Deciding on a thermal scope

D

djones

Guest
thanks for chiming in snake. i'm still not sure about the 'g' thing tho. there is such a thing as constructive delivery. you know, the substance over form thing. if there was intent, we may have to acknowledge the possibility we have a heathen amongst us!

ps - crackdog, i never knew you cared... x0x0x0
 

Ratdog68

LSB Official Story Teller
LSB TURKEY BUZZARD PRESERVATION SOCIETY
SUS VENATOR CLUB
LoneStarBoars Supporter
Totally kidding. I mess with DJ all the time. Sorry for taking it a bit too far. I was boozing it up a bit and lost my filter. Sorry guys.
Glad to hear it was all in good fun, carry on. And, I agree, he IS a good candidate for a little abuse.
 

Ratdog68

LSB Official Story Teller
LSB TURKEY BUZZARD PRESERVATION SOCIETY
SUS VENATOR CLUB
LoneStarBoars Supporter
thanks for chiming in snake. i'm still not sure about the 'g' thing tho. there is such a thing as constructive delivery. you know, the substance over form thing. if there was intent, we may have to acknowledge the possibility we have a heathen amongst us!

ps - crackdog, i never knew you cared... x0x0x0
Don't get any spit-swappin' ideas there dirt clod murderer. :rolleyes:
 

DarkNight

LSB Member
Great info, I've learned a lot already from you guys so thanks for that. I read 640 is better than 336. What makes the 640 better? Does it have to do with resolution/image clarity?
Pulled an all-nighter last night. Called in a triple & could see coyotes clearly with the hd19a but could not aquire with the gen 2+ & ir illuminator.
Another stand saw something that moved like a coyote with the hd19a but could not get in close enough to id. Frustrating!
 

theblakester

Got a black belt in keeping it real.
LSB TURKEY BUZZARD PRESERVATION SOCIETY
SUS VENATOR CLUB
LoneStarBoars Supporter
If everything else was equal a 640x480 would have twice the resolution clarity of a 320x240. Someone might be able to explain this a lot better but it's the number of pixels, 640 along the X axis, and 480 along the Y axis. Previously I had thought 640x480 was 4x better than 330x240 bc if u add the total of each its 4x more. But IR Defense explained that that's not how it works. It's that 640 is two times 320 and 480 is two times 240 x axis by x axis and y axis by y axis.
Then there's the core and it's microns. The LOWER the number of microns, the MORE amt information can fit on each pixel, the better.
Then there's the display screen quality OLED vs LCD (OLED is better). The display screen also has a pixel count, which is also different from resolution(800x600 vs 640x480 usually... this is not generally what people refer to when talking about 640 vs 320 or 336 or 384)
Then there's a bunch of stuff I can't explain. Someone might correct me on the above but I think that's accurate...ish
 

Hard_ware

Here piggy piggy! Deep Deep S. TX.
If everything else was equal a 640x480 would have twice the resolution clarity of a 320x240. Someone might be able to explain this a lot better but it's the number of pixels, 640 along the X axis, and 480 along the Y axis. Previously I had thought 640x480 was 4x better than 330x240 bc if u add the total of each its 4x more. But IR Defense explained that that's not how it works. It's that 640 is two times 320 and 480 is two times 240 x axis by x axis and y axis by y axis.
Then there's the core and it's microns. The LOWER the number of microns, the MORE amt information can fit on each pixel, the better.
Then there's the display screen quality OLED vs LCD (OLED is better). The display screen also has a pixel count, which is also different from resolution(800x600 vs 640x480 usually... this is not generally what people refer to when talking about 640 vs 320 or 336 or 384)
Then there's a bunch of stuff I can't explain. Someone might correct me on the above but I think that's accurate...ish

Yes 2x is right, a 640 using 2x zoom is the same as a 320 unit with the same lens mm.

