atefoto

It never was enabled.
 

for a short while i was using a AMD WX7100 eGPU  but stopped as it was running way too slow (i now understand why..).  changed the base preference in AP to only use the internal GPU that comes standard / build in.   two weeks ran fine, until the past few days.   if i use a plugin ''crash' a simple live filter 'crash'  a small file of a few layers 'crash'. - on and on it goes....i'm at a loss as to what is happening.    I consider myself a rather advanced user of the Affinity apps,  and i feel i can confidently state something is whacked on the software side right now,     wondering was there an update?   

Also for me,  the last two days has been a joke!   it does not matter what function i'm using it will stall and enter 'AP is not responding' - leaving me to 'force quit' i'm on MAC OS 
I've deleted the application, reinstalled and still issues. 
I'm on a spec'd out MAC so can't be down to the computer.
I'm a heavy user of this app, never had issues like this before, what the hell is going on?  i have work to do and need a functioning application ! 
Thanks Rick

Hello DEVs
 
First off, thanks again for great software, I've been sending ~250 portraits through it since I bought it, and I love how the application just simply replaced Photoshop

Over to my "needs".. I really need some features from PS days, and I hope you guys see why.
"Apply Image" in Affinity Photo is not alike PS CC's version of it. I used to use this often to grab an RGB, R, G or B. CMYK or C, M, Y, K or even LAB L, A or B from a layer, a document or a file on the disk, then "apply" it as a mask - inverted or not - on to the selected mask, usually in normal or luminosity mode. I even use it for blending K channel from a dublicate CMYK document in to actual documents RGB composite to enhance faint details which is almost lost. I also use it to mix one of the RGB channels from same document in to the worst channel to enhance the details and remove noise (usually B channel), its called luminosity blending, check out Lee Varis and Dan Margulis photoshop videos. I don't see this is possible in Affinity Photo with the current "apply image", and I wish I had the same blending options in it like default layers so it will be easy to fix a "bad" channel in a image Check video below in this post.
 
I also wish I could set one of the info readouts to a default readout, LAB for example, because I use this more often than RGB or anything else. Left side could be document format. I can change to LAB right now, but it doesnt stick to next image.

 
I love to be able to save my own defaults than white and gra. Every single time I use this I have to go and manually set left one to black and right to white anyway because I rarly use this on other things than masks. It is really annoying

 
After pushing through those 250 portraits since I bought Affinity, I'm really getting tired to recheck "current and below", I never retouch on background or main pixel layers and wouldn't suggest anyone to do it either. Make a new layer then use the healing brush with "current and below" is the safest way to use it anyway, so could it be possible to have persistent settings on these as-well?

 

Check next 5 minutes to see what I mean of luminosity blending:





Thats all for now
 
Best regards,
Aleksander

12 Core AMD Threadripper 1920x with 4.1 GHz in every core. 32 GB RAM 3200 MHz. nVidia 1080 Ti with 11 GB Vram. 
System disk: 1 TB NVMe
Working disk: 512 GB SSD in raid
Affinity with one Window open is more choppy than trying to game while rendering 4k video in DaVinci Resolve, go figure. 
I also have very little GPU and CPU usage by Affinity, like it almost not using the available resources

I forgot to mention, yes, this is on Windows 10 Professional, clean install 1 week ago. Slow before, slow after.

Affinity Designer: Moving a 400x400px image around is choppy, yet 10% CPU utilization and 25% GPU and no disk read/write. Notice the bad utilization.
This happens in all Affinity programs. Not in Adobe programs, not in CaptureOne, not Luminar, On1, etc.
 

Metal / DX12 / Vulkan all offer similar feature sets and performance, though each have their own individual issues...
All I can say is, we're working on it. It is not our intention to have one platform with vastly better performance than another.

@Mark Ingram
Have you guys considered Vulkan? Unlike Cuda it's available on all gpus and I think it offers comparable if not better performance than Metal. 
Don't get us wrong Mark, we appreciate all the hard work you and your team put into these programs but this latest feature fragmentation made us feel little left behind and many would for sure feel better if you guys shared some plans for bringing windows suit of apps on pair with their mac counterparts. 

