00:00.00 archpodnet Okay, welcome back to the arche architectch podcast episode two zero five and we are talking about photogrammetry and gis in human occupied digital landscapes again, take a look down at your show notes for a link to this article and we are talking to Andrew and Sarah the authors of the article. So in the first segment we talked about. A little bit about the why anybody would do something like this but now I want to talk about the how because this is where I think for our audience in arch architect this really gets to be the interesting part and we've talked about photogrammetry a lot on this show. We've talked about drones a lot on this show any longtime listeners know they have to take a drink now because anytime we say drone. That's what happens seems like they come up a lot on this show. Ah so hopefully you're listening at a good time. Yeah, but so anyway you you mentioned you mentioned that you know drone ah drones and and cameras and and things like that have been used to do photogrammetry for for many years in archeology now and even. 00:40.14 Sara Are. 00:43.85 Andrew Reinhard That's what drones do yeah. 00:58.14 archpodnet Just saying drones have been used for many years is ah is a pretty powerful statement. It's kind of weird to think that they still feel nude in some cases but they're very much not in other cases but with photogrammetry and I've done drone photogrammetry I mean you really, you need the you really need the I mean don't need but it's helpful to have the ground control points and then you've got the. 01:00.66 Sara Um, yeah. 01:17.15 archpodnet You've got the spatial nature of the um I guess of the photographs themselves and I've even done it where you take high resolution video and then chop images out of that. It's not quite as good as taking high resolution images but you know we've we've experimented with that and in some cases and ah I'm just wondering you know getting to this. What methodology did you use to get the let's just call them photographs really screenshots of the worlds of fortnnight and no man sky in order to be able to do this actual photogrammetry because you ran you ran these just like you would any other images through metashape to produce your Dms in your 3 d models. So I'm just wondering. You know how did you come up with a way to get these images in ah in an accurate way to be able to do this. 02:01.20 Andrew Reinhard Yeah, um, I'll do a very short answer and then and then Sarah if you could do the detailed answer because this is when I really started to lean on you and I so was emailing you twice a day every day for weeks you know because you know I have access to the games and. 02:02.72 Sara Um, yeah. 02:14.34 Sara Um. 02:18.27 Andrew Reinhard You know when you play a video game. Every video game is different. Every video game will have its own way of of maybe doing a photo mode or photo capture video capture or you have to do it through the consoles native video capture and photo capture software that comes with it and so you know was like well okay. 02:34.90 archpodnet Ah. 02:36.72 Andrew Reinhard You know I have these 2 games they behave very differently with the photo modedes and everything. So what's my camera. How do I use it can I can I fake it out and pretend I have a drone but it's really not a drone. It's it's you know and like in fortnite you actually become the drone. 02:50.95 archpodnet Oh. 02:52.47 Andrew Reinhard Um, so you have this all-seeing eye that you can move around and then a no-man sky you're dealing with like a sphere. It's like a hemisphere and and then your camera's at the top of that and then you shoot around and ah. It was a disaster I was I'm a terrible drone pilot and a bad photographer when it comes to this kind of thing and so Sarah how did you help me because you helped a lot. 03:15.66 Sara Um, yeah, the way that I ah so I wasn't sure if this was going to work out or not and the way that I addressed it was as ah, any other photographry project that I carried on in real life and so um I. 03:26.40 archpodnet No. 03:32.47 Sara Found out that flying drone and then extracting frames um to use Apps Photographs don't really work out I prefer to spend a little bit more time um in taking the pictures but then they're gonna work out. 03:42.19 archpodnet Right. 03:51.89 Sara Much better in the software and so I asked Andrew to just take a picture pause while taking the picture and then move the drone the virtual drone a little bit forward um, taking care and making sure that the next picture will overlap at least at least fifty sixty percent with the previous one. And just go on and on until he would cover the entire surface that we wanted to um, do to photogrammetry for. 04:20.14 archpodnet Okay, and there was no, um, no distortion or anything at the edge of the images I mean in my my experience in the past Sometimes they're a little bit fish eye. You know, kind of on those pictures and things like that did you guys have any problems. 04:30.81 Sara I mean um I mean there's always a distortion. Um, but it wasn't that bad. Um, if I remember correctly it wasn't that bad so it was not like ah a whole. 