Direct Download – Subscribe via RSS – Subscribe via iTunes – Subsribe with Zune
Welcome to Installment 003 of The Custom Integrator Show. First of all, thank you for the feedback and encouragement. It sounds like at least some of you are starting to see where we are going and are getting some ideas on things to consider as a Windows Media Center ecosystem integrator. We finally start digging in a little deeper this week and take a look from the inside out. In other words, we start to get into the Windows Media Center PC itself, some of the requirements you need to consider when matching capabilities to feature sets and deliverables as you analyze the machine’s subsystems. We also touch, once again, on looking to the future and how to start drawing some lines between what you may need to provide for the future versus how far out you want to take that “cutting edge.”
This installment is the lead-in to several others regarding the attributes of the hardware and how to start categorizing the technical architectures within the box. For this one, we look at considerations when evaluating the best location and some of the trade-offs between cooling and noise levels. We also start to touch upon selecting various source components and some of the current limitations when dealing with high definition and premium content. Future installments will take a cursory view of the audio and video, which naturally will start moving us out into connecting up video displays, touch panels for control, and distributed audio.
Larry Greenblatt had posted a question on the blog (http://thedigitallifestyle.com/cs/blogs/custom/archive/2008/09/29/a-little-bit-about-me.aspx) regarding how to take advantage of his new high definition cable service from within Windows Media Center. Although I provided a relatively short response, I would like to start to provide an overview of how we plan to start organizing and covering this type of information because it can become very complex. I have taken the approach of looking at the Windows Media Center ecosystem in much the same way you would look at any complex architecture from the enterprise and IT worlds and came up with my version of a Three Tier Architecture. The current version is US centric (sorry Ian) right now, but we could extrapolate to cover other regions, especially for the sources (much of that is changing for Windows 7 anyhow, so I will wait to update it for a bit). Here is what I came up with:
As you can see, it becomes pretty complex fairly easily, so hang in there Larry, we will get into it fairly soon. It is our hope that we will be able to drill down into each one of these components and look at why you might want to take a certain approach for using one technology or implementation over another. Through it all, we will try to emphasize what about each layer is important to the integrator and why you need to understand the ins and outs of them to really be successful. We are not going to be able to accomplish this in just a few sessions, but we wanted to get you on board with what we are thinking and the direction we would like to take this podcast.
Thank you for listening and welcome aboard the ride. Hopefully, it will not be too bumpy along the way.
4 thoughts on “The Custom Integrator Show Installment 003”
Like a moth to the flame… derek be mindful of complex cuz like ego it will kill. I know this series is suppose to be more remedial to align those out of the know into the known. However the graphic appear to screem out COMPLEXITY, so much so that you even call that out. My comment is simply this. Most of what you are introducing to us is METHODOLOGY based – aka best practices for “getting it down”. To aviod complexity and to temper your content for you audience I would think that the methodology would be less waterfall/heavey process and more agile and iterative based. Meaning Iteration 1. Get a PC footprint in the account and use it to collect digital assets/backup PC’s (bridging of media and IT support), Iteration 2. Integrate it into A/C cable, Iteration 3. Extend the ecosystem to distribute media, Iteration 4. Maybe some automatio, Iteration ….
Anyway… u r the expert I just wanted to call you out re your own guiding principles and goals. I am sure you will cover all these things and that you want to get your refernce out there, I just think that to be succesful with the “cable guys” we need to spoon feed them the content in little doses and aviod complexity out of the gate. It is 12:20 AM, I am a fat finger typist and do not really give a hoot about grammar/spelln as I am more focused on the message and content. So my message is reduce complexity, talk about small, measurable (agile) byte 🙂 sized bits of information… Oh and if that is what you are doing, then simply look at this post as an affirmation that I get what you are trying to accomplish…
Thanks for the input. You are right, we are trying to take the 10,000 foot level right now and the diagram is digging in pretty deep. In trying to break out all of the components involved with the entire ecosystem “package” and come up with a way to discuss the individual technologies in a logical way, I ended up devising an architecture that “modularized” the pieces into something manageable and came up with what you see. I think we are on the same page if we look at just the generalized “three tier architecture” compartments, much like you would with any IT-oriented project. That is where we are now in our discussions.
