DIY Autocue

Autocue [or teleprompter if you come from the USA] are extremely useful for those of us who cannot remember one sentence without stumbling... or who become long winded and boring. I need to do short 3 minute pieces to camera [called stand-ups if you come from the USA] and these need to be short and interesting. Anyway I decided we needed to build an autocue for the office.


I looked around for DIY instructions and found loads of people suggesting using CD cases, for example this one from engadget.com. There was a simple but not transportable autocue from creativepro.com and another simple but not transportable... which also could not allow the camera to be angled correctly for the shot from instructables.com.

Professional autocue units look like this and so that was roughly my target -- make something close to a professional unit for very little money! So... time to design and build my own and then share it here on the net for others to follow:

Having drawn out on paper a rough design based upon a lightweight wooden box with a angled glass running on rods [TV and film cameras seem to love rod based attachments!] The next stage was to get the bits and build it.

I decided to use 4mm MDF for the wood box and use 10 mm square section wood as battens to join the MDF. First step was to cut the MDF and then draw out on the MDF the location of the battens.

Sorry the angle of this photo makes the side wall look really strange.

You will need at least some of the guide lines on both sides of the MDF to enable you to know where to put the panel pins in to hold the battens.

I decided to glue and panel pin the 10mm battens in place. I used the clamp to stop the batten sliding all over the place when I turned over the MDF to tap in the panel pins.

The glue had this tendency to make the batten float everywhere and it was tricky to get the clamp to hold it in place.

When all the battens are in place [there are 4 on each side wall] the side wall looks like this photo. Yes, I know I use too much glue!

Note the 'slot' is for the glass which is added at the final stage. The slot is wider that necessary for a pane of glass as I intend lining the inside with black velvet to reduce any internal reflections.

Professional autocue units use half silvered mirrors [actually 70:30 mirrors] but I cannot find where to get these cheaply so will try with normal glass and hence cutting down the the internal reflections is more important. That's also part of the reason for making it a full hood rather than just a top hood like the professional units.

The other reason for making it a full hood is for ease of construction. The base for the unit is 4 30mm x 20mm battens connected with 4 right angle metal joints. I used the metal joints as a way of trying to make the base square. This is then secured to the side walls and creates a solid structure.

The monitor rests on three of the battens allowing access to the connectors on the bottom. The fourth batten is for to make it a stable structure and give something back and font to take the hood into the rod system.

The dimensions of the hood/monitor cradle were done to fit with one of the LCD screens from the office. This LCD screen has the added bonus that it runs off a 12 volt supply which means that we can even use this autocue on location!

Here is the completed 'hood' with an 80mm hole cut in the rear panel to allow the camera to look through. 80mm is exactly right for our Canon XL1 with the lens hood removed.

Next step is to look at how to mount it on rods. I bought some 14mm aluminium tubes from the hardware store and found that I had a 13mm drill bit. Hmmm... I think I may change to a wood for this mark 1 version and make rods for the mark 2.

The next step was to build the camera riser. This is to lift the camera off the wood 'rods' to make the lens come to the centre of the mirror. I wanted this sturdy and light and found some right angle metal joints and a flat plate at the local builders merchant/DIY centre. I pop riveted them together to make the riser.

This turned out to be approx 4mm too high so I planed down the 'rods' to make it the correct height for the camera.

I addes a second plate to the bottom of the rods as a way of fixing the tripod to the rods. I had to drill out a 1/4 inch hole in both plates and use an extra nut and bolt to secure tripod to rails and camera to riser.

Here is the first camera on riser on rods on tripod test.

I found that I had not noticed that the camera body is offset from the lens so it didn't align properly. I needed to re-drill the mounting hole on the camera riser approx 2cm over to align it properly.

An extra nut was needed for the camera to ensure that the full length of the bolt wasn't used and potentially might damage the nut inside the camera body.

[If you are wondering what the extra box on the back of the camera is, it does the following: 1. Enable a quick release cable to work with the Edirol R-4, taking outputs and inputs from camera to recorder along a single multicore 2. Enable the Edirol and the remote zoom to share the LANC connector of the camera, while only sending power form the LANC to the zoom demand 3. Enable the cameraman to use headphones with a standard 1/4 inch jack plug rather than the 1/8 inch on the camera.]

I decided to try matt black paint instead of velvet inside the hood as it would be very much easier than gluing velvet inside and I can add velvet at a later stage if I find I need it.

Having seen the effect on the inside of the hood I decided that a coat of matt black paint would make the whole autocue look better...

Final steps, buy glass for the screen, and fit and test the LCD monitor.

Here is what it looks like in action. We have two red-heads on for lighting so this gives some impression of how clear it will be in a real-life situation.