If you can't go lower than those settings, you're doing it wrong!

Have you tried a 1x1 px canvas yet?

Sadly not enough for 400x400px image. You should get an iPod or airpods, anything by Apple for your Affinity software. 

I'm on PC, just like atefoto.

Ok, we get that Mac is the primary audience for Affinity, and that your development efforts SHOULD probably focus more on Mac...

However, making Windows the 'issue' of GPU acceleration support is dishonest or amazing ignorant.

Windows NT, since the WDM/WDDM technologies introduced in Vista and newer offer more GPU/Driver abstraction through several frameworks than any other OS technology. The irony of Apple's Metal, is they created it to be MORE LIKE the video technologies in Windows, that offer high performance with better hardware abstraction than OpenGL and OpenCL were capable of providing. 

Windows not only has frameworks like DirectX, but the entire Video stack is managed by the NT kernel, with GPU preemptive multitasking, GPU virtualization, GPU SMP technologies - technologies all controlled and managed by the OS. And since these are managed by Windows, it is a lot easier for developers, as software requires no knowledge, additional coding, or understanding of the technologies to gain use of them. 

Example - even the current versions of Affinity Photo/Designer are getting GPU acceleration features from Windows, unless the Serif developers are purposely circumventing the technology. The easy example is showing a thumbnail of a JPEG or decoding a JPEG using standard Windows APIs are GPU accelerated, and this is pure GP-GPU - and has nothing to do with compositing or rendering using the GPU.

In other words, Affinity should not ever be touching or calling GPU or GPU driver specific code or features on Windows. Not only is it not necessary, it breaks the abstraction and GPU agnostic model of Windows, making the development process harder with buggier implementations. 

This is also why the response about Windows and GPU diversity and GPU driver issues are what held back adding GPU acceleration sounds so crazy.


A software product like Affinity can implement features without regard to the hardware on Windows, an additionally the extra work that Apple's metal requires like managing VRAM and each GPU is completely automatic in Windows.

Even implementing multiple GPU support, like Affinity implements on the Mac, is no more than a using a single GPU context on Windows, and then Windows manages the GPUs, their specific hardware features/differences, threading between all available GPUs in the system, and even handles VRAM between the GPUs and the system by visualizing it to the software. (Affinity Photo wouldn't even have to worry if there was VRAM or enough VRAM or what GPU it was allocated on, as Windows automatically fills in any assumptions/agnostic gaps.)

Simply, these things are abstracted from software through the frameworks, drivers and platform technologies in Windows. They are NOT abstracted on OS X, even with Metal. 

Windows uses a very specific set of agnostic threading technologies for GPU acceleration, which is why GPGPU code has and is scattered throughout Windows itself and the major new frameworks that run on Windows.  If GPU diversity or GPU drivers were an issue, there wouldn't be so man GP-GPU and GPU acceleration features scattered throughout Windows 10, as everything would break.

GPU usage is not hard or fragile in Window.

The side note to this, is that the reason Metal exists, is because when Apple pushed all in on OpenCL - the GPU cooperative multitasking in OS X did not allow widespread use of OpenCL, as OS X and software would collide, limiting the use of OpenCL on OS X beyond specialized software.   Ironically, even with Metal, the same GPU yield/cooperative multi-tasking still exists, which still limits widespread CONCURRENT use of the technology. 

If what I am writing is 'really' new to Serif developers, I could provide a few code samples, that implement similar GPU shader and acceleration features like Affinity on OS X uses in less than 20 lines of code that can run in a UWP App on a Windows PC with any GPU, and it will also work on Windows PCs without a GPU. This same code will run on a Windows PC with 5 different brand/model GPUs, spanning several generations of GPU technologies, and utilize them all equally, without adding or modifying the code. 

I can also provide a similar bit of code in C or C++ to demonstrate how this works at a lower level with DirectX on Windows.

This isn't rocket science, there is no need to use CUDA or OpenCL on Windows.  With Windows and DirectX, you can even use new ML/AI features or full on DXR (ray tracing) features if you want a really fast denoiser for the Development persona.



