NotMyFault

Check the export settings. Is quality 100% or 1%? One is 6 MB (the brighter one), one only 64 KB large (the darker one), and has much smaller pixel size. If you want to compare quality, ensure that both files have exact same pixel size, and viewed at 100%. Which do you rate as „better“?

This doesn’t make practical sense. Shooting images with Cameras has multiple sources of blurriness (lens, motion, sensor noise, de-bayer, …). There is no single pixel resolution, and what have you won by minimizing de-Bayer blurriness without tackling other sources? Either use a Sigma Sensor (excellent at low ISO, but falls apart at higher ISO 1600), or a monochrome Camera and use colored filters to manually combine 3 exposures into color image. Sigma says you roughly get 3x resolution from their sensors (who omit Bayer pattern) https://en.wikipedia.org/wiki/Foveon_X3_sensor

Maybe step 1 is causing the trouble. if you paste something and get parent / child layers, the child gets clipped to the parent. and convert text to curves creates multiple objects from one layer, this can break those parent/child relations. a group of curves (converted from letters) behaves different vs. a text frame with multiple letters Can you upload the actual file ? if possible, remove everything from the file except the layers required to demonstrate the issue.

Any color. It’s a bug used as workaround to solve another bug, without causing any unwanted side effects.

If you want to disable dithering for exported documents, you need to use a trick as workaround: add color overlay fx, end set opacity to 0. on every object where you applied a gradient.

Well, technically shapes have an orientation, even if this is not documented, and you can change the orientation by mirroring the shape. And shapes technically have a start / end point. Add arrowheads to spot where it is (this gets really funny when using handcrafted vector brushes with a vertical structure e.g. multiple colored lines in one brush) This functionality is too minor to waste a feature request. Hoped that someone had found an ingenious trick.

Please try to upload if possible. Otherwise, crop to a small section with relevant colors, flatten, and save as again to a new file of far smaller size.

Same question to you: do you use Serif or Apple RAW (develop assistant)? Apple RAW automatically does lens correction (independent from settings in Photo), so lens correction must be disabled in that case. when I remember it right, this might have changed in V2.2 and automatically gets deactivated and blocked. When using Serif RAW, you can choose if it is on or off.

Hi, i want to use the equivalent of „reverse curve“, but on shapes without „covert to curve“. It seems impossible. The only way to reverse the curve (direction of brush) is to mirror it along one axis, but this will change the shape (for unsymmetrical shapes). The screenshot shows iPad, same challenge on Desktop. Reverse Curve for Shapes.afdesign

I forgot to mention one step: use „save as“ to create another copy. This should be much smaller (about 130MB), 16 bytes per pixel.

This should not happen. Please check „32-bit preview“. Can you provide a screenshot?

Hi, as no one answered I try to step in. A few common causes: color profile issues. color format issues. Exporting directly from RGB/32 to jpeg. RGB/32 can contain lightness above 1.0 and uses linear Gamma. The „downgrading“ to jpeg can lead to clipping in lightness and color gamut. we would need more information about the actual file. Upload it if possible. To reduce file size, save a copy, then flatten, delete all snapshots, and save without history. Provide screenshots of export settings used to create jpeg. which Mac HW, OS version, and Display do you use? in general, try to make a copy of the file, convert to RGB/16, export again and inspect.

Hi, which RAW engine do you use, Apple RAW or Serif RAW?

I don’t understand why you keep raising „new“ reports covering the same topic. The behavior of the color picker for curves is consistent across LAB, RGB, GREY, CMYK. The same for V1 and V2. The same for Photo, Designer, Publisher. It is not rated as bug by Affinity, just works as designed.

There are different ways to do this: a. Bitmap fill Group all your layers containing the clipping objects on the group, use the fill tool. set mode to bitmap fill, and choose the image you want to use as fill b. Blend mode erase Group all your layers containing the clipping objects on the group, set blend mode to „erase“ nest the group to the layer containing the foreground d. Rasterize to mask Group all your layers containing the clipping objects Rasterize to mask apply this mask to the image to get it masked

You images look great, so no urgency to check this. It is more a geek thing for those who want to experiment and optimize every little bit. I think the jpeg files could benefit from some of the sharpening methods I mentioned earlier: try adding an unsharp mask filter, with radius in range of 1 to 2 px. This should be applied selectively to the main object (the birds, not the sky) using a suitable mask try adding a highpass filter, with blend mode soft light or linear light. Keep radius small, 1-2. finally, create a pixel layer by merge visible. Then use tone map persona. By default the effect uses unsuitable slider settings. Reduce the strength from 100 to 0, so the image look’s unmodified. Then gradually change local contrast, usually 10-20%. This really boosts small details and color contrast.

I tried to find this essential function (even available in Word, powerPoint etc since ages. Workaround: using selections (the only place where you find feather etc) Selection from layer shrink selection by 1/2 the intended feather width feather the selection by intended feather width create mask from selection.

I tested with the files you provided. There is absolutely no difference to a simple black and white adjustment. So no good reason to fiddle with channels panel which is a bit clumsy to use. never the less, you need to have a pixel layer. To get one, use merge visible. Then channels panel allows to right-click on a channel called „pixel cyan“, and you can „delete“ the channel (fill with black). the Same can be achieved non-destructively with a channels adjustment, where you choose the cyan channel, at set output contribution to 0. you may merge visible, but this is superfluous if you export to a raster format.

Where do you set the demarcation? For me, IS is always a combination of HW (sensors, motors, tilting devices, or newer piezo-electric actuators without classical mechanics), combined with software. I don’t know pure mechanical IS (except like fluid-damped video tripods). PS has some pure sw based PS (Affinity lacks).

All (except some new ones, see below) Canon and Sigma recommend it in the user manuals, and my tests confirm this. You can see the difference with a 600m lens, e.g. moon shots. The IS is software controlled. The sw assumes a handheld lens, and tries to proactively correct the movements. Humans induce a certain range of frequencies. On a tripod, you get a totally different set of frequencies and amplitudes, which irritate the IS, and produce more jitter. When IS is activated, it continuously corrects, it never says „ah you holding it steady, i stop the motors“. Similar to continuous autofocus. Another reason is battery drain. Only most modern IS detect if a lens/camera is tripod mounted and automatically deactivate themselves.

Nice colorful birds👍🏽

Sorry, these videos look a bit confusing to me. I don’t get the process, I see the steps, but it seems overly complicated to use channels. a simple black and white adjustment layer should do the trick. Or even simpler, why not just skip the B/W conversion. simply adding a fill layer choose your target color, and set blend mode to hue or color?

In affinity, you can use a channels adjustment layer and adjust the settings to remove all cyan. Assuming you are working in CMYK document format. It depends how grey (graphite) areas are encoded, as pure black (K only) or rich black (CMY or CMYK). Pure black will be fine, rich black will do some collateral damage to the graphite colored areas.

Hi, can you please share example file, showing the different states of edit, e.g. original showing multiple colors result after removing blue pencil re-color to cyan There are many ways to achieve what you are asking for, but it depends on several factors which woks best (.e.g. areas where blue and grey are both visible. Do you use only hard pencil lines, or shades?

I get the same visual result when exporting with 1% image quality. But cannot reproduce the results using save as on iPad currently.