bmobius

Dear Serif/Affinity leadership

Shame on you. ALL of you. There has never been a single acquisition of this kind that has benefitted it's users in total. EVER.

Every single creative app that has been scooped up by any kind of greedy, subscription-based, Orwellian corporation has been turned into utter trash. Eventually the Affinity apps will too. Nebulous legally gray features, bloated processes, restrictions on innovation, and so on will reduce your product offerings to mere shells of what they are currently. Simply look at what's happened to the Adobe apps over time to see what I mean.

Or Autodesk apps.

Or Maxon apps.

I loved the fact that Serif was going toe-to-toe with Adobe on their own terms. You all made a space for yourself in a market that is egregiously spiteful towards most artists and design professionals. The sheer fact that so many of us flocked to the Affinity apps, despite the lack of parity of features with similar Adobe products, should be proof enough that there is a deep deep need for professional creative applications that could be easily afforded that didn't rely on invasive, predatory practices like subscriptions.

Sadly now the writing is all over the wall, brightly colored and garish. You didn't give us a definite statement on the subject of perpetual licenses vs. subscriptions. You're making no efforts to ensure that your user base will be protected against the very same, evil practices that Adobe thrives on. You've given all of us the middle finger while licking your lips in greed at the money you've made from this deal. At this point, why didn't you just offer yourselves up to Adobe and let them gut you?

This is just too damn sad

Hi,
I second this feature request and think it also would be good, if transformation tools would be aware if a layer is just a "regular" layer, or a layer containing a normal map.
As soon as you start to rotate a normal map, the stored vector information is not correct anymore. Flipping along an axis can be done, when swapping the rgb channels accordingly afterwards. But rotating values different then 90° or 180° needs to be calculated and there are actually just a few specialized tools out there that do this job.
So painting, creating and being able to rotate normal maps - that would be an awesome feature!

Just like Photoshop so that you can use it for mutliple things. I would use it for Game Development, it would be good if you could draw Normal Maps but also just generate them.

I have an opposite experience : pngs  being converted to tga  have never leaved  any color bleeding because png do not multiply alpha  except for 100 transparent pixels which they make 100% black.   That could be easily workaround with a special layer having 1% of opacity in AP.   That 1% wouldn't allow  black background in pngs and this 1% is  not visible in a game  because of dxt compression which would make it zero anyway.   At least in our game engine that also accepts tga only as textures.
Tga  seems always per-multiply alpha and DO make color bleeding so you have to use something like Solidify plugin  for Photoshop and work with color and  alpha separately.  Imo it makes more troubles than png converting which usually just a click in Xn converter or such.    
It would be nice to have TGA support but I  personally would continue to use png + converter  until AP would have Solidify plugin working  at least.
In general I would say I have higher priority things that make me still paying for Photoshop subscription.
1. alpha stacks using whatever blending math you need
2. "Layer comps"  which allow to switch , pack in channels ( with 3d party script) and save automatically  color/normal/roughness/metal/ao/height/ etc  version of same document using layercomps names as file names ( by script)
3.  Chain links in-between layers that allow to link transformation in between layers from different groups independently . An absolutely necessary thing to do non-destructive something  and AP seems just one step behind .    Wish they would make a checkbox to link same color coded layers  temporary  like  "lock children" .
4. Export color/normal/roughness/metal/ao/height/  packed in different RGBA channels from the same slice
6. Height to normal,  normal to curvature, normal to distortion vector, height to edge detection  and a few other  procedural effects for procedural filter
IMO  those things are much more lacking   for a game artist, not tga at all
 
ps. Right now I have to use Substance Designer where its all and much more  possible  but gosh it's so huge pain in a..., so terribly inconvenient in its every UI part and tool/node while being node based.
Wish a kind of new generation of image editors , vector or pixel based could help to get rid of SD.    Too bad neither Photoshop nor AP could do so.
 

PNG compression is lossless so if some kind of "color bleed" is happening that is a flaw in the software being used, not in the file format.  PNG is much more complex to implement than the relatively simple TGA format, so if someone is coding a custom engine and does not want to use an external library or take the time to write code to support the format, supporting TGA makes much more sense logistically than PNG... though if time is taken to implement the desired support, PNG is technically the superior format.

