# SLAM will never replace terrestrial scanning – Aden Zhang, XGRIDS - Show: 2D Is Against My Religion (2DIAMR), host Michal Gula - Guests: Aden Zhang, Sales Director, Asia Pacific, XGRIDS - Published: 30 Oct 2026 · 24 min · Expo & conference podcast · recorded LIVE - Topics: SLAM, LiDAR, Gaussian Splatting, Reality Capture - Web page: https://www.2diamr.com/watch/xgrids-aden-zhang/ - YouTube: https://www.youtube.com/watch?v=iMXVcd9TUGY - Spotify: https://open.spotify.com/episode/0kxoMgVdgasAyVg9bOF9a2 - Apple Podcasts: https://podcasts.apple.com/us/podcast/slam-will-never-replace-terrestrial-scanning-aden-zhang/id1878399775?i=1000798851747 ## Summary Recorded live at INTERGEO 2026, Aden Zhang, Sales Director for Asia Pacific at XGRIDS, walks Michal Gula and Tomáš Barnáš through the new XGRIDS L3 handheld SLAM scanner, explains why XGRIDS tells clients SLAM will never replace terrestrial scanning, why newcomers struggle with SLAM, and why making LCC more closed was a mistake XGRIDS has since corrected. ## Key ideas - According to Aden Zhang, around 80% of the new products at INTERGEO 2026 were handheld LiDAR scanners; the XGRIDS L3 follows the L2, which according to Aden shipped around 4,000 units the year before. - XGRIDS tells clients that SLAM will never replace terrestrial scanning because of systematic error; according to Aden, the aim is to get as close as possible in a justified range and combine both: terrestrial for points of interest, SLAM to tie the whole site together. - According to Aden, newcomers who scan for 80 minutes without overlap or loop closures get poor accuracy; SLAM takes basic training, and users who know terrestrial registration usually get it right. - Gaussian splatting is not for cadastral survey, according to Aden; XGRIDS positions LCC for reality capture, progress management, site documentation and multi-team collaboration, and talks to large software vendors about bringing splats into their tools. - Aden calls it a mistake that LCC was made more closed and raw data formats were held back; according to Aden, XGRIDS corrected that this year, opened the data formats and made LCC free. ## Questions answered ### Will SLAM replace terrestrial laser scanning? No, according to Aden Zhang of XGRIDS: systematic error makes it impossible, so XGRIDS positions SLAM as close as possible in a justified range, with terrestrial scanning for points of interest. (discussed at 11:38) ### Should a beginner buy a SLAM scanner? According to Aden Zhang, SLAM takes basic training: newcomers who scan 80 minutes without overlap or loop closures get poor accuracy, while experienced terrestrial users usually get it right. (discussed at 13:02) ### Is Gaussian splatting accurate enough for surveying? Not for cadastral survey, according to Aden Zhang; XGRIDS aims Gaussian splatting at reality capture, progress management, site documentation and multi-team collaboration. (discussed at 10:05) ### What is the difference between 16 and 32 channel LiDAR? According to Aden Zhang, the 32 channel version doubles the point density of the 16 channel one; the accuracy is basically the same. (discussed at 18:38) ### Is XGRIDS LCC free? Yes, according to Aden Zhang: LCC used to be paid and more closed, and XGRIDS has since opened the data formats and made it free. (discussed at 21:21) ## Quotes > “SLAM is never going to replace terrestrial. We know that in our bones.” – Aden Zhang (11:43) > “Gaussian splatting is not for your cadastral survey.” – Aden Zhang (10:30) > “People buy this thing, and they go and scan 80 minutes in with no overlap.” – Aden Zhang (13:02) > “We realized that we've made a pretty big mistake there.” – Aden Zhang (21:21) > “For us it's not about the megapixels anymore.” – Aden Zhang (16:21) ## Chapters - 0:00 Aden Zhang, XGRIDS sales in Asia Pacific - 1:16 80% of new products are handheld LiDAR - 2:20 Why the double base plate matters for SLAM - 4:27 A screen instead of the phone mount - 5:22 Bigger camera sensor for Gaussian splatting - 6:54 SLAM accuracy and noise claims - 8:23 Drone mounting, but no aerial LiDAR - 9:44 Splats are not for cadastral survey - 10:54 Should newcomers buy SLAM? - 12:33 80 minutes without loop closures - 14:31 What comes after PortalCam - 15:18 Why scanners skip better cameras - 18:30 16 vs 32 channels: density, not accuracy - 19:37 Closing LCC was a mistake - 22:15 Chinese vendors and the rest of the world - 23:38 Wrap-up ## Transcript _Automatic transcription (ElevenLabs Scribe), lightly edited; times follow the audio._ **Tomáš Barnáš** (0:00): I also wanted to just ask you to give us- **Aden Zhang** (0:02): Of course **Tomáš