[vslist] Compiled answers on: Phosphor persistence and CRT refresh-rates

[Per Magne Knutsen]

In response to my query on VSLIST and CVNET I got a number of good advice on fast display solutions. 
As most people replied directly to me, I have compiled some (cropped) answers below for the rest of you. 

My original posting was:

"We require a display with fast refresh rates and very short persistency. From the literature, many opted
for oscilloscopes such as the Tektronix 608 that uses the fast-fading P15 phosphor. From our vendor,
however, we are told this model is based on analog technology "dead for 10 years".

Can anyone advice on what displays are nowadays used when fast refresh and short persistency is
required? What is, for instance, the persistency of a flashed image on a "regular" PC monitor?

Also, I find no CRT monitors with vertical refresh rates in excess of >160 Hz. Do faster monitors exist?
Again, for how long does the image leave a trace behind when flashed on these monitors?

On newer digital oscilloscopes (that use digital phosphor technology) persistency is tunable but
apparently only down to 50 msec. Are there better performers than this? Manufacturers? Models?"



Per Magne Knutsen

Department of Neurobiology
Weizmann Institute of Science
Rehovot, Israel

Phone: +972 (0)8 934 2604
Fax: +972 (0)8 934 4140
Cellular: +972 (0)55 675 289



*** ON CRTs WITH FAST VERTICAL REFRESH ***

From: Alexander Grunewald <agrunewald@facstaff.wisc.edu>
"I have looked into these issues. I use a monitor: CVX931 made by Totoku. It runs comfortably 
1024x768 @
160Hz, and 800x600 @ 200Hz. I have a photodiode attached to the monitor to keep track of missed 
frames, and
I find that when I run @160Hz, the stimulus essentially is decayed by the next refresh"

--- // ---

From: Andrew Schofield <A.J.Schofield@bham.ac.uk>
"Clinton do a fast phosphur monitor. This is available from CRS Ltd
(VSG people) on www.crsltd.com where you can look at decay
graphs (look under Stereo monitors). CRS also do SONY monitors
with V refresh upto 170Hz (but slow decay)."

--- // ---

From: Laurie M. Wilcox <lwilcox@yorku.ca>
"   I'm sure that you will hear from some of our colleagues suggesting
the "Joyce Display" and/or single-phosphor monochrome monitors (e.g. one
is made by Clinton). The Joyce display can operate at very high frame
rates e.g. 200Hz, is very bright and has little persistence. They are
wonderful - when they work. The problem is that in my experience they
are prone to breaking down, and because they are purpose-built they are
difficult to repair (often requiring expensive shipping to the UK).

The Clinton monitor is a good option. It is made specifically for
Cambridge research systems (http://www.crsltd.com), you can purchase it
from CRS or directly from the manufacturer. Just be careful to ask for
the following monitor model as they don't advertise the model they make
for Clinton. The model number is M20ECD5RE and the type no
is MM313AB . You should be able to get better specs from Clinton or CRS
than I can provide."

--- // ---

From: Cambridge Research Systems Ltd <enquiries@crsltd.com>
"If you are working with insects, CRS can supply a small 15" monochrome
monitor that supports frame rates to about 350 Hz, with a choice of 3
phosphors (white, green and yellow/green). See
http://www.crsltd.com/catalog/joyce-dm5/dp104/ for further information."

--- // ---

From: Kelly, John <jkelly@chmc.org>
"You can contact two companies about short persistence monitors and
video equipment (applications for field alternating stereo)
Cambridge Research Systems (VSG) at http://www.crsltd.com/
Vision Research Graphics at http://www.mv.com/ipusers/vrg/index.html

You might also check out my web site on varying the frame rate of a Matrox video board. That is, the 
video board with a set screen resolution (not the monitor) can determine the fastest frame rate of the 
monitor. However, you must have at least 220 Mhz bandwidth on the monitor.
Go to : http://faculty.washington.edu/jokelly/matroxHz.htm"

--- // ---

From: Bosco Tjan <btjan@usc.edu>
"Look up Clinton Electronics in the US.  They market a monochrome montor with
a very fast yellowish phosphor.  The monitor is rated to run at 240Hz."

