James Ritson

Hey SFCRen, You don't typically export "out" to a RAW format - it's raw sensor information that needs to go through debayering/demosaicing before it can be displayed as meaningful information. When you import RAW files into Photo, they go through this process and become raster images, which you would then save/export as a document or image format (.afphoto, PSD, TIFF, JPG etc). The end of your post regarding availability suggests you've seen an export to raw option in other software, is this the case? What may be happening is that the RAW file is copied (in its original form) along with a sidecar (usually .xmp) file containing changes. This is how DAM software like Lightroom works, where the changes you've made are applied dynamically. In other words, the original RAW file isn't edited or re-exported - it's developed to a temporary working copy until you save or discard it. Hope that helps!

Oh, I completely forgot to mention in the recent new video post, but the Brushes video has also been re-recorded; I've always cringed at the original video with just the raw brush strokes on a white background, so I knuckled down and made it a bit more contextual for image editing: Brushes

Tomall, I replied to your thread in the Bugs forum but I'll put this here as well in case you see it - I've tested my mk2 RAW files on both macOS and Windows and they both open fine. However, I think you're using Apple Core Image RAW as your RAW engine rather than the default SerifLabs - there is currently no E-M1 mk2 support through Apple's RAW engine and it will crash Photo when trying to open them. If you switch back to SerifLabs you should find your images will open fine.

Hello again, just a heads up that we're now focusing on YouTube delivery as well as Vimeo - videos will be available on both platforms going forward. Here's a link to the YouTube channel: https://affin.co/photoyt All videos to date are available so far, and we'll keep parity between the two from now. Thanks!

Hey Rigger, the sources panel is only populated for HDR merges - there's no such behaviour for the Panorama stitching. Currently, merging the images separately and exporting them as a 32-bit format to stitch them is the only way to achieve HDR panoramas. That being said, I believe (although do not take this as gospel) there might be a plan for a future version to identify and merge bracketed shots when stitching panoramas, which may help achieve what you're after. Whether the process could include a populated sources panel, I'm not sure - it would certainly be handy.

Good day all! Got three new videos for you - two are creative and look at long exposure photography, and one looks at using Photo's undo history: Big Stopper Effect Light Painting Blending Undo History Hope you find them useful! Cheers, James

Hi Gary, I picked up the Mk2 too just after Christmas and found that Photo will also open RAW images shot with the High Res Mode - they open to 81MP (10400x7792px) without any issues. How are you finding the Mk2 so far? I previously had the original E-M1 and found it a bit clunky, so tended to use my Panasonic GH4 more (which I had primarily bought for video). No such complaints with the new version though, it feels so much slicker to use. Sensor noise seems to be improved as well, it's much easier to get cleaner results. Anyway, thanks for your comments - I'm also glad that the camera is supported so early on!

Hi Phil, the resampling methods are ordered by sharpness. For photographic purposes you can safely ignore Nearest Neighbour. Bilinear will produce the softest result, whereas Lanczos 3 non-separable will produce the sharpest (the difference between separable and non-separable is very slight). Bicubic is somewhere in the middle and will probably suffice for the majority of your images. One reason to use a softer resampling method might be if your image has lots of high frequency content - from excessive noise to very fine detail like chainlink fences and difficult patterning in architecture. Using a sharper resampling method may exacerbate these parts of the image, and may even produce resampling artefacts (though they are more noticeable in video content to be honest - stills, not so much). As always, experiment! But for a good safe bet with most of your images, try Bicubic. Hope that helps!