04:36.41 Andrew Reinhard Um, me. 04:36.52 archpodnet Sure. 04:41.38 Andrew Reinhard Um, yeah, yeah. 04:47.32 Sara 60 you know that kind of distortion that you see um the parable or anything like that now you know. 04:48.83 archpodnet E. 04:54.40 Andrew Reinhard Yeah, we didn't have to fix too much of that and especially like with fortnite. There's no curvature of the earth. It's flat. You know, ah granted is got topography but but but yeah and you know shooting from the top down and then moving things around and then making sure we had the overlaps. 04:58.66 archpodnet No sure. 05:10.88 Andrew Reinhard You know seem to have taken care of that. Um, and it was fun to kind of hand the files off to Sarah for her to do her magic hence whatever whatever she did on on on her pc ah you seemed to seem to work out. Okay. 05:12.66 Sara Are. 05:16.28 archpodnet Yeah. 05:17.94 Sara Ah, my magic. Yeah yeah, like we tried a couple of oh sorry, yeah no I was just saying that we tried a couple of times and I think the first set of image was um. 05:22.16 archpodnet So regarding that Sarah no no go ahead. Go ahead. 05:36.35 Sara Wasn't working and but the second time like I give I guess more detailed instructions and Andrew was great in taking the pictures and that work out perfectly. 05:47.40 archpodnet Nice You know I'm wondering um with Meta shape I mean obviously a lot of us that are listening to this have put images into metashape before but obviously they're usually images of the earth and or objects and things like that but did it did you have to it. Did it just. 06:01.39 Sara Um, yeah, but. 06:05.18 archpodnet Did it just accept these these digital images that that don't necessarily look like you know anything on the Earth It just accepted them and and did it like normal and did you have to use any special settings. 06:14.74 Sara No, it actually worked I was very surprised but I guess this digital spaces are very very accurate so that they resemble real spaces and the software just picked them up. 06:26.16 archpodnet Um, right? Okay, nice. 06:29.89 Andrew Reinhard Yeah, it was a gamble you know is this one of these things. It's like you ask this? what if question and it sounds really crazy. But it's like it's so crazy it just might work. You know it looks like a mountain. Let's see what happens and and it did it did it. 06:43.80 Sara Um, yeah now I was very very surprised about it because I I mean I don't know how specifically you know the software works like I know what I have to do to make it work. But I don't know what the parameters inside the you know the the algorithm or whatever how they are set up so I don't know the tolerance in my personal experience with real world. The tolerance is pretty high ah more and more with the you know the new updates and so you know. 06:58.81 archpodnet Ah. 07:16.95 Sara Light errors that you might have and they would create would have created um issues in the past like different light. You know if one image is darkened than the next one. Maybe it won't um you know, match them because the the R to B Values of the 2 Of the pixel the same pixel in the 2 different images is different and so it doesn't recognize it. But now it's it doesn't even matter anymore like if it's not completely off even if it's slightly different. It still works. Um, and I think. 07:37.43 archpodnet Um. 07:48.74 archpodnet Nice. 07:52.57 Sara Really think the the trick here to make it work is to give the software. Um, even if it's ah it's a digital landscape Even if it's a fake mountain if you give it enough like angles like if the pictures are take it with the right angles. It will still recreate. Um the shape of the mountain. Even if it's not real real. 08:17.39 Andrew Reinhard Yeah, that took a little bit of doing though I mean I had you you had me go back and shoot stuff again and let's do that again and try it again. Let's try his ways so you know one of the beautiful things about the project is just it. It demonstrates How technology is really iterative. 08:23.81 Sara Um, yeah, yeah, um. 08:34.79 archpodnet Yeah. 08:36.24 Andrew Reinhard And the archeological approach or the digital archaeological ah archaeological approach is it's it's you take steps and it's like okay and you just tweak and you tweak and you tweak and then you finally get it. Hopefully. 08:43.82 Sara You know? yeah I think I I might have asked you to go back and take some oblique images to yeah to kind of complete the set of the Zennet picture. You know the one from above straight from above. 08:47.89 archpodnet Did you? um. 08:52.43 Andrew Reinhard Um, you did um. 09:00.13 Andrew Reinhard Um, yeah, yeah, yeah. 09:02.00 Sara And. 09:02.62 archpodnet So how did you? you know you you produce this let me start over so you produce this this photometric model and you you bring it into a Gis and you've got this. Well first off in Metahape you've got ah you've got a model that has spatial components to it. How did you apply units to that. 09:21.90 Andrew Reinhard This was this This was the the big question is because almost it's like oh my God it's like we get to this point we're like how many meters is this what? what is a foot you know because. 09:22.15 archpodnet Right? because I mean how do you even know what the units are in the online world. 