<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Phillip Caudell on Medium]]></title>
        <description><![CDATA[Stories by Phillip Caudell on Medium]]></description>
        <link>https://medium.com/@phillipcaudell?source=rss-a04dd8abbce4------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*tRGXsPG0dyozm6nwcVpdqw.jpeg</url>
            <title>Stories by Phillip Caudell on Medium</title>
            <link>https://medium.com/@phillipcaudell?source=rss-a04dd8abbce4------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 20:49:08 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@phillipcaudell/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[8 painfully-simple-yet-pratical questions to scrutinise your new startup idea]]></title>
            <link>https://medium.com/founders-factory/8-painfully-simple-yet-pratical-questions-to-scrutinise-your-new-startup-idea-88c3f4dd77f3?source=rss-a04dd8abbce4------2</link>
            <guid isPermaLink="false">https://medium.com/p/88c3f4dd77f3</guid>
            <category><![CDATA[product]]></category>
            <category><![CDATA[ideation]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Phillip Caudell]]></dc:creator>
            <pubDate>Fri, 11 Jan 2019 17:26:40 GMT</pubDate>
            <atom:updated>2019-01-13T12:10:57.973Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GzXTGm6MhDUYD5fQzCllOw.png" /></figure><h3>8 painfully-simple-yet-practical questions to scrutinise your new startup idea</h3><p>Hello my name is Phillip, and I have a really bizarre job: I come up with startup ideas.</p><p>“Oh great, <em>the ideas guy</em>” I hear you cry — but also part of my job is to figure out if those ideas are any good. There’s many different ways me and my team go about doing just that, and in this series I’ll share a few of them.</p><p>In this first post I’ll outline some simple questions you can ask yourself that act as quick filter when you’re considering new ideas. Some of these may be painfully obvious, <a href="http://innovationfootprints.com/wp-content/uploads/2015/07/startup-genome-report-extra-on-premature-scaling.pdf">but seeing as 90% of startups fail</a> because they’re not solving a real problem, they can’t be <em>that</em> obvious.</p><p>Maybe you’re thinking of starting a new business yourself, or maybe me and you have the same job (<a href="http://concepting.foundersfactory.com/join-us.html">hey!</a>) — whatever it is, if product and startup creation is your thing read on.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VvKaLlHVNy0k3jAXv5E_KQ.png" /></figure><h3>1. Do I know anyone else who has the same problem? Are they consciously aware of it?</h3><p>If your idea is worth its weight in gold (from your future exit) then chances are someone else out there — hopefully more than one — is also experiencing the same problem as you. One of life’s best kept secrets is talking to people, and you can unearth a staggering amount of information by engaging in this simple act.</p><blockquote>Hey Jeff, doesn’t it suck when you have to do X every-time Y happens?</blockquote><p>Keep your questions centred on the problem being experienced, and not what your idea is (the most important thing is validating your pain point, rather than your hypothesis of how to solve it — which is almost always wrong at the start)</p><p>How many people do you need to ask to be certain? Well that’s largely dependent on who you think is experiencing your problem. If it’s a problem to do with air traffic controllers, then that’s fairly narrow, so you can probably get away with asking fewer people than if you were thinking of building something a little more mass market.</p><h3>2. Are people “hacking” around the problem today?</h3><p>If the problem is as big as you think it is, people will almost certainly be finding ways around it today. People were using taxis before Uber because the problem of “<em>I need to get from A to B</em>” was a genuine one — Uber built upon that. Maybe people have stringed together a workflow using a series of other products. Maybe they’ve created their own low-fi solutions using something like a spreadsheet. But one thing you can always be certain of if it’s a <em>genuine</em> problem, people will be finding ways to solve it already.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e-VaD6kRYWE53bYzsUJwRw.png" /></figure><h3>3. Is anyone else trying to solve it? What’s wrong with what they’re doing?