• site home
  • blog home
  • galleries
  • contact
  • underwater
  • the bleeding edge

the last word

Photography meets digital computer technology. Photography wins -- most of the time.

You are here: Home / The Last Word / Counts vs DNs

Counts vs DNs

December 31, 2025 JimK Leave a Comment

When engineers of a my advancing years talk about raw ADC outputs, they tend to call them counts. That word choice is not entirely a matter of habit or stubbornness. It reflects how those systems were actually experienced and debugged at the time. In early systems, the output of an analog to digital converter was not an abstract value retrieved from memory. It was a parallel word on a backplane or ribbon cable. If something was misbehaving, you clipped a logic analyzer onto the bus and watched the bits change. As you turned a knob or varied an input voltage, the pattern on the analyzer stepped up or down in integer increments. Those steps were the quantization bins of the converter made visible. What you saw were counts in the most literal sense. The converter was counting which bin the input landed in, and the hardware was showing you that count.

That perspective shaped how engineers thought and spoke. Noise was something that caused the least significant bits to flicker, so it was described as a certain number of counts RMS. Offset was a shift of the entire transfer function by some number of counts. Linearity errors were discussed in terms of missing counts or codes that repeated. None of that language required or even encouraged attaching physical units at that stage. The numbers were understood to be dimensionless indices produced by a quantizer. The mapping to volts, pressure, or mass lived elsewhere, usually in a calibration constant applied later.

The notion of an LSB followed the same logic. An LSB was not primarily a bit weight in some abstract binary representation. It was the width of one quantization bin, the range of input values that produced the same output count. One count corresponded to one LSB step, and adjacent counts corresponded to adjacent bins. That interpretation was natural when you were watching the codes advance one at a time on a logic analyzer display.

Now, direct contact with the converter output has faded. ADCs moved inside microcontrollers and system on-chip devices. The raw codes were no longer visible on a bus but were read from registers or memory buffers. In imaging systems especially, those integers quickly became pixels, array elements, and statistics. In that context, the term data number emerged. It fit a world in which the integer was treated primarily as data to be processed, averaged, and displayed, rather than as the immediate output of a measuring device. Data number is redundant. Once a quantity is a number, it is already data. Adding the word data does not clarify its meaning or origin, it merely reflects a shift in perspective away from measurement and toward representation.

Both terms suppress physical units, and both require an external scale factor to recover meaning. A count by itself is no more a volt or an electron than a data number is. The difference lies in what the term invites you to think about. Counts point back to the act of quantization and to the hardware that produced the number. Data number points forward to the pipeline that will consume it.

If you spent years watching ADC outputs on a logic analyzer, it is hard not to hear “counts” as the more honest description. It names what was actually observed. The number was not yet data in any richer sense. It was simply the converter telling you, one integer at a time, where the input landed.

I am bilingual on this point, but I prefer counts to DNs.

The Last Word

← Hasselblad XCD 35-100E f/2.8-4 summary Is zero a signal? →

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

January 2026
S M T W T F S
 123
45678910
11121314151617
18192021222324
25262728293031
« Dec    

Articles

  • About
    • Patents and papers about color
    • Who am I?
  • How to…
    • Backing up photographic images
    • How to change email providers
    • How to shoot slanted edge images for me
  • Lens screening testing
    • Equipment and Software
    • Examples
      • Bad and OK 200-600 at 600
      • Excellent 180-400 zoom
      • Fair 14-30mm zoom
      • Good 100-200 mm MF zoom
      • Good 100-400 zoom
      • Good 100mm lens on P1 P45+
      • Good 120mm MF lens
      • Good 18mm FF lens
      • Good 24-105 mm FF lens
      • Good 24-70 FF zoom
      • Good 35 mm FF lens
      • Good 35-70 MF lens
      • Good 60 mm lens on IQ3-100
      • Good 63 mm MF lens
      • Good 65 mm FF lens
      • Good 85 mm FF lens
      • Good and bad 25mm FF lenses
      • Good zoom at 24 mm
      • Marginal 18mm lens
      • Marginal 35mm FF lens
      • Mildly problematic 55 mm FF lens
      • OK 16-35mm zoom
      • OK 60mm lens on P1 P45+
      • OK Sony 600mm f/4
      • Pretty good 16-35 FF zoom
      • Pretty good 90mm FF lens
      • Problematic 400 mm FF lens
      • Tilted 20 mm f/1.8 FF lens
      • Tilted 30 mm MF lens
      • Tilted 50 mm FF lens
      • Two 15mm FF lenses
    • Found a problem – now what?
    • Goals for this test
    • Minimum target distances
      • MFT
      • APS-C
      • Full frame
      • Small medium format
    • Printable Siemens Star targets
    • Target size on sensor
      • MFT
      • APS-C
      • Full frame
      • Small medium format
    • Test instructions — postproduction
    • Test instructions — reading the images
    • Test instructions – capture
    • Theory of the test
    • What’s wrong with conventional lens screening?
  • Previsualization heresy
  • Privacy Policy
  • Recommended photographic web sites
  • Using in-camera histograms for ETTR
    • Acknowledgments
    • Why ETTR?
    • Normal in-camera histograms
    • Image processing for in-camera histograms
    • Making the in-camera histogram closely represent the raw histogram
    • Shortcuts to UniWB
    • Preparing for monitor-based UniWB
    • A one-step UniWB procedure
    • The math behind the one-step method
    • Iteration using Newton’s Method

Category List

Recent Comments

  • CarVac on Hasselblad X2DII shadow noise analysis
  • bruce on Input-Referred Noise in Image Sensors
  • bruce on Input-Referred Noise in Image Sensors
  • JimK on More Than Exposure: Understanding Total Light on the Sensor
  • Crack on More Than Exposure: Understanding Total Light on the Sensor
  • JimK on Hasselblad XCD 100-35 on X2D II, Siemens star, corner
  • Tom on Hasselblad XCD 100-35 on X2D II, Siemens star, corner
  • JimK on Hasselblad XCD 100-35 on X2D II, Siemens star, corner
  • Christer Almqvist on Hasselblad XCD 100-35 on X2D II, Siemens star, corner
  • JimK on Hasselblad XCD 100-35 on X2D II, Siemens star, corner

Archives

Copyright © 2026 · Daily Dish Pro On Genesis Framework · WordPress · Log in

Unless otherwise noted, all images copyright Jim Kasson.