 Okay Yes Right Yeah, but we're listening on that. So I believe the Empress is listening to your request. It's being processed process So, yeah, the thing about locomotives is they come in all shapes and sizes, you know, in-scale, HO particular, and there's a lot of combinations that, you know, take a lot of different sound enclosures, I mean, it's a nightmare, but yes, having a, remember, sound decoders are not our primary business, we've done this to support our customers who wanted an Indian solution. So, if you didn't want to be able to call up tech support and have the tech support tell them, well, that's not how sound we're going to go call them when you get this finger pointing. So, we do have a number of customers who understand that it's a medium-forms decoder for a fair price, and they like the fact that if it doesn't work, we are settled with our given a satisfactory answer, versus telling them to call somebody else a tech support. So, it's, you know, people make purchasing decisions based on that, being sure that it's in one family. Not everybody, but there's some people who prefer that background. Any other questions? Go ahead. Yeah, I think I need to get some more background on that. Yeah. Yeah, do you have the latest code in those? Okay, yeah. There's a number of things that are happening that we are tracking down, so I imagine we'll have, we'll look at those issues as we find them. The good news is they're IPL-able, so if and when we find a particular set of combinations of events that cause some issue, as long as we can identify, we can patch it and fix it and get it out to the field. So, you know, as we find things, as we find little glitches, we will fix them. AJ, it's not a glitch, it's not a little glitch at all. No, no, I mean, I'm just saying in terms of, besides a big glitch, then we'll find it and we'll fix it. Let me give you a little background. Five years ago, I was one of the lead people in a large club, a medium-sized club, that was converting from DC to DC. Okay, so we converted and we went with Digitrax because the radio was better than NTE. Okay. Today, all my NTE friends are laughing at me because their radio works in your study. We cannot have a reliable radio operation on Doopla. Okay. On many different maps, not just a single map. Okay. The Dooplax will not stay up if you have a large number of users. If you have one or two or even four or even six users, you can quite often run for two or three hours and have no problem. But the radio just stops listening. Well, we've seen a number of issues in the last, how long had Dooplax out? Three years? Two years? And, you know, we're vexed when we hear somebody's got to lay out on whatever they have in trouble. First off, our first line of defence is tech support. They'll try and get the information in the background. And we've seen in the last number of years, we've seen a number of situations where we've actually resolved it and worked out what's going on. And there's a good example of a club, I think it was in Tennessee, up in Taki. And they had it running fine and all of a sudden we'd just go to hell on a hand bus and get nothing from work. And they got frustrated. I mean, yeah, they got really frustrated. We spent a lot of time with tech support and back and forward testing things. We finally found out what the problem was. And what was happening was that an area where they shared a pole transformer with a welding company on the same leg. And as soon as they knew what was going on, what was happening was the welding currents. I mean, these are industrial welders, not just the little back yard things. And they were putting so much noise back on the common leg of the transformer that it was going into their property and they had no input filtering, which got into everything. Okay, wideband noise and a flat radio. So as soon as we found out there was a sensitivity, I believe they put a UPS on the systems. It's worked flawlessly since that was fixed. So that wasn't anything we did to the DD402Ds or the UR92s. We had a code update about what, nine months ago, a year. We had an issue with names getting changed in the system. And that was a simple glitch once we got the information, we fixed it within, you know, two or three days. So we've got a number of things we've gone through and fixed, you know, we have maintenance releases on code that are problems that we can really identify and track down. We had another gentleman that had a bunch of, I guess, fluorescent lights, a bunch of wires hanging down and he swore up and down that, you know, it didn't work at all. And somebody else actually finally said, we've had this in a large club running beautifully for years. What are you doing wrong? He went over there and took a look and said, this is not right, this is not right. We went through it and they were fixed. So your particular instance, you know, if you talk to the text board and we can't resolve it, then we've got an issue that we're going to take on. But generally, there's four or five items that you can work on and today we fixed them all. Now, there's probably some other little thing in there that we need to look at and we will solve it. Once we can, if we find an issue that is consistent and we can