Honestly I don't even understand what do you mean exactly.  Composite alpha?  Game textures use one single alpha anyway  and a practice to paint something into alpha by hand directly is ill one IMO.  
You could do a separate group of layers with it's own blending math, adjustment layers, vector shapes and hand painting  and then use procedural filter to turn the whole group into composite  alpha values  or any other  of  RGBA channels,  switching it off and on to preview when necessary while restricting alpha input for other layers.     And pack anything to any of RGBA  within the main layer stack in general.   
Same way you could in Photoshop where painting in alpha or even layer masks directly is absolutely not necessary too.   An idea of regular layer masks in Photoshop is kind of outdated technique. Why using totally destructive layer masks while you can use non-destructive group clipping instead with it's own mask stacks with smart objects etc. Same in AP and embedded docs in mask stacks
So I kind of not exactly sure where is the issue. 
Then you could export as png  or tiff and use 3d party converter .   If I am not wrong Batch Photo  one for example could even do it automatically without you doing a click by looking for a new files to convert in certain input folder.  
 I agree, would be nice to have it in AP indeed without any 3d party solution but I am lacking layercomps  and chain links from Photoshop much more.   A necessity to keep embedded doc somewhere in the main layer stack  to keep it accessible and editable after cloning into another layers mask stack is also  annoying   imo 
  So to some extent I agree AP is not ready for gamedev/textures yet , just not because of tga export or alpha channel editing.   To be honest Photoshop is not very well suited for gamedev too so I believe AP could do better actually.
For example they could implement a kind of edge wear mask system  based on depth/heigh and normal input imported from Zbrush within "procedural filter". Like     Substance Designer does.   The math in use is simple like 2x2 and in fact you could do same in any image editor   but with proper live filter interface it could be so much more convenient.
Each time I do something in Substance Designer I feel myself like a robot drawing something with a huge long robotic arm holding a brush .  AP seems so close and needs so few minor touches to be able to do all the same  but with much more natural artistic feel of an instrument and a canvas behind your fingerprints.

@Andy Somerfield,

The main reasons TGA is used over PNG in game development are that TGA is a very simple format, which makes it quick to implement, and related, TGA image data can be stored as raw image data or lossless compression as required. The most significant difference is TGA stores data as "dumb" RGBA channels, whereas PNG stores this as RGB and Transparency, for most cases this is fine, however the alpha channel in games is often used to store non-colour data, where each channel represents something entirely different. The functional difference here is 0% opacity gets exported with static colour data (either white or black) not the 'real' colour of the image, I'm not sure if this is actually defined in the spec of png or it is just the way it is implemented, however. Premultiplied alpha would also be a problem if this was actually exported this way, however as stated earlier this is not in the spec for PNG and it doesn't appear to be used in AP either.

This is also the reason that alpha channel editing is requested (and very useful, there are workarounds for that, however using adjustments).

Also, while I understand your reason for wanting to limit the exporters, it seems a little shortsighted to say "ask the other end to support png" when we are not talking about a single application not supporting it but many.

Overall I like AP a lot and there are many features that I use that photoshop simply cannot comprehend (inverting a 32bit/channel image for example), however I still cannot use it for many purposes as it is limited to the idea that images are designed to be viewed as-is which in most of my use cases is not true.

Edit: Also related, DDS support is also requested for the games industry, as it is widely used due to it being able to be decompressed on GPUs, and allows for mipmaps to be directly included (and edited, in special circumstances).

likewise
 
I have...  in the specifications.  PNG is lossless and is defined to support a non-premultiplied alpha channel when transparency is provided.  If you are suffering from unintended "color bleed" that is either due to an error in how you are using the file or a bug in the implementation.
 
PNG and TGA files are alike in that they do not support multiple layers.  The image will obviously need to be flattened to be saved in those formats.  How were you expecting any differently?
 
Here we go again...  the official PNG specification is here: http://www.libpng.org/pub/png/spec/iso/index-object.html#6AlphaRepresentation
 
To quote from the specs:
 
 
Bingo.  I agree that it should be supported, but let's try to get the facts straight on PNG as well.