09:23.39 Sara Um, right right? Yes, this was the funniest part. 09:33.56 Sara Um, exactly. Um. 09:39.23 Andrew Reinhard You know this stuff is like totally made up and so um, we we had to believe that with both of these games that they were designed for an average sized adult human being um I don't know you know less than two meters tall 09:40.11 archpodnet Um. 09:56.31 Andrew Reinhard And let's just go with that now a Noman sky they have units which equate to meters and so you can kind of do a and I actually did this is that you can do a like a field survey. 09:59.93 archpodnet Oh. 10:08.83 Andrew Reinhard In the game and you can have a starting point you set a little beacon and then you start at the beacon and you walk into onto the horizon to a point and when you finish you can see how far you walked. It's like oh okay, that's cool, but not so much in fortnite fortnite. You just kind of had the had to make it up and take a best guess. Um, so. 10:18.34 archpodnet Right. 10:26.89 Andrew Reinhard Yeah, dealing with units of measure in video games does not really exist I mean they have it under in the underlying code right? You know because you have to have things make sense when you're doing the design but they don't really impart that to the public. It's just part of their gaming mandate or whatever. 10:30.75 Sara Um, yeah. 10:34.47 archpodnet Ah. 10:41.49 archpodnet Well I noticed I noticed in the article you read one of the I don't one of the updates or additions of fortnight had Indiana Jones there's a known quantity Harrison Ford has a predictable height. So yeah. 10:49.53 Andrew Reinhard Yes, I thought precisely and it you know that's one of the wonderful things about archeology is that it's you know all of it's got so much synchronicity and so you like having Indiana Jones show up while we're doing this project. 10:51.89 Sara Um, yes. 11:09.15 Andrew Reinhard It was just a godsend. It was like oh my God Yeah I can't even believe it. 11:10.65 archpodnet Ah, ah you know it speaking of distances too. 1 thing I would wonder and I don't know if you were able to suss this out or not because they're like a no man sky. There are some flat plains but a lot of it's like rolling hills and things like that that distance measurements back I'm I'm familiar with that from you know playing this a little while ago. Um. 11:13.40 Sara Um, this. 11:28.92 archpodnet Do you know if that's a linear distance like line of sight distance from you to the target or is that taking into account the landscape the topography. Okay, right? okay. 11:37.67 Andrew Reinhard No, it's it's line of sight kind of as the crow flies ah distancing so it's going to be so pretty accurate, but not 100% accurate and I don't know that we would be able to get 100% accuracy at least not yet anyway until we figure that out. 11:49.80 Sara Are. 11:49.90 archpodnet Um, okay, right right? So back to the games. What made you choose the oh sorry I think I cut you off Sir go ahead. 11:56.49 Sara Um, yeah, but then you know. 12:02.46 Sara And Nonravo I we just wanted to add that. Um, once you accept that is going to be an approximation in distance and and that was not even you know it wasn't the goal of the article to be precise precise we just wanted to. See if the project was possible or not but you know once you approximate the distances we just marked them in a way that I could see them in the pictures and instead of using coordinates from I don't know total station or Gps or whatever that I didn't have I just scaled the model. 12:21.85 archpodnet Um, yeah, um. 12:38.30 Sara With those distances. Um, and I treated it as a regular you know a regular Cloud point that I that needed to be scaled. 12:39.60 archpodnet No. Ah, yeah. 12:48.95 archpodnet Okay, so regarding the online worlds of fortnite and no man sky. How did you guys decide really I guess Andrew how did you decide what you were going to map for this that would make a good representation for testing your your guys' theories. 12:59.53 Andrew Reinhard Um. 13:05.85 Andrew Reinhard Yeah, um, you know with with fortnite. Yeah, we knew that we were doing the project during a season which meant that the island itself wasn't going to change for those of you who don't play fortnight you play on a place called battle royal island which is this kind of crabshaped island. And every three months things happen to it. You know there's a volcano or then there's an implosion in the middle of the island goes away and the climates change and stuff like that and so you know being able to document it one season and comparing that to the next is always something very interesting. Um, but but for for me I wanted to play over several different rounds because the rounds are always the same. You start you land as a player on the island you find out other players to eliminate and in order to protect yourself. You build things and so I wanted to go and play a bunch of rounds and then be able to record what people were being what people were building and where they were building them. And in relation to other landmarks other landscape features to each other to see if we could draw some conclusions potentially and so you know that was that was pretty easy to do because people just build and you just film the building. Um, and then you map it now for no man sky because it's procedurally generated. Um, you never know quite what you're going