[Music] Good morning. You're watching the Aussie BIM Guru. And today we're looking at face analysis grids in Dynamo. So, 

essentially we've looked at how to build a feasibility study in Revit and Dynamo and Power BI in a previous series. Feel free to watch that if you haven't already. It's on my channel in a playlist. But today we're going to look at how to 

convert our results from the sun study into a face analysis study instead. So essentially we're taking some results from sensors within our study and we're going to be - we've already done this part, so we've sent sunrays to our sensors to count how many hours they're receiving 

using vectors and times of day and a Ladybug package. And we've ended up with a bunch of results on a facade. But obviously this doesn't look very nice. It's all just orbs with color based on a filter, so it's not really suitable for a 

report for example. So what we're gonna do in this study is convert this into a face. So we're not going to use any custom notes today, but we are going to be using Dynamo. So you want Dynamo, you're also going to need some results 

on a surface. So these can be like a family, like the way I've set mine up, it can be a lot of different elements that can hold data which you can tell Dynamo to push back onto a surface. So I'm going to show you how to do that today and the 

surface will need to be on a mass face analysis doesn't work on anything except for massing from my experience there might be other categories in Revit that can handle these studies but I haven't found them yet so for versioning I recommend that we

use Dynamo to point X at the very least but ideally 2.0 point 3 is a good version I know that Ladybug doesn't work in 2.1 at the moment so probably two point over went 3 and 2019 should work

without further ado let's get started and jump straight into my model so I'm essentially picking up where I was during part 9 of my series so I have a model here with a bunch of phases with results upon them and your room you might remember these are adaptive

components placed at a point so it's that's my computer that so essentially it's a an adaptive component placed at a point and within this adaptive component there's a bit of information stored so one of these things is the number of Sun hours that the sensor received in this

parameter here and also the comment one and a comment two are basically telling you which study it's a part of and which side of the building it is you can see here this is the north side of my building for example because you can only do one face at once using face

analysis so we're gonna isolate these sensors based on those parameters in our script yeah well let's jump into Dynamo so I've just got a fresh script here and I'm in manual mode at the moment so the first thing we're gonna do is get the

family types and we're gonna find out the detector family or whichever family holds the data that you want to convert to a surface. So mine is myself, my sensor, in this case, it's called detective, but I'm going to get all the family instances by family type, so it's just

know here the fifth one down, if you feed this in, this will give you every instance of that in your entire project. So if I run this, I should get a list of about 1500 sensors. Obviously, I don't want to use all of these sensors, so I'm

getting some data about these elements to filter it down. So I'm gonna get parameter value by name, and I'm gonna run this twice in order to filter down my sets, firstly based on the option that they're related to and then the face that they're related to. I'm gonna get a

string as well, and I'm just gonna go for my comment field. And if you were building this for Dynamo player, you'd make this an input. I won't build for Dynamo player today, but this is the first input you use. You basically set your two parameters as an

input, and then the two values that you're filtering for is inputs as well. And then the other input would be the type of family for the sensor and the face you're analyzing, but I'm going to pass through those today just to save a bit of time. Okay, so what we need to

check is, does this parameter equal a certain value? And the value that I'm looking for in this case is option one, be. This could be any value in your study, maybe you don't need to filter your data twice, maybe you only need to filter it

once maybe you don't need to filter it at all maybe you're looking at all the elements but essentially these should be the things against the surface that you're analyzing so we're going to use a boolean mask in order to filter our data down so we're gonna take those family

instances and based on their value we're gonna filter them down so essentially everything that goes out of our analysis now belongs to study 1b I think at the moment I've only done one study so we should expect that all of them are gonna

be in there but if your future proofing your script and you're basing your feasibility study on the method I showed you this is the way to do it so what we're gonna do now is just take another set of that essentially and filter it

down one more time so we're gonna take our in list I'm going to filter by comment too in this case and I'm gonna filter for North Tower in this case so now we're just getting on the North Face so now I will actually get some level of

filtering on my sensors or as a result so we'll take our in list and more mask based on that so now I should only get the sensors on my North Face so you can see now we are actually are getting rid of quite a lot of sensors in our study