Barnáš** (0:02): a pr- a brief introduction of, yourself- **Aden Zhang** (0:04): Of course **Tomáš Barnáš** (0:04): and the position. **Aden Zhang** (0:05): Thank you, Thomas. Yeah. So, my name is Aden, for those of you who don't know me. I am the sales director, for the Asia Pacific region of XGRIDS. So I look after Japan, Korea, Australia, New Zealand, Southeast Asia, that region. I've been at XGRIDS for four years. Me and Sunny joined back to back, one week apart. So yeah, one of the older, older, older sales team members. **Michal Gula** (0:28): Uh- **Tomáš Barnáš** (0:28): So pr- **Michal Gula** (0:28): wh- which year of the INTERGEO is it for you? Sorry. **Aden Zhang** (0:31): I think it's my f- fifth year, because I worked at Green Valley before. **Michal Gula** (0:35): Okay. **Aden Zhang** (0:36): So yeah, I was also here twice with Green Valley. Yeah, yeah. So my first year- **Michal Gula** (0:39): But as X-, XGRIDS- **Aden Zhang** (0:42): Third **Michal Gula** (0:42): as a brand? **Aden Zhang** (0:42): Third. **Michal Gula** (0:42): Third, yeah. **Aden Zhang** (0:43): Yeah. So Stuttgart, Frankfurt last year, and then this year, Mu- Mu- Munich. **Michal Gula** (0:46): Mm. **Aden Zhang** (0:47): Yeah. **Tomáš Barnáš** (0:47): And I want to say, to say that we are pretty lucky to have you here if you are, responsible for Pacific, but since you're fifth INTERGEO, looks like this- **Aden Zhang** (0:55): Yeah **Tomáš Barnáš** (0:55): is something what you always take care of. **Aden Zhang** (0:57): I think it comes down to, like, I grew up in New Zealand, so obviously I... a- and I have a survey background, so I can, you know, I can present the, products quite well. that's why the company keeps bringing me back every year, even though I say, "Let, let me take a break," you know. **Tomáš Barnáš** (1:11): Yeah, yeah. **Aden Zhang** (1:13): Yeah. **Tomáš Barnáš** (1:13): You can maybe, also tell us about- **Aden Zhang** (1:17): Yeah **Tomáš Barnáš** (1:17): the new device you brought to us. **Aden Zhang** (1:17): I would, I would love to. I would love to. So I think this is the year where INTERGEO, there's the most new products, right? And I would say probably 80% of them are handheld LiDAR scanners. that's kind of the way the market has been going for the past two and a half years. handheld LiDAR is kind of the something everyone is trying to get into. XGRIDS, obviously we're not the pioneers, but we are one of the early adopters. and this is, our L3. It follows the heavy success of our L2, which we shipped around 4,000 units, last year. and we are very happy to bring it to INTERGEO this year, if you can see that. ergonomically, we've made a few changes. A slightly, larger FOV horizontally and vertically. 360 by 285 now. before it was 270. And then we've have a two base plate, set it up. So with SLAM, I, I'm sure I don't need to tell you guys, but we... obviously we suggest loop optimization, so loop closure. **Michal Gula** (2:13): You wouldn't believe, but we're not doing this only for us. We also have some audience. **Tomáš Barnáš** (2:17): Oh, yeah. **Aden Zhang** (2:17): Yeah. Of course. **Michal Gula** (2:18): So feel free to explain it like- **Aden Zhang** (2:19): Yeah. Yeah. Yeah **Michal Gula** (2:20): in full range. **Aden Zhang** (2:20): Yeah. So for SLAM, to start and stop at the same position is very important. you have something to reference your starting point, and then ending, and then that kind of ties the whole accuracy together. so that's why we designed, this double base plate. so this one you start with, and then you leave it on your starting point. and then you take it off into this one, and then you go and do your scan. while you're doing your scan, you have the backside, which has the GCP collector. also with the, front mount, takes the weight off your hand, 'cause this is a little heavy. 2.5 kgs, but it takes the weight off right off. you go around. Next update, we work with Sony on this, but- **Michal Gula** (2:59): S- sorry. But, uh- **Aden Zhang** (3:00): Yes, please **Michal Gula** (3:00): can you explain, like, why would you, like, remove this during the scanning? **Aden Zhang** (3:05): Yeah. one, the first design is that, the GCP collector is now on the back, on the small base plate. **Michal Gula** (3:11): Okay. **Aden Zhang** (3:12): And then also, uh- **Michal Gula** (3:12): So you, you just need to, like, point it on the, on the- **Aden Zhang** (3:16): Yes **Michal Gula** (3:16): target, yeah? Or- **Aden Zhang** (3:16): Correct. On, on the ground, on the wall, or upside down in mine shafts, for example. And then this serves as kind