Hello,
just a quick question: are there any plans for a rework of the windows version UI of SA?
I currently own latest version, but when I bought it I didn't realize that those nice screeshots I had been looking at were related to the Mac version.
Then when I started the program on my computer I must say I was a bit disappointed.
The looking and the placement of the componets feels off, but It's not just that, it's also how they feel...
I mean, at a first look it seems like a quickly assembled and unoptimized software from the 90's...
On the other hand I know that under this face SA can be very powerful...
The fact is I am a bit worried that being the UI not hardware accellerated then this could slow down the entire program operation?
As of now, when I use SA, I'm usually left with this feeling that I don't know wheter it is running as good as it could run or not...
Or maybe this aspect doesn't affect performance at all?
Anyway, having bought this program 5 months ago and not having seen any updates during this time in this area, I just wanted to draw some attention here and try to understand a bit further what the plans are, if any.
If time has to be dedicated to more important fetures than this then please don't wast too much time replying me. Just a couple of lines would do!
Thanks a lot in advance.
Replies
Ever since we did the mac only to mac-windows cross platform port, we have used the philosophy of having the interface controls and general appearance use the default generic UI the platform provides. That behavior is provided by the cross platform framework we built the interface on top of.
That standard appearance is hardware accelerated by the framework under the hood for the various platforms.
On the mac, if the computer appearance behavior is setup for dark appearance display, then the interface adapts appropriately. I'm assuming the 'dark mode' screen shots you were referring to is associated with this behavior. We're just displaying the base platform appearance behavior.
When we released V5.5, there was no real equivalent display appearance behavior on the windows side. If you manually adjust your windows appearance preference settings to display dark mode interfaces appropriately, then the program will adapt accordingly. I know there are windows users who have setup their machine so studio artist displays in dark appearance mode. I thought we had a post on how to do that, i can try to dig it up if you are interested.
I'll add your vote to the list for providing better dark appearance display options on windows in the next major release.
The plans are definitely there to get retina display support fully working on mac and windows in the next major release.
We had to pull back from that a little bit in the V5.5 release at the last minute due to some problematic inconsistencies on the windows side, so that might be part of what you are referring to.
Our release cycles go though stages. There is a period when we do various release updates, and then at some point that dries up and nothing changes until the next major release. We're in that stage of the cycle right now, so that is why there have not been any point updates in the last 5 months.
I have to be honest, the 'standard' visual appearance and behavior of some standard windows interface components drives me nuts. Clunky user interaction behavior behavior with terrible visual design (the mac is a lot better at this kind of thing). So at some point in the future we may change our philosophy about always using that standard platform appearance and behavior. Switching to full custom display appearance and behavior.
> Our release cycles go though stages. There is a period when we do various release updates, and then at some point that dries up and nothing changes until the next major release. We're in that stage of the cycle right now, so that is why there have not been any point updates in the last 5 months.
Synthetik 7 months have passed since you wrote this, and one full year has passed since I bought this software, and not a single update has been released so far.
When I bought Studio Artist I read that I was entitled to updates for one year, but neither minor or major updates were released during this period of time and if they will be released shortly I will not be able to benefit from them.
I understand that development goes in cycles, but when cycles are so slow that nothing is released for one full year, then I think it's a bit of a stretch to say to your users that they are entitled to "one year of updates".
This slowness also communicates a certain degree of uncertainity about the development of this software... So at this point I really need to ask you: is there a Studio Artist development roadmap we can consult with ETAs for the next minor and major releases?
In case it was more of a coincidence that I bought the software in a long period with no updates, I hope that in this case you will close an eye and be kind enough to at least let me grab the next release...
Looking forward to hearing something from you on this aspects.
Thank you.
As i pointed out when V5.5.5 was released (the 5th update in the V5.5 release cycle that was spread out over 7 months), that would be the last update in the V5.5 release cycle, so there would not be any additional updates until V6 came out.
We're currently working on V6. We had hoped to have it out by now, but it's taking longer than we had originally expected for a number of different reasons. That is certainly not unusual for us, or for the software industry in general.
