Timing issues - Serial Response Box

[Vaaal]

Hi Paul, thank you for your reply. 
I am totally not stuck with USB. Actually the machine that I am going to 
use has a serial port. I just wanted to be sure that the pooling rate with 
the serial port and in general the latency is better than with usb port. In 
this case I read from the e-prime file that the latency is around 1.73 with 
a low standard deviation, which should work fine for my purposes. 

The problem with normal keyboard is that I usually record what I call 
"movement time". This is the time between the releasing of a button until 
the pressing of a second button. These two buttons are really close, so the 
reaction times are low. Since the keyboard pooling rate is about 125Hz (8 
milliseconds), I found out that all my movement times are multiply of 8 
millisecond. This means also that they are all even. As you can imagine, 
this outcome is hardly acceptable in any journal. 
Plus, since I am performing RT experiment in really short latencies, I 
would like to use a more professional equipment that a simple keyboard, at 
least because I want to know how much error I am putting in the measurement.
So I suppose that after all the Serial Response Box is the best choice.
Thank you very much,
Valerio

On Tuesday, 30 July 2013 13:12:23 UTC+1, Paul Groot wrote:
>
> Hi Valerio,
>
> You best option with eprime is to use the serial button box on a *NATIVE* 
> RS232 port. I.e., keep away from USB to RS232 converters if timing accuracy 
> is critical. RS232 ports use a baud rate to specify throughput (in this 
> case configurable to 9600 or 19k2, if I remember correctly). Specific to 
> the serial response box is the so called CPS-setting, which defines the 
> number of bytes are send to the computer per second. This defines the 
> resolution data flow (i.e. refreshes per second) and is 800 by default. 
> Native RS232 ports are interrupt driven, so latencies should be minimal. 
> Although PST doesn't give any details about this, I think that EPrime has 
> some built-in intelligence to increase the accuracy of the reported RT's. 
> (The data stream has a fixed CPS-rate, so it is possible to fix for small 
> 'hiccups' in the interrupt handling.) Such a mechanism could also minimize 
> RT-variance when using a USB converter.
>
> If USB is your only option, you should have a look at the following page:
> http://www.pstnet.com/support/kb.asp?TopicID=1835
>
> Also, I personally think that the 1ms requirements are overrated for 
> almost all experiments...
>
> cheers
> Paul
>
>
> On 30 July 2013 13:42, Vaaal <valerio.... at gmail.com <javascript:>> wrote:
>
>> Hello to everyone, 
>>
>> I am struggling to get accurate reaction time without starting put my 
>> hand in engineering stuff, since I don't have the required knowledge to do 
>> like that. 
>> I would like to buy a Serial Response Box, but the technical details in 
>> the page are really not enough. I am particularly worried about the serial 
>> port connection. 
>> I know that the pooling rate of the usb port is normally 125 hz, that is 
>> 8 milliseconds. I am able to overclock that value in order to have a 
>> pooling rate of 1000Hz, which is 1 milliseconds, which is quite fine. But, 
>> I am still not sure of the accuracy of my measurement, just because using a 
>> normal keyboard can add some latency in the process of transmitting 
>> information.  This is why I am willing to buy a professional device.
>>
>> The problem is that I don't know the technical specification of a serial 
>> port. There is anything similar to a pooling rate of a serial port? If it 
>> is so, how much it is, and there is any way to overclock it? Basically how 
>> much accurate is the response box? 
>>
>> The other option is to use DirectIN High Speed Button-Box here 
>> http://www.empirisoft.com/directin.aspx, which actually uses a USB port 
>> that I can overclock. BUT since I use E-Prime I would like to buy a piece 
>> of hardware that is designed to be used with E-Prime, just to avoid 
>> unnecessary headache in the future.
>>
>> So generally my question is: what are the limit and the advantages of 
>> having a serial port? Do you think that the Serial Response Box is really 1 
>> ms accurate? Do you use any other equipment? 
>>
>> Thank you for your reply,
>> Valerio
>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "E-Prime" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to e-prime+u... at googlegroups.com <javascript:>.
>> To post to this group, send email to e-p... at googlegroups.com<javascript:>
>> .
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/e-prime/efd28746-7911-4f5f-9ef9-87da1954f8b1%40googlegroups.com
>> .
>> For more options, visit https://groups.google.com/groups/opt_out.
>>  
>>  
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups "E-Prime" group.
To unsubscribe from this group and stop receiving emails from it, send an email to e-prime+unsubscribe at googlegroups.com.
To post to this group, send email to e-prime at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/e-prime/906a9319-230d-4ff7-8c43-f363dc9fc912%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/eprime/attachments/20130730/1de1eace/attachment.htm>