of a point to remember where you started. So you leave that on your starting point, walk around wherever your site might be, come back. Because for now we have users who, you know, use paint or chalk or a piece of rock to remember where they started. Now it comes designed that there is a remembering point. **Michal Gula** (3:37): does the starting and ending point needs to be like a, with a centimeter precision, or- **Aden Zhang** (3:43): No. No **Michal Gula** (3:43): it's enough to come, like, within the- **Aden Zhang** (3:44): Two meters **Michal Gula** (3:45): two meters? Yeah. **Aden Zhang** (3:45): Within two meters is fine. **Michal Gula** (3:46): Okay. **Aden Zhang** (3:46): It's always been about two meters, but it just serves as like a, as a marker. **Michal Gula** (3:49): How, how many of these do, do we have in a, in a package? Because I just suppose it will get stolen. **Aden Zhang** (3:55): That is actually a really good point. Y- you're the third person who's brought that up. we ship with one now, but, our CTO is here. I will speak to him. I think maybe two or three would be a better option. But I- **Tomáš Barnáš** (4:04): I think it's- **Aden Zhang** (4:05): Yeah **Tomáš Barnáš** (4:05): very important. **Aden Zhang** (4:05): Yeah. **Tomáš Barnáš** (4:05): It's the first thing you will... It doesn't have to be stolen, but it can be lost. **Aden Zhang** (4:09): Yes. **Tomáš Barnáš** (4:09): It can... It's... Anything can happen, and then you are struggling. **Michal Gula** (4:12): But stolen mostly. I, I, I mean-... if you are not, like, scanning somewhere, like, totally out of the people- **Aden Zhang** (4:18): Yeah **Michal Gula** (4:18): somebody will go and- **Aden Zhang** (4:20): Yeah, yeah **Michal Gula** (4:20): just grab it- **Aden Zhang** (4:21): On a busy road or something **Michal Gula** (4:21): or kick it or whatever, you know? **Aden Zhang** (4:23): Yeah. **Michal Gula** (4:23): Yeah. **Aden Zhang** (4:23): Kids or something, you know. **Michal Gula** (4:24): Yeah. Yeah. **Aden Zhang** (4:24): Just always happens. Yeah **Michal Gula** (4:27): So what, what's like the... If you had to pick one feature that is new compared to L2, like, what's the, what's the top priority or what's the most important thing that changed compared to L2? **Aden Zhang** (4:40): Can I say three? **Michal Gula** (4:41): No. **Aden Zhang** (4:43): Ah. **Tomáš Barnáš** (4:45): The three. **Aden Zhang** (4:46): I would probably say the... Honestly, I would probably say the screen, because a lot of our feedback was that the phone mounting becomes a problem. because of that connection after around 30 minutes, this, it tends to lag. The, the O- OS system we spent a long time perfecting, and it's really, really smooth. we're really excited for, for Michal to come over tomorrow and really try it hands-on for himself. **Michal Gula** (5:06): Yeah. **Aden Zhang** (5:06): It's super smooth. now you don't need to connect to anything. If you are using the RTK outdoors, you just need to give it a hotspot from your phone. uses very little, little data, yeah. So everything's self-developed. So that, that's probably the biggest one, but the camera is also an amazing upgrade, yeah. **Michal Gula** (5:22): What's the resolution right now? **Aden Zhang** (5:23): so it's 4,000 by 3,000- **Michal Gula** (5:25): Yeah **Aden Zhang** (5:25): after processing. and then, in real time we can get 12 megapixel, yeah. but it's a 1.3 inch, one over 1.3 inch CMOs that we co-developed, with some really big names in the, in, in, in, in the industry. more dynamic range, 10 bit, which really allows us to tweak the data how we want for Gaussian splatting and point cloud colorization, in the post-processing scheme of things. **Michal Gula** (5:48): How else will compare in, in terms of Gaussian splats, to PortalCam, PortalCam? **Aden Zhang** (5:52): I think this will be, quite interesting for everyone, and it's actually the best system that we've had for Gaussian splatting. and that really just comes down to one, the more powerful LiDAR giving you more accurate Gaussian points, Gaussian positions, and then obviously the larger sensor for camera. so we did a few scans- **Michal Gula** (6:08): Is it also with the larger resolution or just like sensors are larger, so- **Aden Zhang** (6:13): Larger sensors, so more light- **Michal Gula** (6:15): Okay **Aden Zhang** (6:15): and then more range that we can tweak, yeah. **Tomáš Barnáš** (6:17): So compared to PortalCam, you have there like two high quality cameras in the front- **Aden Zhang** (6:20): Yeah **Tomáš Barnáš** (6:21): and then, lower res of