I have been reading this book for a while now. FYI, I know nothing about the device and the SDK, while totally being interested to learn it. I have bought this book and I am gradually learning things that are new to me. I don't know why you are fustrated. I used to get fustrated when I cannot understand the code or cannot run the code. But this book has made extra-ordinary work by pulling things that are rarely found over internet. Believe me, I would recommend this book always for a newbie.
I think probably you might not be very competent on technologies used on the book. I dont know if you are aware of WPF or other ongoing technologies which are widely used on the book.
Let me go a little on your findings and let me try out them myself.
1. You seem to have managed to build InfoBox sample, that means you have a real device with you and it is connected. 2. It does not give complete picture of Skeleton Tracking. Well, it is true. I think you need to read the book further to get the entire idea. This chapter overlooked on it intentionally I think, because there is a separate section for the same. 3. Hmm, yes, there are sections which are intentionally cut down to decrease the page count. I know it might be hard for a beginner like you to code it completely. But probably this book is not intended to give insights on actual programming, and I think you need another book to learn WPF and/or windows programming before you can go with this book. This book is intended only for Kinect SDK. 4. Page 169. I dont know why the code sample Failed for you. Did you added proper namespaces? As I see the code uses Linq, and probably you need to add that namespace in addition to the SDK ones. Can you specify your error message so that we could get an idea what you are missing (probably if you cannot fix it yourself). 5. SensorChooserUI is not a part of SDK, it is a part of toolkit. I think the book already states how to use Kinect developer toolkit in Page 205.
I think you should read the book again before you comment again.
@Simon did not see Amits explanation and your follow up comment. Sorry about that. I have already edited my text on my comment, But I will request you to edit your first comment since it is not appropriate and looks bad when anyone visits this thread. I am glad to meet you online.
Amit clearly said that in HASH operations, HASH table data gets spilled down to temDB if memory is short on box. You are hearing only what you want to hear. You are taking DBCC statements too far and everyone in this thread has clarified that so I would not like to comment on that. When you warm up the cache, how do you measure the time taken to read data from the disk? This is what I think the speaker meant by Physical IO. Even if I would have spoken to somebody this is what I would do to demonstrate actual physical IO. I cannot measure an IO when some page is loaded from cache. He, in his session, wants to measure that time only. Thus, DBCC DROPCLEANBUFFERS is needed in his demo.
And your testing with VSTS DB is really a not a good thought either. When I want to show query/index performance rather than application performance, why should I include VSTS? If you test from VSTS DB (which DBAs dont generally do) you are adding more noise from the client & network latency. You need to test raw performance of SQL DB engine and this is what the session showed.
What do you mean by show it in windows and record in notepad?? Funny it is that you work in Infy; very surprised !!! Really want to test SQL performance?? Use tools like SQLIO...
Last but not the least, your email id is incorrect. I didnt want to spoil this thread and get this discussion in personal by sending a mail few minutes ago, but it bounced back. Fake users like yourself de-motivate all the good efforts of the MVP community. I request Channel 9 to delete your comments.
@Nauzad Kapadia and @Jerald Fernandes: +1000 from me. Practically speaking, I always try to motivate speakers on the community. And on this Demo, there is really no means of criticisms. Demo oriented session really touches mind of attendees. Even though theories are important, but as an audience, I generally like these kind of demo heavy sessions more than theories.
Just after looking at the first comment, I got an impression that the session might not be upto the mark. Simon rightly marked that DBCC FREEPROCCACHE is not safe to run in production.
But why do you need to comment such a crap on the very top of the comments section, when the Author clearly said before running the command that "If you are tired of your job, run this in production". This clearly makes sense.
The demonstration which uses these statement is about "choosing right indexes", and the demo is absolutely fine and really fruitful to the DBA's.
I really like this session the most, its absolutely a stunner from Amit. These important tricks will really guide me during my work.
@Simon, Sorry, last time I didn't see your comment and responded. I would recommend to please edit this comment as Amit already explained. It actually gives bad impression to the thread when the first comment looks like this. I personally feel sad when I see such a nice session gets such criticisms.
@Dinesh Do you have a profile ? Can you show the content of your blog and your online activities before speaking something about the speaker. Well, people like Amit turns out to be a god in SQL server. Even when I speak about internals of CLR I always hide few things, and it makes perfect sense. If you have spoken anywhere before, you would have come accross to this too. It makes sense to speak in abstraction otherwise it would be difficult to catch up with the audience, but it is important not to speak false statement. I am quite aware that the speaker didnt made any false argument. Dinesh, I dont know why you need to hide yourself if you are sure about something...
I really enjoyed this session and I appreciate Amit for spending time and delivering a session for TechEd NA. Because of these sessions, Teched gets its glory. Cheers mate.