</h3><p>It seems counter intuitive, but if another business is tackling the same problem as you, that’s probably a good sign. Just because someone else is doing it doesn’t mean they are doing it well — what it does mean is that someone else believes in that problem enough to have also created something to solve it. The important thing is to understand why they haven’t solved it, or where they’re falling short on solving it. Perhaps your solution is novel enough to differentiate you from your competitors. Which leads nicely onto…</p><h3>4. Why would I be the right person to solve it?</h3><p>As it says on the tin: what gives you an edge over everyone else. At Founders Factory we call this our “Unfair Advantage”, and it’s usually something that our investors provide to give us a head start: be that unique market insights, or an existing customer base. When we were coming up with <a href="http://brb.travel/">brb.travel</a> we were able to leverage insights from easyJet that would have been difficult to get on our own.</p><p>For you it might be deep domain expertise in a particular market or industry that you know inside out. It may be some proprietary data set that you have access to. But think why you’re the right person to do this, and what will give you a leg-up over everyone else.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dfhFzw1a3FchfIHB2gntEA.png" /></figure><h3>5. Could I build this myself?</h3><p>This question always forces me to think about what’s involved in actually delivering upon an idea. I’m fortunate in that I’m both a designer and a developer so it’s relatively easy for me to ‘eyeball’ something, but even if you’re not, think about what off-the-shelf tools you could use to string together an MVP. And if you’re able to do it yourself, it’s obviously less risky to just go ahead and try it without worrying about expensive developers.</p><h3>6. How am I going to acquire customers?</h3><p>Thinking about this at the start will not only make you think about where your potential customers are (and how you can speak to more of them), but also get you to consider things you could build into the product itself. For example a referral system, or some type of sharable user content. Most people go straight to thinking about paid media, but there’s so many other alternatives such as targeting Facebook groups or online forums that don’t cost a thing.</p><h3>7. Is my idea heavily dependent on another company outside my control?</h3><p>Being dependent on another business is — well — risky business. Do you need access to 3rd party data from an API you don’t control? Are you building on-top of Facebook? If those businesses decide they don’t like you anymore, what would the consequences of that be? Think ahead and try to mitigate wherever possible: are there backup suppliers you could use? Could you get the data yourself?</p><h3>8. Is there any growth left?</h3><p>Finally, and perhaps the hardest question to answer: <em>is there any significant growth left in this market</em>? Unless you have a crystal ball you’re just ‘guestimating’ like the rest of us, but it’s something you should at least be thinking about. You can probably tell if you’re at the tail-end of a trend: are there signals to suggest that it’s going to keep on going, or is it looking like a bubble about to burst?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RdbByKLhkz6KnJs-8rQgiA.jpeg" /></figure><h3>Something I probably wouldn’t ask myself just yet: What’s the business model?</h3><p>If you can comfortably answer all the above, you can nearly always find a way to monetise it. Sure it’s a good thing to think about, but it’s hard — next to impossible — to accurately plan upfront. Notice how all the questions are centred around the problem or customer need, not your solution. You have so much to figure out about your product still; the exact features, how things will work, what that’s worth to customers. And it’ll almost certainly (and should) change as you learn more.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/397/1*UeJd1_BU0EJrdW-N5TPrdQ.png" /></figure><h3>Do you agree?</h3><p>So those are 8 questions I ask myself when thinking about new ideas. What do you think? Anything I’ve missed, or something you disagree with? Come at me on <a href="https://twitter.com/phillipcaudell">Twitter (@phillipcaudell)</a></p><h3>Find this interesting? Come join us!