Pixels are on the LCD or OLED screen
detectors are on the microbolometer sensor
A b/w (8 bit) pixel can only be made to change from black thru 254 brightness levels(shades of grey) to white.
A b/w (10bit) pixel can only be made to change from black thru 1022 brightness levels(shades of grey) to white.
So there is no changing amount of data placed within a pixel, a pixel is a single unit that can only have its brightness level adjusted.
Having a 800x600 display will have to have a 640x480 detector re-sampled to up-convert the lower quality detector to fit the higher quality screen. A decision will have to be made in firmware as to how the sensors arrays 640 detectors will be resampled to the displays 800 pixels. The microbolometers individual detector can also only contain 8 bit or 10 bit information. The display will have to multiply 1.25 by the number of detectors to have the 640 fit into 800, but since you can't have 1.25 of a pixel it has to be resampled.
detector pitch is measured in microns, a smaller number = more detectors per inch but that does not matter cuz they are not sold by the inch but sold by the array size IE 320x240 or 640x480 etc etc
Smaller core = smaller lenses this saves the company money as germanium is $$$$$
640 core = 75% more detectors over a 320 core, but also has a 75% larger FOV (if using the same mm lens) so resolution has not changed. 75% more detectors and 75% more FOV resolving power has not changed.
The ratio of detectors to FOV is the same.
The detail is sharper on the display because you squeezed the larger FOV with more detectors that go with it on to the same LCD or OLED screen, so dots per inch of information goes up.
But the rabbit got smaller vs a 336 core with same lens, so did everything else didn't it?
Now the trees and critters are 1/2 the size but everything looks sharper.
This detail improvement is based on as you make things smaller the image gets sharper or more dots per inch.
Place a quarter at 100yds and view it with a 336 75mm core, then view it with a 75mm 640 core.
640 core will not see it unless you zoom in at 2x, which is the same as a 336 core at 1x.
Resolving power is based on the detector pitch and the lens mm, the size of the detector has nothing to do with resolving power only how much the FOV. Larger detector larger FOV with the same lens.
Lets use say a 20 degree FOV using a 640 detector, the detector will see the 20 degrees of image which will be stretched across the 640 detectors on the sensor. Notice how the pitch does not matter, only the FOV and number of detectors.
Now if you had a 50 micron pitch detector you would need a lens double the size compared to a 25 micron detector, or 4x larger then a 12 micron detector would need.
The micron size has nothing to do with the clarity, but has everything to do with the size of the detector and the size of the lens . The sensitivity of the detector, the amount of LWIR that passes thru the lens material, the quality of the display screen, as well as the firmware that cleans up the information that is detected are what makes one unit stand out over another.

A lot of magic happens to make these awesome images the thermal units we have available to us today happen.
All the thermal MFG's have stepped up the game and the images the units produce show it.
Hopefully prices will continue to drop.
 
Last edited:

theblakester

Got a black belt in keeping it real.
LSB TURKEY BUZZARD PRESERVATION SOCIETY
SUS VENATOR CLUB
LoneStarBoars Supporter
Yes 2x is right, a 640 using 2x zoom is the same as a 320 unit with the same lens mm.

Pixels are on the LCD or OLED screen
detectors are on the microbolometer sensor
A b/w (8 bit) pixel can only be made to change from black thru 254 brightness levels(shades of grey) to white.
A b/w (10bit) pixel can only be made to change from black thru 1022 brightness levels(shades of grey) to white.
So there is no changing amount of data placed within a pixel, a pixel is a single unit that can only have its brightness level adjusted.
Having a 800x600 display will have to have a 640x480 detector re-sampled to up-convert the lower quality detector to fit the higher quality screen. A decision will have to be made in firmware as to how the sensors arrays 640 detectors will be resampled to the displays 800 pixels. The microbolometers individual detector can also only contain 8 bit or 10 bit information. The display will have to multiply 1.25 by the number of detectors to have the 640 fit into 800, but since you can't have 1.25 of a pixel it has to be resampled.
detector pitch is measured in microns, a smaller number = more detectors per inch but that does not matter cuz they are not sold by the inch but sold by the array size IE 320x240 or 640x480 etc etc
Smaller core = smaller lenses this saves the company money as germanium is $$$$$
640 core = 75% more detectors over a 320 core, but also has a 75% larger FOV (if using the same mm lens) so resolution has not changed. 75% more detectors and 75% more FOV resolving power has not changed.
The ratio of detectors to FOV is the same.
The detail is sharper on the display because you squeezed the larger FOV with more detectors that go with it on to the same LCD or OLED screen, so dots per inch of information goes up.
But the rabbit got smaller vs a 336 core with same lens, so did everything else didn't it?
Now the trees and critters are 1/2 the size but everything looks sharper.
This detail improvement is based on as you make things smaller the image gets sharper or more dots per inch.
Place a quarter at 100yds and view it with a 336 75mm core, then view it with a 75mm 640 core.
640 core will not see it unless you zoom in at 2x, which is the same as a 336 core at 1x.
Resolving power is based on the detector pitch and the lens mm, the size of the detector has nothing to do with resolving power only how much the FOV. Larger detector larger FOV with the same lens.
Lets use say a 20 degree FOV using a 640 detector, the detector will see the 20 degrees of image which will be stretched across the 640 detectors on the sensor. Notice how the pitch does not matter, only the FOV and number of detectors.
Now if you had a 50 micron pitch detector you would need a lens double the size compared to a 25 micron detector, or 4x larger then a 12 micron detector would need.
The micron size has nothing to do with the clarity, but has everything to do with the size of the detector and the size of the lens . The sensitivity of the detector, the amount of LWIR that passes thru the lens material, the quality of the display screen, as well as the firmware that cleans up the information that is detected are what makes one unit stand out over another.