panoramatic. **Aden Zhang** (6:23): Yeah. **Tomáš Barnáš** (6:24): So here you kind of unified the image quality, right? For the entire 360. **Aden Zhang** (6:28): we didn't unify the image quality. What we did do is we built HD enhancement into the system. **Tomáš Barnáš** (6:33): Okay. **Aden Zhang** (6:33): So actually now you can, stop and during the scan you can take an image using that front camera, and then that will be used in the process- processing without having to upload another file from a camera or something. Yeah. So the PortalCam is built in to get small words. This one you use a built-in, camera for HD enhancement that automatically processes it. **Michal Gula** (6:53): did you come with also with some changes on, changes on the software side or... **Aden Zhang** (6:59): Yes, of course. so the SLAM algorithm is the next biggest one that I was gonna say. so we have a white paper where we did against 1,600 benchmarks. we are marketing this as a 4.5 average millimeter accuracy at 40 meters. So that's across, 40 different scanning sites, each of them around 2,000 square meters. obviously, the accuracy will go a little bit down, if the size of the scan- scanning site gets larger. But even at around 5,000 square meters, we can still uphold around a 6.5 millimeter accuracy. so that's the Lixel Studio side of things. And then for, LCC, like I said, the HD enhancement. **Michal Gula** (7:36): What would be the noise of the data at, at li- let's say at 10 me- 10 meters range from, from the scanner? **Aden Zhang** (7:42): the, noise, the biggest change would be for reflective services. It hasn't actually changed that much for like, your normal walls and stuff like that, but for pipes, for road signs, it's changed dramatically. before we were seeing about a three centimeter noise, around road signs, but now, we're seeing around maybe just under one centimeter noise. So that's quite a big change for, you know, anyone doing, AI, classification or- **Michal Gula** (8:10): has it... I think I saw it in a video that, it has a RTK, uh- **Aden Zhang** (8:14): Yes **Michal Gula** (8:14): built in? **Aden Zhang** (8:14): Built in. **Michal Gula** (8:15): Yeah. **Aden Zhang** (8:15): Yeah. Yeah. So this is the built-in. slightly, better than the survey grade for the L2 Pro, so... And it's built in, yeah. **Tomáš Barnáš** (8:23): I'm often getting the question about like, okay, now, now PortalCam and your XGRIDS devices are pretty well known and, many time people are using drones or getting the combination with the unaccessible areas. **Aden Zhang** (8:36): Yes. **Tomáš Barnáš** (8:37): And that's not like super streamlined. Like are- streamline. Are you more thinking or is it for you interesting to get into that market a little bit more or do something for the drones as well, which would be like from your production? **Aden Zhang** (8:50): Yeah. we don't intend for the system to be, replacing any aerial LiDAR systems, but the system, does come with drone mounting. So these are actually ports that are left for drone mounting, and it goes on the M300, 350, and 400, which is the biggest improvement, and it does go straight into the SkyPort system. So you don't need to, you know, use another configuration or something like that. **Michal Gula** (9:14): Well, you know, nobody is expecting like this multi-beam LiDAR configuration- **Aden Zhang** (9:18): Yeah, yeah, no **Michal Gula** (9:18): for, for drones because it's different purpose. **Aden Zhang** (9:21): Exactly. **Michal Gula** (9:21): But, but the help of just being able... When I'm scanning the building and just the help of- **Aden Zhang** (9:26): Getting the roof or something **Michal Gula** (9:26): me being able to get a roof- **Aden Zhang** (9:28): Yeah, yeah, yeah **Michal Gula** (9:28): and everything else, that, that helps, right? **Aden Zhang** (9:29): Yeah. Our product m- product team was like, "Should we do like three returns?" And I was like, "We w- " And the, the sales team was like- **Michal Gula** (9:35): The, th- **Aden Zhang** (9:35): "No, that's not what we're bui- **Michal Gula** (9:36): That's not the purpose, yeah **Aden Zhang** (9:36): that's not what we're doing here." **Michal Gula** (9:37): Exactly. **Aden Zhang** (9:37): Yeah, yeah. So I think we're a company with a quite a clear focus on what we want to be and what we're not, so yeah. **Michal Gula** (9:44): Is there... because when you first came or when I first noticed XGRIDS, it was very clear like we are here to do the Gaussian splatting with the LiDAR. Is this, during the time, is this somehow changing or is this still like the main focus of, of