</h3><p>If you love this sort of thing, we’re hiring! My team (<a href="http://concepting.foundersfactory.com/">The Product Concept Team</a>) is hiring: this could literally be your job. Check us out!</p><p><a href="http://foundersfactory.com/">Founders Factory</a> is a startup that builds startups. We build 13 new startups from scratch each year, and accelerate 35 more. We’re backed by the likes of L’Oreal, easyJet, The Guardian, Aviva and M&amp;S who tasked us with uncovering and creating the next big thing through our Incubator and Accelerator programme.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=88c3f4dd77f3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/founders-factory/8-painfully-simple-yet-pratical-questions-to-scrutinise-your-new-startup-idea-88c3f4dd77f3">8 painfully-simple-yet-pratical questions to scrutinise your new startup idea</a> was originally published in <a href="https://medium.com/founders-factory">Founders Factory</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to build your very own large LED Jumbotron video wall — and go insane in the process!]]></title>
            <link>https://medium.com/@phillipcaudell/how-to-build-your-very-own-large-led-jumbotron-video-wall-and-go-insane-in-the-process-8c515a49a07d?source=rss-a04dd8abbce4------2</link>
            <guid isPermaLink="false">https://medium.com/p/8c515a49a07d</guid>
            <category><![CDATA[raspberry-pi]]></category>
            <category><![CDATA[jumbotron]]></category>
            <category><![CDATA[electronics]]></category>
            <category><![CDATA[led-matrix]]></category>
            <category><![CDATA[led]]></category>
            <dc:creator><![CDATA[Phillip Caudell]]></dc:creator>
            <pubDate>Wed, 18 Oct 2017 20:44:59 GMT</pubDate>
            <atom:updated>2018-03-27T17:55:37.568Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*hhhcO5lvvFAEuVUFq1EUFQ.jpeg" /><figcaption>Behold! (This thing doesn’t photograph well, but trust me: WOW!)</figcaption></figure><p>Hi, Phill here! We recently moved office and needed something to dazzle guests when they entered our lobby (other than the tasteful-yet-blinding yellow walls). We debated all sorts of things from a large 3D printed model of our logo (edgy) down to some nice plants (pretty). In the end, we ended up building a large LED video wall. Obviously.</p><p>So how do you end up building a large LED video wall? One Saturday I was building a train departure board from some LED panels I had acquired (as you do) and wondered if it could be cranked up from 2 panels to LOTS more panels — 48 panels, in fact — to create a large video. Wouldn’t that look great when people entered the office I thought. After all, how hard could it be?</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*5rya4aEtOwd0PXsZCdnpLQ.jpeg" /><figcaption>My initial ‘train’ departure board I had cobbled together with some LED matrix panels and a Raspberry Pi</figcaption></figure><p>A word of warning before you continue and decide you want to build your own:<strong> </strong><em>this project will literally destroy your life</em>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*C9p2LHj4B1kLJcBeUs8soQ.gif" /></figure><p>I’ve blown up components, shocked myself, cut myself, got metal shards in my eye, <em>and</em> had large metal frames fall on me 🤕. Needless to say this isn’t for the faint of hearted. Going from 2 panels cello-taped together showing text, to 48 panels showing 60 fps video is quite a bit different.</p><p>This is by no means and exhaustive guide. I don’t confess to be an expert on building LED screens — this is more of a general gist than an exact recipe.</p><p>Okay now with the warnings and the “<em>I told you so</em>” out of the way, here’s what you need to know.</p><ul><li><strong>Cost:</strong> About £1,000 in raw components. Then about the GDP of a small country if you factor in labour.</li><li><strong>Build time:</strong> Honestly I lost track. I started the project about 3 months ago having no idea what I was doing, and only doing it in my spare time — so maybe you’ll fare better.</li><li><strong>Size:</strong> 2 meters width by 1.2 meters height</li><li><strong>Resolution:</strong> A whopping 384 x 256. (It sounds <em>way</em> more impressive if you say that’s 98,304 <em>individual</em> LEDs you’re addressing)</li><li><strong>Number of panels:</strong> 48</li><li><strong>Number of Power Supplies</strong>: 3 x 5V 60amp transformers</li><li>Some electronics know-how. I’m by no means an expert, but I know 60 amps * 5v = 300 Watts. Just about.