there is no reason to record more than two channels. It's really quite a cool and neat trick that has other fascinating applications with other stereo microphonic techniques.
One of the interesting things I did recently was this film documentary. It included a great rock 'n roll track. But the vocals Fighting with the Dialogue. So Utilizing a Similar Scheme to MS, Where I Only Applied the MS Technique, between 300-4000 Hz. Everything below 300 Stayed in the in Phase Mono Center Image. Everything above 300 and up to 4000, Had the Stereo Image Widened by Quite a Bit. This through the Lead Vocal Basically Out Of Phase. This Allows the Vocal to Surround the Dialogue and Appears to Come from behind the Monitor in Stereo or in the Rear Speakers through 5.1 Decoding Simply from a Stereo Mix. This Land Was Able to Keep the Music Level Fairly High throughout the Entire Documentary Never Colliding with the Dialogue That Was Mono Center. So I Turned This into a 5.1 Surround Mix from a Stereo CD and Mono Dialogue. All Utilizing Filtering, MS and Special Processing of the Dialogue Track.
You understand how to start playing with phase and timing, you can make all sorts of cool things happen and/or disappear. And when the documentary may be listened to in mono, the lyrics but not the rest of the music, nearly entirely disappear but not by 100% but damn close. Which keeps it ever from conflicting with the dialogue even in mono. In the surround mix you are surrounded by the music while the dialogue is always front and center. That doesn't come from just equalization alone. Not possible. If you don't do what I did, the music track would always be low and tinkling in the background. I didn't want that. I wanted the music to be as prominent as the dialogue without the lyrics getting in the way so phase tricks lots of phase tricks.
Beam me up.
Mx. Remy Ann David
I fiddled with MS at a "Brigadoon" dress rehearsal, tonight. Since I am limited with the software. I did it in analog with a Y cable. I didn't make one; I bought it. It dawned on me, after replying above that I can use the standard Y cable and invert polarity with a button on the 3124+. I mixed the channels using the mixer on the API unit and recorded it on to a Marantz HD recorder. The recording environment was not ideal (nor was the mic placement). My main purpose for the evening was to help with sound issues, not to tweak recording setups for the wind ensemble concerts. What I recorded, sounded OK with headphones. Tomorrow, I'll listen more critically.
I didn't have to by new hardware for the mic stand. I had every thing I needed. Here's a picture:
At the next evening wind ensemble rehearsal I'll be more scientific than I was tonight about testing and tweaking things.
Well, you got those microphones mounted correctly. Basically, the idea is to only listen to that Mono Center cardioid microphone when you are placing that microphone. Because that's the full picture. When you find the right placement for that one single microphone and then bring up the sides, you'll probably get what you want.
Yes that's fine utilizing the Y cable into the 3124 that way. I have the oldest 3124 in the world, serial number 1/2 as it was the second prototype that Paul at API to my friend who owned the company, made. And then serial number 11. Mine have no phase inversion switches on them. So I either use an XLR phase reversal barrel on the Channel 3 side microphone. I also have some pre-built XLR Y cables. One of which, I simply reverse the phase of one of the two male XLR connectors and marked that cable as such. And that too works great. The only problem with it that way, is that you cannot change the stereo width, after-the-fact, without converting it back to MS, in the software. And when you are just working with headphones, you're taking a big risk that way. Because what sounds right in the headphones won't necessarily translate right, to speakers. Otherwise, it is what it is.
So to circumvent that problem, here's what I recommend you do. Do what you're doing so that you can monitor it properly when you are recording it. But instead of taking the 3124 mixer stereo output, just take the M direct output from the 3124 and the one in phase channel 2 side channel direct output to your other right channel record input. This way, you have the MS, isolated for decoding after recording. And then you mix it with your monitor speakers. That will give you flexibility that you won't have any other way.
You've got the right idea.
Mx. Remy Ann David
Thanks.... I'm looking forward to next the chance for fiddling. I'm limited with software. Adobe Soundbooth CS4 doesn't have a decoder. I need to get some software that does.
Unfortunately, I really don't know anything about Adobe's Sound booth? I still can't quite believe they don't have a method to decode MS? I mean this is basic stuff. It's been in most all software since the beginning of software. But it's possible you might be right? So, get a free download of Audacity. Look for an old shareware copy of Cool Edit 96, you know... from 15 years ago. It's free. They all have MS decode capabilities.
Hey! Wait just a cotton pickin' minute...You and I are in Herndon Virginia, I'll give you my shareware copy of Cool Edit 96. And then you can buy the beer.
Mx. Remy Ann David
Fun gig recording for Herndon; thank you for sharing. I record district events throughout Loudoun and Fairfax, and I also record collegiate wind ensembles. e.g. [[url=http://[/URL]="http://www.youtube…"]Frank Ticheli Symphony No. 2: III. Apollo Unleashed - YouTube[/]="http://www.youtube…"]Frank Ticheli Symphony No. 2: III. Apollo Unleashed - YouTube[/]
If the meters never peak higher than -12 dbFs on the Marantz, you are doing fine. You can raise the level in post. If you aren't doing anything to edit the recording, you might consider recording at a higher level and/or putting a compressor between the API and the Marantz. Definitely explore post-production options. Cutting the concert into a listenable CD and adding a little reverb can go a long way.
With regards to your sample, pay attention to the balance between left and right and pay attention to the balance of the ensemble.
Thumbs up on having supportive boosters!
Thanks for the comments. Great video. I love wind ensemble music; and can't get enough of it. Wondering - What was your setup? The video shows, what appears to be ORTF with additional mics on the left and right of the stage in front of the ensemble and one to the right of the tympani.
Cutting CDs: We have been doing that for the families of our ensemble members. Though the recordings sometimes are not so good, the families like them. I'll be doing the recording for Herndon HS for the next couple of years; and I'd really like to take some big steps towards improving the quality of recordings.
Left/Right Balance: Normally, I do do some tweaking after the concert. The recording above was relatively untouched with post production. I wanted feedback on the "raw" recording thinking that if I can improve the actual capture during the concert, then I will have an easier time in post production, which is a workflow I'd like to learn more about.
If you are doing the FCPS District 12 Chorale concert, I hope you wouldn't mind my introducing myself to you (my son will be there). I'd love to see the equipment you use and to see your set up.
Thanks, again, for the comments.
Just how many of us are meandering around Herndon Virginia? LOL I mean we should all get together for some pizza? After all it is the holiday season and we should meet each other? We could even do a late-night over at the 24-hour diner on Elden Street? I'll show you my equipment. I could even show ya how I'd set that Herndon high school job up. And we'll use all three 414's. And heck, I could bring out a fourth? Which was basically the way I used to do those jobs over there anyhow. And on that precisely the same equipment. Do you have any tall, collapsible aluminum microphone stands? Mine are like 12 feet high fully extended. Are you using a multi-conductor snake or, just individual long cables?
202-239-7412 and please leave me a message on my answering service.
Mx. Remy Ann David
Neumann sold a matrix transformer for use with M-S recording techniques. That is probably no longer available. The schematic of a M-S setup is on line at the Neumann site. (Look under the SM-2 microphone documentation.) However, classic audio transformers (i.e. UTC LS series) can be configured for M-S. The transformers may also block RF on mic lines if the recording venue is near a radio station.
Remy...Haven't set foot in Herndon HS for over 30 years. There was a band director there who was also an audiophile.
I recorded a performance there a LONG time ago when analog tape was the medium...