In conclusion... Windows makes the GPU brand/mode/driver agnostic to software, so blaming the GPU/Driver variation/complexity in the PC world is just NOT TRUE.  (The only time the GPU matters is if the developer CHOOSES to use a GPU specific technology like CUDA, and even then, creating an agnostic fallback is really simple.)

So... 
We get that iPad and Macs are the primary market for Affinity, but don't create conspiratorial arguments or place the blame on Windows for the differences between Windows/Mac.

If Serif wants and needs to put the work into Metal and Macs, that is fine, just say so.

 

I wouldn't be this harsh then, but I'm not sure if it is worthy the comparison of what they show in the Keynote. The keynote demo was too amazing for me to go back and open this Windows app, kinda frustrating when you regularly work with very big images or many layers.

I'll give them a clap for making RAW edit and HDR editor better, a lot more fluid, but not nearly fluid compared to anything else.

Okay then, I'll stop sobbing about Winblows PCs and their issues with MAC Native apps.

Good luck and we cheer for you DEVs.
I'm out. Peace.
Aleksander

Related to OpenCL, the last I heard was, that in the future, OpenCL should be part of Vulkan as OpenCL-V in order to gain even wider support. Recently, OpenCL Next was announced for 2019 with some new information (see for example here).

Sure, but there are still valid options on the table..

On the other hand, what are Affinity suggesting we users do then? Drop our $2-4000 Windows PCs and buy a $5000 MAC because there are no other rendering engine that Affinity want to use? I'd like that, but I ain't got any $5000 for a MAC.. It's kinda cheaper to just use another program then, yet I stick with this syrup to save some bucks from subscriptions.

I don't think OpenCL is going anywhere soon. Too many applications depending on it, and I think even Skylum use OpenCL in their new apps as-well, but thats my 5 cents.
I hope you manage to find a solution because it is not a good experience so far. I really mean it when I say rendering a 4k video and playing Battlefield 4 is a more fluid experience than having only Affinity Photo open and a 400px image..
 
Best Regards 
Aleksander

 

As Mark said, they're still "researching it".

One more "small" thing"

I made an RGB selection of the whole composite, then I thought "Hmm, I need to add more contrast in to the mask to make gray almost black"...

Then I tried adding a curve to the mask just like I've been doing for 15 years in PS CC, to adjust the mask, but I don't think it is possible in Affinity Photo? Am I right?

In thats the case, I request this function too

In Photoshop I was using CTRL+M or CMD+M on layer masks to edit the mask itself, it was extremely useful to make better selections and controlled adjustments.
 
Best regards,
Aleksander

Thanks for your data, I'll set aside 1 hour on Friday to make same measurements - as long as no clients "need" more images.
 
Best regards,
Aleksander

+1

Hello DEVs
 
First off, thanks again for great software, I've been sending ~250 portraits through it since I bought it, and I love how the application just simply replaced Photoshop

Over to my "needs".. I really need some features from PS days, and I hope you guys see why.
"Apply Image" in Affinity Photo is not alike PS CC's version of it. I used to use this often to grab an RGB, R, G or B. CMYK or C, M, Y, K or even LAB L, A or B from a layer, a document or a file on the disk, then "apply" it as a mask - inverted or not - on to the selected mask, usually in normal or luminosity mode. I even use it for blending K channel from a dublicate CMYK document in to actual documents RGB composite to enhance faint details which is almost lost. I also use it to mix one of the RGB channels from same document in to the worst channel to enhance the details and remove noise (usually B channel), its called luminosity blending, check out Lee Varis and Dan Margulis photoshop videos. I don't see this is possible in Affinity Photo with the current "apply image", and I wish I had the same blending options in it like default layers so it will be easy to fix a "bad" channel in a image Check video below in this post.
 
I also wish I could set one of the info readouts to a default readout, LAB for example, because I use this more often than RGB or anything else. Left side could be document format. I can change to LAB right now, but it doesnt stick to next image.

 
I love to be able to save my own defaults than white and gra. Every single time I use this I have to go and manually set left one to black and right to white anyway because I rarly use this on other things than masks. It is really annoying

 
After pushing through those 250 portraits since I bought Affinity, I'm really getting tired to recheck "current and below", I never retouch on background or main pixel layers and wouldn't suggest anyone to do it either. Make a new layer then use the healing brush with "current and below" is the safest way to use it anyway, so could it be possible to have persistent settings on these as-well?