the company? **Aden Zhang** (10:01): I would s- **Michal Gula** (10:01): To be the, the, the top one doing LiDAR with the, with the Gaussian splats. **Aden Zhang** (10:05): It's still definitely a very big part of our focus. Our R&D team, is still lar- largely revolving around LCC, but what we're seeing now is that we need to go and educate the users. before, before 3D Gaussian splatting can be adopted largely into third-party software, it's not going to be majorly adopted by end users. So that's why we're in talks with Esri, with Autodesk, all of these big players. That's what our next step is. Educating what- So Gaussian splatting is not for your cadastral survey, not for, you know, this kind of use case. But for reality capture, for progress management, site documentation, multi-team collaboration, this is what we want. So we're s- in specific talks with software teams from those companies to really bring LCC and 3D Gaussian s- splatting to light. and that's still very much our focus at XGRIDS here. **Michal Gula** (10:53): when I'm speaking to, to guys that are new to laser scanning, so, like, they have zero- **Aden Zhang** (11:01): Yeah **Michal Gula** (11:01): experience with the terrestrials or with the SLAMs, and they are thinking about buying a SLAM. I'm very hes- hes- hesitating to, to advise them to do so. **Aden Zhang** (11:11): Yes. **Michal Gula** (11:12): most of the times I don't advise to- them to buy SLAM because they are missing the basics of- of how lasers kind of work, so of how overlaps works. **Aden Zhang** (11:22): Yes. **Michal Gula** (11:22): Of how registration works- **Aden Zhang** (11:24): Yes **Michal Gula** (11:24): and so on. how confident... Be- because you just said that you need to educate the, the market more. How confident are you, selling the SLAMs to, to totally newcomers into a scanning business? **Aden Zhang** (11:38): Yeah. So, at the booth today, we've had, like... Our sales strategy is we tell clients, like, "SLAM is never going to replace terrestrial. We know that in our bones. systematic error, it's just not doable. There's too many moving parts. But the thing that we can do is bring it as close as possible in a justified range." So terrestrial, if you're doing 30 setups, for example, 10,000 square meters, that's... we're never gonna reach that level. But for example, if we can do it in a 2,000 square meter and we still can get five millimeter, relative accuracy, that's what we're kind of telling the clients here. The next thing is terrestrial for your points of interest, your pipes, and then XGRIDS to kind of put everything aligned together, all the points overlapping in between so you can get a large scale view of the entire site. That's what we're positioning and how we're kind of educating the point cloud users on where XGRIDS is heading or what XGRIDS is doing now. **Michal Gula** (12:32): W- where... Usually where I have the problem is that, once somebody's using, even for the first time, once they are using TLS, I can always, like, look at the registration. I can tell, "Now, go back there and do one more scan here." **Aden Zhang** (12:46): Yeah. **Michal Gula** (12:46): Like, exactly here. With SLAM, especially if it's not just one loop, like, on a one- **Aden Zhang** (12:52): Yes **Michal Gula** (12:52): journey, but you need to do, like, several laps and, and just connect those data, I'm scared to - **Aden Zhang** (13:01): Yes. You're, you're- **Michal Gula** (13:01): for the newcomers. **Aden Zhang** (13:02): Yeah, yeah, you're totally correct. Like, people buy this thing, and they go and scan 80 minutes in with no overlap, no- **Michal Gula** (13:08): Yeah **Aden Zhang** (13:08): you know, no glue cl- many loop closures in between. They come back, and why is the accuracy so... But that's why we're lucky to have so many great partners, 'cause they actually go and educate for us. And I think more people now every year understand what SLAM is and what it's capable of. So I think it's moving in the right direction, and with the right education, right partners, I think- It's, it's definitely doable. **Tomáš Barnáš** (13:26): I think this is, this is very important point you mentioned, and, that they are educating the market. It's the same thing, like we did a lot of trainings for photogrammetry, for example. **Aden Zhang** (13:35): Yes. **Tomáš Barnáš** (13:35): And the biggest, mm, point of error I saw when DJI came with, M300 or 350, and everybody thought like, "Oh, I'm getting a gr- great machine which has a great resolution, so I will get a great result." But you still need kind of like understanding and basic- **Aden