A lot of magic happens to make these awesome images the thermal units we have available to us today happen.
All the thermal MFG's have stepped up the game and the images the units produce show it.
Hopefully prices will continue to drop.
Ya that's basically exactly what I said hahaha pot-AE-toe, pot-AH-toe....
 

znztivguy

LSB Member
Then all things being equal..one should aim also for resell value and coolness factor lol.....I'm for the products used by various organizations around the world BAE OASYS and INSIGHT....
Flir comes a distant third.
Buffering on the IRD products has been known to be suspect causing rates of fail that are higher currently than the units fielded by the military.
I guess that goes for most of the commercial products as well. Most of us would never shoot enough to push the various products to that degree anyway.
If I had to go with Commercial I would go with the IRD....they say that they use the 12 micron BAE OASYS micron chip.
Thanks to the above write up by hard ware, he clearly explains that the
"The micron size has nothing to do with the clarity, but has everything to do with the size of the detector and the size of the lens"
Taking that into account compare the 17 micron OASYS thermals with the current crop of 12 micron commercial products you still will see overall, that the milspec OASYS and Insight UNITS still kick butt and are higher rated for recoil, not to mention better built.
 

Hard_ware

Here piggy piggy! Deep Deep S. TX.
Then all things being equal..one should aim also for resell value and coolness factor lol.....I'm for the products used by various organizations around the world BAE OASYS and INSIGHT....
Flir comes a distant third.
Buffering on the IRD products has been known to be suspect causing rates of fail that are higher currently than the units fielded by the military.
I guess that goes for most of the commercial products as well. Most of us would never shoot enough to push the various products to that degree anyway.
If I had to go with Commercial I would go with the IRD....they say that they use the 12 micron BAE OASYS micron chip.
Thanks to the above write up by hard ware, he clearly explains that the
"The micron size has nothing to do with the clarity, but has everything to do with the size of the detector and the size of the lens"
Taking that into account compare the 17 micron OASYS thermals with the current crop of 12 micron commercial products you still will see overall, that the milspec OASYS and Insight UNITS still kick butt and are higher rated for recoil, not to mention better built.


The IRD units control firmware does not appear to be very refined, I suspect they are using interrupts to jump in and detect the rotary encoders. Debounce times and delays can cause problems when switching fast if not properly handled in the firmware.
When and where you jump in makes all the difference as to interrupting what was being done, this is reflected in the menus that do not populate correctly until you switch screens etc etc. Just go slow and let the processing catch up:eek:
If IRD were to hook up a parallel video processing engine internally so that recording or outputting HD 720p or 1080 video would make unit even better. But without adding additional video processing power the unit acts like it would lock up if it trying to output any digital video streams.
The image rendered is very nice. I want to get a IRD patrol next, but need to sell several units to fund it first.
 
Last edited:

DarkNight

LSB Member
The IRD units control firmware does not appear to be very refined, I suspect they are using interrupts to jump in and detect the rotary encoders. Debounce times and delays can cause problems when switching fast if not properly handled in the firmware.
When and where you jump in makes all the difference as to interrupting what was being done, this is reflected in the menus that do not populate correctly until you switch screens etc etc. Just go slow and let the processing catch up:eek:
If IRD were to hook up a parallel video processing engine internally so that recording or outputting HD 720p or 1080 video would make unit even better. But without adding additional video processing power the unit acts like it would lock up if it trying to output any digital video streams.
The image rendered is very nice. I want to get a IRD patrol next, but need to sell several units to fund it first.
So what thermal scope(s) are you using hard ware? Did you start out with less expensive models & upgrade over time?
 
D

djones

Guest
popcorn_zpsbzyipy5c.gif
 

Hard_ware

Here piggy piggy! Deep Deep S. TX.
So what thermal scope(s) are you using hard ware? Did you start out with less expensive models & upgrade over time?
Flir, Armasight, IR hunter, and DRS.

Armasight uses the Flir Tau cores, IR Hunter uses the BAE core, and DRS uses a m6000 core they developed back in 2006 used on small military drones.

I currently like the image of the IR Hunter units.
The DRS 640 with a 62mm lens is in a remote controlled go light that I mount on the roof of my truck and controlled by a hand held remote I built, so no computer is needed to control it.
 
Last edited:

JPK

LSB Active Member
Like that, eh DJ?

I was hog hunting the last couple of nights out of a 4wd van with a remote controlled 640 camrera mounted on the roof. We could scan huge fields while sitting in the van, pretty plush.

JPK
 
D

djones

Guest
that sounds like comfortable way to hunt. we might explore something like that when we hit our 90's and don't feel like hiking for miles in blizzard conditions just to LOOK for hogs.

team OinK isn't just about looking good while blasting hundreds of hogs. a lot of hard work goes into the organization giving interviews, fundraisers for the OinK foundation with proceeds mostly going to needy cowboy cheerleaders, and numerous other worthwhile activities. i'm not sure you could even keep up with Oink following him around changing out his weapon sight's batteries.

here's a peek at the conditions for the upcoming event, which will stretch the limits of man, machine, equipment and hog. you might be safer staying curled up in the good times van sipping hot cocoa and reading up on recipes from your paula deen cookbook.

forecast.jpg
 

RattlesnakeDan

San Antonio Texas
LSB TURKEY BUZZARD PRESERVATION SOCIETY
SUS VENATOR CLUB
LoneStarBoars Supporter
No freezing rain? Wimpy!
 
Top