6
$\begingroup$

Like the question says, JW sensors are stated to be 4 mp for near-IR (also 1 mp for mid-IR). I'm curious why this low a resolution was chosen.

"Webb uses two different types of detectors: mercury-cadmium-telluride (abbreviated HgCdTe) "H2RG" detectors for the 0.6-5 μm "near-infrared" and arsenic doped silicon (abbreviated Si:As) detectors for the 5-28 μm "mid-infrared". The near-infrared detectors were made by Teledyne Imaging Sensors in California. "H2RG" is the name of the Teledyne product line. The mid-infrared detectors were made by Raytheon Vision Systems, which is also in California. Each Webb H2RG detector has about 4 million pixels. The mid-infrared detectors have about 1 million pixels each."

The answer could be as simple as, it really was state of the art for space-environment reliable sensors made of those materials, at whatever time the devices they went in were fixed and commissioned, and NASA investigated and concluded they wouldn't have been practical to update any time after that.

Or maybe more really wasn't needed for its mission, though that seems very unlikely (when is less resolution ever desirable in cosmology!)

But I don't know, and I'm curious.

$\endgroup$
6
  • 1
    $\begingroup$ Maybe to do with the materials used? webb.nasa.gov/content/about/innovations/infrared.html In all likelihood, your 'could be as simple as' would be the most likeliest. "New and Better Detectors for the Webb Near-Infrared Spectrograph" is a 2014 publication that seems to be looking at updates but I cannot find a link. $\endgroup$ Feb 7, 2022 at 16:46
  • 2
    $\begingroup$ Remember that not all megapixels are created equally. $\endgroup$ Feb 7, 2022 at 17:28
  • 3
    $\begingroup$ The main imager has 40 MPix, not 4. Resolution is limited by the optics. $\endgroup$
    – asdfex
    Feb 7, 2022 at 19:13
  • 1
    $\begingroup$ @uhoh so who's correct; asdfex that says 40 megapixels or the OP that says 4 megapixels? $\endgroup$
    – Sheldon
    Feb 7, 2022 at 23:08
  • 3
    $\begingroup$ @Sheldon Looks like the 4mp is per sensor unit, but they use multiple units per instrument. "Each Webb H2RG detector has about 4 million pixels" - then fig 3 shows them in a block, and table 1 reports that NIRCam has ten of them, so 40mp total. The other near-IR instruments have 2 and 3 units, plus 3 of the lower-resolution detectors on the far-IR unit. $\endgroup$ Feb 7, 2022 at 23:20

1 Answer 1

4
$\begingroup$

Cheap customer digital cameras are sold by the marketing using high pixel counts. More pixel than possible by the resolution of the lens, the signal noise ratio and the sensitivity.

The JWST is used by experienced scientists prefering a pixel count aligned to the telescope resolution at much longer wavelengths than visible light. Digital images without redundant pixels allow the transmission of more images per hour.

$\endgroup$
4
  • 1
    $\begingroup$ Thanks! So far, this just says to me that mass market cameras have marketing-style pixel counts, and scientists prefer good quality and more "valid" pixel counts, for data reasons. But doesn't actually answer the question. Can you address the OP question? $\endgroup$
    – Stilez
    Feb 8, 2022 at 20:33
  • 2
    $\begingroup$ the high density consumer camera chips use mass produced CMOS technology processing rather than the custom Mercury-cadmium-telluride H2RG deep infrared chips needed by JWST. So JWST's chips don't benefit from some of the improvements in density that consumer chips have gotten. Anyway, JWST is focused on optimizing sensitivity and low noise in the infared rather than density. From what I saw they arrange the chips in two different 2x2 arrays which have16 megapixels each, plus other chips. $\endgroup$
    – Sheldon
    Feb 8, 2022 at 22:45
  • $\begingroup$ I remember a very patient salesperson letting me unbox two digital cameras and take pictures to actually compare them rather than trusting megapixels. Back in the day the higher megapixel one was better, not sure if still true today. But it felt that I could downsample 2:1 and not lose anything. $\endgroup$ Sep 3, 2022 at 0:58
  • 1
    $\begingroup$ @KevinKostlan I had once (in 2004) compared a 6.3Mp compact digital camera image of flowers with a ~3Mp larger body fixed lens digital camera using the same scene. Zooming in the detail was better and sharper on the 3Mp camera, and they were in a similar price range. Really made me realize lower pixel count with larger optics is better than higher pixel count with smaller optics. $\endgroup$ Apr 19, 2023 at 2:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.