A closer look at Backscatter’s 2026 strobe test: corrected charts and additional measurements

Backscatter (BS) recently published a comparison of more than thirty underwater strobes. It is an impressive amount of work and contains genuinely useful data. We have now gone through that data in detail, repeated several measurements ourselves, and published our findings here:

https://www.retra-uwt.com/blogs/news/a-closer-look-at-backscatters-strobe-test-retra-uwt

We found several issues that we think deserve discussion:

• Guide numbers are plotted on a linear scale, which makes differences between powerful strobes appear larger than they are in photographic terms. We replotted the data in F-stops.

• The beam viewer uses different exposure settings for some strobes and does not normalize their centre brightness, so it should not be treated as a calibrated beam comparison.

• There are almost no error bars, variance figures or trial counts, making it difficult to tell whether smaller differences are meaningful.

• We could not reproduce the reported Retra Maxi color-temperature behaviour or its reported Boost recycle time.

• The Isotta RED64 guide-number result appears inconsistent with BS’s own beam images and flash-duration data.

There is also a commercial context worth acknowledging. BS is both a retailer comparing products it sells and the owner of a strobe brand included in the comparison. That does not automatically make the data invalid, but it does make transparency about methods, uncertainty and commercial interests particularly important. This commercial focus is also visible in the content BS produces for its website and social media. The commanding majority of its promotional effort is focused on its own branded products.

Including many competing brands in the test is commercially useful for BS. It creates the impression that they have a complete overview of the market and can offer neutral advice on almost every available strobe. Many customers come to BS precisely because they are undecided and want guidance. Once BS is accepted as the authority with all the data, it becomes much easier for them to steer those customers towards their own products. This does not automatically mean that the data is false, but it is an important reason to examine how the testing and presentation may support that sales strategy.

The same standard applies to us. Retra manufactures two of the strobes discussed in the test, so we are not a neutral third party either. We have disclosed that interest and included our methods, measurements and reasoning so they can be challenged directly.

Our intention is not to dismiss BS’s work. We would like to see the questionable readings checked, the charts improved and more information about measurement uncertainty published. If anyone finds an error in our analysis, please point it out. We will correct it.

I’m interested to hear how others interpret the data, particularly the color-temperature results, the RED64 reading and whether guide-number charts should be presented in F-stops.