to get so I wanted to find a settlement that was simple I didn't want to do like a big complex or anything at least not yet I just wanted to find something that had a simple structure that was anchored in the landscape and then I wanted to see if we could you know measure it if we could place it in the landscape and. 14:37.39 Andrew Reinhard And and map it and so I looked at about 5 or 6 different candidates. You know I've got a kind of a gette that I keep because I'm a big no man skyhead and so it's like oh well, we'll go to this planet and this person lives there and so you know check to see if they're home. Um, and they weren't so you know we. We went and we got our got our video done got our photos done. 15:00.74 archpodnet Nice You know I'm wondering if ah, so you guys were going into these worlds and and taking these you know just doing this and I know there's a section on ethics in the in the paper there where you talk about you know, reading the terms of service and being conscious of you know the fact that there's other real humans in there. And and things that they created and and and along those lines plus Also the game designers I mean these are proprietary environments that they get the game designers I mean technically own and can shut down at any point point in time but they do have provisions within their tcs to um to. 15:28.00 Sara Um, yeah. 15:37.74 archpodnet Allow for you know online play and and you know things like that which which made this possible but I'm wondering. Did you guys contact or do you think you'll contact in the future. The actual game designers themselves because they I mean they have all the source material they have they have what they know the measurements are they have you know probably. You know maps of all these areas and and a lot more information that could ever be gained by 1 person going in there and doing a study on this have you thought about involving them in this anthropological study so to speak the bigger picture. 16:06.86 Andrew Reinhard Um, um I have and ah, um, actually when when the archaeo gaming book was getting close to publication. We needed to get permission. 16:07.87 Sara Yeah, we talked about this. 16:17.50 archpodnet Um. 16:19.50 Andrew Reinhard For the cover image which was a screenshot from a procedural world in no man sky and so it took me about two months to actually get a response from hello games and they actually did and the the Ceo actually was the one who responded which really blew me away to give me permission to go ahead and use the image I'm like wow this is awesome. Um. 16:30.77 archpodnet Um. 16:38.28 Andrew Reinhard They are super busy and like trying to write Epic games. You know to get info about fortnite. Yeah I have a feeling that we'd be signing Inda's out the wazoo and we wouldn't be able to necessarily publish on that even though we would have the information and and. 16:52.85 archpodnet Right. 16:55.48 Andrew Reinhard Yeah, they don't really have a lot of time for archaeologists. It might be a curiosity Um, but you know that's that's normal. Um you know because it's a business you know for for these Alea games. Anyway, I mean for if we were to do more of an indie game or something like that. Ah you know like you know working with maps and season. 17:04.77 archpodnet Right. 17:15.30 Andrew Reinhard You know, for example, then we could write and we probably get a response and and actually have a collaboration which is something that would be super awesome because it connects the the company with the community with the archaeologists and then we can do some public archeology there. Um yes so Sarah I don't know if if you've got. 17:25.42 Sara Are. 17:34.15 Andrew Reinhard Experience or if you've tried to reach out to some of these different places that you've been interested in. 17:39.61 Sara Um, um, I've tried um and I tried twice and once I had a a very good response. Um, but it was more like ah as you said, an Iny um production. And they were very interested and thrilled that I used um their game and it talked about their games in my research, um, the others the other company. Never even responded to me. 18:06.11 Andrew Reinhard Yeah, that that's pretty typical. It's not that they're being rude. It's just they're super busy and you're low priority. Yeah. 18:08.81 archpodnet Um, okay. 18:11.80 Sara Um, yeah, yeah, so I don't know I don't know I guess that we will have to make a to make a compelling case. Um for them to be interested. You know, maybe if it can be something for business exactly like you know, um. 18:12.38 archpodnet Okay, all right? Well let's take a break What? m. 18:21.17 archpodnet Ah, yeah. 18:22.78 Andrew Reinhard Um, does it make the money and they're interested. 18:28.21 Sara Ah, Ubi South opened um, historical department for their games. Um, and that became like another um business for them. So if we can make a similar compelling case. Maybe we could start a collaboration that would be great. 18:29.67 archpodnet M. 18:40.25 archpodnet Um, um, prompt. 18:43.62 archpodnet Yeah, okay, all right? Well, that's a good point to take a break and we'll come back on the other side and and talk about a little bit more big picture and next steps for this back in a minute. 18:44.50 Andrew Reinhard Um, yeah, yeah.