OnsetDelay and resolution problem?

Mon Sep 30 19:07:30 UTC 2013

Hi David, 
thank you for your reply. 

Today I keep trying to solve it and finally I did it!! I am so happy about 
that :D 

So, if anyone had the same problem, this is the reason why this happens and 
the solution. It's extremely easy and trivial.

The reason why with high resolution the onsetDelay is big it's because 
E-Prime has to "draw" more pixels. But if in your slide you just have a 
little stimulus in the middle (example a little black X, with a white 
background) you don't actually have to draw ALL the slide, but just some 
portion around the middle of the screen. To do that, I just had to adjust 
the value Height and Width of the Slide (to something like 10% and 10%). In 
this way a new slide will be put in front of the previous one, but the 
slide will be small and e-prime will not waste time drawing pixels that are 
the same as the previous slide. Since all my previous slides are white in 
the background, this works like a charm. In this way the software has to 
draw less pixel, and the onsetdelay decrease dramatically: I reached, with 
the right value of Duration (multiple of 1000/refreshHz) a onsetDelay of 
0!!!
You may have some problem if your slide have different backgrounds, but I 
assume that, knowing how it works, there is probably some way to go around 
the problem.

David, thank you again for your help.

Hope this is helpful.

Valerio

On Monday, 30 September 2013 19:44:51 UTC+1, McFarlane, David wrote:
>
> Valerio, 
>
> What range of Durations do you use for FP?  What happens to the 
> OnsetDelay for LANDC if you set the Duration of FP to something long, 
> e.g., 1000 ms?  What happens to OnsetDelay of LANDC if Duration of FP 
> is constant instead of varying?  What happens to OnsetDelay of LANDC 
> if it uses the same image every time instead of varying? 
>
> Just trying to explore more factors to see what is critical 
> here.  Maybe start with a simple trial design with constant, long 
> Duration for FP, and constant image for LANDC.  If that fixes the 
> OnsetDelay, then vary those factors systematically until it breaks again. 
>
> In addition, take a look at the new GeneratePreRun feature, in 
> particular the "TopOfProcedure" setting. 
>
> ----- 
> David McFarlane 
> E-Prime training 
> online:  http://psychology.msu.edu/Workshops_Courses/eprime.aspx 
> Twitter:  @EPrimeMaster (https://twitter.com/EPrimeMaster) 
>
> /---- 
> Stock reminder:  1) I do not work for PST.  2) PST's trained staff 
> take any and all questions at 
> http://support.pstnet.com/e%2Dprime/support/login.asp , and they 
> strive to respond to all requests in 24-48 hours, so make full use of 
> it.  3) In addition, PST offers several instructional videos on their 
> YouTube channel (http://www.youtube.com/user/PSTNET ).  4) If you do 
> get an answer from PST staff, please extend the courtesy of posting 
> their reply back here for the sake of others. 
> \---- 
>
>
> At 9/29/2013 03:18 PM Sunday, Vaaal wrote: 
>
> >Hi, 
> >I am trying to design an experiment with precise timing measure. In 
> >order to make my problem clear, I have to give some details about my 
> >experiment. 
> > 
> >The order of the slide is 
> >CROSS - FP - LANDC - BLANK. 
> >The "timing important" frames, however, are only the last three. 
> > 
> >CROSS, FP and LANDC have "PreRelease" on (same as duration), the 
> >forth one doesn't have any prerelease. This is because the forth 
> >frame is a response frame, so the prerelease would be useless. 
> > 
> >The CROSS slide lasts 1000 milliseconds and has prerelease. FP 
> >duration, however, changes everytime (reading from an attribute 
> >"Duration" in the trialList). 
> >LANDC lasts 200 milliseconds. LANDC is a slide with an image object, 
> >and the image is in a directory specified by another attribute 
> >("Image") in trialList. 
> >BLANK wait for the subject response. 
> > 
> >All the subject but LANDC are blank slides. 
> > 
> >The refreshAlignment is 50%. 
> > 
> >I am not interested in the timing of the CROSS slide. 
> > 
> >When I test my experiment on my office computer, the delay of FP and 
> >BLANK is around 0. The delay of LANDC is more variable (from -8 to 
> >8, usually). The resolution in this computer (LCD) is 1920x1080 at 
> >60Hz. I can accept a delay of max  +-((1000/60))/2 ms. This means 
> >that for a refresh rate of 60hz, it is fine for me if the slide 
> >appears +- 8 ms before/after it is supposed to appear. Ifigured that 
> >FP and BLANK have a delay of 0 because the duration of their 
> >previous frame is a multiple of 16.66 ms. This seems to be the case: 
> >the CROSS slide duration, as said before, is 1000 (and 16.66*60 is 
> >999.6). The duration of LANDC is 200 (16.66*12=199.92). However, 
> >LANDC doesn't have a delay of 0 because its previous frame, FP, 
> >doesn't have a multiple of the refresh rate (and I will not change 
> >that: the duration of FP changes constantly, according to the 
> >Duration attribute). 
> > 
> >The onsetDelay of 0 for the two frames is a lucky coincidence, and 
> >it is not required to be zero. As said before, it could be anything 
> >between +-(1000/refrashRate)/2. 
> > 
> >Now, MY PROBLEM IS that when I try the experiment on ANOTHER MACHINE 
> >the onsetDelay behave in a problematic way. 
> > 
> >When the resolution is 1600x1200 70Hz, I have almost perfect result 
> >when the refresh rate is 75Hz or lower. If the refresh rate is 85 or 
> >100Hz, I have a OnsetDaly of approximately 11-12ms for all of the 
> >three slides (FP, LANDC and BLANK). Why this happens??? 
> > 
> >I could just stick with the 75Hz or lower rate, you could say. In my 
> >experiment, however, I highly prefer to use an high resolution. I 
> >tried a lot of different combination of  resolution/refresh rate, 
> >and all the higher one lead to a dramatic (10/15 ms) increase of the 
> >onsetDelay for all the three slides. 
> >In particular I tried= 
> >2048x1539, 75Hz and 60Hz 
> >1920x1440, 85Hz, 75Hz, 70Hz and 60Hz. 
> >1920x1200 70Hz and 60Hz. 
> >1920x1080 70Hz 
> > 
> > 
> >I also tried with 1600x900. In this case, the results were good 
> >(around zero for CROSS and BLANK, within a reasonable range for LANC). 
> > 
> >So, it seems that, REGARDLESS of the refresh rate, at high 
> >resolution the onsetDelay increase. 
> > 
> >Does anyone have any idea why this happens? Does anyone have any 
> >suggestion about HOW COULD I USE HIGH RESOLUTION WITHOUT ONSET DELAY? 
> > 
> >Just for your information: the second monitor is a CRT Sony GDM-F520 
> >(specs= http://www.docs.sony.com/release/specs/GDMF520_sp.pdf ). As 
> >you can notice, the recommended resolutions are 1920x1440 85Hz, 
> >1800x1440 85Hz and 1600x1200 85Hz. For some reason, I cannot set the 
> >resolution to be 1800x1440. 
> >Anyway, as you can see I tried a lot of combination without success. 
> >I really need my resolution to be high (>= 1920x1200). The refresh 
> >rate is not so important, but since it is a CRT monitor I would 
> >prefer it to be higher than 72Hz. 
> > 
> >I really don't know what to do :( 
> >Any help is highly appreciated. 
> >Best Regards, 
> >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+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/6df1daa5-1ed2-45de-afd6-bf53f6c104b8%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/20130930/7864682e/attachment.htm>