|
Frequently Asked Questions
Researchers, academics and technical staff from around the world are equally concerned about timing and often
contact us with specific questions. By analysing trends, we have put together this FAQ that should help answer many of your questions and allay any concerns you may have.
General
The Curate’s Egg*
Using the Kit
Technical
Bench Test Equipment Comparisons
Training & On-site Demos
Contacting Us & Ordering
G E N E R A L
What is the Black Box Toolkit and why should I use it?
Researchers within the behavioral sciences regularly make use of computer-based paradigms. Most assume that the computer accurately
presents stimulus materials and records responses as programmed. However research has shown this expectation is often misplaced.
Presentation, synchronization and response timing errors can be caused by many factors. Whether you make use of a commercial experiment
generator or write custom software, you are likely to succumb to such errors. Inconsistent timing can lead to spurious conditional effects, poor replicability and bad science.
The Toolkit has been developed to help researchers address timing issues related to their own paradigms running without modification on their
own hardware. The toolkit operates as a virtual human programmed to respond to stimulus presentations and generate appropriate responses
very accurately. Typically sensors are hooked up to a second PC or Mac to detect stimulus presentations. A generation interface is used to
feed a response into the paradigm under test. The toolkit measures the timing of all stimulus and response events extremely accurately,
allowing the researcher to compare timing measured by their own equipment with the actual timings.
Easy-to-use software allows graphical control of monitored stimulus events and responses to generate. A virtual 8-channel scope is used for analysis. Cursors allow measurement of timing between any two points. Top
What can I measure with the Black Box Toolkit?
The kit offers various sensors that can be attached externally to a second computer system running an experiment. These sensors detect when
a visual stimulus is displayed, audio is played, a TTL signal is received, or when a human presses a remote response button. Easy to use
software allows the kit to be programmed to record activity and to respond to predefined patterns of stimuli as if responses had been made by a
virtual human. The difference being that the kit records all activity and generates responses with sub-millisecond accuracy. By comparing what
the kit “saw” or “heard” and the response times it generated with timings recorded by the experiment, presentation and response timing errors
can easily be spotted and corrected. All testing can be done without modifying your experiment, in-situ, on your own equipment by virtue of the use of external sensors.
As standard the kit comes with two opto-detectors for monitoring two screen regions on either CRTs, TFTs or Data Projectors and two Active
Switch Closure leads. Switch closure leads are used to close buttons or keys on your existing response devices. However the kit is capable of
very advanced testing with its optional modules. The kit offers a total of 8 digital input/output lines. All 8 can be used for input or a 4:4
input:output split can be used for stimulus-response testing. A summary of all the interfaces offered is given below:
- Up to 4 opto-detectors for monitoring screen regions (using to add-on modules)
- 2 high performance digital microphones for timing auditory stimulus presentations
- 2 high performance digital tone generators for triggering remote voice keys
- A 4 button BBTK response pad for accepting responses from human participants. Active Switch Closure leads from the pad allow you to
trigger up to 4 buttons/keys on your own response device. When using the pad four other inputs are available for monitoring screen
regions, sound presentations, +5V TTL signals etc. so that the kit times both presentation and response times independent of the computer running the experiment. Hence you can always be confident in your timings
- 2 Active Switch Closure leads for triggering buttons/keys on your own devices
- 2 Passive Switch Closure leads for monitoring when buttons/keys are pressed on remote response devices
- 2 +5V TTL inputs for receiving input from external equipment such as MRI scanner sync pulses
- 2 +5V TTL outputs for triggering external TTL equipment
As the kit connects to a PC using a standard parallel port you can also use the kit as a standard response device if your experiment generator
can utilise the parallel port. For example you could use the BBTK microphones to act as Voice Keys with your own experiment.
In short the kit can be used to test the timing of virtually any experimental paradigm either by acting as a very accurate virtual human pre
-programmed to respond to stimuli. Alternatively it can take over the presentation and response timing duties of your own hardware meaning you
can be sure of sub-millisecond presentation and response timing in situations that are traditionally impossible, e.g. video clip presentation and response timing. Top
How common are conditional biases that affect presentation and response timing?