</li><li>Patience. This thing took a long time to assemble, not to mention waiting for parts to arrive from China 🇨🇳. Hopefully this post will speed things up and you can order everything in one go (as opposed to ordering a bit, seeing if it fits, ordering more, etc)</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*OhqCqXHfIsVMXJsjTzNs-w.jpeg" /><figcaption>The Pi is cute, but doesn’t scale well (I think we can all relate)</figcaption></figure><p>For my initial LED board I was using a <a href="https://www.amazon.co.uk/gp/product/B01CD5VC92/ref=as_li_tl?ie=UTF8&amp;camp=1634&amp;creative=6738&amp;creativeASIN=B01CD5VC92&amp;linkCode=as2&amp;tag=phillip0a-21&amp;linkId=faca64b7da31d70a3ccf1462faf28382">Raspberry Pi 3</a> with a couple of <a href="https://www.amazon.co.uk/gp/product/B06XNJZN89/ref=as_li_tl?ie=UTF8&amp;camp=1634&amp;creative=6738&amp;creativeASIN=B06XNJZN89&amp;linkCode=as2&amp;tag=phillip0a-21&amp;linkId=b72ef19f969721561d77ae8bbfbeaf10">LED Matrix Panels I got from Amazon</a>. I was using a fantastic library by Henner Zeller (<a href="https://github.com/hzeller/rpi-rgb-led-matrix">rpi-rgb-led-matrix</a>) to drive the panels from the GPIO on the Pi.</p><p>The problem came when I tried scaling up the number of panels. <strong>When it got to about 28 panels, things started flickering</strong>, the refresh rate was poor, and anything with motion looked like crap. It became clear the Pi wasn’t going to cut it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*FVIV0MmljRSBLluWiJW2NQ.jpeg" /><figcaption>Using a breadboard and jumper cables is totally what a professional would use in production. Fact.</figcaption></figure><p><a href="https://github.com/hzeller/rpi-rgb-led-matrix/blob/master/wiring.md">Henner has a pretty good guide on his Github</a> if you don’t plan on going past 28ish pannels. You can still follow the rest of this guide for the frame and power solution.</p><p>Ordering 46 more panels from Amazon wouldn’t have been that cost effective, as it would have been about £27 per panel, so I needed to get my bulk order on — plus the pannels were fairly small in size.</p><p>Thankfully you can find these panels from sites like <a href="http://aliexpress.com/">Aliexpress</a> / <a href="https://www.alibaba.com/">Alibaba</a> or <a href="http://dhgate.com/">DHgate</a> (where I placed my order). My supplier is no longer — urr–supplying, so you’ll need to have a little search around for your own.</p><blockquote>The panels I used were <a href="https://www.dhgate.com/product/yosa-led-display-p5-32-64-pixles-320-160mm/375715254.html">P5, 32*64, 320mm x 160mm, 1/16 scan, RGB, HUB75</a>. Catchy. That got the cost of a panel down to about £18. Nice.</blockquote><p>After waiting about 2 weeks for them to be shipped from China, they arrived!</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*vb4HuTrmoUVz1kH_jZHJDA.jpeg" /><figcaption>So. Many. Panels.</figcaption></figure><p>It’s worth noting that the panels should also come with power and data connectors (IDC 16 Ribbon Cable).</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*nJgBXFoB4q9Bh4vnHrV8Ew.jpeg" /><figcaption>Power cables should come with the panels</figcaption></figure><p>I have a lot of regrets on this projects, but by far the biggest is the frame. Oh man, how hard could it be to fix 48 panels together?</p><p>The panels came with large magnetic flat head screws that can hold the weight of the panel when stuck to a metal surface. So we designed a frame of horizontal metal bars at regular intervals to stick them to.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*p0PuaFNnefNgp4bFpbXCKQ.jpeg" /><figcaption>Dianogal bars were added to stop it wibbly-wobbling (technical term)</figcaption></figure><p>And this would have worked quite well, <em>if</em> it weren’t for the fact lining up the bars to match the exact position of the screws is incredibly fiddly, and requires <em>very</em> precise measurements and drilling. Even the smallest drift in measurements meant the screws wouldn’t make contact, and the panels wouldn’t stick.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*RNcaKzVJowHbkzVbH5geCA.jpeg" /><figcaption>Drill baby, drill</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*u2p-BA-xYWF1ng9N1DkX9g.jpeg" /><figcaption>Me after drilling and cutting the frame to the wrong measurements. Again. Don’t do that.