Hey, guys,
my firefly MV B/W worked but i can not set up frame rate in touchlib, i installed flystream driver1.0a12 and manual scanned the WinXP SP2 system, WinXP found the hardware and install the driver, in AMCAP.exe i could adjust the frame rate up to 60FPS, but seems it didn’t works for touchlib, do you guys know how to adjust the frame rate in touchlib? my frame rate in touchlib shows 15FPS, thanks!
I believe the only way is to use the ROI feature of the camera. When I punch in some numbers there the framerate jumps to ~110fps. I’m having some difficulty adjusting it to the correct settings. Will let you know if I can get it right.
Hi, GW,
did you use Flycapture driver or Flystream driver? i can not find the option in Flystream driver for ROI, and i checked the PTgrey website they mentioned Development SDK kit driver could use ROI, did you use that too? thanks!
Hey, guys,
my firefly MV B/W worked but i can not set up frame rate in touchlib, i installed flystream driver1.0a12 and manual scanned the WinXP SP2 system, WinXP found the hardware and install the driver, in AMCAP.exe i could adjust the frame rate up to 60FPS, but seems it didn’t works for touchlib, do you guys know how to adjust the frame rate in touchlib? my frame rate in touchlib shows 15FPS, thanks!
I believe the only way is to use the ROI feature of the camera. When I punch in some numbers there the framerate jumps to ~110fps. I’m having some difficulty adjusting it to the correct settings. Will let you know if I can get it right.
Hi, GW,
did you use Flycapture driver or Flystream driver? i can not find the option in Flystream driver for ROI, and i checked the PTgrey website they mentioned Development SDK kit driver could use ROI, did you use that too? thanks!
I used the FlyStream, then you click on “Format 7” inside the PGR settings box that appears. I don’t believe you can use the FlyCapture with Touchlib.
I just got the FlyStream to work in Vista. It seems that you have to install in the specified default directory for it to work. I kept installing it in D: Drive and it did not work and when I installed it in C: drive, it works like a charm.
So wait, and please forgive my ignorance if I’m wrong, but the touch screen in the movies that were posted was created with only two lasers, as opposed to using a ton of LED’s?
I was wondering if you could briefly go over calibrating the ROI feature of the FireFly, I’m having dificulty understanding how the offset works. Everytime I try and Format 7 in my own coordinates I always get a very small percentage of the actual screen, for example, the top right corner and the touches are offset. I would love to get this format calibrated, the 120fps is incredible!
The way ROI works is that it creates a ‘window’ within the full area of the CCD (752x480).
So for example if you want to capture frames at 320x240, you would set the width=320 and height=240.
Now, this would result is captured frames of right resolution, but the ROI origin is still at (0,0).
This means that the 320x240 window within full 752x480 is positioned at (0,0) (top left corner).
So we must move it so that is centered within full 752x480 camera resolution.
Therefore the new origin is:
X=(752-320)/2=216
Y=(480-240)/2=120
This goes for any other resolution you might chose.
Hope this answer your question.
~Alex
Thanks for the reply Alex.
I punched in these numbers, and after reducing the size my of screen I was able to fit it all into the 320x240 capture size. The distance from by FireFly to the screen is now about 6 feet.
In order to fit in a screen as large as yours, is your camera very far away? Or are you able to trick this by using a 150 degree fish eye lens to fit it all in there?
Are you using the original microlens that came with Firefly MV camera?
I used both 90 and 150 degree lens and this reduced the camera distance a lot.
I was wondering if you could briefly go over calibrating the ROI feature of the FireFly, I’m having dificulty understanding how the offset works. Everytime I try and Format 7 in my own coordinates I always get a very small percentage of the actual screen, for example, the top right corner and the touches are offset. I would love to get this format calibrated, the 120fps is incredible!
The way ROI works is that it creates a ‘window’ within the full area of the CCD (752x480).
So for example if you want to capture frames at 320x240, you would set the width=320 and height=240.
Now, this would result is captured frames of right resolution, but the ROI origin is still at (0,0).
This means that the 320x240 window within full 752x480 is positioned at (0,0) (top left corner).
So we must move it so that is centered within full 752x480 camera resolution.
Therefore the new origin is:
X=(752-320)/2=216
Y=(480-240)/2=120
This goes for any other resolution you might chose.
Hope this answer your question.
~Alex
Thanks for the reply Alex.
I punched in these numbers, and after reducing the size my of screen I was able to fit it all into the 320x240 capture size. The distance from by FireFly to the screen is now about 6 feet.
In order to fit in a screen as large as yours, is your camera very far away? Or are you able to trick this by using a 150 degree fish eye lens to fit it all in there?
Are you using the original microlens that came with Firefly MV camera?
I used both 90 and 150 degree lens and this reduced the camera distance a lot.
~Alex
Using the stock microlens that came with the camera I have great blobs, but there has to be a large distance from the camera to the screen.
I ordered a Swann 150 degree lens from Frys and gave it a shot tonight. It did significantly reduce the distance needed to fit the screen in, but the blobs were inconsistent. Towards the edges I would get feint or no blobs at all.
Maybe I should give the 90 degree a try and see if it does the same thing.
Alex , did you notice this with the 150degree lens as well?
I would be willing to help out and commit my fixes if I get the svn access, sure. Currently I’m using DSVideoLib in my config file and changes that I made pertain to this input method.
Regarding the smoke app, I didn’t make any changes there, but being a native windows app that links the touchlib, I think that the faster blob processing somehow affects the smoke app speed. Maybe the fact that I removed the SLEEP(32) from the touchlib resulted in smoke rendering speedup, I’d have to look into this.
But one thing is sure, the touch response if definitely way faster than before.
~Alex
Its been a while since you’ve posted the fixes, you could try sending whitenoiz a PM if you still want SVN access. I will be checking out your dsvl fix soon and commit it.
Just incase anyone’s interested in buying a Firefly MV I’ve had a recent quote. You can actually buy a single black and white/colour Firefly MV, but you must purchase a dev kit with it for $300. On top of that you’ve also got to spend a whopping $100 for shipping.... those lead lined parcels sure don’t come cheap!
I’m pretty sure this is common knowledge on the forums but just incase I thought I’d share it, particularly as the results using it are extremely impressive!
I finally got the FireflyMV to work, and it’s fast! But, the laser modules I got from Alex only seem to have a range of 6-8 inches. It seems they defy the laws of light, because they fire straight and level for this time then seem to bend up hugely, which obviously they don’t. I thought they were bouncing off the acrylic, but they can’t be bouncing that extreme. It’s very odd.
I finally got the FireflyMV to work, and it’s fast! But, the laser modules I got from Alex only seem to have a range of 6-8 inches. It seems they defy the laws of light, because they fire straight and level for this time then seem to bend up hugely, which obviously they don’t. I thought they were bouncing off the acrylic, but they can’t be bouncing that extreme. It’s very odd.
When adjusting the lases there are two things you have to worry about. This is true because now you are dealing with a plane of light.
Lets break this adjustment process into two steps:
Step 1 Lets assume that you removed the line lens from a laser module. The laser module is now generating a single dot. In this step you need to adjust the laser so that the beam going from it is parallel to the surface.
Step 2 Lets assume that you now have the line lens on and that your laser module is still adjusted following the Step 1. Now you have to turn the line lens so that the line that is being generated by the line lens is fully parallel to the surface.
As you can see, for the lasers to be used in LLP method, you need to adjust them in two directions.
I think in your case, you accomplished the adjustment described in Step 1, but your line generated by the line lens was not fully parallel to the surface resulting in strange reflection with increasing angle as you were going from side to side of the laser line beam.
