AI generative Fill in Affinity

[R C-R]

43 minutes ago, CM0 said:

We don't know yet from where they will source the funds.

i think that the chances Adobe will just absorb the costs for this is approximately zero.

[nezumi]

8 hours ago, CM0 said:

Adobe can track it because it is their own data, their own model and you are using their platform to create the images. They will likely count a generation as a use.

Since I completely switched to Affinity (and I am not planning to go back in the near future) I dont know how it works. I thought generation is done locally on your machine but for what you say it would suggest its done on Adobe servers? Its always online thing?

 

[DigitalVisuals]

7 minutes ago, nezumi said:

Since I completely switched to Affinity (and I am not planning to go back in the near future) I dont know how it works. I thought generation is done locally on your machine but for what you say it would suggest its done on Adobe servers? Its always online thing?

 

Its not done local, there is need a very big database with several models. For Generation you need strong and fast servers.
Midjourney works with Discord where the Bot is working. 
But we will see what the future will bring.
 

[lphilpot]

FWIW I have no interest in AI at all from a large scale generative viewpoint. But small scale inpaint / fill / retouch / editing? Absolutely. And sharpening / denoise? Absolutely, 10,000%. I don't want to add significant visual elements with AI, but having some intelligence when 'restoring' previously obscured textures, structures and such would be not only a huge time-saver but a definite improvement over trying to manually regenerate and match them in terms of the end result. Beyond that? I've seen nothing so far that interests me.

But if Serif are truly not interested at all in exploring AI, then IMO that's to their (and our, as users) detriment. Kinda reminds me of back in the '90s when Bill Gates initially thought the Internet was overrated. But one forum remark from one person is hardly an official corporate stance, IMO.

[carl123]

5 minutes ago, lphilpot said:

But one forum remark from one person is hardly an official corporate stance, IMO.

Serif won't tell you (or the competition) what they are working on.

Look at "Studio Link", no one saw that coming.

[R C-R]

2 hours ago, lphilpot said:

But if Serif are truly not interested at all in exploring AI, then IMO that's to their (and our, as users) detriment.

All they have said is the developers are not planning on adding any AI features at present.

[albertkinng]

In my experience, I find that I don't require additional features or AI implementations. Instead, I use third-party apps that assist me in accomplishing my design goals. There's no need to criticize Serif for not having the specific features you desire. Personally, I employ four different apps for color management. Ideally, one color management tool would suffice for all designer needs. However, since that isn't the case, I purchased the necessary apps to effectively complete my projects.

[lphilpot]

1 hour ago, R C-R said:

All they have said is the developers are not planning on adding any AI features at present.

Hence my statement, "But one forum remark from one person is hardly an official corporate stance, IMO."

[R C-R]

11 minutes ago, lphilpot said:

Hence my statement, "But one forum remark from one person is hardly an official corporate stance, IMO."

I'm not sure what point you are trying to make. It doesn't say anything one way or another about their interest in exploring AI.

[lphilpot]

That point wasn't about AI itself, but Serif's stance on it. I pointed out it's not worth trying to draw long-reaching conclusions based on one person's forum post. IMO YMMV whatever

[nezumi]

I feel like we are standing over the dead body of the horse that used to be that conversation screaming GO BOY!, GO! We said everything there was - couple times around.

Not trying to shut this conversation, but is there any other angle we can explore here? Personally I am way more interested in whats the plan for 2.2, what WILL be there. Rather then keep on discussing usefulness of features we know that will not be there anytime soon.

Side note - it always baffles me how quickly people find absolutely necessary to use new stuff. Stuff they didnt knew is coming just few months ago. Some new feature is teased and immediately somebody screams "I NEED THAT!". Really? You NEED something you didnt even knew is coming? Your project - that you are making now - depends on feature that few days ago wasnt existing? Thats some amazing planning man. Good thing that feature showed up - you wouldnt be able to finish your project otherwise.

And then you see the project, and you know it would take 10 minutes in Photoshop 4... 🤣

[Latens]

Not going to add my voice here whether I'm for or against.
But to put things in perspective:

 

[CM0]

26 minutes ago, nezumi said:

Side note - it always baffles me how quickly people find absolutely necessary to use new stuff. Stuff they didnt knew is coming just few months ago. Some new feature is teased and immediately somebody screams "I NEED THAT!". Really? You NEED something you didnt even knew is coming?

Not sure why that is surprising. You see demonstration of feature that makes easier work you do everyday. Wouldn't it be more illogical to state I don't need that?

Sometimes a new feature also can make possible something that you had never considered; however, you can immediately imagine how you would put that to use. If a new feature didn't elicit that kind of response in either scenario, then it is probably feature that shouldn't be prioritized.

[nezumi]

The problem with these I have is that these are not the places and not the people from real world. What is the point of picture if then you change everything about it? Everything now will be fake? "Look, me on the top of the Eiffel tower, on the bicycle, wearing Armani suit, drinking coffee!"? Or "check out photo of my grandma - she looked nothing like this but AI improved the quality and blue eyes are looking better on her anyways! And here me with her on one photo - isnt that amazing? She died before I was born!".

I dont know - technically its amazing stuff. You can make a redneck use suit. Very cool for some advertisement stuff. On the other hand now you dont need any skill to confuse the hell out of everybody. Which means all the stupid people have now tool to express their stupid ideas with ease. Before some effort was required which was surely great filter for lazy and unskilled people to polute internet with yet another meme. Now floodgate is open. I doubt that is good in any shape or form. But hey - "progress"...

I guess what I am trying to say - its another razor given to bunch of monkeys. Man will shave with it, but what monkeys will do - I guess we will see on social media soon...