Unfortunately they have existed in every paradigm we have looked at regardless of platform, operating system or whether a commercial
experiment generator has been used or not. In all cases the researcher has been unaware of the presentation of response biases. More
worryingly on many occasions there has been a conditional bias caused by the equipment itself. For example in cross-modal studies it is
common with trails where audio is present to have markedly different response timing characteristics. This can account for apparent statistically
significant experimental results leading to spurious conclusions. In the majority of cases timing errors can be improved by use of the Black Box
Toolkit. Either the experiment parameters can be altered or post-hoc statistical manipulation can be carried out where response data has a low variability and constant absolute error. Top
Which academic journals are requesting that authors validate their timings?
Currently no journals are requesting that authors specify that they have tested the timing characteristics of their study. However this will be
changing in the near future with journals such as Behaviour Research Methods, Instruments, And Computers setting the trend by requesting
that authors submit their software, scripts and data for inclusion within a repository. Whilst we accept that academics follow best endeavours on
many occasions it is useful to know the timing limits on a given study and how the author verified timing accuracy. In addition within academic
publications it is often difficult to unpick the contribution to timing which may have come from a variety of sources. By testing the timing of
computer based paradigms this will help ensure that successful replications are more likely. In fields like priming there is still an active debate
on the direction of findings and in general researchers are seeking out ever smaller and smaller differences between conditions. In both cases
timing validation is crucial in order to avoid statistically significant results due to equipment variation. Top
Absolute error or variability in response devices: which one should concern me the most?
Absolute error in response devices is tolerable in some instances so long as the variability is low and other events are not reliant on fast
processing of responses, e.g. if the next stimulus won’t be presented until the response has been registered. Ideally both should be as close to
zero as possible, but low variability is the more important than absolute error in relative terms. Top
Have you got any benchmarking software so I can see how good my computer system is likely to be?
Yes – we have a calibration utility which assesses the sampling rate the kit will be capable of on your own hardware. This is available as part of the FREE trial. Top
How can your software/kit get accurate timings when my experiment generator can’t?
Our data capture software runs in real-time priority mode on Windows 2000/XP. This means that we can be sure we have exclusive access to
the CPU and are not interrupted. Unfortunately experiment generators cannot run in this mode as nether the keyboard or mouse can accept
input and screen updates do not appear. We also supply a small utility which checks that your host PC is suitable for use with the kit. Top
Can I test the timing of my Mac or Linux based experiment? What about other platforms?
Yes. You can test the timing characteristics of any experiment on any platform. This is because the kit behaves as a virtual human and operates none-invasively. Top
T H E C U R A T E’ S E G G
What’s the most accurate experiment generator for presentation and response timing?
Some experiment generators are strong in one area and weak in others. In an ideal world one might consider using a mix of experiment
generators and custom written software. However to discover in which areas a particular piece of software is strong you will need to use the
Black Box Toolkit to confirm this. In addition you should bear in mind that the same piece of software can perform very differently on two
computer systems. Again this is a prerequisite for using the Black Box Toolkit. Top
Which brand/model or computer is the most accurate?
Unfortunately all computer hardware varies and so a single best system cannot be identified as there are some nay variables to be considered.
Manufacturers typically change the components they build into computers systems on a batch-by-batch or month-by-month basis. You should
consider computer systems as being commodity items. So two identical looking systems may actually be very different internally. The way the
operating system is configured can also have an affect as can installed software, drivers and virus checkers. Again this is a prerequisite for using the Black Box Toolkit. Top
What’s best for visual presentation: a CRT, TFT or data projector?
Depending on the context you wish to use the display device in and paradigm in question the balance varies. For example for RSVP the only
realistic option remains the CRT. As long as the researcher is fully aware of the mechanics of how specific displays work and their likely effect
on timing this remains key. One way to help ensure that the effects physical hardware characteristics are kept to a minimum is to make use of
a tool such as the Black Box Toolkit. Obviously some paradigm do not require such accurate display timings. Again this may be a mitigating factor in choice of display device. Top
What’s the best response device?
Depending on the experiment you are running you may be limited by your choice of response device. Under ideal conditions you should use the
manufacturers response box or pad as this will have a small known delay within its own hardware. However some mice can have an equal or
better variability as compared with the experiment generators manufacturers own devices. Combined with a small absolute error they can offer a
viable alternative costing many hundreds of times less in some cases. Typically much of the delay a response device contributes to response
timing is due to inherent delays in the hardware itself. For example a typical keyboard adds about 30~40ms where the variability is due to a
10ms scanning loop where keys are checked. This rule also applies to mice. You should bear in mind that manufacturers’ of input devices, e.g.
mice, keyboards, sound cards are under no obligation to make the millisecond accurate. So long as they work as intended and are electrically
safe they are suitable for sale. The on-line gaming community are quite fanatic about mice as for them fast responses are equally important. A
search for “fastest mouse response for gaming” on Google will give you a good insight into timing characteristics of widely available mice. You
should bear in mind that manufacturers are liable to change components within input devices even though they may look the same externally.