</figcaption></figure><p>After drilling and fixing all the horizontal bars to the vertical, the 4th, 5th and 6th row of panels didn’t make contact, and wouldn’t stick.</p><p>Argggh!</p><p>So I re-drilled one row at a time and physically placed the panels on to check they were making contact.</p><p>Even then when I stood everything up right the 4th row <em>still</em> didn’t make contact. It was just every so slightly out. So I used cable ties to support the 4th row.</p><p>I mean this frame <em>works</em>, but if I was starting the project again I would definitely use taller bars so there’s more wiggle room for the magnets to stick onto. (It also means once it’s built it’s impossible to move it without taking the panels off, otherwise they’ll just fall.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*ZOnBqwJ8qegVekiZMesmBw.jpeg" /><figcaption>Then the panels stick on with magnets. Neat (until you move it and they all fall off)</figcaption></figure><p>So the manufacture says each panel has a maximum draw of <strong>18 watts,</strong> and the panels require<strong> </strong>5 volts<strong>. That gives us a draw of 3.6 amps per panel </strong>(Watts / Voltage)<strong>. </strong>So for 48 panels we’re gonna need a 172.8 amp transformer. Now I’m no electrician, but suddenly this thing sounds very dangerous.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*d9Bo9W7RrGlz74VH3nTPpg.jpeg" /></figure><p>The highest amp transformer I could find was a<a href="https://www.amazon.co.uk/gp/product/B01DDTYMIU/ref=as_li_tl?ie=UTF8&amp;tag=phillip0a-21&amp;camp=1634&amp;creative=6738&amp;linkCode=as2&amp;creativeASIN=B01DDTYMIU&amp;linkId=03e01dc4ac6b837de1e2b342dac95226"> 5V 60A transformer from Amazon</a>, so <em>three</em> are needed to give us 180 amps.</p><blockquote>Don’t do what I did, which was get overly excited and want to try out more panels than the transformer could handle. I don’t know what I expected to happen, but quelle surprise, it overheated and never turned on again. Don’t do that.</blockquote><p>There’s a few ways of doing it, but I found the simplest way was to have it ‘star out’ from the transformer. An individual transformer powered 16 panels (2 along and 8 down).</p><p>If your cables have little metal hooks, clip them off and strip the plastic back. I then fed cables from multiple panels (4 wires that then branch off to 8 panels) into a terminal block.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*OwUrmp_H3dkMHSzT98l8AA.jpeg" /></figure><p>I then fed a wire from the 2 terminal blocks (upper and lower panels) to the transformer. Repeat for the other transformers. And it should be as simple as that for power!</p><p>Finally you’ll need to hook up the data cables between panels. That’s just standard IDC 16 ribbon cable that should have come with the panels.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*f9-YsFT0F_TNZGiKhl0spw.jpeg" /><figcaption>Power and data cables linking panels</figcaption></figure><p>I mentioned earlier the Raspberry Pi wasn’t sufficient to power that number of panels, so what is?</p><p>In the end I ended up turning to some pro hardware that’s used by actual manufactures of these displays. The names vary, and there’s a couple of brands, but they’re generally called ‘<em>LED sender and receiver cards</em>’.</p><p>The sender card plugs into a computer or other HDMI input. An ethernet cable then connects the sender to a receiver card, which in turn connects to the LED panels via ribbon cables.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*F4fZYbwJoc4wohGn-C3O9A.jpeg" /></figure><p>I ended up ordering:</p><ul><li><a href="http://www.ledcontrolcard.com/linsn-led-card-c-18/linsn-ts802d-sending-cardlatest-linsn-sd802d-sending-card-p-542.html">Linsn TS802 LED Card (Sender)</a></li><li><a href="http://www.ledcontrolcard.com/linsn-led-card-c-18/linsn-rv908t-led-receiving-cardlinsn-rv908t-led-receiver-p-468.html">Linsn RV908T LED Card (Receiver)</a></li></ul><p>These are considerably more pricey than the Pi, clocking in about £140 in total with shipping and tax, but the results are dramatically better. You’ll also need a dedicated computer or other device to drive.