NotMyFault

Well, destructive has several possible meanings. It is technically correct that Photo will not change raw files. But you can use develop persona on e..g. Tiffs and save them - with destructive effect to the affected pixel layer. In my perspective this is destructive in comparison to adjustment layers as the settings applied in develop persons will destructively change the rgb pixel layer. If you go insane on the exposure and apply, you cannot re-enter develop persona and tweak exposure to re-gain out-blown highlights. If you use an exposure adjustment in photo persona instead, it would be non-destructive as you could re-adjust the sliders any time later and re-gain out-blown highlights. So please keep these different concepts of “destructive” in mind: 1. Photo cannot save files as RAW Format, so will never do any harm to them (non-destructive by being forced to save a copy in different format) 2. Develop Persona (and Tone Map, Liquify etc) are destructive to the pixel layer. You cannot re-adjust the settings you have chosen before hitting apply. (destructive). Re-opening Develop and applying inverse adjustment is possible, but could lead to loss of details 3. adjustment and filter layers and some other functions are truly non-destructive. 4 Applying adjustment or filters directly is destructive to pixel layers using undo / history does not count for this discussion

If you want to fully eliminate ghosts, you could use HDR merge instead of simple stack. This allows to automatically remove ghosts. Just uncheck the actual HDR merge option. If you need the stacking for other purposes, e..g. smoothing the water or noise reduction, you may process the picture with both methods separately and combine and mask required parts in a later step.

Hi, as a workaround you can use Affinity Photo for editing if you use (free) Canon DPP to convert the CR3 files to Tiff first. This can be done quite easy as batch processing. If you have lots of pictures to edit, this will safe you time as you can apply basic corrections (e.g exposure, lens correction, picture style, white balance) to many pictures in one step. I personally prefer to develop my CR2 files to tiff in Canon DPP as this provides more consistent results.

Hi Chris, this is a good question. Some YouTubers (beside Affinity and James Ritson) covered this question in more detail. There are 3 distinct types of functions: Functions only available in develop persona E.g. Lens corrections like rotation, horizontal / vertical correction, refinement Functions only available in photo persona Most of the good stuff like advanced adjustments, filters, using masks Functions available in both personas like Curves, brightness, exposure My basic advide would be to start in develop persona with "global" corrections (affecting the overall picture, no masks) Exposure, highlights / contrast, Lens correction Detail Refinement and Noise removal And the switch to photo persona. Why? If the exposure is off you need to correct it before any other step, e.g. reducing out-blown highligts, or raising general exposure. Lens correction and Detail refinement is not available in Photo persona, so it should be done here. Detail refinement and noise reduction are kind of opposite, so they need to be adjusted in combination. But all these edits in Develop Persona are destructive. So after having corrected the beasics, it is advisable to switch to photo persona as soon as possible, as this will allow more non-destructive edits and using more advanced / fine-granular edits.

Do you really say 98x98 px? This is next-to-invisible. You Profile snap has a size of >500x500px, so the picture is blown up by factor 5 or more, this allone leads to blurryness. Google search gave me a recommanded sizes of 800x800. Could you try again with 800x800 px?

One important preparation: the pixel layer has to be rotated by a angle before applying equations.

Hi, applying a very simple equation (skew effect) took about 5-10 seconds on iPad Pro 2018. Picture used is 4000x3000 rgb8 jpeg from GoPro 3. This seems quite slow to a live skew effect in transform menu. x=x*2*a+y*b same operation is instant in non-beta.

Hi @DonPenny, if you want a "perfect" mask with only 0 / 1, you have several options which might work better than "classical" masks: 1. Start with a new (empty) mask layer, and paint the mask using a sharp round brush (fill, opacity and hardness all set to 100%, color 100% white ). 2. Use a suitable vector shape with fill color set to white, atop you pixel layer. Then sent it to "mask to below" 3. Use the pen tool to create any shape you need. You may "convert to curves" a vector shape from step 2 for additional fine tuning. Regards, Timo

Hi Mangolife, there are a lot of good guides how to achieve best quality. Your chosen size seems a bit small. Example guide (I'm in no way related to this website, just the first random link i found): https://dustinstout.com/facebook-image-sizes/ Besides choosing the right export format, size & quality settings, check that all objects and text are pixel-aligned in affinity photo, otherwise thei can get blurry.