Testing input devices is again a prerequisite for using the Black Box Toolkit. Top
I can connect my response device by either USB or serial. What interface is best?
Research has shown that although interfaces offer theoretical differences in transfer speed the actual response speed is largely determined by
the response hardware itself. That is, how quickly the devices internal microcode scans for button or key presses and transfers data. This is
illustrated by the fact that a mouse that offers both USB and PS/2 interfaces actually changes response characteristics when the interface is
changed. That is, simply by swapping the interface two different pieces of microcode are used. Each has its own timing characteristics
independent of the interface and should be checked. Again this is a prerequisite for using the Black Box Toolkit. Top
PC/MAC/Linux... Which is best?
As with experiment generators one cannot recommend one platform over another as each has its own specific strengths and weaknesses. Most
modern operating systems are multi-tasking and all have the inherent risk of inaccurate timing. All operating systems interact with the software
and hardware drivers running on them which makes any generalisable prediction impossible. Top
Soundcards... Which one should I buy?
As a general rule of thumb do not use soundcards that are integrated into the motherboard, e.g. AC97 compliant chipsets. These are known to
have poor timing characteristics. Generally more expensive soundcards have lower start-up latencies. Typically cards that have ASIO drivers
generally offer better quality hardware. Again you will need to check start-up latencies using the Black Box Toolkit. For computer based
musicians latencies are crucial. A search on Google for “sound card latency” will help you identify inherently fast cards. You should be looking
for cards with “low latency”. For an acceptable "feel" when playing a sound, the latency should be in the range of 10 ms or lower. Typically most cards have a much higher latency. Top
Which graphics card?
Typically all cards that support Microsoft DirectX 8.0 or 9.0 should be capable of very fast display/buffering times. AGP is a prerequisite,
preferably 8x. However you should bear in mind that some drivers are better than others. Once you have checked a paradigm with the Black
Box Toolkit you are advised not to update any system drivers without retesting. In addition some drivers offer advanced options such as tick
boxes that allow you to sync the card with the refresh signal, e.g. ATI’s Catalyst drivers. Various settings can alter display performance and
should be checked if you have problems. You are advised to consult various gaming websites as again display performance is crucial for
gamers. A Google search for “fast graphics card for gaming” will help you identify good cards and those with known problems. Top
Microsoft Windows or DOS based experiments?
For hosting mission critical studies DOS can be a better bet as it is single tasking and good DOS based experiment generators (ERTS etc.)
can hit the hardware directly and obtain better presentation and response timings. However, these days obtaining hardware that is DOS
compliant, e.g. VESA compatible graphics cards and specific model ISA soundcards can be hard. To help ensure a similar level of accuracy you should check your own paradigms using the Black Box Toolkit. Top
Are faster processors better than slower ones?
Not necessarily as there are no guarantees that a faster system doesn’t have other “factors” that mitigate against accurate timing. For example
drivers, add-in cards, installed software, background tasks will have an effect on your ability to obtain accurate presentation and response
timings. New Pentium 4’s with Hyper Threading may also display strange results as they emulate a dual processor system. Testing has
suggested that such systems whilst having a very high clock rate, e.g. 3.0Ghz, may actually sample at half the rate expected as each virtual
processor receives alternate clock ticks and a given piece of software typically runs tied with its affinity to one processor and hence half of the clock rate. Top
*What does the term “Curate’s Egg” mean?
A curate is a term used in various Christian religions who possess an ordained ministry to describe a priest who is not a parish priest but
operates in effect as his or her deputy. Some larger parishes may have more than one curate. Most curates are eventually raised to become a parish priest in another parish as the older priests retire or die.
The expression "a curate's egg" meaning something that is partly good and partly bad and thus not wholly satisfactory, arises from the
publication of a cartoon in Punch on 9 November 1895. The cartoon, drawn by George du Maurier and entitled "True Humility", featured a timid
looking curate taking breakfast in his bishop's house. The bishop says "I'm afraid you've got a bad egg, Mr Jones". Apparently trying not to
cause offence the curate replies, "Oh, no, my Lord, I assure you that parts of it are excellent!" http://www.punch.co.uk/shop/bigprod.asp
?id=pu00083
Top
U S I N G T H E K I T
What’s do the terms Host and Remote mean?