about 1200 of them are being excluded from my study so that's great okay because we're gonna be applying this to our face so I can take a little bit of time if you have too many things okay so we're gonna do is just build a little

naming algorithm based on these two things that we've picked so we're just gonna say that I'm naming algorithm is X plus I'm a colon space ' plus y because we're gonna build a name for our surface analysis study so let's just make it our option and then

our so we're gonna get option one B semicolon North Tower and you'll see why I do this in a second okay so what we want to do then is get the location of all of our sensors so if you look for element location and just go down to get

locations we get existing elements location we're getting at the location of a lot of sensors so I'll just do it  and there you go that's my sensors in Donovan and then we're gonna have to get our sites so we're gonna do a

get face node will select face and I'm just going to temporarily minimize this window so I can go into into Revit and get my face so I'm gonna pick this face here so this is my study face and this is where my analysis group will be

applied to okay so the first thing I'm going to do is they're going to find the closest point to that face so we're going to use closest to point two so from that face to each of these points imma gonna lace it to longest so that

every single point finds out the closest point and essentially we should get a set of points on that face because these sensors were actually pushed away from the face just a tiny bit so we need to go back to my surface and find a point exactly on that surface so it

understands the parameter to be applied on the study at that point on the surface, okay? From there, we need to actually go and we need to map some UV coordinates. I'm gonna show you why now, so by gonna use a node to create our face analysis study. So we're going to

look for face analysis display, so there's a few sets of nodes that relate to face analysis study. In this case, we're gonna use by face points and values, so our face points are essentially these points that we've mapped onto our face, and our surface is

the surface here. So we're gonna get the best thing, we're gonna get a view. I'm just gonna freeze this node for now because we don't want to run this yet. So we're gonna get current document, I'm just gonna search for it. That's quite a

lot of custom packages with current nodes. So the current document and active view, it's always good to make sure that you're using the native active you node where possible, obviously, that means is less custom packages required to run your script. For quite a common node, some

packages do require you to use their version of the current document because they convert their, their language into something native to their package, such as a awkward for example. Anyway, I digress. So we're now going to go and feed in our surface as our test surface. You can see

that now we need sample locations, so we have a fools. So our samples should be, sorry, sample locations, they're gonna be a little bit tricky. So we need to find the UV parameter at those points because if you check sample locations, it's basically saying the locations at which

you want to create the analysis values, but it's not receiving them as points; it's receiving them as UV map points instead. So what we need to do is we need to run a list map upon our points, so we're going to find out the UV parameter

at point, my test surface, so UV for ammeter at point. So for every single point on our surface, we want to find out its UV. So our surface is our input, but we need to list map this because we want to do every single point. So we're going

to use a list map, and we're going to be running in each point at a UV to this face, and then we're gonna run this, and we should get a UV every single point as a result. There we go, and then we just want to flatten this

list, and it's important that you keep the list structure in the same order throughout the script because we're going to go back and pull our values from my senses as well. Okay, so that's our sample locations, but now we need our samples. So this section is quite easy, so

what we need to do is get our ammeter. So in this case, this is my Sun hours. So I'm just gonna get another get parameter value by name, and I'm going to go to my senses after I filtered them, and I'm gonna search for the Sun parameter value

which in my face was AVG_data_somehow_Sun hours one. So essentially, we should get a list of values now that I can speed through as my sample values, which will guide my surface analysis grid. Okay, so that's my samples, and then all we need is a name

description and a unit type so for name and description I'm just going to use my name and any algorithm that I built earlier and for unit type in this case I don't have any specific units that I've used it's just a number so what you can do in this case is

actually just input the value null but you could use a unit of measurement if you knew you're using a specific unit like meters for example or volume so it just comes down to your study obviously okay so at this point we're actually good to run the study so I'm just gonna

unfreeze and we're ready to go so we've got our face nominated we've got the lost answers ready what I might do is just close and reopen to refresh all my notes and I'll run the study so if I get run it should apply all this data into a

surface analysis and hopefully you should see something like this rivet analysis display face analysis display is basically the nice way of telling you it works and if we go into rivet straight away you can see that I've got some results coming up and that's