If we had a subscription model where people were paying by the month, then we'd be doing incremental releases on a regular basis. But that isn't the model we're currently working in. Maybe that is the model you are more used to with other companies? If our goal was just to extract as much money out of people for using our software as we could get then that is probably the model we'd be working in today. But it isn't. The one we currently feel more comfortable with is one where once people buy the software they can continue to use it forever without having to continue paying us money. My take is that people actually appreciate that aspect of how we operate?
But with that kind of model we can't continue to add features for free forever to Studio Artist. Once we go into a freeze period, everything we add after that freeze cutoff is a part of the next release. We make a point of not working in some artificial release schedule where every year on the dot we release a new update, even if it only has a few actual new features in it. I could certainly point out a few companies that work that way, so that they can squeeze more money out of their customers on a regular basis.
This is the way we've done things since the company was started. So it's not like we're doing anything different. We work on it until we're comfortable that we think it is finished as a new version. That process will take as long as it takes.
It is a little frustrating for me to be honest, because i would love to see people using new features sooner rather than later. But it's the tradeoff we have to make if we aren't going to charge people on a monthly subscription basis.
I'm not aware of any official company policy that promises a year of free updates. I am very aware of people mentioning what i just stated above, that a typical release cycle for us involves an initial new release and then some additional update releases associated with it as we fix any issues that are uncovered, and add some additional new things that seem like they would be cool for people to work with. I'm sorry if there was a miscommunication associated with what i just described. You are welcome to email sales to state your case if you feel you were mislead in any way. We really try to be reasonable and honorable with how we treat people.
I feel like we offer a lot of value for the amount of money people pay for the software. Especially when you compare how much Studio Artist costs in contrast to the cost of things like filling up the tank of your car with gas for one example. A commodity that is very quickly gone, while Studio Artist continues to be useful for years and years.
And i also feel like we offer a lot of value when we do new releases. There are always a ridiculous number of new features. And if people don't want to update when we do a new release, they have that option. Of waiting until later, or of skipping a version entirely if they feel it's not offering them any value personally.
So that's all i'm going to say on that topic.
As far as windows ui concerns, yes, i understand where you are coming from. And i'm more than willing to discuss that in whatever level of detail you want to get into. Because of the history of Studio Artist being mac only for the first haf of its existence, and because everyone here tends to use macs for most of our work, windows does end up being a second class citizen to some extent.
As i continue to get more frustrated with apple, maybe that will radically change at some point? The ai infrastructure on windows with access to advanced Nvidia cards and associated software api underpinnings seems much more robust for some of the future things we've been looking into.
I think the Windows Surface devices are really great in some respects, and i'd like to see us take more advantage of what they have to offer. Our existing interface is very oriented towards the big display desktop computer with an external tablet model that it was originally conceived in. The Suface tablet can run software designed for that older model (which was awesome for Microsoft to let happen), but the control layout and how you work with some features could be restructured to be better suited for the tablet interface experience.
Are there specific things on windows you would like to see? Or is it more of a case of just wanting better dark mode and high res display support? How do you personally use Studio Artist? 2D painting, video processing, image effects, abstract art,or? That kind of feedback is always very useful for us.
As a software developer myself I think that if I mentioned the Agile work methodology you wouldn't look at me as if I was speaking Greek, would you? :)
https://en.wikipedia.org/wiki/Agile_software_development
https://en.wikipedia.org/wiki/Release_early,_release_often
The reason this methodology became popular and became adopted by so many companies, together with the "release early release often" philosophy is that it has proven to be better both for the companies (in terms of reduced costs and better organization) and for the customers (which are allowed to provide feedback in real time and guide development towards success).
Releasing often, has nothing to do with being forced to charge users monthly... I could mention countless examples of companies having a fast paced release cycle but only charging a fixed price...
For example around this time last year I also bought NicePage (a web authoring software which is becoming very popular for web frontend development). Well, they only charge me once and the purchase covered one year of updates and their release cycle was impressive, week after week adding new features or fixing bugs and actually responding and adapting to the community.
