Streaking when applying fertilizer

[germschwartz757575]

Mod Version

2.4.7.0. Pre release version

Farming Simulator 25 Version

Latest

Game Mode

Singleplayer

What happened?

Was applying nitrogen (uan 32%). To a field of planted corn and sugar beats and it left streaks. Also checked and it did it with potash too

Steps to Reproduce

1. Load game on French decks 4x. Map was modified with latest version of soil installer
2. Was on year 3 planted the fields with the fendt momentum with 0-40-60 in it. Forget the name think I got the numbers right (poliska I think)
3. Waited a year for crops to grow tried adding nitrogen 32% with the fast tool bar from os modding and it left streaks
4. Looked and it also did it when applying dry potash with the fendt fertilizer spreader
5. Potash was dry both were in manual with variable rate turned on

Log Snippet

Turned off computer filing report so I don't forget

Other Mods Loaded

Lots. All 5 base mods for realistic fertilizer worker wages mod market dynamics moisture mod. Not seasonal crop stress place fences and buildings anywhere. No auto drive or course play

Anything Else?

No response

[github-actions]

Hey @germschwartz757575, thanks for opening this issue! 👋

@TheCodingDad-TisonK has been notified and will take a look as soon as possible.

While you wait, please double-check the following:

• You are running the latest version of Soil Fertilizer (visible in the mod manager)
• Your issue includes relevant lines from log.txt
  • Log location: Documents/My Games/FarmingSimulator2025/log.txt
  • Search for [Soil Fertilizer] - paste the relevant section in your issue
• You have searched existing issues for a similar report

Important: If this issue is missing a log snippet, it may be closed without investigation. A log is almost always required to reproduce bugs.

[nitroSkirata]

Also getting streaking with the base-game Bertould sprayer, applying UAN-28 and liqiuid potash

[mbakker96]

When using the sky X50 (DLC) it spayers the full width, but skips the center part. Looks like the tractor width.

[Zocker-Club]

I can confirm this problem. I use the standard Hardi 24m sprayer.

[TheCodingDad-TisonK]

Confirmed, reproduced in the code, and re-fixed. Thanks to @nitroSkirata, @mbakker96 and @Zocker-Club for confirming it on a Berthoud, a Sky X50 and a Hardi 24m. Four different sprayers ruled out a machine specific cause straight away, which is the reason this got found quickly rather than slowly.

This is a regression from the fix for #735, so it is our own doing. Since #735 the mod paints the exact cells you cover instead of the whole field average. The painter marked cells at the boom positions it sampled on each tick. At working speed the distance between two ticks is wider than the cell grid, so whole rows between the samples were never marked. That is the striping. @mbakker96's Sky X50 case, full width applied but the centre skipped, is the same gap seen through a different boom layout rather than a separate bug.

The painter is now a swept quad. It fills the parallelogram between the previous boom position and the current one, so there is no gap between ticks and no double coverage where passes meet. Three things worth saying about it:

• It self heals a tick that failed to register, rather than leaving a permanent stripe.
• The dose is divided by the swept strip area, so the total product applied is unchanged. You are not getting more or less than before, it is just spread across the cells you actually drove over.
• Coverage marking and painting are now separate, which is what let the overlap go away.

Status right now: built, tested and open in PR #783, waiting to be merged to main. It is not in v2.5.0.0, so it lands in the next release after that merge. I will not call it shipped until it is.

One thing still owed on our side, and I would rather say it than let you find out: this has not yet been observed in game. The tests cover the geometry, not the look of a real field. So when the next release lands, a screenshot of a pass on the same French Deck field would be genuinely useful, especially with variable rate on, since that is the combination you reported.

[HStein72]

Did a quick test with current development version 2.5.0.4, built from github. Now it changes the cell value to the absolute maximum for that category, even with minimum application. Used manure with very small working width. Its also still creating strips on the map, despite 100% coverage using GPS.

Can test more later and on clean save, if required.

[sasha-rf]

Thanks, this is a genuinely useful regression report and I can confirm from the code that it is real, not an artifact of the test build.

The re-fix scales the per-cell dose by dividing the field average by the area of the strip it paints. The strip area is computed from the boom width and the travel between frames, and for a tool with a very small working width that strip area gets tiny. Dividing by a tiny area pushes the dose up hard, and the value map caps the cell at the absolute maximum for the category. So a minimum application on a narrow tool lands at the top of the range. That matches your screenshot exactly.

The strips remaining despite 100 percent coverage is the other half of the same picture, the strip geometry on the map is still not matching the covered cells.

Both halves are on the build side now, this is a regression of a fix that only just merged, so it is tracked as such rather than as a new request. No date promised, but your report went straight to the top of the queue because it is a correction to work that already shipped.

If you can, tell us the exact spreader model and working width you used, and whether you were in singleplayer or on a server, it saves a reproduction step.