because I set my default analysis study sample to just a new surface analysis display so I probably to explain surface analysis settings first I'm just going to hide all my senses so this is sort of what a surface analysis display looks like you can see on my dynamic points to

run at the moment so I might just go turn those off because they're sort of getting in the way so I need to turn off closest point to and I'll turn off my face as well okay so you can see that we've got a gradient which is ranging

between a certain color range so you can see over here we have a little legend that's popped up as well but the way that this is all being triggered is through the analysis display settings in the view properties and I actually set previously a default analysis display

style usually this would be set to none so you wouldn't see anything straight away that you would need to come here to edit and then you can create a surface analysis study here under analysis display style you can see my study ranges between seven hours and zero and

there's no units applicable to the study obviously you can turn it on and off at will and this is a view specific study it won't appear in every single view on your project just the one that you conducted it in what we're gonna do is

create a new style so the first thing we can do is just have a look at some of our basic settings I'm not just zoom in so it's easy to see how these things impact the study most of you probably haven't actually seen this area of Revit

before so if a new technique for most of you it was quite new to me until recently so one studying you can do is just add grid lines and essentially this will show you the surface analysis division of how the study was conducted by samples so you

can see exactly how many samples were taken to generate the study you can also change the color of the grid you can change the transparency of your study as well I don't usually like changing the transparency because it will actually change the transparency of the face that

You run the study on, so it looks a bit strange because usually you see through it. It doesn't make a lot of sense. I haven't noticed what contour does yet, to be honest. I haven't noticed that have an impact on my studies. If anyone knows,

feel free to leave it in the comments or it's probably on Google too. If anyone is interested, let's just make a grid black for now so that it's more obvious. Okay, so what's important now is to go to color, and you can either work on a gradient or

a range. I'm gonna just work on a range today because I prefer ranges myself. So I'm gonna just add, we'll add six rows, so I'm just gonna add six rows, and you'll see that now we have just six steps in our gradient. I could immediately apply

this and just accept the colors. What I'm going to do is set all my values, so I know my minimum is zero, so my first value I need to specify is one, two, three, four, five, six, and my maximum is seven. So if I apply that, you'll see that

I've already got a seven-stage color scale. Obviously, you can change these values as well, and you can do a gradient as well. A gradient goes between the colors, or as a non-gradient typically is more separate between the samples. So if I turn off my grid lines, you'll notice a

difference, a bit more. So a gradient is more smooth, a range is more stark. So you can see that that's much more clear. I prefer a range because I think it's more accurate, and it represents the true samples, whereas a gradient is a little bit

misleading because you haven't actually tested the whole way along that gradient, so if you're working with very few samples it can be quite misleading. What I'll do is I'll just add an extra value for zero now. We have full control over our samples, so for my study I used

colors a little bit like this, so I start with blue. I'm gonna work my way through greens and eventually I'm making my way up to yellow, so I go through greens, I go up to an orange pale yellow and a bright yellow. I don't quite use those, I just

a bit more or less and you'll see that as you change those all your samples change as well, so pretty cool, right? You can also turn off the legend if you don't want to see it and you can modify some properties about it, so you can

change the text sizes present in the legend and you can change the rounding of the figures and that pretty much customizes everything. You can change the height of your color samples as well. I haven't actually tried that, but I assume it just makes everything a little bit

different. So by default it might, the CMO a default I guess if you don't specify, so if you go 0, you see I've got a fair bit of control over that legend in terms of its proportions and its size, so pretty neat. And so I found this a really

handy technique because what I do from here is basically I conduct multiple studies and I fix my view to a specific angle, so what I'll do is I'll just, I'll turn off my legend and get that out of the way and I'll just show you what I

usually do so what I'll do now is I'll lock my view orientation and then typically I just crop my view down to a specific size and I pretty much just build up a couple of images to composite in Photoshop cool so I've got one study

there I'll see it's not quite the same color as I used before um you can also see I might just turn the grid on I think it looks a bit yucky without the grid cool let's just turn that on looks a bit more technical cool it looks much better and

you can pretty much just export these out as image passes and composite them in an image software such as Photoshop easily I just convert these to say usually an increment of two so I'll just pick 2048 by 2048 as a PNG save it somewhere

let's just call that one we'll just start breaking up Photoshop in the background and then what you can do with Dynamo the great thing like this is you can just go and run another study so I can just pick a different face and a different set of senses so I'm just going to

change pick a new face and I can say West Tower instead of North Tower run my study again and there we go now I have all my samples in that face as well and I can just export that I'm going to offer you turn off my turn off my mask as well if