Same for me, Windows 10, AMD Ryzen 7 1800X, GTX 1080, performance with HW accell is badbeta 1.9.0.876

Hi Timo, see file attached. Just a fill layer to show the masking effect of a vector layer with a gradient. The trick for the gradient is to use opacity instead (or in combination with) with color. You can adjust the gradient at any time. Grüsse aus Hamburg. Untitled_6.afphoto

Hi Timo, i fully share your frustration about masks being second class citizens in Affinity, but dislike your way to express it which feels a bit aggressive to me. i would suggest to use vector shapes or fill layers and gradients when possible. In contrast to masks, they remember gradient settings and can be re-edited. In many cases, they can be used as full replacement for masks. In other cases where you really need to continue with a mask, just copy, and rasterize to mask, and preserve the (unselected) vector as option to re-start later. Second, you could create swatches for gradients, so you can always have a prepared starting point instead green field. After using Affinity for about 2 years i learned to find good workflows where i initially struggled badly with the UI before. At least, Affinity remains constant so you do not need to relearn with every new release. Best, Timo (another one

On the right, you can see and adjust the opacity of the selected layer. On the top left, you can see and adjust the opacity of the selected tool. Yes, it could be confusing. Especially when you have the “color” tool active. Too tempting to get distracted by the wrong opacity field.

It seems you have opacity at 0%. Could you try again after increasing opacity set to 100%?

In my view the blur tool is just fine. In most cases, you could use better techniques like blur filters and masks instead of the more basic blur tool. YouTube will provide tons of good training videos to learn a more elegant workflow. Most important for beginners, and highly decorated veterans switching from Photoshop, as you need to get rid of legacy thinking workflows first. There are a lot of traps a beginner can fall into, like not selecting a pixel layer, using wrong settings, or simply trying to use Photoshop Workflows which will lead to hell in Affinity Photo. If you really want to use the blur tool, you might want to start with “merge visible” first. This will guarantee a fresh and usable pixel layer and thus sorting out a lot of complications.

Can somebody please explain how "fill history" work? Help file did not show any hints.

Hi, when using a mask layer applied to a member layer of a live stack group, the mask behaves strange: observed: If the mask value is 255, the pixel is inlcuded and shown. If the mask value is bewteen 0 to 254, the pixel is not shown. Expected: Values between 0 and 254 should lead to an pixel with accoring alpha. You can try at your own by adjusting the slider of the levels adjustment. You can see that it only affects which pixels are shown (when reaching alpha 255). I might be wrong with my expectation live_stack_and_mask.afphoto

Hi, when i try to move the mask from the red rectangle to the group, the AP beta crashes immediately. live_stack_and_mask.afphoto

Hi there, i tried to use a live stack to merge several images in a HDR style. Images are exposure bracketed. I do not understand how alpha channels work in combination with live stacking: If i apply a mask to one of the pictures, the mask is effective to filter out unwanted parts. But if i try to adjust the alpha channel of the mask, e.g. with help of a levels adjustment, this does not show any effect. Is this inteded behavior, or a bug?

Hi, when i open develop persona rotate by -2,5 degree try to change settings in grid manager or guides manager, all settings of develop persona get lost/resetted in visible picture, while slides stay as before. In this example, picture gets un-rotated (Step 9 in attached file, Windows Step Recorder) Issue occurs with current beta and 1.8.5.X Recording_20201202_2223.mht

AP beta 1.9.0.864 Focus Merge in combination with HW acceleration (Nvidia 1080) still crashes the display, even another tab which was open in AP got corrupted by lots of artefacts. Used preprocessed TIF, not RAW .

Thank you, haakoo. This hint really helps! But you need to stay with the mouse exactly over the decimals field (4.7px below). It would be great to have this option while the mouse is over the slider, or the whole area of this control.

Hi, a would really appreciate if i could see the histogram while adding adjustments and change parameters.. currently 1.8 can show metadata / histogram, but it will always disappear (no matter if pin is on/of) when trying to select / add adjustments.

Please allow to cancel DEV Persona with one click/key (no extra confirmation)

Hi, i would love to have an option to adjust the sensitivity of any slider. E.g. Gaussian blur has a default scala from 0 to 100, in increments of 0.1 pixel For me, it is next to impossible to exactly move between 0,9 1.0 1.1 via mouse, as the mouse jumps nearly almost by 0.3 or more. using keyboard to enter exact numbers is no help, as i want to use mouse to smoothly adjust and see the effect immediately until i got the visibly best result. Adjusting mouse settings to slower is impractical on 5K screen. Tried different settings like dynamic on/off etc. Different options would help: Separate slider for values between 0 and 3.0, Clickable option to limit input range to user selectable range (e.g 1.5 to 2.5), allowing precise and smooth scrolling between these values. Clickable option to limit slider scaling between 90% and 110% of last selected value