At some point, however, one still needs to understand all of the factors involved. What I was trying to convey to Larry is that it is tough to get from A to Z in a few half-hour talk sessions, but I wanted him to see what is over the horizon. We are not at the point of evaluating and designing a solution for the various source options like his high-def cable setup yet, but I think we at least have a path and plan for getting into it now. As you know, implementing and testing all of the minutia involved with delivering a solid solution takes years of hands-on experience. Trying to convey that experience requires a methodology of starting at the top and peeling away the layers as you move down the stack. With what I put out there, I exposed the stack already. I still think it will provide a useful model for conveying where we are going over time though.
My thought is that I will use the blog space for defining all of the terms associated with the individual compartments and components on the diagram. Once we are all have a common understanding of the terms and what the technologies are, then we can start to look at how to evaluate the options when addressing each one of them. We then can start wrapping “the glue logic” around them as a way to delivering a solution. I think using the diagram and written text will be a good complement to discussing how to work with the pieces via the podcast and will save a lot of “hot air” trying to explain all of them verbally. I do have another one that starts to detail the glue logic and technologies that it takes to integrate the components on this diagram, but I will save that one for a later time…..
I wanted to make a post to clarify something that I said about three quarters of the way through Installment 003 (yes, I actually listen to them myself to see how it went). It was regarding TV tuners and using USB tuners versus PCIe-based tuners. I made an off-the-wall statement that inferred using USB-based tuners would see congestion because of bandwidth limitations of USB. What was I thinking?!??
First of all, the CableCARD Digital Cable Tuners currently are all USB-based, so obviously USB can handle that. Also, if you do some simple “worst case” math, four streams at even 25 Mb/s is not going to eat up the 400+ Mb/s of USB 2.0.
What I was thinking in my mind was an installation we saw recently where they used four USB-based ATSC TV tuners mounted on the wall at the DEMARC connected together via a USB hub to a port on the Windows Media Center PC. They ran Windows Media Center TV Pack 2008 to enable the four tuners their homemade box. They were having problems with pixelations and stuttering in addition to taking too long to switch channels. It turns out they had used some fairly inexpensive tuners. The biggest problem was that their signal levels too low, which was obvious when running the WMC signal strength wizard. It appears the tuners were not as “efficient” as those used for similar internal PCIe cards because replacing the external USB ones with internal PCIe tuners cleared up the problems. I guess my mouth was engaged before my brain when I was trying to describe something to look out for when evaluating tuner types. One of the drawbacks to a “live” show.
Anyhow, the point I was trying to make is that you need to look at every component in the chain and optimize each piece for the application. When looking at tuners, try to go with the best and highest bandwidth ones you can. We have seen other issues with some of the USB-based tuners. Not that they are not good, but they usually sacrifice quality, sensitivity, and bandwidth for size and convenience. If you have a choice, the internal PCIe tuners usually are of a better quality and will give you better results in the end. Sorry for the confusion.
Installment 003 digs into some of the deeper aspects of delivering HD content from cable cards and over the air sources. While some of this is “baptism by fire”, most is typical of the potential issues currently involved. The assumption here is that while Windows Media Center is very reliable at delivering content from SD cable/satellite and OTA/ATSC tuner sources in addition to serving up “on demand” content for instant viewing and, of course, home video and pics. While the skills required to successfully design and build a WMC PC combine strong experience in PC HW/SW and at least a general familiarity with A/V electronics engineering. Successfully building a WMC PC is NOT for everyone. While someone with good A/V skills can successfully install Home Theaters, it’s usually wise to purchase the WMC component – at least initially. It will be difficult or impossible to avoid the desire and requirement for HD content, however. Selection of an off the shelf (or customized) WMC PC does require a good understanding of product offerings and the customer’s expectations. Integration and automation obviously require a fairly strong understanding of the fundamentals to be successful. This installment is worth hearing twice. Nice work, Derek. Looking forward to # 004. Thanks!