IMO  Substance Designer is just one more image editor.  They have a nice baker inside still missing a key feature of being able to bake from UDIM textured hi-res objects.    And no it's not a shader editor, maybe partly with its pixel processor but in general  it's  just a node based image editor.    
  Problem is   while it's super flexible master of all trades, a kind of a swiss knife it's basically same inconvenient as a swiss knife usually is.    I appreciate its node based flexibility but  I prefer a simple hammer ,  convenience over flexibility. They are not the same things. 
All its basic node/tools are monstrously, unbelievably inconvenient starting from transform gizmo, svg vector tools  or curve editor.   I've been asking for scaling around a transform center for years.   HLS node is almost unusable and you really start to appreciate AP "picker"  in curve editor once you try it in SD.  It lacks so many all those tiny conveniences we all have accustomed in something like Photoshop years ago already.     You have to spend months to make more or less usable tools for yourself there and with a next update half of them are getting glitchy.
It's beside the idea of doing everything procedural itself .   Which is crazy thing IMO.     Once you try to use it as a mixer with hi-res photogrammetry scans  it turns slow as hell.
I am looking forward to their new Alchemist  but from what I saw  so far I am not very impressed either.    It's just unnecessary extra buildup over same Substance Designer approaches.        They just don't understand what artist friendly things are.  I bet it's because they mostly listen what various technical artists tell them . That folks usually have codding background experience not an art one. 
I believe we need just a few "specialized" things in AP  and it could be actually better than what Allegorithmic is doing. For art guys at least.  
 
ps.     I'v been using SD for years already and half of a time you have to care about documenting what you are doing there. Make comments, write special on screen remarks  because if you don't  the document turns into undecipherable Gordian knot . A week later you couldn't understand a thing in your own files among all those "exposure parameter" and functions.    A totally crazy thing either about their UI
It's so much a relief to be able to simply  pick things of screen in AP or switch some layer on and off quickly to recall what it's doing.  Weirdly impossible thing in SD  while still perfectly possible  in Blender node system for example.( M for bypass a node)   Also a thing I requested for years, still "impossible"
 
 

Painting in alphas or even layer masks is not bad  nor good,  it's just a bit outdated approach    
People have been whining about Photoshop layer masks not being live, auto-updating and non-destructive for years.   Now after introducing  clipping groups  couple years ago  (since CS6) you could use smart objects as masks  stacked in it's own mask stack order and blending math   providing basically same flexibility as node based image editor.         I don't use regular layer masks in my psd files at all anymore.       
You could do same in AP,  well not same but close.
In a word you can paint to alpha whatever you want. You just don't need to do it directly in channel dialog  and could use regular layer stack for this in much more flexible and non destructive way.   Procedural filter could make any layer group to be Alpha  or any other of  RGB , swap in between  channels, etc.     Just look at it closely .
In my opinion AP needs to overcame still existing gap to Photoshop regarding non-destructive, 100% adjustable work style absolutely necessary for texture art  first  and then make all that tiny extra adornment  like tga etc.

Thanks guys,
An interesting discussion - kirk23 - you make a good point about PNG optimising away colours when alpha is zero. This is a strong enough argument to justify the inclusion of TGA as an exportable format - so we will implement it.
A
 
.eb: Do not make ad hominem attacks on other users - it makes you look like a child and if you continue I will remove you from the forum.

THIS! All of this! Even if you only consider .TGA a legacy format, there is still a sizeable number of artists working with this venerable format. It would be a great addition to have .TGA export included either natively OR as a plugin. I too am also a huge fan of .DDS and the way it handles mipmaps and I would be delighted as can be to see that implemented as well.

Getting back to the point, Affinity Photo is SO CLOSE to being a perfect replacement (GENERALLY SPEAKING) for Photoshop, at least from a game development / game artist standpoint. Other talented artists that I currently have the privilege of working with are eagerly watching to see how AP continues to develop. Many of them have their hands on their wallets, ready to scoop up a copy just for their own use.

Lastly, as AP can effectively import .TGA, how  hard would it be to also allow for export? I encourage Serif to give this format some serious consideration and move to implement as soon as possible in one of their upcoming updates!

I didn't want to push the envelope on feature requests but I too agree, DDS would also be awesome to support but if I had to choose I'd still prefer TGA was added into Affinity Photo first.
If its not too much trouble all Affinity team needs to do is implement an exporter for TGA and its the end of the story really. Other than the fighting going on between a few members in this thread (which I hope affinity team does not write this thread off due to that) I hope the points made by several members of why we want TGA support will suffice as good reason to have it added.
If Affinity Team does not want TGA as a core exporter because it does not fit their vision maybe have it as a plugin. This way users could choose to install and anyone who sees TGA as 'bloat' won't see it unless they use the add-on. I'm pretty much looking for any alternative or compromise we can get to have TGA a reality in Affinity Photo.

The .TGA format is still used widely in the game development industry. This is especially true when regarding older engines still in use by a number of game dev studios. I get that folks using products such as Affinity Photo are using it for, well, photo manipulation. If that's all Serif wants Affinity Photo to be, then it's understandable that implementing .tga export and better alpha channel control isn't necessary.