</p><p>You’re going to need to cut your own ribbon cables, as otherwise each row won’t reach the card. I bought some <a href="http://uk.rs-online.com/web/p/flat-ribbon-cable/2899874/">16 way flat ribbon cable</a> from RS components, as well as matching <a href="http://uk.rs-online.com/web/p/idc-connectors/6741101/">IDC connector heads</a>. You just need to clamp down the cable between the heads of the connector, and voila, custom ribbon cable!</p><p>The receiver card has some very tiny markings numbering each slot, starting from 0 to 11, so your top row goes to 0, second to 1, and so on.</p><p>You’ll need to wire in power to both the sender AND the receiver card. For the receiver card I used the spare lane from one of the transformers at the back of the display. I got a seperate 2.5 amp plug for the sender card, as it was going to be some distance away from the display (in our comms room, connected through a ~150ft ethernet connection).</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*5neLOSFZObrEHcd79wh6bg.jpeg" /><figcaption>The display is powered from our comms room in a 3D printed custom cardboard box enclosure</figcaption></figure><p>The bad news: you’re gonna need to configure the card from a Windows installation. The badder news:<em> the software was designed by the dude who made the original Geocities template</em>. The good news is you’ve made it this far without injuring or killing yourself. The final horcrux.</p><p>By the way I’m assuming you’re using the same cards and same panels as the rest of this guide, otherwise you may need to tweak some of these steps to fit.</p><p>Start off by connecting the USB and DVI to your computer. I was using my Mac with Parallels to run Windows. <a href="http://linsnled.com/linsn-download.html">Download LED Studio</a>.</p><p>Open up LED Studio and go to Option &gt; <strong>Software Setup</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*bbEjcSH_ZOMfu65Pi92HTw.png" /></figure><p>Then without clicking anything type the letters “<strong>linsn</strong>”. A dialogue should pop up asking you to input a password.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*C3Z-g56S_yB39rUGR80c4g.png" /><figcaption>Ooo a mysterious dialogue that can only be revealed by knowing not one, but TWO passwords</figcaption></figure><p>Now the password is “<strong>168</strong>”. Why? I have no idea, but you should now have access to some advance settings (aka the ones we need to configure the card). On the tabs along the top, select “Reciever 1”. Then in the “<strong>Load capacity setup</strong>” enter the width and height of your display in pixels. So if each panel represents 64 pixels, and we have 6 going along horizontally, then the width would be 64 * 6 = 384. Same for height, 32 * 8 = 256. Then click “Send to receiver”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*YcSWlNsOLfDKTfxeQyHktg.png" /><figcaption>The cockpit: this is Jumbo001 we’re ready to taxi, over.</figcaption></figure><p>And them boom! That’s it, you <em>should</em> be seeing something on your display.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*K6smzM5UavLE2a-W5GamhQ.jpeg" /><figcaption>Sam being ecstatic that after 3 long months of my bitching it works!</figcaption></figure><p>Honestly, it could be so many things. A lot of this was just trial and error, so like I said, have some patience, step through things methodically, and have a play around with settings if you’re not getting a good picture.</p><p>A quick word on what’s actually driving the content. It’s a Mac Mini running a custom app. It has a JSON file that can define playlists of content that can:</p><ul><li>Play things on specific days</li><li>Play things at a specific time</li><li>Play videos, photos and gifs</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*hxCPu2HNGfLC4Ff_gONPpQ.png" /></figure><p>It also has a Slack channel where you can DM it messages and gifs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*vdz90mPW1ytefT3M2QD0hg.gif" /><figcaption>HELLO!!!!</figcaption></figure><p>Anyways, that’s it! If you decide to go ahead with this, good luck! Happy Jumbotroning.</p><p><em>Originally published at </em><a href="https://medium.com/@incubation.ff/how-to-build-your-very-own-large-led-jumbotron-video-wall-and-go-insane-in-the-process-beef53363ba1"><em>medium.com</em></a><em> on October 18, 2017.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8c515a49a07d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[We built an award-winning product to help the visually impaired in under 48 hours.]]></title>
