Home Page >>> Backstage Area >> Newsletter > Special Reports- Laser F/X '99

Laser F/X On-line Newsletter - Special Reports

Monday, May 31 1999

Scanner Wars

    Scanning systems were a major topic of hallway discussions at Laser F/X '99. With three new offerings, one of which promised much higher speeds, there was a lot of intrigue and drama in the air. From the E-mail sent out days before Laser F/X '99:

The Catweazle scanners from MediaLas delivered on their promise of fast scanning and offer the lowest cost ILDA 30K scanning in the industry at $1,000 US$ retail. They are shipped configured in an X/Y mount and with pre-tuned scan amps that incorporate their own power supply so implementation is simple and easy.

30K Catweazle scanner from MediaLas

The Cambridge 6210 galvos were said to be able to project 60K images and there was a great deal of discussion about them before the demo during the Brewster Awards. The demo proved unsatisfying to many as it consisted of a module playback.
While it was very artistic and certainly put the scanners through their paces, the playback did not provide any proof of the speed claims. There was much skepticism and laserists preferred to wait until they could see the ILDA Test Pattern before rendering their opinion.

Cambridge Technology scanners and amps on display at the Trade Show

What to look for in the ILDA Test Pattern

What the assembled laserists wanted to confirm was the most important part of the ILDA Test Pattern, which consists of a circle around a square (test pattern images courtesy of Patrick Murphy).

Another aspect of the ILDA Test Pattern is the width of the scan angle is during the test. To meet the ILDA 12K standard, the test pattern must meet the circle-in-the-square test at 15 degrees optical scan angle. To meet the ILDA 30K standard, the angle must be "about 8 degrees optical or less". It is generally agreed that scanners faster than 30K must show the ILDA Test Pattern at around 8 degrees.
There appears to be some confusion, because the acid test of scanners is not whether they can merely display laser frames running faster than 30K. Just about any scanner can accept signals running faster than 30,000 points per second. Even most parts of the ILDA Test Pattern can look acceptable at faster-than-rated speeds, except for the circle-in-the-square test. But the circle-in-the-square test is the main element of the ILDA Test Pattern! (Incidentally, it is not a perfect test, since it measures scanner response at just one point on the scanner's performance curve. But it is the only well-defined and agreed-upon test in the laser show industry.)
In summary, laserists are looking to see how fast a set of scanners can correctly reproduce the circle-in-the-square of the ILDA Test Pattern -- when the circle just fits inside the square -- while being projected at a around 8 degrees scan angle -- scanner-to-screen distance is 7.1 times the width of the projected test pattern.

Testing the Cambridge 6800s and 6210s

At the What's New seminar, the 6210's were put through their paces using the ILDA Test Pattern. A number of technical problems prevented them from reaching their full potential. These included prototype driver cards that were not yet fully optimized to 60+k, and a smaller-than-optimum power supply with only basic heat sinking for the driver amps.
Discussions with Bob Ash and the Cambridge representatives led to the conclusion that when the 6800's were introduced they were only 24K scanners. In short order they were tweaked to 30K with some users currently running then at 40+K tuning. Bob Ash and Cambridge believe that with a bit more work on tuning and power supply issues, the 6210 may reach the 60K speed that Cambridge's computer analysis indicates is possible.
The surprise of the conference was Pangolin's announcement and demonstration of their after-market modification of the popular Cambridge 6800 scanners to reach ILDA 50K at 7 degrees. Pangolin demonstrated two 25K shows running simultaneously from one scanning set as well as the scanners flawlessly reproducing the ILDA test pattern at 50K. The shows were running from a standard Pangolin QuadMod32 board (not the new, faster PCI-bus board). The scanners were four-year-old Cambridge 6800H's (not the newer 6800HP).
The Pangolin software is adjustable in integer increments so it was not possible for them to show the scanners running at exactly 50K. They were able to demonstrate the ILDA Test Pattern at 48K with the circle just outside the square and at 55K with the circle well inside the square thus validating the contention that the scanners were tuned to ILDA 50K.