characterize it, we will fix it. But we have a lot of layouts, big ones, running dozens of throttles, not a hiccup. Anybody here got a large layout link with a lot of users and they're perfectly happy to do this? Anybody here with a layout to do this that has trouble? Right, I think, again, what level of discussion we have with text support on what your issues are and how they're occurring? I haven't personally talked about that on the board. Okay, I think that's a first step I'd request that you actually discuss on. It's in right hands of David Clark. You know David's name. I work on David's layout and I work on other different types of layouts in the same layout. So I get my input through David. I haven't talked to them myself. But I've been working with people like franchise and I've been working with a lot of smart people. And we're not solving these problems. We've gone through all the various things that everybody says we have to do. And the bottom line is, you know, if you have eight or ten users out there, you've got a couple hours. And then you start having throttles that go back. We're getting better, okay? But we still have the throttles that just stop responding. They don't do anything. You push a button, nothing happens. And the only fixed board is a bowl of batter, but a vacuum. By the way, a collar switches on the throne. Okay. I thought it was switch work that you could install in your throat in your case that you have that. You know the questions? Okay, go ahead. How do you come up with a solid set of arrangements? I asked this question in a while. When was mobile? Is that a year ago? Was that, yeah, okay. Yes. So what do I tell them? The truth. What's the truth I must know? You're never going to get that answer. That's not the truth. I bet he's looking for a solid state to power management. And we'll listen. We'll listen. We'll listen. Oh, okay. Okay, another question. But it's not that I have to kill you. Yeah, we don't do that. Okay, what do you have? I know you have it. 144-some, the code, put it into, it's got the plugs on there. Put it into the Unicode mode. Speed on DC without the decoder. The mode mode is 50 steel miles per hour. Okay. What's the decoder? If it becomes quiet. If we want to keep decoders, we're going to supply them. Yeah. How long can you see everything with the decoder? Yeah, the concept of animal mode conversion in decoders is great in theory, but of course, until you have enough altars to run the decoder, it's not going to work very well. So, I'm hearing you having trouble with animal mode conversion, is that what you're saying? No. I can take a decoder out of the run mode mode in 50-scale miles per hour. Okay. If I want to decoder it, it sounds great, by the way. Speed can be degree, forward top speed is 5 miles per hour. Okay, Dave, do you have any input on that? Yeah, if it's a bit. On your second one. Yeah, what you could try is, depending on the armature inductance of that motor and that locomotive, you could change CV9, I think you could change to 0 or 255. 255. That'll lower the frequency. And that will actually, typically, where you're running a high inductance motor, the load, the current through the armature gets very constrictive or very used at the high frequency because the default comes out of the factory at about 20 kHz. If you drop that down to 1 or 2 kHz, it'll be a tremendous difference in torque and that will run your speed back up. What's the locomotive? Is there a different motor style there? It's a little bit different motor. I need to take it, I'll talk to Matt about it. We've already had a little conversation about that. Yeah, the easiest way to fix that torque forward is to go forward with the frequency, which is the physical characteristic of the motor. It's something that the kona can do about anything. Basically, we have to drop the pulse width frequency. Change CV9 to 255. Yeah, that'd be your first test and that generally will make it pretty snappy if there's an issue with inductance on the motor. Okay. Anybody else? Andrew, can we go to 2nd while I change tips? Okay. DD400. Okay. The DD400, the upgrade on the DD400 is changing it to the 402 series. No, there's been no changes. That product is where it's at and if you wish the new features of the 402, we can do a factory upgrade there for you. So we can change the DD400 to a DD402. So it adds extra things. I'll forgive what they are, but there's a number of extra things. You get 28 functions and we'll grab that once it ends. Okay, we have one more question, I believe. Okay, good. Okay. Yeah, we'll send Rich. Yeah, we'll send Rich today, but do you have a laptop that you can do a site scan on? Yeah, if you have a PR3 or an MS100, we can get one of our guys to bring it over. All you can do is do a site scan because you get to a place like this and you're going to see probably a couple of channels chewed up with Wi-Fi minimum. We only have 16 channels available and sometimes you'll get very solid traffic. A lot of people are there with laptops and stuff and surfing and downloads and stuff. We saw that in, I guess, we did the first public showing the DD402s in Louisville and we had a couple of channels that were not usable, just totally unusable. So we did a site