Hey animationhalo, I'm able to address some of your points: 1) Are you on Windows? If so, there is an issue with some actions not being recordable (whereas they are on the Mac version). The developers are aware of this and it will hopefully be addressed soon. 2) There is an offset filter, it's called Affine (see Filters>Distort). 3) There are multiple tools - there's a Perspective Tool and Mesh Warp Tool (the perspective tool can do single and dual planes) - you can find them on the tools panel on the left (above the Zoom Tool). There's also a live version of the Perspective Tool so you can use it non-destructively. 4) Are you able to elaborate when you say not useful? Are you using the Move Tool? Thanks, James

The above link refers to focus merging/stacking - I think ve2cjw is instead referring to live stacking using operators. ve2cjw - you can use File>New Stack and add your images. Photo will then align and place them into a live stack group. You can change the operator; you'll probably want Maximum (which does the same as setting Lighten to all images if you had them individually as layers). See the Maximum Stacking video for a bit more information: Once you've created the live stack, you can also ungroup the images using Arrange>Ungroup, then shift-click to select them all and change their blend mode. You should find Lighten achieves the same result as the Maximum operator, but you can experiment with other modes like Average. I'm getting round to doing a "big stopper" effect tutorial and a light painting blending tutorial soon, which will use similar techniques to what you're after here. Hope that helps!

Hey, thanks for the suggestion - you are correct, the shadows and highlights adjustments tend to compress tones rather than boost/recover them. In the meantime, if you're interested I can suggest a couple of alternative approaches: 1) Use the Shadows / Highlights filter rather than the adjustment (in the Photo persona). See the Shadows/Highlights video for more info: 2) Remove the default tone curve and bring the highlights slider all the way to the left - this will recover a huge amount of highlight detail and allow you to craft your own tone curve. There's a video called Custom Tone Curve that covers this: Hope that helps!

Hi Kim, I shoot with the same lens and I haven't had any issues (the video tutorial content was shot with the 60mm Macro too). Photo should compensate for the focus breathing when aligning the images - you'll end up with blurred borders but you just crop those out of the final image. Can you give more detail about what's going wrong?

Hi Tomall, that's strange as I have the same camera and I'm opening RAW files just fine (including the 80MP high res ones). Are you on Mac or Windows?

Hey foxdavis, yes, you're more or less along the right lines - open your 360 image, use Layer > Live Projection > Equirectangular Projection. Pan around to an area you want to edit, select your tools and do your edits. Once you're finished, remove the projection via Layer > Live Projection > Remove Projection. Then you just need to export the original unmapped image for use in an external 360 viewer. I would mainly recommend using the Inpainting Brush and Clone Brush for retouching seams and blurred areas. If you're having trouble getting the tones to match and it's not looking seamless then maybe try the Healing Brush, as that will blend the source and target tones. Check out these videos as they will detail using live projection, making multiple edits, merging multiple layers down and exporting: 360 Live Editing 360 Advanced Editing 360 Retouching Also, as you mentioned real estate, check out the video on making multiple copies and repositioning each one: 360 Multiple Views You're in good hands as some of the tutorial content above was shot on a Ricoh Theta S. There's a bit of quality sacrifice for the convenience of a single lens, one-click system, but it's not too bad and should deliver acceptable results. Hope that helps!

Hey Guyon, the easiest approach is to use Merge Visible from the Layer menu with both frequency layers shown: this will create a new composite layer. You can then hide and keep the frequency layers in case you want to go back and re-edit them later. Hope that helps!

Yes, it seems to open them fine, at least the ones created from an E-M1 mark 2 (only just got it the other day) - confirmed that they open on both Windows and Mac versions. They're 80MP as opposed to the 50MP JPEGs that are also created - very happy with the quality. I can't wait to try it out with some landscape/architecture shots when the days start to get a bit longer again!

To add to MEB's response, the developers are aware of this issue. There is a big disparity at the moment: for example, I have some high res 80MP shots taken with the High Res Shot mode on the new Olympus E-M1 Mark ii. These develop and load in less than 8 seconds on my Windows box - macOS actually takes longer! Whereas on the other end of the spectrum, users are finding that 16MP raw images can take noticeably longer to load. There is a new customer beta available (version 51) - please could you try this version and see if you note any improvement? Link is here: https://forum.affinity.serif.com/index.php?/topic/33310-affinity-photo-customer-beta-15151/

Michail, raster information in .afphoto documents is stored completely uncompressed. If you're working from raw files, those are developed to 16-bit. 16-bit uncompressed raster information is quite large! If you don't need the extra precision, you can try converting to 8-bit via Document>Colour Format before saving your document. If you've already saved it, try Save As to create a fresh copy with no redundant data. Additionally, when you develop or begin work on an image, an initial Snapshot is created so you can always revert to the original image. This takes up space too, so you may want to delete it. If you go to View>Studio>Snapshots to display the panel, you can then click on the Background snapshot and delete it (the bin icon). Hope that helps.

