OK, a few weeks ago I made mention of my desire to fill a milk jug with 100° water put it on the ground and see how far away I could spot it with my thermal device. 

I finally perform this test tonight. I filled the jug with 110° water and put it in the edge of a soybean field where the dirt was powder fine and 100% grass/weed free for maximum contrast and clear viewing. 

My current weather conditions were 66°F, we had a north wind all day, 61° dewpoint and 84% humidity. 

I had two thermals with me tonight. 

1. Flir Vue Pro 19mm/640 mounted on the roof of my truck and viewed from an inexpensive windows laptop with a 1280x720 display. 

2. IRD Mark2 35mm/640 scope. 

I put the jug out in the end of the field and drove away until I could no longer see it. 

I intend to go back tomorrow in daylight and verifiy the distance where I left a marker on the ground when I could no longer see the jug with the Flir Vue.

For tonight, I'm going to estimate that I could just barely see or detect a spot on the screen at approximately 450 yards away without using zoom. 

With the Vue set to 4x (max) digital zoom, I could make out the jug at just a bit past 600 yards. I attempted to record, but for some reason my Flir Vue will not record while zoom is used. It still records, is just does not record the change from base magnification to either zoom setting. 

At this point I took out the IRD to have a look and could see the jug clear as a bell with out using any magnification. 

Then I drove to the far end of the field which is approximately 1200 yards away and again without using any magnification was able to just barely see the jug. Unfortunately, I forgot to bring the DVR that goes with the IRD scope so there's no video for it it either. And even if there was video, I would be surprised if it would indicate a hot spot in the field 1200 yards away anyway. 

So I mentioned doing this test to my local thermal dealer/hunting buddy of mine and he was all excited about wanting to go do this with me next time and do some comparison between other brands of thermal devices. 

It would seem that with the IRD I am getting basic "detection" with less pixels that industry standard criteria of 3.5 pixels.  If my calculation is correct a 12" x 12" milk jug (which is actually slightly larger than a jug) would be less than one pixel at that distance. 

Likewise with the Vue, at 400 yards it should be right at one pixel?

A milk jug is about a raccoon sized animal, and if the milk jug were moving around I have no doubt that I would have spotted it with either of these two devices at those maximum ranges listed above. But to be fair I might not have noticed the tiny black dot at 1200 yards away had I not known exactly where to look for it. But I still saw it. 

I wish the display play on the Vue were more sharp like the IRD. I think may be experiencing some video loss because I'm using a laptop with a USB-to-video adapter on my computer.

I plan on doing this test again with the Flir PTS233 whenever it shows up, as well as a couple of the pulsar offerings and an IRD 60mm mark3. 

I would love to see someone who has access to all the various thermal devices perform a similar test under the same conditions, perhaps using an iPhone video if the eyepiece, etc. 




  

So it sounds likes the IRD was superior?

I love tests like these. Thank you. Far FAR better than comparing spec sheets. Far better.

I don't know that the IRD is necessarily superior to the Flir Vue, but it did let me see the jug at a distance beyond what I though was capable. 

Thats what I was trying to explain with my setup for my Vue. Laptop and using a USB video adapter probably results in some video loss. When you record with the Vue, it records a digital video, not analog. I know my Vue will become a 640 Boson camera as soon as they are available, since Boson will output digital image across USB.  Less cables on my dashboard also. 

OK, so just got back from checking the distances and the range was a lot farther than I realized at the time.  The Vue was capable of seeing the jug on 1X at approximately 525 yards,  but on 4x I could see it all the way out to approximately 900 yards.

I would like know what happens to that pixel densities calculation for minimum detection when using digital zoom? 

Do you recalculate based on half the FOV with each higher zoom setting? 

Thanks. 

Simply sounds like the IRD has twice the lens size of the Vue and therefore sees twice the distance just like it should.

BTW, this thread sux without pictures.

I know. I know. My iPhone camera is on the blink, and I forgot my DVR. Haha.
 
When I get back where I can connect my laptop to Wi-Fi, I can get some screen captures and video. It just doesn't demonstrate any zoom so it won't show that maximum distance that I was able to see on the monitor. 