survey and we picked the widest channels and it worked fine. We had, I guess, 109 thrones running. We had probably a dozen or two DD400s. We had another dozen or so DD402Ds. There's a whole lot of every throttle we've ever made running on the layout at the same time. So we've seen sites where they are very noisy. They've got a lot of background, Wi-Fi networks and we have to cooperate with the other networks. We don't generate a lot of traffic but the other networks can do a tremendous amount of high-end duty cycle traffic. OK, go. Question. A lot of server work is not being used in the army. Is it up to a proprietary person like Microsoft to try to develop an interface so that the load of that will come in and control it? Oh, you're talking about server mode as a switch position? I'm working on it as a relay, replace the torus connection that would turn the server off and on to the server loader. I have a demonstration bar. I just use an off-on switch. OK, when you say server mode, are you meaning like a little model aircraft RC server motor? Yeah. A servo system, not just a motor. Yeah, a servo. A servo motor controlled by a servo board? Correct, yeah. A programmable memory that can program the upspeed and the downspeed independently? Yeah, we've been requested that we'd have something like that. I believe somebody in Taiwan on about five years ago needed a server motor driver. I believe that. A lot of good commutative, a lot of naturally-to-use computer programs. Yes, that's great. I've never found one. I'm not sure if they agree with that. I believe that as well. I've seen a couple do that, so maybe we can dig that out and let you know exactly. OK, so is that what you got? Yeah. Did you try to score? At least one fellow had a peculiar problem hoping that he could be solved. He had several different radio files and you were 91s and they all worked well except for UT 4R, which worked fine. At least one other layout that he would have liked for his own layout. Do you have any ideas of why this is so? Maybe you want to talk with Dave and go over what issues might be with the UT 4R, I believe. So basically, it was too high a power or it was causing a problem with the reception. Is that what I'm hearing? Apparently all those other radio files of different types worked. I guess it's 300 work. The UT 4Rs only worked Yeah, maybe you are. At this it's another call play out somewhere else. Yeah, we need to adjust that one. We'll just say Dave. Again, just call back Dave and we can do the best thing. We can resolve that. Yeah, we can solve that. OK, anything else? Go ahead. I'm just curious how many devices can you handle? Let's Steve push. Steve. How many can you handle? Great. There's not that much to that. Not about bandwidth. I'm not sure about the bandwidth. So if it doesn't keep the network occupied it's only connecting and sending the message back when it gets off the net. And then it's changing its frequency when it does that. So assuming you don't have a lot that's using the same channel you could get quite a few products connected. You're also going to have a bottleneck going back to your question. The local net backbone rate is a fraction of what the 802.11 modulation schemes give. So they're going to send little chirp packets and they're going to be way slower than that part of local net. So I don't suspect that will be a problem. If your question is, can you have 400 thrones hooked up and doing a little bit of traffic you could I guess I guess the number of IP addresses issue is 255. I was just concerned that it would be a case where 5 4, 7, 1. No, from the source I've had with Steve it looks like stack sizes, the resources available on the little modules adequate for probably a dozen or two. So we would expect that that's a reasonable number. OK. Generally my PR3 works pretty good and I'll use it occasionally to look it up and get the interface brought up and download some sounds. But sometimes I'm not sure if I'm doing things in the long order I'll go set some CVs and then I'll think I'm downloading the sound when you now play it back. Nothing was downloaded. I'll find that I've got to go reset some CVs and I don't know maybe the lesson I should learn is when I boot it up and it's made the connection and it's reading things that just download the sounds without using it to change the number of times. Normally by recycling or resetting some of the CVs I'm always able to begin to download the sounds and it works pretty good. Some of the mode changes we're changing from running a sound stream down to CV programming there may be some issues there. There's a number of worker threads and a lot of things going on inside the box and each operating system has each computer running say Windows which is what's running under can have some severely different latencies and scheduling between the different threads. So it is possible because they're asynchronous to get a little bit out of whack. So if you get some problem facing isn't reconnected and as you say if you be changing modes a lot you can get into some pretty scurvy stuff. I haven't seen a lot but I know that if there's an issue you just reconnect and please everything up so it's a more of a Windows issue because we're trying to do a