But then again, wouldn't Serif love to have additional customers, in game development, attempting to move away from Photoshop find a home with Affinity Photo? Wouldn't that also, may, just perhaps convince them to make use of the other Affinity apps such as Designer (for creating UIs and such)?

Most of our management team and product research team at our studio do not relish the the monthly outlay of money to maintain multiple Photoshop CC seats. Affinity offers a superior product at a vastly superior pricing model. Those of us that have adopted Affinity Photo out of our own pockets, who also see the need for .TGA and better alpha channel control wouldn't be asking if .png and .tiff did what we needed them to do. This is a matter of not only workflow speed but also the need for legacy file format interaction.

Lastly, and this is important, most of the other products used with legacy formats such as .TGA also do have .PNG. and sometimes .TIFF import. But when you state that we should hit up a number of multiple vendors and tell them that their apps all need better PNG and TIFF support, it kinda comes off as dismissive. Those of us, in a number of other threads in the Affinity forums no less, who use Affinity Photo and Affinity Designer are asking that our image editing app of choice (read: Affinity) export .TGA so we can use them with a host of other apps.

Allow me to furnish an theoretical situation. Say we have 12 other apps that use the .TGA file format. Management doesn't wish to spend money purchasing licenses for a number of other apps for updated or "superior" file formats. Doing such would seriously impacted limited funds. Along comes an image editing app that represents as a real alternative to a subscription based software. Management thus concludes that purchasing 15 seats / 15 licenses of Affinity Photo and Designer would be far more cost effective in the long term. The Affinity apps need just two things to be viable for that. those two things are .TGA and alpha channel manipulation.

The funny thing about the above theoretical situation is that this actually happened. 

So look at it this way. It's a win for Serif to be flexible with their exportable file formats. Serif gets even more customers moving away for your competitor and those same customers get a more feature complete application for use in their professional field.

I love Affinity Photo. I will keep using it every single day. Just please take a moment and ask the Affinity development team to put this request into the Affinity suite.

The .TGA format is still used widely in the game development industry. This is especially true when regarding older engines still in use by a number of game dev studios. I get that folks using products such as Affinity Photo are using it for, well, photo manipulation. If that's all Serif wants Affinity Photo to be, then it's understandable that implementing .tga export and better alpha channel control isn't necessary.

But then again, wouldn't Serif love to have additional customers, in game development, attempting to move away from Photoshop find a home with Affinity Photo? Wouldn't that also, may, just perhaps convince them to make use of the other Affinity apps such as Designer (for creating UIs and such)?

Most of our management team and product research team at our studio do not relish the the monthly outlay of money to maintain multiple Photoshop CC seats. Affinity offers a superior product at a vastly superior pricing model. Those of us that have adopted Affinity Photo out of our own pockets, who also see the need for .TGA and better alpha channel control wouldn't be asking if .png and .tiff did what we needed them to do. This is a matter of not only workflow speed but also the need for legacy file format interaction.

Lastly, and this is important, most of the other products used with legacy formats such as .TGA also do have .PNG. and sometimes .TIFF import. But when you state that we should hit up a number of multiple vendors and tell them that their apps all need better PNG and TIFF support, it kinda comes off as dismissive. Those of us, in a number of other threads in the Affinity forums no less, who use Affinity Photo and Affinity Designer are asking that our image editing app of choice (read: Affinity) export .TGA so we can use them with a host of other apps.

Allow me to furnish an theoretical situation. Say we have 12 other apps that use the .TGA file format. Management doesn't wish to spend money purchasing licenses for a number of other apps for updated or "superior" file formats. Doing such would seriously impacted limited funds. Along comes an image editing app that represents as a real alternative to a subscription based software. Management thus concludes that purchasing 15 seats / 15 licenses of Affinity Photo and Designer would be far more cost effective in the long term. The Affinity apps need just two things to be viable for that. those two things are .TGA and alpha channel manipulation.

The funny thing about the above theoretical situation is that this actually happened. 

So look at it this way. It's a win for Serif to be flexible with their exportable file formats. Serif gets even more customers moving away for your competitor and those same customers get a more feature complete application for use in their professional field.

I love Affinity Photo. I will keep using it every single day. Just please take a moment and ask the Affinity development team to put this request into the Affinity suite.

The .TGA format is still used widely in the game development industry. This is especially true when regarding older engines still in use by a number of game dev studios. I get that folks using products such as Affinity Photo are using it for, well, photo manipulation. If that's all Serif wants Affinity Photo to be, then it's understandable that implementing .tga export and better alpha channel control isn't necessary.