I just want to isolate that for Photoshop sake can't because it's a part of the mass sorry I stand corrected don't look sport no PC images and you just take the other one so what's the same in that case I'm just gonna call that

two and then in Photoshop I can just go and grab my first pass and my second pass and all I have to do is just a little bit of sneaky less using obviously I'm being quite rough I was applies you take a bit more care

than I do probably better ways to manage this with layers and masks as well there you go so you get like a full full study across the face of your building so that that's a much more effective way of communicating the results of the study

using face analysis and so hopefully that sort of helps give you a new technique in Revit and Dynamo that you probably want to wear off and it helps you get better results for your clients and yourself so thanks for watching today there'll be more videos on

feasibility and future I'm not sure when the next one will be but there are some topics I have mined up so if you're interested in this series and they'll be more for you so thanks for watching today and I hope you enjoyed this tutorial if you've got any comments or

feedback feel free to leave it in the comments if you're gonna need feature requests I'm always happy to hear them as well getting a lot of collaboration from my followers which I'm really enjoying and so thanks for watching today and if you're not already

following up subscribing feel free to do so and I'll see you in the next video thanks take you you

Face Analysis Grids using Dynamo!

Aussie BIM Guru

Dynamo의 얼굴 분석 소개 및 개요

Dynamo의 얼굴 분석 그리드에 대한 호주 BIM 전문가의 동영상을 소개합니다.

Revit, Dynamo 및 Power BI를 사용한 타당성 조사에 대한 이전 시리즈를 참조하세요.

태양 연구 결과를 얼굴 분석 연구로 변환하는 방법에 대한 설명입니다.

초기 태양 연구 결과의 시각적 표현에 대한 비판.

다이너모 사용에 대한 언급과 표면에 대한 결과의 필요성을 언급합니다.

사용된 모델 및 적응형 구성 요소에 대한 설명입니다.

매개변수를 기반으로 한 센서 분리 설명.

패밀리 유형을 가져오고 매개변수를 기반으로 데이터를 필터링하는 단계.

이름별 매개변수 값을 사용하여 센서 데이터를 필터링합니다.

부울 마스크를 사용하여 연구 및 얼굴에 따라 데이터를 필터링합니다.

Dynamo에서 얼굴 분석 연구 만들기

표면 분석 연구를 위한 네이밍 알고리즘을 구축합니다.

Revit에서 센서 위치를 가져오고 스터디 면을 선택합니다.

얼굴에서 각 센서 위치에서 가장 가까운 지점을 찾습니다.

'얼굴 분석 디스플레이' 노드를 사용하여 얼굴 포인트와 표면을 지정합니다.

UV 좌표를 매핑하고 목록 맵을 사용하여 각 지점에서 UV 매개 변수를 찾습니다.

샘플 값으로 태양 시간 매개변수 값을 가져옵니다.

학습의 이름, 설명 및 단위 유형을 지정합니다.

연구를 실행하고 Revit에서 결과를 확인합니다.

표면 분석 디스플레이 설정 사용자 지정

표면 분석 디스플레이와 그 모양에 대한 설명입니다.

뷰 속성에서 분석 표시 설정에 액세스하고 수정합니다.

격자선, 색상, 투명도, 윤곽선 등 설정에 대한 자세한 설명입니다.

분석 디스플레이에서 색상 범위 및 그라데이션으로 작업합니다.

여러 연구를 위한 포스트 프로세싱 및 워크플로

이미지 내보내기를 위해 뷰 방향을 잠그고 뷰를 자릅니다.

이미지를 내보내고 포토샵에서 합성하기.

다양한 얼굴에 대해 여러 연구를 실행하고 결과를 내보낼 수 있습니다.

여러 연구 결과를 포토샵에서 합성하여 종합적으로 볼 수 있습니다.