Cambridge 6210 Update - 03 July 1999

On the IRC chat of 16 June, Phil Barrows of Cambridge Technology was our special guest to discuss scanner issues and particularly the new 6210 scanners. During the chat, Phil stated that he had validated the 6210 scanners could reach speeds of 60K+ and had scope photos to prove it. Here are the photos and Phil's comments on each photo:

Photo #1: Who: Mike Thanos, Perry Bakalos, Steve Sequeira, Martina Casey, Phil Barrows What: 2x6210, 2x67821, w/3mm X/Y mirrors for 60° Optical, peak to peak, X & Y Position signal response When: 6/11/99 Why: Small angle DC step time, ~1.6° (Opt.), ~150uS

Photo #2: Who: Mike Thanos, Steve Sequeira, Martina Casey, Phil Barrows What: 2x6210, 2x67821, w/3mm X/Y mirrors for 60° Optical, peak to peak, X & Y Position signal response When: 6/11/99 Why: ILDA test pattern, ~4° Optical, -3db frequency response 5.347KHz=64.164Kpps

Photo #3: Who: Mike Thanos, Steve Sequeira, Phil Barrows What: 2x6210, 2x67821, w/3mm X/Y mirrors for 60° Optical, peak to peak, Y-axis Position signal response When: 6/11/99 Why: Similar set-up as #2, -3db amplitude measurement

Photo #4: Who: Mike Thanos, Phil Barrows What: 2x6210, 2x67821, w/3mm X/Y mirrors for 60° Optical, peak to peak, Y-axis Position signal response When: 6/11/99 Why: Similar set-up as #2, same as #3, except lower point output rate, beginning amplitude measurement.

Photo #7: Who: Steve Sequeira, Phil Barrows What: 2x6210, 2x67821, w/3mm X/Y mirrors for 60° Optical, peak to peak, X & Y Position signal response When: 6/25/99 Why: -3db frequency response, ILDA test pattern, ~4° Optical, 5.025KHz=60.3Kpps in ILDA test pattern terms. Note the second measured peak precedes the second vertical measurement bar, making 5.025KHz of –3db response (or ILDA 60.3Kpps) a conservative measurement.

Photo #8: Who: Steve Sequeira, Phil Barrows What: 2x6210, 2x67821, w/3mm X/Y mirrors for 60° Optical, peak to peak, X & Y Position signal response When: 6/25/99 Why: Same set-up as #7, scope X/Y

By themselves, Photos 2 through 4 do not explicitly document the phase relationship between X & Y, nor do they prove the scan retains a sine shape, as is required for a true -3db measurement and proper use of the ILDA test pattern. These shots alone require some measure of faith that the test is properly executed.

Photos 7 & 8 were questions touched on during Wednesday's chat. I believe Greg (and perhaps Bill) made a few remarks on the subject, though I don’t have my log. These could have been captured in the first session, but I was pressed for time and did not think to take these specific shots.
So, Photos 7 & 8 remove that subjectivity. These two photos also show that not only is the scanner -3db response higher than is needed for ILDA 60Kpps, but also that the circle response still exceeds the boundaries of the square. I could have touched up the tuning a bit, but felt it unnecessary. For the record, the pattern on the wall looked considerably better than Photo 8.

Having said all that, it is also worth saying this: "Cambridge Technology makes no recommendation for or against regular use of the 6210 at ILDA 60Kpps at this time."

Pangolin TrueK 50 Scanners Update - 07 July 1999

These oscilloscope photos show four-year-old Cambridge 6800 scanners with the Pangolin TrueK 50 modification. The ILDA Test Pattern is being output at 52,083 points per second, at an angle of 6 degrees optical.
This is slower but wider than the Cambridge-tuned 6210 oscilloscope photo #8 (60K at 4 degrees). We believe the 6800 scanners we used could also be tuned to 60K/4º. However, we did not re-tune as we wanted to show performance typical of a TrueK 50-tuned system.
The photo below points out some features of the TrueK 50 modification.

Here is the entire ILDA Test Pattern on the scope, so you can see all features. Careful readers will note that this is the correct orientation. The original Cambridge scope photo had the X axis flopped; Pangolin followed this in making our close-up photo, to make comparisons easier.

Links - Quick Links to Report Pages

In order to present as many pictures as possible, we have broken this special report down into a number of pages to speed access. Most of these are large pages with many images so please be patient while they download.

Laser F/X '99 Report - Introduction and background
Friday 28 May - Behind the scenes and the crew
Saturday 29 May - Pangolin School and Open House
Sunday 30 May - Trade Show - Trade Show info and pictures
Sunday 30 May - Brewster Awards - The Banquet and entries
Monday 31 May - Seminars and LaserFest
Scanner Wars - Faster than 30K scanners debut at Laser F/X '99
Candid Camera - A selection of 20 photos (very large page)

DISCLAIMER: Some of the information in the Backstage area is provided by the persons or companies named on the relevant page(s). Laser F/X does NOT endorse or recommend any products/services and is NOT responsible for the technical accuracy of the information provided.  We provide this information as a service to laserists using the Backstage area. 

[ Introduction | World Scan |  What's New | Special Reports | Virtual Visits | General Articles | Digest ]