@CM0 Theres significant difference between "I WANT" and "I NEED". Between "I can now do it faster" and "I wouldn not be able to do it before".

Although when I think about it, in many cases AI will made things for people who otherwise wouldn't be able to do at all... So yeah, this might be true. They are still unable to do it themselves but can get it done for them with a single click.

[TariqMK]

8 hours ago, carl123 said:

Serif won't tell you (or the competition) what they are working on.

Look at "Studio Link", no one saw that coming.

This is a scenario where silence will hurt their business. Not the same.

This is less about 'new features' than it is 'catch up'.

[R C-R]

Just to stir the pot a bit....

https://www.safe.ai/statement-on-ai-risk#open-letter

[nezumi]

We have software that can predict next word from database and other that can generate pictures. I think danger of extinction because of it is pretty slim

I was raised during cold war. They need more to scare me then picture spitting software or chat.

[Twolane]

What was another topic that went on and on for years? Oh, right. Linux something something. How much easier Linux running on a PC is to operate, or some such nonsense. I see they're still trying to claim how easy it is for Linux to run Affinity products. Yes. Of course it is. Grandma and Grandpa have all bought Linux-capable machines and are running all three programs like it was a fresh, invigorating spring breeze. Bwaaaahahahaha.

But I digress. It looks like we have another/more thread(s) with the same idea, this time with AI something something.

2.1 shut down many of the complainers and the complaints. New complaints/complainers have come along, and here we all go again. Good luck, people. Whine away something something.

[kaffeeundsalz]

21 hours ago, nezumi said:

What is the point of picture if then you change everything about it? Everything now will be fake? "Look, me on the top of the Eiffel tower, on the bicycle, wearing Armani suit, drinking coffee!"? Or "check out photo of my grandma - she looked nothing like this but AI improved the quality and blue eyes are looking better on her anyways! And here me with her on one photo - isnt that amazing? She died before I was born!".

All of what you describe here is perfectly possible to accomplish without any use of AI image technology, it's just harder. But I think you know that. I agree with you that AI will make heavy image manipulation more accessible to anyone and that this is not necessarily a good thing. But please let's not forget that Serif explicitly targets professional users with their Affinity Suite. The praise of AI comes from both worlds, but for different reasons: Non-professionals may see their capabilities extend, and in part dramatically. Professionals on the other hand will be able to do much more in much less time, and this totally makes sense from a business perspective. The less time you spend on an editing job, the more time you have to do another one.

Now, I don't say that this is the solution to all our problems, and I also don't say that it's not a good idea to be skeptical about AI. But I do understand why people from many different backgrounds say that they "need" this – be it to be able to do more (which, by the way, even applies to many professionals) or to save time. It's just that when you think about it, saying they "want" this might often be the more honest approach.

[CM0]

14 hours ago, R C-R said:

Just to stir the pot a bit....

https://www.safe.ai/statement-on-ai-risk#open-letter

Yes, these are the topics I currently write about. FYI, just published something new today. This is more light hearted humor from a counter perspective. 

https://dakara.substack.com/p/researchers-discover-reality

Something more in depth and serious if you are interested.

https://dakara.substack.com/p/ai-and-the-end-to-all-things

FYI, all image composites, edits and illustrations done with Affinity for all articles :-)

[Celia at Grafted Works]

All I'm saying is that it's lucky that Photo 2.0 came out before Photoshop revealed their AI generative fill.

[albertkinng]

20 hours ago, Return said:

But to put things in perspective:

Essentially, if people start believing that computers can accomplish everything by simply typing in commands, our livelihood will become much more challenging. Although this perception doesn't reflect reality, it's the only thing that matters to clients in our industry. As a friend of mine noted, we'll need to work harder, not less, due to the emergence of artificial intelligence. Essentially, we'll need to spend more time and effort to surpass what others can achieve with a single click. This will make it difficult to earn a living since we'll be faced with two challenges: first, clients may demand lower prices since they assume we're using AI, and second, we won't be able to charge more for using it.

Prove me wrong.

[CM0]

3 minutes ago, albertkinng said:

we'll need to work harder, not less,

It is interesting to note that this has been the trend essentially with all technological advancements. The promise is we will work less, we can relax and take more time off. However, it has been precisely the opposite that has happened. As you state, it increases competition as a side effect. There have been posts on AI forums about those deeply involved already burning out from the exhaustion of trying to keep up.

[albertkinng]

24 minutes ago, CM0 said:

deeply involved already burning out

Let's not overlook the fact that they're not compensated for their hard work. As a freelancer for over 25 years, I still find myself staying up late and spending an excessive amount of time working on challenging projects for clients, despite having access to incredible tools. On occasions, clients will visit websites like lexica.art and send me examples of what they're seeking to accomplish, leaving me to match that level of quality over several days for pennies. Meanwhile, they assume that I'm merely leveraging AI technology more effectively.

[Granddaddy]

1 hour ago, CM0 said:

Yes, these are the topics I currently write about. FYI, just published something new today. This is more light hearted humor from a counter perspective. 

https://dakara.substack.com/p/researchers-discover-reality

Immediately your article reminded me of Isaac Asimov's 1958 story "The Feeling of Power" in which one man discovers graphitics. He learns to calculate with pencil and paper. Doing arithmetic without computers obviously has military significance:  it makes manned missiles possible.
https://archive.org/details/1958-02_IF/page/n5/mode/2up?view=theater

Even older is E. M. Forster's 1909 tale about the rediscovery of outdoor reality beyond one's room when "The Machine Stops."
https://archive.org/details/e.-m.-forster-the-machine-stops_202008

I hasten to add that while I purchased my copy of "If" from a newsstand in 1958, Forster's story was published before even I was born. I know of it only from an anthology I bought many decades ago.