Host is use to refer to the Microsoft Windows 2000/XP based PC that has the kit physically plugged into the parallel port and runs the BBTK suite of applications.
Remote is used to refer to the PC/MAC/Linux system that is running the experiment you are interested in checking the timing of. Remember that the Remote could also be other specialist equipment, e.g. MRI scanners. Top
What’s the difference between Passive and Active Switch Closure?
Passive Switch Closure is where a flying lead from your response device switch is fed into the kit. So when you press a button on your
response device its timing characteristics are also recorded by the kit in addition to your own experiment (response onset and duration are recorded).
Active Switch Closure is where a flying lead is attached to your response device switch. By programming BBTK applications such as DSCAR
(Digital Stimulus Capture And Response) the kit can close the switch of your response device as if it has been pressed by a human participant. Top
How do I check whether two stimulus types are synchronised in a cross modal study?
To check the synchrony of a visual and auditory stimulus attach one or more opto-detectors to the remote screen. Then place one or BBTK
microphones next to the remotes speakers. Run DSC (Digital Stimulus Capture) on your host PC. Now capture a sequence of presentations.
When your capture is complete use the data analyser to check for synchrony by means of the measurement cursors. Top
I need to present a series of images in rapid succession how do I check a different image is presented on every frame?
Pre-prepare two or more images with 16x16 pixel white reference blocks superimposed on them. On each image the reference block should be
at a different location. You should place a black reference block at the position where the other image has the white reference block. Set the
background of your experiment to black (this helps the sensors detect images for testing purposes). Attach two or more opto-detectors over
where the reference blocks will appear. Run DSC (Digital Stimulus Capture) on your host PC. Now capture a sequence of presentations. When
your capture is complete use the data analyser to check whether stimulus images appeared on alternate frames as intended. Top
I
’d like to make use of video clips in my experiment but need accurate response timing relative to various scenes in the clip. How can I use the toolkit to ensure I get millisecond accurate timing?
Traditionally all experiment generators are very poor at timing when video is playing. By using the BBTK response pad together with DSC
(Digital Stimulus Capture) you can obtain sub-millisecond accurate presentation and response timings. Use a video editor to super impose a
16x16 pixel white reference block at the start of each scene you are interested in. Other scenes should have a black reference block so that
they don’t activate the opto-detector. Attach an opto-detector to the screen of the remote PC. Typically we do this by using a smaller opto
-detector which can be attached unobtrusively to the bottom right of a screen. Attach the BBTK response pad to the BBTK expansion port.
Attach the Active Switch Closure lead(s) from the pad to the primary response button of your own response device if needed. Run DSC (Digital
Stimulus Capture) in response pad mode. When your capture is complete use the data analyser to check for the occurrence of the reference
block on the scene you are interested in. You can then use the cursors to measure the reaction time between the appearance of the reference block and a response button being pressed. Top
My chosen response device is adding to response timing. What should I do?
The first thing to do is establish the contribution the device is adding. Once this has been established you need to determine its variability and
decide whether the level of variability is acceptable within your current study. Next you need to decide whether the absolute error that the device
is adding is acceptable. You should bear in mind the absolute error of a response device may also affect when a stimulus presentation is terminated. Top
I’ve found an error in the synchrony between my visual and auditory stimuli. What can I do to correct this?
Typically you will need to re-test the paradigm after making corrections to the stimulus onset times. This can involve various manipulations
specific to the software you are using, e.g. in E-Prime you may need to cache images using the canvas object or alter the timing mode etc.
Alternatively you could also manipulate the stimulus materials themselves. For example if a sound stimulus file is played for longer that it
should once you know how long extra it was played for you could use a sound editor such as SoundForge to shorten the duration so that it
matches the correct duration when presented by the experiment generator. Another example may be where you need to insert silence into a
sound file start to ensure that the actual sound comes out at exactly the same point as a visual stimulus. You may also need to alter the
refresh rate or other settings on the system you are using to present data. Again this is a prerequisite for using the Black Box Toolkit. Top
I want to know the contribution my mouse or other response device adds to response time. How can I do it easily?
