How to record a technical demonstration video

Yesterday I wrote an article that highlighted the new series of technical videos that various community members were putting together for Citrix. Since then I've received several emails from people asking how these videos were recorded.

Yesterday I wrote an article that highlighted the new series of technical videos that various community members were putting together for Citrix. Since then I've received several emails from people asking how these videos were recorded.

In this article, I'll describe exactly what you need to do (and what equipment you need) to make a videos like these. If you make your own videos, send me an email and I'll get them posted.


The good news is that you only need to invest about $500 to make good quality videos.

From a software standpoint, you need software that records the screen activity as well as your voice. I like Camtasia Studio because in addition to recording voice, screens, (and if you want) live video, it also has a basic editor that lets you cut out stuff, add titles, boxes, animations, links, etc. It can also export your video into almost any format using any codec. The price is right at $299.00 too!

For the Citrix videos I linked to in the intro of this article, I did everything in Camtasia with no other software. If I messed up, I paused for a few seconds and then just re-said what I was trying to say. Then after the whole recording was done, I looked for the big sections of silence and deleted the parts that I screwed up. You can also use Camtasia to mix together multiple videos or to cut and paste segments from different recordings.

A lot of people think that once you have Camtasia, you can just plug any cheap mic into the “mic” input of your sound card. While that will technically work, you will get much, much better results if you buy a better quality audio capture card. Fortunately due to the popularity of podcasting, a lot of audio hardware vendors have starting selling podcasting “kits” that include a decent mic, a high-quality USB audio interface, and some software. In our case we don’t need the software from the kit (since we’re using Camtasia), but we can buy a kit for the mic / USB interface combo.

One of the most popular kits on the market is the M-Audio Podcast Factory for $180.

Podcast Factory

If you’re getting some “popping” sounds from the letter “P,” you can go to Radio Shack and buy a foam pop / windscreen that you can stick over the end of the mic for about $5.

Finally, it’s probably a good idea to grab a decent pair of headphones so that you can listen to your voice as you record it. This will help you make sure that you’re not “clipping” (exceeding the limits of the input) or fading out too quietly. While it’s possible to adjust the audio level throughout the recording with the Camtasia software, it’s much easier and faster just to get the levels right when you make the recording.

You should get some of those big “over the ear” style headphones so that they block out some of the background noise. It doesn’t really matter which ones you get—pretty much anything will do. You can get these as cheap as $20 nowadays, and anything over $50 is not needed for this kind of work.

The Setup

You’ll plug your mic and your headphones into the M-Audio box, and then that box connects to your computer via USB and essentially becomes an external soundcard. In the audio world for Windows, there are two types of drivers: ASIO and WDM. The M-Audio box will be able to operate in either mode. You should use the ASIO drivers as they are better quality. You should be able to control this via the control panel once you install the drivers.

You’ll also want to make sure that you configure the Camtasia Recorder to use the M-Audio interface. You can configure this via the Camtasia Recorder software screen. (Tools | Options | Streams tab | Audio Capture Device) All the remaining default settings should be fine.

I usually change the resolution of the computer that I'm recording from to 800x600. That way the final video will fit nicely on a screen that's 1024x768.

When you’re done recording, you’ll end up with a .CAMREC file that you can import into Camtasia Studio to edit.

When you’re ready to output (or "produce") the file for the public, Camtasia has several settings and options. WMV format is probably the best for these purposes. Once you choose the type, you’ll get to a screen where it asks what “profile” you want to use. A profile is a collection of settings that control the output of the file.

You’ll probably want to make two versions of the video—one for the web and one that’s lossless. I’ve created two downloadable profile settings files here:

Unzip the two .PRX (Windows Media Profile) files to the "Windows Media Profiles" folder under the Camtasia install folder. They should show up automatically after a Camtasia restart. (You’ll access these when you produce your video, File | Produce Video As | WMV | Next | Profiles drop-down list) You can click “edit” here to import and export profiles.

Branding the Video

When you make your video, you'll probably want to package a logo or some other branding element into the video so that people know what they’re watching. When branding videos, there are two schools of thought:

  • Create a little HTML web page wrapper that surrounds the WMV file.
  • Embed graphics or branding elements directly into WMV file itself.

Each technique has its advantages and disadvantages.

Putting branding elements in an external HTML wrapper is cool because it’s easy to change these elements at any time. (Just change the HTML and you’re done.) The main downside though is that the since the actual WMV file itself doesn’t have any branding, if someone does a “right-click, save as…” to save the video file directly, they’ll have the video without the branding.

Therefore the preferred method is to embed the branding elements directly into the WMV video file. (Of course the downside to this is that you’ll have to re-render the WMV if you ever change the branding.)

The good news is that Camtasia can do this via it’s built in “watermark” functionality. The idea with a watermark is that you can stick a logo or something in the corner of the video that will hover over all the content (just like the local TV stations do). The problem of course is that since our videos are only 800x600 resolution, we really can’t afford the extra screen space to be consumed by our watermark. Fortunately, we can still use Camtasia’s watermarking functionality in a different waay.