lot of things in that box a lot of different types of processes running through a single COM port even though it's USB it in effect it looks like a COM port and Windows didn't do a real strong job of that so it can be a child and the best solutions that you suggest and that seems to be every time I've had a catapact in the last five years When you're setting the CVs you've got to reset the problem there Well that depends whether the decoder got confused or the PR3 got confused over the USB interface so there's a possibility of problem both places The software has actually been remarkable with SOLID that you run across a lot of different versions of Windows 98 and Run Under Windows 95 and Run Under XP and Run Under Windows 7 so it's actually a prudely simple piece of code that doesn't make any fancy football work but that seems to work quite stably across a lot of different platforms and I'm talking about Windows platforms not across multiple computer platforms but basically if you have something squirrelly generally focus on that one task run it and it generally works on that's what I was thinking Any other questions? I think we've achieved a little So anyway, come to the booth you've got other questions that come up, sir OK, well anyway as you come to the booth for the next couple of days and talk to us and bring up any other issues you have and as well I mean Zana wants to know what things you want SOLID state oh SOLID state power and control I think we've got two people you want there let's go to what else more speakers, I believe I heard anything else I was going to address the DD402 issues with Radio Duke players as well Probably read that on the programming track sample the functions sample functions if they're all working in terms of sound or in terms of wired function leads that's I guess technically it's possible but we'd have to know that you've actually put a lead load on the lead and ransomware will detect that so that's probably a little bit more problematic than the sound if you start to talk about LED light loads which are weird as an incandescent so some of those current models would be quite tough to weed out from a motor so the NMRA standard designed, I think it was a 60mAh current difference typically at a minimum of at least 60mAh difference and you'll have a lot of LEDs at only 5 or 20 so I think on the phase of it it would be a pretty tough thing to do so unfortunately you're back to manually okay let's look up the decoder and let's try and see what works and it's a really good idea anytime you get a new decoder is just check it out on the bench to make sure it works we run everything through a computerised test facility so everything is theoretically 100% tested and we've had a pretty good run of luck it still doesn't mean that you can't have infinite mortality because we changed to a no-lead process which is a lot harder on electronics okay that's just a known outcome running a no-lead process the extra 10 or 15 degrees centigrade causes a greater number of infant mortality events that's just pure fact that's throughout the consumer and also the all the hydeic industry has found that informally so yeah we'd rather be doing lead but of course we're in California so I can't talk about lead anyway so basically I don't watch that am I going to have to be to use a booster like the new back CDs on a program track or with your system the PTD uses that is that a sound track? yeah that seems to be the most popular fix that I see from people I had to put one of those is that for sound decoders or for just the new back CDs on the program track did you track the coders or QSI I got a tsunami I got a tsunami yeah I think the reason they did that was because some of the command stations are a little sensitive and that's a correct answer they've built a little program called a read back booster or something and that's the right answer to have because that will be and I think it's going to be a lot of these sound decoders use way more juice than a regular decoder a regular decoder can operate on a 5 or 10 milliamps of track current the tsunami in fact once they've used their poles at about 100 milliamps at 2 or 3 volts and even if you invert that down from 12 that's still that big current draw and so your quiescent and non-sound load is greater than the difference of the NMR8 specified for the current by recollection so we can wind up the current but the trouble is if you're testing more sensitive decoders and something goes wrong there's going to be a puff of smoke more easily so that's always been a point of tenderness in terms of trying to deal with a wide range of current draws and still be something that is benign to a mis-install decoder and they do happen that's why we've always got to agree load power programming track for that reason so it's forgiving for a mis-installation so if you want to ride high current decoders I would recommend using the booster and then if you're running something that's not a sound decoder then you can go back to the regular programming track and you will have less likelihood of having a mis-wire and install something missed causing decoder to be damaged that's a personal opinion and I think that's good okay come see us at the booth thank you