Zhang** (13:54): Basic knowledge **Tomáš Barnáš** (13:54): and, and, and this is the same thing with, with SLAM- **Aden Zhang** (13:57): Yes **Tomáš Barnáš** (13:57): where I would say it feels so simple that you are just thinking like, "I just take it and do whatever I want, then, then it's great." But you still kind of like need to get a basic education. **Aden Zhang** (14:06): Absolutely. That's what Mi- Michal said. If you don't have a basic understanding on how I- ID- ICP registration or how to, you know, point pairing, matching, stuff like that, it's, it's... it does take some training to get SLAM correct. But if you are familiar, for example, you're a 20-year TLS user, you transi- transition to SLAM, you're, you're probably gonna get it right because you know the, the physics behind it to, to see. **Tomáš Barnáš** (14:28): That's important. Yeah, yeah, yeah. **Aden Zhang** (14:29): Yeah, I would say so as well. **Tomáš Barnáš** (14:30): Yeah, yeah. **Aden Zhang** (14:30): I would say so. Yeah. **Michal Gula** (14:31): I would bet like a lot of money that today you have been already asked several times about the PortalCam and whether there is coming a new generation or- **Aden Zhang** (14:42): Yeah. **Michal Gula** (14:44): Am I correct? **Aden Zhang** (14:45): What, what I mean- **Michal Gula** (14:45): That's the question. **Aden Zhang** (14:46): I mean, I think, it's, it's natural. It's natural that something will come. I, I don't know if it will be a PortalCam 2 or it will be a new product. I mean, I do know, but- Just to tease you guys a bit. I think, it's definitely going to be better, but, I think the LiDAR will kind of be the next thing that we kind of switch up a little bit. Obviously, the sensor as well, but also the LiDAR. we kinda wanna make some tweaks there and play around it and kind of experiment. but not this year, but it's, it is coming. **Michal Gula** (15:18): I was asking this, several other SLAM companies, like, or even terrestrials, like, I want to understand why is it so difficult or why you companies don't use like the really top-notch sensors for, for the pictures. I understand it's heavy on the data, but you can still limit this with the firmware- for some period of, period of, of the time, right? But just to allow to collect as good data as possible, you know, from, from the field. **Aden Zhang** (15:49): Yeah. **Michal Gula** (15:49): And then decide what to do with them. **Aden Zhang** (15:50): we've been doing that for years. Like our XSPIN or whatever proprietary format that we've been using, it, it really shrinks the image size. And then from us, we realize that for the good Gaussian splatting and good point cloud quality, it's not all about the megapixels. But you're in- right in saying that there's a large base of users who need high quality or as high quality image as possible to get, you know, your onsite verifications and, you know, reference against real site, real site conditions. but for us, it's just... Right now, I think for us it's not about the megapixels anymore. It's, it's more about, you know, what's the right balance to get the best point cloud colorization in, in Gaussian splatting LiDAR. **Michal Gula** (16:31): So that, that's the case because you are l- like looking for a balance between, the, the quality and the price- **Aden Zhang** (16:39): Yes **Michal Gula** (16:39): to keep the price of the final product at some level. **Aden Zhang** (16:42): Of course. **Michal Gula** (16:42): Right? So did I, did I understand perfectly? **Aden Zhang** (16:44): Yes, you are. **Michal Gula** (16:46): we are speaking about ranges of several thousands, of, of euros. **Aden Zhang** (16:50): Yes. **Michal Gula** (16:50): for Port- PortalCam it's from three and a half or, or something like that, right? **Aden Zhang** (16:54): Yeah. Something about four. Yeah. **Michal Gula** (16:54): Okay. So, so that's perfectly fine. My problem is with the higher tier scanners, where we are at 30,000, 40, 50,000. And with DLS, like s- 60, 70. **Aden Zhang** (17:04): Yeah. **Michal Gula** (17:05): like adding better sensors at additional cost of one or two thousands- it doesn't really make a change. **Aden Zhang** (17:12): No, it doesn't. **Michal Gula** (17:13): So why not to implement better sensors there? **Aden Zhang** (17:16): So that's a very good question. That's a very good question. I think, from a ROI point of view, the companies who are placing the s- the systems are above 30, 40, 50,000. Their, revenue or, on the each system is actually really, really high. so th- actually, they can afford to use better sensors. **Michal