But then again, wouldn't Serif love to have additional customers, in game development, attempting to move away from Photoshop find a home with Affinity Photo? Wouldn't that also, may, just perhaps convince them to make use of the other Affinity apps such as Designer (for creating UIs and such)?

Most of our management team and product research team at our studio do not relish the the monthly outlay of money to maintain multiple Photoshop CC seats. Affinity offers a superior product at a vastly superior pricing model. Those of us that have adopted Affinity Photo out of our own pockets, who also see the need for .TGA and better alpha channel control wouldn't be asking if .png and .tiff did what we needed them to do. This is a matter of not only workflow speed but also the need for legacy file format interaction.

Lastly, and this is important, most of the other products used with legacy formats such as .TGA also do have .PNG. and sometimes .TIFF import. But when you state that we should hit up a number of multiple vendors and tell them that their apps all need better PNG and TIFF support, it kinda comes off as dismissive. Those of us, in a number of other threads in the Affinity forums no less, who use Affinity Photo and Affinity Designer are asking that our image editing app of choice (read: Affinity) export .TGA so we can use them with a host of other apps.

Allow me to furnish an theoretical situation. Say we have 12 other apps that use the .TGA file format. Management doesn't wish to spend money purchasing licenses for a number of other apps for updated or "superior" file formats. Doing such would seriously impacted limited funds. Along comes an image editing app that represents as a real alternative to a subscription based software. Management thus concludes that purchasing 15 seats / 15 licenses of Affinity Photo and Designer would be far more cost effective in the long term. The Affinity apps need just two things to be viable for that. those two things are .TGA and alpha channel manipulation.

The funny thing about the above theoretical situation is that this actually happened. 

So look at it this way. It's a win for Serif to be flexible with their exportable file formats. Serif gets even more customers moving away for your competitor and those same customers get a more feature complete application for use in their professional field.

I love Affinity Photo. I will keep using it every single day. Just please take a moment and ask the Affinity development team to put this request into the Affinity suite.

I too would love to see TGA support come to Affinity Photo. The program is really great and my #1 feature after 1.7 would be to have TGA export support still. As others have mentioned I primarily use Affinity products for game development/creation and sometimes I need the direct TGA output. I am fine using 3rd party PNG to TGA converters but I do hope one day I won't need to anymore (for sake of workflow and speed) and Affinity can have the built in export function with alpha and 24/32 bit options.
 
 

One more voice for Targa export (with alpha channel).

There's a few topics on this aready and they said no to TARGA support being in 1.7 but now 1.7 is out I'd also like an update on this.

The more supported file formats for exporting the merrier, I say!

I'm also a game dev needing TGA support with alpha for masking & hueing my 3D objects, I would like to add my voice as well!

Not at all thanks for confirming this

As game artist myself I would say an ability to save tga in export persona would be nice indeed  but until Solidify filter( making edge padding within transparent pixels) for Photoshop would be working in AP ,   tga would still be pretty useless.     
I would save png anyway which are doing padding automatically and  then convert them into tga in Xnconvert or something.   So AP needs that free "Solidify" be compatible first or have its own working.
So ideally we need not simply  tga export   but an option to make color padding or rather un-premultiplied    tga export.
Surprisingly   a decade ago Photoshop 7 had such option for tgas for some time  but then "fixed" it because many users considered it a bug.
 
As of  alpha channel manipulation AP already have everything you need actually.  "Procedural texture" filter could  make what exact channel each and every layer or layer group would be adding to.    Just push "+" button there and it makes default zero expression  and appoint that zero to every channel you don't need to be visible.     That way you could pack any layer  to a channel it should be influencing only.  Not only alpha  just any of RGBA.  
  You still have to make grayscale values transparent before through "blend option"  curve editing dialog  since AP couldn't interpret gray-scale intensity as alpha directly
 It's all a bit too complicated  but instead of making "composite alpha" editable  I would rather want they would support alpha stacks using all available blending methods/math not only multiplying   and  "Layer Compositions"   from Photoshop so we could set and switch in between color/roughness/alpha/ normal/height  representation of a document and export them in export persona as same sized slices.

+1.
Working in the video game and movie industry mainly as environment artist . 
Most of my textures need to be exported as .tga as it is still used as main texture output in a couple of engines .
I really want Adobe out of my system as fast as possible , this would be a huge win for affinity to have more export options .