[Spike 5.03] capture XY positional data from a PC mouse

[ThibaultV]

Hello!

First of, I apologize if Spike 5.03 is no longer supported by CED and questions asked on old software are deemed irrelevant, but I found manuals and upgrades for Spike 5 on the CED website, so I assume support is not yet discontinued for Spike 5?

We have Spike 5.03, a Power 1401, and wish to capture XY positional data from a PC mouse (or a very similar device). But MouseTalk is only available from version 8 and onwards, and I was unable to find it as a standalone for download or to know if it would be compatible with version 5 at all:
http://ced.co.uk/ensearch/search.php?zo ... =mouseTalk

Is it possible to capture XY positional data from a mouse in Spike 5 without going through the 1401?

If not, do you need two BNC cables for the X and Y signal to plug in the 1401? or would it be possible to capture the two data, two values, through one BNC cable?

In either scenarios (through USB to laptop, through BNC to 1401), what do you think would be the delay between mouse signal to the final spike 2 data file? Would this delay be stable or vary a lot? a 10 ms uncertainty on the timestamps is already big for our experiment.

Thanks in advance for your help!

We are using an MRI compatible joystick that uses a 932 interface:
https://www.curdes.com/mainforp/interfa ... -932b.html
We are not even certain yet that the XY positional signal can be converted to BNC (it probably can), but this is a question for Current Designs

[Tim Bergel]

Hi Thibault,

Support for Spike2 version 5 is over, in the sense that if you find a fault we will not release an update that fixes it. But we will still support your use of it as best we can - that won't change.

No, you would have to use the 1401 to capture position data with Spike2 version 5. You would need two BNC cables, one for the X voltage and one for Y, connected to two 1401 ADC ports. And a different/adjusted joystick that generates two analogue voltages for the X and Y outputs.

A USB-laptop delay cannot really be predicted, it would be a few milliseconds, but will vary immensely according to the situation - you would have to measure it. However things are very precise with a 1401 - the time at which an ADC sample is taken is known to an accuracy of a few tens of nanoseconds.

The delay from input voltage -> new data in Spike2 is, of course, significant as it uses USB. I believe runs at 2 to 5 milliseconds or thereabouts, though of course your PC will affect this a lot. And the delay can jitter, so this will be a problem. The solution (assuming that what you want is for the system to 'react' fast to some action by the subject) is to use a sequence, which runs inside the 1401 and has immediate access to incoming data, to handle this. That will give you a reliable timing accuracy on the order of 200 nanoseconds (if you run the sequence fast, which you can with a Power1401) or better.

I hope this helps, please feel free to get back to me if you have more questions.

[Greg Smith]

I presume that you know that 'modern' versions of Spike2 can capture PC mouse XY positions using the Mouse Talker that comes as a Talker example. The timing accuracy will be whatever MS provide, likely no better than the screen frame rate.

As regards support, we will always try to help you, but as regards bug fixing we promise to do this (where it is possible) in the current major release (version 10) and in the previous major release (version 9). We do occasionally fix bugs in older versions, but it is gets more difficult as versions get more out of date as we do not necessarily still have the software tools to deal with ancient code. We add new features to the current major version (10), and sometimes retrofit them to older versions for reasons of compatibility.

Spike2 version 5 was first released in 2003 and was built with truly ancient versions of the various software tools we use... the Help system it uses is no longer supported by Windows... I could go on. Versions 5 (and even 4) still run on my Windows 10 box, but they are nothing like as nice to use as version 10.

[ThibaultV]

Hello Tim, Hello Greg,

Thanks a lot for your input!

We have found a technical solution and will likely plug two BNCs on the 1401, as to have the signal in real-time. Windows delay uncertainties would be an issue for us. Thank you for these details Tim.
>>> A USB-laptop delay cannot really be predicted, it would be a few milliseconds, but will vary immensely according to the situation - you would have to measure it. However things are very precise with a 1401 - the time at which an ADC sample is taken is known to an accuracy of a few tens of nanoseconds.

Thanks again for your help