I talked to the glass merchant about half-silvered mirrors. They don't have the 30:70 mirrors the professional units use, but they do have partially reflective glass used for the outside of buildings. I decided to try regular glass before buying the more expensive glass.

The regular glass produces a double immage because the light from the monitor reflects on both the front and back surface of the glass, so I will get the glass merchant to try and get a piece of the glass for outside of buildings and try that.

Having said that, I found that the text was perfectly readable and the unit usable with the regular glass.

I also found that the matt black paint was not quite as matt as I would like allowing some internal relections. However again this didn't impact its use too greatly.

I tested a number of programs for the prompting and eventually decided on Presentation Prompter. Current version is 4.2. It isn't perfect, but produced the smoothest scrolling of all the programs I tested. It costs $65, which is a reasonable price for the quality of the software.

OK, one last problem. How to control the autocue. The answer is Salling Clicker. Salling Clicker is a remote control program that allows your mobile phone to control your computer. It costs $23.95. Very good value for money. It will control Keynote as well, so I will use it for many things.

This means my Palm Treo 650 can become the remote control and just by rocking my thumb back and forth across the 5 way button I can control the speed of the autocue. But... of course there has to be a but... I had to write the AppleScript to make Salling Clicker do this! Salling Clicker uses AppleScript to do all of its remote control. Here's where to find the Salling Clicker script I wrote http://download.rsdt.org/Presentation-Prompter.html

Edirol R-4

This isn't a technical review, but a user review. It comes from the first shoot I used the Edirol on - one week shooting single camera. We did 22 interviews, recording 7 seminars and one mini advertising piece to camera.

Summary

The Edirol R-4 worked faultlessly throughout the shooting. There were features lacking compared to a regular mixer like the Shure FP33, SQN 2S or similar [noteably stereo pan controls in the case of the Shure, which I have used quite a bit]. The quality of the audio was excellent, with one exception - the noise on the mic amps when using high gain was poor.

Break away cable
I made a 'break away' cable for the R4 to go to the Canon XL1 we were shooting on. Using a break away cable made it quick and easy to connect the XL1 and the R4. Since we were doing quite a lot of interviews [22 in 5 days] this was essential.

The break away cable took the audio output from the R4 [using the phono ouputs - if you're American you'll call them RCA outputs] to the XL1. It took the headphone output from the XL1 to an extension headphone jack at the end of the breakaway cable.

It also took the L-terminal [called LAN-C by other manufacturers] from the XL1 to the R4. Because we use a Manfrotto zoom demand, the signal was split in the breakout box between the R4 and the zoom demand. Of note is that the L-terminal signal has tip as serial data, ring as power and then earth. Because the zoom demand need power the power is sent to the Manfrotto but not to the R4.

The break-away system used a cheaper multi-way connector, rather than the industry standard Hirose connector. The break away cable worked very well except on one occasion when the XL1 would not control the R4. We think it was probably because the connector quality was not as good as the Hirose. It worked again within a few minutes.

Noise on the mic amps
We were shooting some seminars - this involved taking a line feed from the PA mixer plus suspending a couple of Shure SM57s over the audience for audience effects. I would have preferred to use condencer mics, but I was borrowing from the PA people. When I wound in enough gain for the audience effects the mics appeared to be a little noisy. What that shows is the signal-to-noise ratio on the R4 is not as good as it could be.

However if I was using condencer mics with a higher sensitivity and higher output I don't think I would have noticed it. When I used the Rode NTG1 or the AKG lapel mics which are all condencer mics there was no noticeable noise on the mic amps.

Having said that, when I mixed the audio back at the dubbing suite using our ProTools system and the noise was not noticeable, it was only noticeable during times the audience was relatively quiet while recording. So in a real programme making situation it's not a problem.

Signal routing
Recording the seminars showed up a limitation of the firmware of the R4. What I would have liked to do is send the PA output [coming in on channel 1] to the left output and the effects mics to the right output. If I had the effects mics on channels 2 and 4 I could have done that, but because input 1 was set to line level input I could not use input 2 for a mic.

So I put the effects mics into inputs 3 & 4 and then didn't route them to output at all, but just recorded them onto the R4. This meant that everything was on the R4 but the camera recording [standard backup when you are using a separate recorder] only had the main PA output.

A simple upgrade to the firmware would help a lot for this sort of thing.

Firmware improvements
What would be good is to be able to select Left, Right or Both for each input independently. The selection of which output each input went to was rather strange and looked like it was thought through by an engineer not an operator! I realise this is version 1 of the firmware and these sort of limitations are common in version 1 releases. We have a version 1 XL1 too but there are no firmware upgrades for it!

Another little gripe, probably not fixable in firmware, but would be REALLY useful is to be able to change the monitor settings when in record pause.