The best way to do this is to create an extra 25-pixel tall “strip” along the top and bottom of the video. This will essentially mean that we’re taking 50 pixels of vertical height for our branding strips. To compensate for this, we configure Camtasia to render the videos at a resolution of 800 x 650 instead of 800 x 600. If we center (do not stretch-to-fit) our 800 x 600 video on an 800 x 650 rendering, then we’ll end up with a 25 pixel empty strip along the top and a 25 pixel empty strip along the bottom.

Then all we have to do is tell Camtasia to put our logo strips in those blank spaces. To do this, we first need to create the branding graphic file. We need to create a single GIF file that is 800 x 650 pixels. We can put whatever we want in the top 25 pixel and the bottom 25 pixel strips, but we must leave the center 800 x 600 area empty (and configure our graphics software so that area is “transparent”). Here's how your overlay GIF should be laid out:

Then in Camtasia, we render the 800 x 650 video with our 800 x 650 GIF overlay, and the 800 x 600 transparent section lets the video “show through.” To do this in Camtasia:

On the video size screen, choose the Custom Size, and enter 800 for the width and 650 for the Height. Ensure that "stretch to fit" is NOT checked.

  1. On the video options screen, check the box that says "include watermark."
  2. Click the "options" button.
    1. Browse to the path of the overlay GIF
    2. Ensure that no boxes are checked in the "scaling" section
    3. Click the center box in the "position" section

Final Steps

The final thing you should do before you render the video is to check / change the metadata. This will control what words scroll across the bottom of the user’s media player when they watch the video. You want these to say something like “Created by Brian Madden” instead of “Created with Camtasia Studio,” etc.

In Camtasia, you can change this via the Production Wizard (which is the same set of dialog boxes you use to add the watermark.)

On the video options page (in fact the same page where you add the watermark), click the “Options” button in the “Video Info” section. Here you can specify the metadata for the video.

IMPORTANT: Camtasia “remembers” the options you used last time, so make sure you change the title and other metadata of each video before you render it.

Other Tips

  • Record everything at 800 x 600 so that the videos will look good on 1024 x 768 screens.
  • Remove the clock from the taskbar on the computers that you’re recording since you’ll probably be skipping around as you edit together the final video, and you don’t want that little clock in the corner letting everyone know how much you suck at recording stuff!
  • Stop any background items that might pop up on your screen. (IM or email notifications, etc.)
  • Stop any background processes that may take resources. (For example, Google desktop continuously indexes new content, and this can cause what sounds like static or pops to be recorded into your videos. Killing the Google processes fixes this.)
  • Record a short sample and move it into Camtasia and produce it as a WMV to make sure that everything sounds and looks right.
  • Rendering takes a long time—potentially several times as long as the actual video—so be patient.

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Hi Brian,

Good tips... I love the one about the clock, indeed you see strange timewarps in some of my video's.

One additional tip is to record a windowed RDP session, RDP has several advantages:
Camtasia Recording can be a resource hog, expecially combined with virtual machines on the laptop/workstationBecause you are recording a windowed desktop, other applications cannot interfere by popups if you place it strategicallyWith RDP (or ICA) you can customize the Desktop size in detail. My favorite resolution for the recorded desktop is 610 x 990 as this perfectly fits in a browser on a 1024x768 Monitor. The commandline is: mstsc  /console /h:610 /w:990 /v:DemoSystem
PS: For a size perspective it seems the best format to produce your video is flash, the quality is fenomimal in comparison to its size. You only need to up the sound quality, default it is a tad grainy.

PS2: Is Camtasia 4 worth the upgrade from version 3?
Big time. RDP recording is the way to go in many cases. That is why RecordTS was created in the first place! :-)
I'm sure TSrecord is a cool tool, but my comments were in respect to recording a Windowed RDP desktop with Camtasia.
I myself have used Camtasia also. Great product, but I prefer Qarbon's Viewlet Builder. Has great features, cost is not bad and does a great job of creating the webpages and Flash files easily.
Check them out.
I dugg it.
I know Jeroen! The main question is if you are going to record an RDP session why use Camtasia that is a resource hog? RecordTS in this particular case (RDP session recording) is way more efficient/scalable. :-)
Camtasia 3 is "good enough".  Version 4 Adds some features I find useful.  Especially some audio enhancements that clean up the sound on my cheaper mic.  Also zoom&pan is good.  And note: while it can work on Vista it crashes all the time so don't upgrade your recording PC like I did!
Check ObserveIT Terminal Session Recording capabilities enable continuously recording all Terminal Server sessions, Citrix or any other remote control software such as VNC, NetOp etc, in a compressed and searchable format. It is an enterprise ready solution (1000 monitored servers = 25GB per year (