Blogcast how-to: Recorder Settings for Windows Media Encoder

Numerous people have asked me what exact settings I used to recently record the series of VSMT blogcasts.

The software I used to record the demonstrations was Windows Media Encoder, which is freely downloadable on (Note that this also works correctly under Windows Server 2003. You may need to perform the download from a Windows XP machine and copy it across to your server.) Yes, there are many other products available on the market, but remember Windows Media Encoder is 100% free!

The settings I list here gave a reasonable balance between a reasonably decent playback experience while trying to limit the bit-rate to around 100K. I'm not saying the following settings are definitive though - if you know better, please provide feedback.

You can also download my settings file here if you want a quick way of setting it up. Save the .wme file from the ZIP download to "My Documents" and use File/Open when you start Windows Media Encoder to use those settings.

The most important thing to do is to adjust your display settings. I used 800x600 at 16-bit colour depth.

Start Windows Media Encoder and cancel the wizard
Click the Properties button

On the sources tab

  • Enter an appropriate name
  • Ensure Video is checked and "Screen Capture" is selected
  • If appropriate, click the configure tab next to video to select which window or region is to be recorded. I tend to find full screen capture most useful (the default)
  • Ensure Audio is checked

On the Output tab

  • De-select "Pull from encoder"
  • Check Encode to file and enter and appropriate filename (with a .wmv file extension)

On the Compression tab

  • Under Destination, select "Web server (progressive download)"
  • Under Video, select "Screen capture (CBR)
  • Under Audio, select "Voice quality audio (CBR)"
  • Under Bit rates, select "97Kbps Bit rate , 10 frames per second, Output same as input"

Note that the screen shot you see here will change when we edit the settings in the next step
Click Edit

General Tab for Compression settings

  • The Name should be "Voice quality audio (CBR) / Screen capture (CBR)
  • Select the appropriate language (unless you are american)
  • Ensure under media type for audio, mode is CBR and the Codec is
    Windows Media Audio 9 Voice
  • Ensure under media type for video, mode is CBR and the Codec is
    Windows Media Video 9 Screen
  • The Target bit rate box should display 97Kbps (97000bps)
  • The video format should be custom, and neither check box for interlaced
    or non-square output should be selected

Select the "97K" tab

97K Tab for Compression Settings

  • I tend to up the buffer size from the default of 5 seconds to 8 seconds
  • The video smoothness is changed from 50% to 100%.
  • All other settings are left at their default settings.

Click OK to return to the compression tab

Back on the Compression tab

The destination, video and audio drop-downs should all now read "Custom".
The details box will contain the following information

Audio encoding mode: CBR
Video encoding mode: CBR

Audience: 97 Kbps
Audio codec: Windows Media Audio 9 Voice
Audio format: 4 kbps, 8 kHz, mono
Video codec: Windows Media Video 9 Screen
Video bit rate: 86 Kbps
Video size: Same as video input
Frame rate: 10 fps
Key frame: 8 s
Image quality: 100
Buffer size: 8 s

On the Attributes Tab

Enter appropriate information to tag your output file.


I kept all other tabs at their default settings. Happy blogcasting!

Edit by John: 3rd Nov 2005 - Rehosted download file.

Comments (13)
  1. Anonymous says:

    Okay, this post is for me (well, I guess there is a little heads-up for you too). I found this post months…

  2. Anonymous says:

    I was asked by a customer, or rather a colleague of mine took the query who passed it on to me, how you…

  3. Karan Mavai says:

    You mention that it is most important to reduce the display settings down to 8×6@16, but don’t explain why. Is that in order to reduce the hit on the CPU as WM encoder is very process intensive?

  4. John Howard [MSFT] says:


    It’s important for a few reasons – first to stop frames being dropped due to the large amount of information if the colour depths or screen size is larger, and also to ensure that the final file-size is kept to as reasonable as possible. Yes, encoding is very resource intensive both disk i/o and CPU.

    There’s not many demo’s which should need more than 800×600. It also means that when users are viewing blogcasts on a 1024×768 display, they can run Windows Media Player and still see the recording at its full size.

    Hope this makes sense.


  5. castie bloggs says:


  6. castie bloggs says:


  7. Karan Mavai says:

    Thanks John … makes sense

  8. luc says:

    John, a big THX. I was looking for something like that.

  9. Don Smith says:

    John, thanks for posting this entry. I’m planning on doing a number of BlogCasts and you’ve helped make it a lot easier.

  10. Brian Tucker says:


    I really appreciate your post. I have noticed that when I am in the office, the blogcast looks fine, but when I am home the video is off. How do you compensate for those people that may not have a great video card?



  11. John Howard [MSFT] says:

    Hi Brian. Nice website you’ve put up by the way 🙂 By coincidence, someone inside Microsoft asked me only that question a couple of days ago, and as I result I was investigating with a variety of machines I have at home. It seems the answer lies in the video card drivers – one particular card (pretty old now) although it didn’t give spectacular viewing, the latest drivers made a huge improvement. Also, have you confirmed its not a bandwidth issue by saving the .wmv locally rather than pulling it off the net? Just a couple of thoughts – hope they help.



Comments are closed.

Skip to main content