One thing that was somewhat irritating is that the R4 doesn't go E-to-E when in stop. What I mean is that in stop you hear nothing at all... What I would like is that in stop you hear the input, so levels can be set etc. What this means is that you have to press Record/Pause to hear the audio inputs for setting up levels. Not a problem in some ways, but it means you alwasy record a small amount of audio when you start up.

Final firmware improvement that I would like is to have a mode for recording each shot as a separate file. This is actually the most important one from a user point of view. At the moment the R4 sets markers in the file - treating each scene as a separate file with markers for each shot. However, we use a Matrox RT system for editing and each shot is a separate file, so matching shots is a pain with the R4.

I know they are talking of Time Code for version 2, and that would be interesting for some filming, but not generally helpful for us. We use ProTools for the sound mix after the video edit, and although ProTools can use embedded time code within broadcast wav files for placement on the time line, it would not help in this instance as the time code is related to finished programme not original shoot.

Location of controls
Sadly its obvious none of the design team actually tried out pre-production models doing live shoots using the carry case. There are a number of controls placed in really irritating positions:

• Power On/Off
• Limiter On/Off
• Phantom Power On/Off

I want the Power and Limiter controls on the front panel not the top. Reaching in when the R4 is in its operational carry bag to turn it on is a pain!







Similarly I want the Phantom power switches above the mic/line switches so I can get at them easily.

Which brings me on to the operational carry bag - I had to cut it to enable the break-away cable to work, so that I could connect in the phono outputs to the cable and so the cable could sit at the bottom of the R4. This also had the effect of pushing the R4 up by about 1 cm which meant that the headphone jack was not so easily accessible as was the mic/line switch for inputs 1 and 2. I will think further about this, but probably eventually buy a Portabrace case for it, having designed the case to do what I want.

Meters
The meters are strange on the R4. They are like no other meters I have seen in my life. They are sort of peak reading VU meters. I say 'sort of' as the scale is wierd and the behaviour wierd. That having been said, I am used to swapping between all sorts of equipment and it didn't take me long to get used to them. So although they are decidedly wierd, they are not a hindrance to use of the equipment.

I would really like standard PPMs as meters. Since this is all firmware it shouldn't be difficult. In fact I don't know why the meters are so strange considering they are firmware.

The meters were easily visible in bright sunlight, but I was somewhat concerned about the effect bright sunlight might have on the LCD screen after a problem a few years back when I damaged a notebook computer by leaving it near a window over lunch and as the sun moved it caught the bottom corner of the screen and it was never the same again!

Batteries
We were using rechargeable NiMH batteries and the R4 behaved well. I had bought 2 complete sets expecting problems, but we easily good a whole day of filming out of each set. We didn't have the R4 on continually, but in a real life situation it used batteries conservatively.

Warning
You cannot 'hot swap' between external power and batteries. It crashes the R4 as we found out the hard way!

L-Terminal/LAN-C
The L-terminal connection works well, allowing hot plugging. What this means is that if you want to you can bring camera and R4 together, press the start on the camera so both start recording together than unplug to film away from the R4 and reconnect and stop both in sync. Thus it sort of works like an electronic clapper-board. Now of course if there's some way to send the signals over a radio mic channel...

Limiter
The limiter appeared to work well, I haven't seen the specification, but it did generally stop 'going into the red'. The only problem was the location of the control. It was difficult to turn it off for taking levels and on for recording. If the R4 was sitting alone that would not have been a problem, but in the carry case it was.

Effects
I never used any of the effects. There were two reasons for this. Firstly I normally do any corrections to the sound on our ProTools system after editing and secondly, trying to set them up in the carry case was impossible!

USB
USB transfer worked faultlessly to my Sony Viao notebook while on location. And... for those of people like us who use Linux, it worked perfectly on Ubuntu Linux, with the OS recognising the connection and opening the folder with no problems.

File formats
The BWF wave files that were recorded on the project imported into ProTools easily, however the 'markers' that are supposed to show start of shot were not displayed. Because I had recorded the audio on the camera as well it was a reasonably easy task to match audio in Adobe Premiere, but it really would improve things no end for the R4 to record each shot as a separate file.

Interal speakers and mics
The internal speakers seemed a little quiet, so not really useable. Why, oh why, would anyone ever want internal mics [omni-directional] in something like the R4?

Additions/changes
Apart from the layout changes and firmware upgrades, the main addition I would make is a return audio monitor and switch so I could integrate the 'off tape' feed from the camera. I'd change the phono [RCA] output connectors for XLR, so we could feed the output to line if needed, and I'd add a Hirose connector so the break-away can come straight in without a messy cable spider!

Conclusion
Overall I would highly recommend the R4. If you want to compare it to the Nagra V or the DEVA II I am sure you would find limitations - but in 'bang for the buck', as the Americans put it, the R4 wins hands down.

Oh, and I'd love to be on the design team for the R4 mk 2.