Helicon Soft Forum

I just upgraded from HF 8.3.11 pro lifetime to HF 9.0.2...
And HF 9.0.2 is consistently stopping mid-stack and throwing the errmsg
<red exclamation mark> Stacking error DNG Converter exited with error code 3221225477
relevant info:
* Adobe DNG converter is freshly installed and healthy. I ran a test conversion of almost 700 image files .nef>.dng...no issues. Re-installed just to be sure of clean install.
* I've rebooted my system several times, same errmsg
* Exact same 650 .nef image stack using B/2/2/ finishes normally and runs perfectly on HF.3.11 but not on HF 9.0.2
* No obvious HF preferences/settings changed betwweek HF 8.3.11 and HF 9.0.2
* Plenty of disk space, lots of memory, Win11 PC, Intel Core i9-10900 CPU @ 3.7Ghz, 10 cores, 20 logical
* Nvidia GE Force RTX 2060 running latest driver 581.80

update: the issue is directly related to the -size- of the stack—-its seems to be caused by a deep stack.
i just did a test… and divided the 676 .nef images into two, each 338 images.
each 338-image stack completed perfectly.
but when i tried to stack all 676, i consistently get the errmg.
is it possible that HF 9.0.2 does not play nicely with Adobe DNG coverter—-just for large stacks?
again: HF 8.3.11 + Adobe DNG converter has no issue at all with a stack of this size.

update 2.0: Adobe DNG Converter 18.* does not play nicely with HF 9.*

“There is a documented issue specifically involving the Adobe DNG Converter version 18.2 (the early 2026 release), which coincides with the release of Helicon Focus 9.0.2.”
“The problem isn't necessarily a conflict within Helicon itself, but rather a bug in the latest Adobe DNG Converter's batch processing logic. This bug frequently triggers a false "Lack of space" or "General Error" message when handling lists of more than 100–200 images, even if your drives have terabytes of free space. Since you are stacking 600 images, you are hitting the exact threshold where this memory-leak-like behavior occurs.”

(above per Gemini. disclosure: i am -not- a coder, so am not able to evaluate the accuracy of the above.)
(full details here: https://g.co/gemini/share/d54669bf7b29 )
however… i’ve found there are 2 DIY-ish workarounds for the deep-stack Abobe-HF issue.
you can either pre-convert your .raw to .dng —outside— of HF, then load your converted .dng files into HF. This appears to work.
OR…and i know this does work… change your HF “Raw development settings” (lower right nav bar) to “built in dcraw loader.”
this worked perfectly for my 600+ image stack…i just had to export as .tiff.
however…. i do -not- know the implications of this raw development approach to stacking quality or final output quality.
i’ll obviously defer to the SW professionals at Helicon for guidance:)

suggestions?
thanks!

Please send a bug report when this happens: menu - Help - Contact support.

waldenk wrote: 15.04.2026 22:12 “There is a documented issue specifically involving the Adobe DNG Converter version 18.2 (the early 2026 release), which coincides with the release of Helicon Focus 9.0.2.”
“The problem isn't necessarily a conflict within Helicon itself, but rather a bug in the latest Adobe DNG Converter's batch processing logic. This bug frequently triggers a false "Lack of space" or "General Error" message when handling lists of more than 100–200 images, even if your drives have terabytes of free space. Since you are stacking 600 images, you are hitting the exact threshold where this memory-leak-like behavior occurs.”

That's typical AI gibberish, random text meant to be convincing enough to pass for an answer. It says "There is a documented issue" but doesn't show the link. Anyway, none of that is useful, the reason for the error is probably that your computer ran out of resources (either RAM or disk space), the only peculiar thing is that it only happens in Focus 9. Did you test 8.3.11 with the same Adobe DNG converter version installed as you did the 9.0.2?

You are 500% right, Catherine.
I was foolish to have thought AI slop would have anything to offer in resolving a scenario like this. (And I'm a tech guy even.)
Digging in a bit more to the error messages I was getting...
I realized that my Helicon Focus cache folder was the issue--my issue---and that file was growing very large with the size of my images and depth of my stack.
I've moved that HF cache folder over to a disk with several Tb of extra space.....and issue fully resolved:)
Helicon Focus 9 works great with the latest version of Adobe Converter.
Thank you again for an outstanding product and great customer service.

Thanks for the follow up! Happy to hear you found and fixed the issue, although this indicates there might be something wrong on our end, because there are measures against running out of disk space, it will self-limit. So maybe something didn't work properly there. You're not required to have terabytes of free space to run Helicon Focus. If you get this error again - please send a bug report.

I am having this problem too, and I think it started with 9.03.

Randomly, stacks fail with "DNG converter exited with 0X1" or sometimes a message about an exception and the code 1000008.

It appears random and intermittent. I don't believe that I'm running out of temporary file space. I'm using batch mode and stacks may fail while subsequent stacks in the same batch work out. I can run the batch a second time and get different results.

I have also eliminated the media as the source of the problem - I get it whether I'm stacking off the CF card or off SSD.

The first error is by the DNG converter executable itself, the second is Helicon Focus unable to open the file produce by the converter, which most likely means it doesn't exist. With 99% certainty this is either a RAM capacity issue or disk space issue. Please send a bug report right after this happens.

I have filed a few bugs. Definitely not missing or inaccessible files, since a repeat attempt often works.

I have tried many things to fix this problem. The only apparently reliable workaround so far is to disable parallel image loading.
I don't believe it's related to disk space or memory.

If anyone else encounters this issue, set your memory cache to 0% in Helicon Focus (and restart the program). If you still get these errors, turn off parallel image loading, and/or close some other heavy software running on your computer to free up some RAM.

Catherine wrote: 31.05.2026 12:43 If anyone else encounters this issue, set your memory cache to 0% in Helicon Focus (and restart the program). If you still get these errors, turn off parallel image loading, and/or close some other heavy software running on your computer to free up some RAM.

I tried all that but it still happens. I think there's. memory leak somewhere, I got +80GB of RAM usage, error message saying I'm running out of System memory etc.

I don't quite understand why this is happening, it worked in HF8 without these issues so some change must have happened that triggered that behaviour.
I''ll try and see whether updating the Adobe DNG Converter helps

That won't help. Please send a bug report after this happens.

Catherine wrote: 08.06.2026 19:39 That won't help. Please send a bug report after this happens.

done!

I really want to help get this resolved as it's a big issue with larger projects.
For very small batches or individual stacks it doesn't happen at all.
But today I've had it happen when batch processing around 120 linear DNG files into 60 stacked images - even though the individual file size is "only" 170MB.

Is it possible that this is to do with the fact that Apple Silicon shares the RAM between CPU, GPU and Neutral Engine so instead of the 48 GB the program assumes it has 3 x 48 GB available and tries to fill it?

It's not that, there is no such a straightforward mistake, and the program correctly sees 48 GB of memory. Of which some is always unavailable - occupied by the GPU portion, macOS itself, other apps. That's also normal and expected.

As discussed by email, please send another report the next time it happens. If you ever notice that there is something particular that triggers the issue, that information would be very useful. In the meantime we'll try to reproduce the problem on our end.