Gula** (17:37): But, but clients would pay for it, you know? **Aden Zhang** (17:39): Yes. I think so. I think so. **Michal Gula** (17:40): If you tell me I can have, I don't know, I w- I'm using RTC360 from, from Leica. **Aden Zhang** (17:44): Yeah, yeah. **Michal Gula** (17:44): If they tell me like, "You can have, much better sensors for, for the imagery- **Aden Zhang** (17:49): Yeah **Michal Gula** (17:49): at the additional cost of 3,000s," I would happily pay that at, at the cost- **Aden Zhang** (17:54): Jump, jump on board **Michal Gula** (17:54): of, 60,000 of the scanner- **Aden Zhang** (17:56): Yeah **Michal Gula** (17:56): I would happily add- **Aden Zhang** (17:57): Yeah, yeah **Michal Gula** (17:57): pay more. **Aden Zhang** (17:58): It seems like peanuts compared to the actual price, right? **Michal Gula** (18:00): Exactly. **Aden Zhang** (18:00): Yeah, yeah. **Michal Gula** (18:00): Exactly. **Aden Zhang** (18:01): Yeah. good question. **Michal Gula** (18:03): And, and I think it would, be the same for, for this, for example. the LT, free Pro will be at, what price range? **Aden Zhang** (18:10): entry-level price will be around 27,000- **Michal Gula** (18:13): Okay **Aden Zhang** (18:14): euros. Yeah. **Michal Gula** (18:15): So are there going to be like different tiers or...? **Aden Zhang** (18:18): there's a 16 channel, and then there's a 32 channel that I have here in my hand. No 300 meter anymore. **Michal Gula** (18:23): What will be the price difference? **Aden Zhang** (18:25): the 32 channel 120 is around 36,000, I believe. Yeah. **Michal Gula** (18:30): Okay. C- can you maybe quickly, like explain to our viewers, like what's the difference? Like why, why do you differentiate between 16 channels and- **Aden Zhang** (18:37): Yeah, sure **Michal Gula** (18:37): 32 channels? **Aden Zhang** (18:38): so basically, there's beam that gets shot out. We call them lines. there's a few different ways of saying it. That's basically how many original lasers are being shot out, and there's a mirror that spins around that, does arrays the LiDAR. 16 just means there's 16 original lines of poi- of LiDAR, and then 32 means there's 32. and as you expect with the 32, it's double the point density of the 16 li- line. the accuracy is actually basically the same. The main difference that you're paying for is a, a point cloud density. Yeah. So those who need, you know, pipes or, you know, very perfect edges or stairwells, stuff like that, your high density is gonna help you a lot, even with manual drawing or extraction. **Michal Gula** (19:18): So it's, it's not about the speed of, of capturing the data. **Aden Zhang** (19:21): Yeah. **Michal Gula** (19:21): It's about the density, about the quality, yeah? **Aden Zhang** (19:23): Yes. Absolutely. **Michal Gula** (19:24): And maybe not e- even about the quality. Really about the density. **Aden Zhang** (19:27): Really just about the density. Yeah. And in turn, the... If you're smoothing algorithm and your soft- your company has a good algorithm, then the, it transitions to the, the data quality. **Tomáš Barnáš** (19:38): If you look at the software side, like before, we started the, the podcast, you just mentioned that you have now a plugin for Unreal Engine. **Aden Zhang** (19:46): Yes. **Tomáš Barnáš** (19:46): Then you mentioned Esri. Like, from a software perse- perspective, are you looking at it like you wanna be very close solution or you, you are opening API or making plugins to, to try to actually spread your software also for direct integration to other solutions? **Aden Zhang** (20:02): I think definitely open has been like... But we, we definitely made mistakes. So last year we made, you know, LCC more closed, and then we didn't give out raw data formats and, you know. But we realized that was wrong. so I think that's where we corrected ourselves this year. Everything is now proprietarily open. You can extract any, you know, file format that you need or raw data format that you need. and we'd lo- we'd... we want to work with these, these other third-party software, because that's not where our strength is. where our strength is in providing good quality data and as open of a SDK format as possible. And then on top of that, adding a little bit of value, through our functions to our clients. Yeah. **Tomáš Barnáš** (20:39): That's close. **Michal Gula** (20:39): I, I, I, I like what you said that you did a mistake because, like I'm the advocate of this- **Aden Zhang** (20:45): Yeah. **Michal Gula** (20:45): that you need to, you need to understand the market that, like, okay, if you have a small com- **Aden