Testing response devices can be done in two ways. The first is with DSCAR. You will need to tack a flying lead to the switch of your primary
response device. This is known as an active switch closure. In short the kit will be able to close the button of your response device as if a
human participant had pressed it. If you are using a visual presentation you should superimpose a 16x16 white block on to each image. Set the
background of your experiment to black (this helps the sensors detect images for testing purposes). Attach an opto-detector over where the
reference blocks will appear. Run DSCAR and design a sequence that will detect a visual stimulus and generate an output on an active switch
closure line. For example you could construct a sequence that waited 300ms after detecting a visual stimulus then switch closure that lasted for
500ms. Once constructed run DSCAR with your experiment. Each time a visual stimulus is detected a switch closure will be generated. Be
examining the reaction times recorded by your experiment you will be able to determine the absolute response error and its associated variance
. In this case reaction times should be 300ms and last for 500ms. Any errors can be largely attributed to your response device. Top
How do I plot the absolute error and variability of my response device?
Depending on whether you have used DSCAR (Digital Stimulus Capture And Response) to simulate responses or the BBTK response pad to
actively trigger a response device the method will vary slightly. If you used DSCAR you will have entered a fixed delay which will have elapsed
before a response was simulated. This is a known delay and should match the response time recorded by the experiment on the remote
computer. Any deviation is the absolute error. Response times from the remote paradigm can be pasted into Microsoft Excel and then
subtracted from the known target response time. Variability in the response device can be calculated by using the standard Microsoft Excel statistical functions such as STDEV etc. Top
Alternatively if you used DSC (Digital Stimulus Capture) together with the BBTK response pad and Active Switch Closure lead tacked to the
primary button of your response device then the method will be slightly different. In this scenario a real human will have responded to a stimulus,
e.g. a bitmap by pressing a response key on the BBTK response pad. Simultaneously via the Active Switch Closure flying lead from the pad the
standard response device will also have been triggered. Each response will typically have been generated at a different time. This means that
you will need to use the BBTK data analyser to work out the actual response time that was recorded when the pad button was pressed. This
can be done using the measurement cursors. The first cursor should be place on the leading edge of the stimulus event and the second on the
leading edge of the pad button being pressed. A response time can then be read from the status bar. You will need to do this for as many
responses as you wish to analyse. In effect you have know obtained an exact known delay and should match the response time recorded by
the experiment on the remote computer. Any deviation is the absolute error. Response times from the remote paradigm can be pasted into
Microsoft Excel and then subtracted from the known target response time. Variability in the response device can be calculated by using the standard Microsoft Excel statistical functions such as STDEV etc. Top
I’m running a paradigm in which participants make a response based on the position of one or more images on screen. How do I
simulate a response based on the activity of one or more screen regions?
The kit can monitor for up to four active screen regions. By programming DSCAR (Digital Stimulus Capture And Response) the kit can be
instructed to react to activity on any of the four regions by means of the OR function. Alternatively patterns of activity can be AND’d together to
provide for addition functionality. For example in the “OR” condition you could use two opto-detectors to monitor for top right “OR” bottom right
and simulate an Active Switch Closure that closed a button on your response device. Stimulus images appearing at either OR positions would
trigger a simulated response. In the “AND” condition if using the same two positions then both bitmaps must be visible simultaneously before a simulated response is triggered. Top
How do I test the response latency of my voice keys?
Assuming you are making use of visual presentations and a vocal response to trigger a voice key super impose a 16x16 pixel white reference
block on to each image. Set the background to black. Attach an opto-detector over where the reference block will appear on the screen. Now
place a BBTK tone generator next to your voice key. Run DSCAR and design a sequence that will detect a visual stimulus and generate an
output on the line with the BBTK tone generator. For example you could construct a sequence that waited 300ms after detecting a visual
stimulus then generated a tone that lasted for 500ms. Once constructed run DSCAR with your experiment. Each time a visual stimulus is
detected a tone will be generated. Be examining the reaction times recorded by your experiment you will be able to determine the absolute
response error and its associated variance. In this case reaction times should be 300ms and last for 500ms. Top
Why do I need to check every paradigm. Can’t I just benchmark the PC/MAC the experiment is running on?
Unfortunately not. Benchmarking a PC/MAC in traditional terms, e.g. to obtain a MIPS rating, doesn’t tell you how the system will perform in the
real world. In short it is a synthetic benchmark. In much the same way benchmarking one paradigm doesn’t tell you how another will perform on
the same hardware even if the two paradigms appear similar. Any settings changes in the experiment generator or third party software are likely
to have an effect. In addition is it hard to account for human error when defining paradigms without checking presentation and response timing.