If you're saving into the original directory you'll have to click Authorise beforehand (see the top right) - this is due to sandboxing in macOS. Hope that helps!

Stuart, are you previewing at 100% or larger when using these filters? If you're at a lower zoom level, Photo may be using a mipmapped (lower resolution) version of the image for performance reasons - therefore the preview will look different from the final result that is applied to the full resolution image. There shouldn't be a huge difference, however. Could you perhaps try the same filters at 1:1 zoom and see if this solves the issue? (And if not, please let us know!) Thanks, James

Yes, pixel and image layers will be rasterised for export so any cropping will become destructive. If at any point you want a crop to become destructive anyway (useful for inpainting so it won't sample areas you've just cropped), right click the image/pixel layer and choose Rasterise. Hope that helps!

Hi Mike, these are greyed out because you're editing in 32-bit float colour format (known as unbounded). Pretty much all of the filters were adapted to work in float (as opposed to integer), but these filters don't make sense mathematically. Somewhere along the line either you've switched colour formats whilst editing in the Photo persona, or you've got the raw Develop output set to 32-bit instead of 16-bit. If it's the first, you can simply go to the top menu and choose Document>Colour Format>RGB 16-bit. If it's the latter and you don't want this occurring in future developed raw files, you can change the output format through the Develop Assistant. This video on 32-bit Raw Development should help you with that (just reverse the step to go from 32-bit to 16-bit): https://vimeo.com/192599184 Hope that helps!

Hey Fred, You can optionally pre-process raw files and save them as JPEGs or TIFFs. This gives you two advantages: 1) Faster processing of images since Photo doesn't have to decode each raw before merging/stacking. 2) External handling of raw processing if you prefer the results that other software offers (for example, Photo doesn't have camera picture profile support at the moment; emulations of in-camera profiles like Vivid, Portrait, Natural, etc). You don't have to pre-process though. Just start a new HDR merge and add your raw files. Noise reduction is already checked, which will denoise the merged result. You may want to experiment with the strength though, as I've found the default value of 40 can be a little aggressive. 10-20 seems to work better for most low ISO HDR work. Hope that helps!

Hi Fotis, chromatic noise and luminance noise is present in all raw images. However, most raw processing software will perform some degree of noise reduction by default; chrominance noise is usually removed automatically because it looks ugly and appears to degrade the image quality. Previously, Photo didn't do any automatic noise reduction, leading many users to believe its raw handling was sub-par and that it actually "introduced" noise to raw images. This wasn't the case; rather, Photo simply exposed how raw images can look before any filtering is applied to clean them up. The 1.5 update now analyses the image and performs (by default) colour noise reduction to give the user a better starting result when working with raw images. The patterning of the noise is likely due to the demosaicing of Affinity's raw engine. There's always room for improvement here (I believe improved demosaicing may be planned for a future update), but the automatic noise reduction should remove the patterning for a cleaner image. I've looked at several different raw engines and they all appear to take slightly different approaches. I believe one technique is to actually smooth the image of both chrominance and luminance noise, then dither it with some artificial, smoother luminance noise (usually Gaussian) to retain some texture. You can reproduce this in Photo: try a base setting of 25% colour noise reduction and 50% luminance noise reduction, then check the Add Noise filter and try a value of between 4-10%. You'll end up with a texture that looks more like grain than digital sensor noise. Hope that helps!

Are you using a high DPI screen? (Or do you have Windows scaling set above 100%?) If you zoom to 100% (try Ctrl+1) in Photo then zoom to 100% in your other apps like Irfanview is the scale the same?