--- // ---

From: Frank H. Durgin <fdurgin1@swarthmore.edu>
"VRG (vision research graphics, based in the US) makes a P46
monochrome monitor offering very short persistence and high refresh
rates (240 Hz). They may be a good bet: http://www.vrg.com/VWprices.htm"


*** ALTERNATIVE DISPLAY TYPES ***

From: Greg DeAngelis <gregd@cabernet.wustl.edu>
"On a regular CRT, the red gun is by far the fastest and I believe
that the green is slowest.  I don't know how short a persistence you need,
but you might stick a photodiode on a CRT and measure for your self.
3-chip DLP projectors have essentially no persistence because they
are totally optical, but I don't think you will find refresh rates above
100Hz."

--- // ---

From: Peter Carras <pcarras+@pitt.edu>
"If you do not need precise color and many levels of contrast, perhaps you could use an LCD (Liquid 
Crystal
Display) monitor. The disadvantage is that the brightness of the pixels does not decay; they stay on until
they are turned off.  The advantage is that, theoretically, they can be turned on and off at a very high
rate.  For a monitor with digital inputs, it should be possible to drive the display as quickly as you
video card can send it information... If you do not need to modulate the intensity of spatially extended 
images,
consider using one or more light-emitting diodes, perhaps optically combined with a steady surrounding 
image."

*** ON PHOSPHOR DECAY ***

From: Janus Kulikowski <janus.kulikowski@umist.ac.uk>
"Even faster is phosphor P4[whitish] but technology is dead. All
colour displays may have different persistence of each phoshor;
you CANNOT generate 'white' without checking if nder dyamic
condition it stay white.Sales' description is unreliable."

--- // ---

From: Christopher Tyler <cwt@ski.org>
"In response to your query, I don't have specifics on current models,
but I wanted to comment on your evaluation of manufacturers specs.
The common P31 phosphor has a decay time of 25 microsec, which sounds
fast, but spec this is only down by10%.  If you ask for a 1% time
constant, it is about 25 MILLIsec, and leaves a visible trial when
the spot moves on the screen.  Thus, the 1% criterion is much more
relevant to human vision, with its log units of dynamic range."

--- // ---

From: Peter Carras <pcarras+@pitt.edu>
"To measure the actual persistence of a CRT monitor, put a photocell (or photodiode connected to an
amplifier) on the face of your monitor, and observe the resulting waveform on an oscilloscope.  When we 
did
this on a new 70 Hz Barco monitor, the phosphor signal decayed by 50% in about 7 msec, and by 90% 
in about
12 msec.  (The Barco is not particularly fast, but it is very stable over long periods of time.)  To make
sure your photocell is reading accurately and not low-pass filtered, place it next to a light-emitting
diode powered by a square wave signal generator."

--- // ---

From: Steve Suter <ssuter@academic.csub.edu>
"We have an old DataCheck electrostatic monitor that we drive at 256
frames/s. We collected data on phosphor persistence using a single video
frame presentation five or six years ago because we had video frame
presentation synched to light shutters in order to route alternating frames
to left and right eyes to create binocular disparity. What we found is
this, with a P4 phosphor:

It peaked at about 150 microseconds (all data from frame onset). It was
reduced to 50% of peak at about 260 microseconds. It reached 10% at about
700 microseconds. It was still at about 3% at 1000 microseconds."

--- // ---

From: Gregg Podnar <Gregg_Podnar@ux4.sp.cs.cmu.edu>
"CRT display persistance decay must be fast enough to prevent
smearing of a moving image from one frame to the next.  Although
there is some persistance, it is something like two or three stops
down as the next scan starts.

You may consider a display made with micro mirrors
    <http://www.dlp.com/dlp/home.asp>

As there are no phosphors, the display time will not persist
beyond the frame time.  The scanning may have an impact on
your research, but if you are concerned with maximum display time
(not minimum), then it may be a good choice."