Often presentation errors are undetectable by the experimenter but go on to affect results when running with real participants. Top
T E C H N I C A L
How does the kit connect to the host PC?
The kit connects via a standard EPP compliant parallel port (all LPT ports support this mode). This negates the need for expensive timing cards. Top
Environmental Considerations
When using the opto-detectors you are advised to work in a room illuminated by natural lighting. Artificial fluorescent lighting for example can
inadvertently activate the sensitive opto-detectors. Typically this will appear as mains frequency hum when you analyse your data (regular
spaced peaks at 50/60Hz). Indirect artificial lighting is generally suitable or as an alternative mask out the back of the polycarbonate opto-detector with opaque tape. Top
Will your kit work in our MRI scanner room?
Yes. We have special fibre optic based modules which plug into the kit and can be cabled into the scanner room. Please contact us for a
quotation based on your needs. Top
Is there a Mac version of the software?
No and given the cost of development there are unlikely to be an versions of the kit that can be hosted on platforms other then Microsoft Windows. Top
We’ve built some custom response equipment. How do we link to the 5V TTL interfaces offered by the kit?
For details of how to link to and from the 5V lines of the kit see the technical manual. In short you can use the two powered-in lines (typically
used for the BBTK Microphones) to accept 5V TTL signals and the 5V powered-out lines (typically used for the BBTK Tone Generators) to send
them to your own equipment. We accept no responsibility for any damage caused to your own equipment through incorrect wiring etc. Our
consultancy service can help build custom links for you. Please contact us for a quotation based on your needs. Top
How many screen regions can I monitor at once?
The default toolkit ships with two wired opto-detectors that can be used for monitoring screen regions (TFT or CRT) or data projector projections.
If you wish to monitor up to four regions two additional external plug-in opto-detector modules can be purchased. If you wish to monitor stimulus
images in a MRI scanner environment then external plug-in fibre optic based modules must be used. Again up to four modules can be used simultaneously. Top
How many Active Switch Closures can I trigger when using the BBTK’s response pad?
As the BBTK response pad has four buttons these can be mapped via Active Switch Closure leads to up to four buttons or keys on your own response devices. Top
How do I adjust the sensitivity of the kits sensors?
Basically you should take the view that sensors should be adjusted in-line with the performance characteristics of a human. Simply, place the
sensors where you’d expect a human to be looking or listening. Then using the line testing utilities of the kit display your stimuli and adjust the
sensitivity tuning pots until then activity lights on the just Black Box illuminate. Top
Can I use the kit with third party experiment generators or software I wrote myself?
Yes. The kit uses a standard parallel port connection. So in theory any package that can access the parallel port should be able to read and set
the state of any of the lines. For example you could use one of the BBTK microphones as a voice key to trigger an experiment generator. Top
How do I link the kit up to my MRI scanner sync pulse so that I can relate presentations to actual slices?
The kit provides two 5V powered-in lines. Normally these are used for the BBTK microphones but can also be used to accept scanner sync
pulses. You should alter your scanner sync pulse characteristics so that it is long enough to be detected by the kit (0.10~1ms). When you
capture sequence you will also capture the scanner sync pulse as well as visual presentation timing and response data. In the data analyser
you will be able to see the relationship between the scanner sync pulse, the visual presentation and any responses (using either the BBTK
response pad or passive switch closure monitoring). With this additional data you should be able to correlate brain activity on any give slice with exact presentation and response data. Top
B E N C H T E S T E Q U I P M E N T C O M P A R I S O N S
Can I use the kit like a traditional digital oscilloscope?
No. The Black Box Toolkit only supports binary states on each input and output line, either on or off, +5V and 0V, or switch closure active or not
active. In addition there is no live moving scope display as with a traditional scope. Instead data is analysed graphically once collected. You can
think of the kit as offering the same functionality as a binary state 8 channel digital scope. Eight channel digital oscilloscopes are extremely
complex and extremely expensive. For example a Tektronix TDS8200 costs $10,000’s of dollars. Top
Can I use the kit like a traditional pulse/function generator?