Zhang** (20:51): Yes **Michal Gula** (20:51): company of, of one person scanning, okay, they pro- they are probably within one brand- **Aden Zhang** (20:56): Yeah **Michal Gula** (20:57): doing scanning. As long as you have, like, five to 10 people and more, they usually have different brands- **Aden Zhang** (21:03): Yes **Michal Gula** (21:03): different sensors, different scanners, and they want the interoperability. They want to work from, with, with one data sets within others. They are not like, like locked into one environment. **Aden Zhang** (21:15): For sure. For sure. **Michal Gula** (21:16): And that's, very bad for them if you try to lock them- **Aden Zhang** (21:20): Yeah **Michal Gula** (21:20): within one just software. **Aden Zhang** (21:21): Absolutely. Yeah, absolutely. We realized, yeah, that we've made a pretty big mistake there. So we, we corrected that, spoke to the CEO, spoke to the- **Michal Gula** (21:30): Mm **Aden Zhang** (21:30): product team, and they was like... They were more than happy to change it. And as you guys know, LCC came out when PortalCam came. There was a charge, there was a charge to use it, but now we've made that completely free as well. I think it comes back to educating. We don't want to stop at, you know, 5,000, 10,000 users. What we want to do is we want to spread 3D Gaussian splatting into 100,000, millions of users. That's where actually it makes sense, you know. There's no point in charging 2,000, 3,000 people and using a software and then not actually expanding the reach of the technology. **Michal Gula** (22:01): Well, Gaussian splat, sorry, they are really opening a totally new addre- addressed market. **Aden Zhang** (22:06): For sure. For sure. **Michal Gula** (22:06): So it, it makes sense you wanna, like, cover the market- **Aden Zhang** (22:09): Yeah **Michal Gula** (22:09): and, and, and absorb it, not, not to be just like on, on a small portion of- **Aden Zhang** (22:14): Yeah **Michal Gula** (22:14): of it. **Aden Zhang** (22:15): Exactly. **Tomáš Barnáš** (22:15): I believe that it... this is also very appreciated from market to hear that, you also learn and listen the market, and actually you probably figure out from the market feedback that it would be good to be more open. And I, I really, like, personally appreciate it that you can openly speak about it. **Aden Zhang** (22:31): Thank you. Thank you. We try. It's... Sometimes it's difficult to work with Chinese companies. You know, I like you. It's, growing up as a Kiwi is like, is difficult for me as well. But understanding where, what we can do better as a Chinese company, hearing about European users, you know, American users, users from Japan, and kind of trying our best to actually s- spread our technology worldwide, I think that's what's been interesting for us. **Michal Gula** (22:54): You know, the, the truth is that, the, the Chinese market is so competitive that- **Aden Zhang** (22:59): It's too competitive. **Michal Gula** (23:00): that it's, it's much more difficult for you to succeed in a Chinese market- **Aden Zhang** (23:04): Yes **Michal Gula** (23:05): compared to the world. **Aden Zhang** (23:06): For sure. **Michal Gula** (23:06): So if, if you are the number one or in the top three in China, the rest of the world is like, cherry on the top. **Aden Zhang** (23:12): Yeah. **Michal Gula** (23:12): Right? I mean... But, but it's, but it's even more important to understand the rest of the world. **Aden Zhang** (23:18): Yes. **Michal Gula** (23:18): Because the, the, the rules and the, the markets can be very, like, very different- **Aden Zhang** (23:22): Absolutely **Michal Gula** (23:23): compared to, to China. **Aden Zhang** (23:23): Absolutely. Yeah. Yeah. Yeah. There's maybe 20, 30 Chinese companies here. If you go to a Chinese expo domestically, there might be 60 doing the same thing, so- **Tomáš Barnáš** (23:32): Exactly **Aden Zhang** (23:32): it's like, you know. The 20 here are probably the 20 better ones. **Tomáš Barnáš** (23:35): Mm. Mm. Mm. **Aden Zhang** (23:35): Maybe. Yeah. So. **Tomáš Barnáš** (23:37): Perfect. **Michal Gula** (23:38): Amazing. **Tomáš Barnáš** (23:38): Thank you so much for coming. **Aden Zhang** (23:39): It was a really, really pleasure. first time meeting, Thomas, and Michal, nice to see you again. You look great. Yeah. Yeah, yeah. **Michal Gula** (23:45): Thank you. **Aden Zhang** (23:46): Yeah. We'll have Michal at our booth tomorrow, and really excited to have him try this thing hands-on for us. Yeah. **Tomáš Barnáš** (23:51): Of course. Looking forward to it. **Aden Zhang** (23:52): Yeah. Thank you, guys.