            <link>https://medium.com/founders-factory/we-built-an-award-winning-product-to-help-the-visually-impaired-in-under-48-hours-eb6115ca8ec7?source=rss-a04dd8abbce4------2</link>
            <guid isPermaLink="false">https://medium.com/p/eb6115ca8ec7</guid>
            <category><![CDATA[hackathons]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[design]]></category>
            <dc:creator><![CDATA[Phillip Caudell]]></dc:creator>
            <pubDate>Wed, 07 Dec 2016 11:09:41 GMT</pubDate>
            <atom:updated>2018-03-27T17:56:04.157Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>We built an award-winning product to help the visually impaired in under 48 hours. Here’s how we did it.</strong></h3><p>We spent the weekend at <a href="https://techcrunch.com/event-info/disrupt-london-2016/disrupt-london-hackathon-2016/">TechCrunch Disrupt’s 2016 London hackathon</a><strong>. </strong>It was a pretty intense 48 hours of brainstorming, plotting, researching, eating, building, testing, bit more eating, topped off with a presentation to a few hundred attendees in the room. Gulp. But we won a prize. The special IBM Watson award!</p><p>This was the first time we had donned our branded hoodies and stormed a hackathon as a team, so we were pretty happy with the outcome. Here’s some more details about what we built.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C1uIrel9tppPnBnntNT3NQ.jpeg" /><figcaption>Photo credit: Farah Kanji</figcaption></figure><h3>Inspiration</h3><p>One of my family members has an eye disease called Glaucoma which damaged the optic nerves in their eyes. The damage has resulted in severe vision loss, meaning they struggle to find everyday items around them. With advancements in computer vision, it seemed reasonable that we could replace their eyes with an artificial one.</p><h3>What It Actually Does</h3><p>AEye is an artificial eye. Using a combination of the microphone and camera on the device, someone is able to ask where an object is and be guided to it. Moving around the room will result in a “hot” or “cold” reading.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FJjOI-BBcWFE%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DJjOI-BBcWFE&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FJjOI-BBcWFE%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/253dfaddc0835ecb66564cf65ce9abfa/href">https://medium.com/media/253dfaddc0835ecb66564cf65ce9abfa/href</a></iframe><h3>How We Built It</h3><p>Split into two teams: 3 people concentrating on the mobile app &amp; the UX design; 2 concentrating on the backend and visual recognition training.</p><p>We have trained multiple image classifiers on the Watson visual recognition service. This can give a reliable match when the image contains the object we’re searching for. This service is wrapped by a Python web service, which maps the desired object class to the associated classifier.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FLDwiIAzD-Do%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DLDwiIAzD-Do&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FLDwiIAzD-Do%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/d273dc04cd20d66a30cb084884100e1a/href">https://medium.com/media/d273dc04cd20d66a30cb084884100e1a/href</a></iframe><p>The iOS app uses the Watson Speech to Text service to translate voice input from user. Once the user selects a class to search for, we give aural feedback (“warmer”, “colder”, “found it”, etc) as the user moves the phone’s camera around the room or surface.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HkEwuTZbVo4EehuLUuPEeA.jpeg" /></figure><h3>Tech We Used</h3><ul><li>Swift</li><li>Python</li><li>IBM-Watson</li><li>Radix</li></ul><p>So yeah, without a doubt the next step would be to add celebrity voiceovers. Jeff Goldblum perhaps?</p><p>Looking forward to the next Hack! 😬</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*O4wcL4saDoISGKggsp_yQg.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=eb6115ca8ec7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/founders-factory/we-built-an-award-winning-product-to-help-the-visually-impaired-in-under-48-hours-eb6115ca8ec7">We built an award-winning product to help the visually impaired in under 48 hours.</a> was originally published in <a href="https://medium.com/founders-factory">Founders Factory</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>