No. The Black Box Toolkit only supports binary states on each output line, +5V and 0V, or switch closure active or not active. However the
function generation part of the Black Box Toolkit is more advanced than a traditional function generator. With the kit each event can have any
onset and pulse width and events can occur on any line. A sequence of up to 32,000 can be pre-programmed. Traditional generators do not
allow you to vary onsets and widths on an event-by-event basis. Traditional two channel generators are fairly limited as compared with the Black
Box Toolkit in this application. For example you would need two HP Agilent 8115A Dual Channel Pulse Generators at a total cost of over $10,000. Top
Can’t I do what the Black Box Toolkit does with a digital oscilloscope and signal generator?
No. In effect you would need an eight channel digital oscilloscope and two dual channel signal generators. This setup would cost around $30
,000 and in this area would by cost prohibitive and the ultra high sampling frequencies would go to waste. In addition making the oscilloscope
talk to the signal generator on an event by event detection basis would be extremely difficult. Remember the Black Box Toolkit can respond on
an event-by-event basis to stimuli and generate patterns of responses as if it were a signal generator. All parameters can be set easily and
quickly and results are displayed on a virtual eight channel scope. For example you can vary patterns that are responded to, delays before
responses occur and the characteristics of each response. Bench equipment does exist that can do something similar such as the Agilent
81132A which is used for testing ABS braking systems in cars. The can be used in stimulus-response mode but costs over $20,000 with add-in
cards and then only gives the user two channels. Again varying all parameters on an event-by-event basis is impossible. In addition the
complexity of using such equipment together is simply too great an overhead as compared with the simple graphical interface of the Black Box Toolkit. Top
What’s the typical sampling rate of the Black Box Toolkit?
Relative to bench equipment the kit samples at relatively low rates. Sampling rate is dependent on the host PC clock speed. Typically on a
1Ghz CPU rates are around 48kHz plus (48 samples per millisecond) up to around 100kHz on faster PC’s. This gives us enough accuracy to be
confident we are measuring and generating events with sub-millisecond accuracy. Top
T R A I N I N G & O N - S I T E D E M O S
Our department would like you to come and give a talk. Would you be willing to do this?
Yes. As standard we charge a daily rate of £250 plus expenses. Talks and visits typically include a grounding in the academic timing literature,
a live demo of the Black Box Toolkit, a Q & A session and an interactive forum where we can advise on your specific paradigms. Please feel free to contact us to discuss your requirements. Top
We’d like to integrate some best practice guidance into the way we teach undergraduates in the use of experiment generators. Could you advise us?
Yes. Typically we help construct high quality teaching materials which can be delivered to students during research methods courses. There are
worked examples with common experiment generators and resources which highlight common errors and workable solutions. As part of the
experience we favour one on-site visit where we can demonstrate errors and take part in Q & A sessions with students. Please contact us for a quote. Top
Our study is time critical and we need to have validated results together with independent timing certification. Do you offer a consultancy service?
Yes. We offer a service whereby we can visit on-site, test your specific equipment and paradigm and produce an independent timing certification
report. As standard we charge a daily rate of £350 plus expenses. Usually such testing is combined with on-site training for organisations which
have recently purchased a Black Box Toolkit. Please feel free to contact us to discuss your requirements. Top
C O N T A C T I N G U S & O R D E R I N G
How do I order?
Usually you will place an order via our world-wide distributors PSD. You will require an official purchase order from your organisation. PSD can
handle various methods of payment and are used to shipping to any country in the world. You will need to contact them for a final quote that includes shipping and appropriate taxes.
If you wish to take advantage of a multi-pack discount, would like to take advantage of on-site training or consultancy you will need to contact
us first. If you are purchasing hardware then we will pass details on to PSD who will contact you with a final quote that includes shipping and
appropriate taxes. Black Box Toolkit Ltd will handle any training or consultancy services directly as this will require a separate purchase order.
If you are unsure about the ordering procedure please contact PSD in the first instance. Top
Do you offer multi-pack discounts?
Yes. We have priced the Black Box Toolkit very competitively both in terms of cost and the functionality it offers. It is priced at approximately
double the cost of a standard response box and we hope the organisations that are serious about the quality of their science will purchase more
than one Black Box Toolkit in numbers approaching the quantity of response boxes they own.
For individual quotes please contact us directly. Actual order fulfilment will be via our world-wide distributors. Top
I need to talk specifics. Who do I talk to you or your distributors?
If you need to talk about kit functionality you should talk to us directly. We will then advise you based on your query. Please note we are unable
to give specific advice on timing unless you are one of our kit users as this will count as technical support. Top
|