Seriously though, it would be nice for a vendor to take all their wares out one night and perform a similar test at the same time using the same method of recording the data. 

It be nice to see how far away each of these thermal devices could potential see a small hot object.  I'm thinking to do this test properly would require two jugs. One lying on its side and one standing upright like mine was last night. 

Thanks. 

I shoot hot water jugs at 200 yards in the back yard with my thermals for fun!

That sounds fun!!

This wasn't supposed to be a comparison or a "slight" against my Flir device.

Just a demonstration of their capabilities under very good conditions.  

When I get home and can upload video from the Vue, I will also try and put file on google drive so you can download the actual file and not the youtube "watered" down (compressed/lossy) version. 

Im not sure it's going to matter, but watching the video on my laptop now, I can see the dot all the way out to 900 yards on my laptop even though there is no digital zoom available on the recording. Remember video from the Vue records to the sandisk card at a higher digital rate versus crappy analog live video output. 

I need to figure out a better way to improve my dashboard viewing capability and am open to suggestions. 

Unfortunanlty, I stopped recording when I could no longer see the image in my laptop. I should have kept video on all the way to 1200 yards. 

I dont know. This seems like a pretty neat test to me and also seems to indicate that "detection" can occur much farther than I calculated. 

Here's a youtube video and I hope a google drive copy of the actual video that came straight off the Vue camera.

https://youtu.be/P3xjx_T0Y9I

Flir Vue file

At 2:42, I stopped at the cross levee half across the 1200 yard wide field, which based on where I placed the jug would have been between 500-600 yards away. On the google drive file, you can still see the dot when I stopped the last time near the end of the video approximately 900 yards away. I'll admit, the detection of the jug seems to come and go on this video the farther away I drive from it, but it still visible at times including after that last NUC just after 5 minutes.

During the video, I am not messing with the settings on the camera. The Vue is doing its auto NUC. 

I hope the google drive link works?

It works great, and that was a cool and useful thing to see. Thank you.

Vue Pro can stream live analog video, but it is only 8-bit (standard NTSC/PAL).  It can capture 14-bit TIFF files (stills or sequences), but not display the stream live.  For that, you'll need a Tau2 with a CameraLink adapter board.  14-bit data does not have AGC, color palette or digital zoom applied.  Only 8-bit data does.  Are you using the analog video output of the camera and running it through an analog frame grabber to your laptop?  If so, you may be comparing the 14 bit (??) stream from the 12u BAE core to the 8 bit analog output of the 17u Tau.  Both VP videos appear to be 8-bit.  Admittedly, I don't know much about the IRD design so perhaps both are 8-bit?  I just don't know which video stream they use to drive their display.  You can try recording a TIFF sequence from the Vue Pro.  It won't be as smooth of a recording (Vue Pro can only capture a TIFF at about 2 frames per second) but you'll get better resolution.

Also, the sky and the top of the vehicle are wreaking havoc on the Vue Pro's AGC.  You can try redicing the ROI down a bit, but that'll only take out the sky.  You'll need to physically adjust the camera to remove the lower portion.

I'm using the analog video output > RCA video jack to USB > laptop > VLC program to watch live. 

I can also record the analog with VLC. 

For this video, I used the internal digital video recording the Vue has built in via micro SD card. 

Edit: I generally keep the ACG set to 75%

If you record using the TIFF file format, you can adjust AGC in post processing using ImageJ.  Makes for a really nice image.  Doesn't really test the real-time capabilities of the camera though.

  I'm sure I've never used the time lapse photography and that setting unless that's a default setting that up later around with one time? Thanks. 

I suppose a better test would have been to approach the milk jug from the 1200 yard point at the end of my test strip and drive toward it to see when I can first spot it. 

I was hoping this test might prompt some discussion on how far away you could see something like a coyote or a hog which is obviously bigger than a milk jug full of hot water. 

My best detection range so far was on a group of deer at 2060 yd away, per Google Earth Pro.
Temp was 40 F
Humidity was pretty low for central La, don't know the exact number but there was just a slight north wind.

Using a IRD MkII    2.5 native mag,  35mm lens, 640x480 res