 

Check next 5 minutes to see what I mean of luminosity blending:





Thats all for now
 
Best regards,
Aleksander

Thanks for the sample files. I've spent a good amount of time running some tests and this is what I came up with:
New Stack with 2 images - 47 seconds New Stack with 9 images - 03:55 3. New Stack with 17 images -  07:45 So pretty much 40-47 seconds per image
New Panorama with 2 images - 51 seconds / Rendering took 43 seconds New Panorama with 9 images - 05:02 / Rendering took 47 seconds  New Panorama with 17 images - Inconclusive. My entire PC after 30 minutes was unresponsive and I had to do a hard reboot.  Tone Mapping Persona with these images always took over a minute but as I work with the software daily, this behaviour is what I would describe as typical but that does not mean I don't think it couldn't be improved. 

I downloaded the jpegs, made 3 sets of 5 images. Each set was put through HDR. This gave 3 images
as output. The 3 HDR output images was treated in panorama mode to get the single final image.
No other editing was done except slight exposure correction by curves at the right top quadrant area for clouds.
The HDR processed within 3 minutes (per set of 5 jpegs).
The second image is a quick hand manual exposure blending of just two images.

I can confirm that sometimes when using sliders the whole UI becomes unresponsive (to the point that standard window decorations start to show). It usually happens not long after launch and gets better over time.

Hello guys,
 
Thanks for developing this nice application.
 
I have some issues merging stacked images from "new stack" option. I try merge 20  images and it doesn't start doing anything even after 15 minutes, but the RAM usage is over 30 GB, all JPGs only. Force quitting program from task manager is the only way to solve it. The stacking in any of the menu options takes too long in stable or BETA, almost 5 minute just for a simple stack process.
I also have issues with Panorama and HDR stacking like in previous stable edition. Aligning HDR brackets will sometime lock to skies and move the landscape, you know how that story ends. The HDR noise reduction removes important details and replace them with big artifacts or just massive blotches of black/nothing.

The "tonemapping" function - or your RAW converter - have a long way to go. I get the cringe every time I need to go in there, and the results I get can easily get better surpassed by cellphone apps. It feels like I'm pushing a JPG every time I'm in that converter and not a RAW file, the details instantly clip to black using "black point" setting, but you did some magic with the latest "highlight" and "shadow" slider, it feels better. Worst part is the crazy lagg I get from pushing any of the sliders, mouse and system freeze for seconds before slider jumps to "next" spot, even freeze my Spotify playing. This is a 12 core Threadripper 4 GHz with 32 GB RAM, 1080 Ti and NVMe SSDs, so it should be able to handle a basic RAW converter, I got CaptureOne Pro v11 and no problem there with a lot more tools and stuff.

I don't find any way to tune the Panorama merge function like in PtGui, I wish it had more options like that program. I don't know if it is possible to tweek the perspective after the merge, but there should be more features while having the image as layers, thats when you got the most power to fix things.

There is one function I miss from PS CC and I hope you can add similar function, because this is invaluable when using a Wacom tablet like I do. Just FYI, I replace mouse buttons on my pen, so this works great for me:
In brush tool (PS CC), hold (Left Alt), Hold (Right mouse click), move mouse up and down to change size of any brush tool. Move mouse side to side to change the hardness.

Don't take the critic too hard, I know you're not Adobe and I probably won't get any of the things I'm asking for, but I hope you guys spend more time on the features above. The stacking features and RAW converter/tonemapping should really be top priority for you to fix. I just left Adobe but I feel forced to go buy PtGui because this stacking you got going on never works for me.
 
 
Best regards
Aleksander

Yeah, when using PS CC I used ALT+Right mouse click on my Wacom's thumb button, so I could easily change the hardness and size of the brush. It is very useful when working with dodge and burn on portraits and such - even in general use.

I know I can bind brush size to my Wacom buttons, because I do this now in Affinity, but it is not as fluid and intuitive as using the PS CC method.