I understand that your approach can be different and of course I am far from judging it, but please don't make it look like it's a required condition in order to not charge monthly fees. Because it's simply not true...
If you are more comfortable working this way ok, but please take a second to examine your own words:
> As i continue to get more frustrated with apple, maybe that will radically change at some point? The AI infrastructure on windows with access to advanced Nvidia cards and associated software api underpinnings seems much more robust for some of the future things we've been looking into.
Yes, NVIDIA AI support is great on Windows and the AI field it's traveling at the speed of light, new creative tools coming out everyday... yet you guys are sadly held back from all this exactly because of your long development cycle, prohibiting you to jump and ride this amazing train...
I also remember that approximately one year ago, when Disco Diffusion came out, you Synthetik (the man, not the company, sorry I forgot your name) were here on the forum discussing all the great way these new AI algorithms could be integrated in Studio Artist.
Well, one year later, what has become of that? What steps forward have been taken into this direction?
Meanwhile, Disco Diffusion has now been outperformed by Dalle-2 and Stable Diffusion, a Stable Diffusion plugin for the open source Krita software was released, Stable Diffusion v2 was released 2 days ago and is targeting 3D generation, Blender people are rubbing their hands, together with so many other commercial companies and open source projects... Reaserch is advancing like crazy... And... you guys are frustratingly locked away from all this while still at the stage of fantasizing about how great it could be if you only had the chance to switch to a more supported hardware & software ecosystem...
And this brings me back to my main point... You saying that faster release cycles would forcefully involve monthly payments looks just like an excuse that is keeping you from making new and braver choices... Choices which not only would make your customers happier, but could also allow you to live on the very bleeding edge where AI computer vision and digital art are magically converging at, these days...
You know what? The shear truth is that I secretly hope that Studio Artist could become THE TOOL that brings it all together, AI + ART + Human in the loop, the tool that perhaps everyone is unknowingly expecting to be released...
So I wonder... What if, you guys at Synthetik decided that right now is the perfect time to become aware about what Studio Artist could actually become... (if only it started moving at a different pace)?
(I hope I've touched the right strings... All the best ❤)
Yeah, i am familiar with the Agile development ideas. I guess one thing to think through is that what might be going on internally here or with interaction with a select set of alpha testers is different than how we are interacting with the existing customer base that only use the finished release software.
But i do appreciate the points you are bringing up, so i'll give it all some thought. We could certainly do better.
You mention Stability. One of the factors slowing down finishing up the next Studio Artist release is that i'm spending a very large amount of time very focused on the ai things you mentioned. There are a ridiculous number of free things people can use to run the existing Stable Diffusion models today. So i'm not sure what the real benefit is for us to just copy what a literal army of other people are already doing. We'd really like to put together a unique spin on it if possible.
I work with existing Studio Artist and generative ai systems every single day, using the 2 systems working together in the context of some larger scope art project to test out different ideas and approaches, and to make sure i'm on top of everything that is happening in this area. So there is certainly nothing preventing people today from incorporating existing generative ai image synthesis systems into a Studio Artist workflow, and vice versa.
Could we integrate that better, sure. We're working on making that happen, although a lot slower than i would like. Part of that is not enough time in the day, part of that is waiting for certain api issues we'd like to take advantage of to solidify. We could certainly lament that i'm not a superhuman, but i'll avoid that whole discussion. It will be nice when ai code systems get developed enough to change the whole approach one could take to build creative tools like Studio Artist.
I could also point out that if you are really following the whole generative scene, then you would realize that the specific model architecture Stable Diffusion uses is already out of date (although most people are probably not aware of that yet). There are unified approaches to generative latent diffusion that are really better approaches in many ways.
So yes, the whole generative ai scene is exponentially exploding. Everything 3 months old that people are so fixated on today is already out of date.
Maybe I'm one of the few users who visit the forum here frequently to hear new news about the new version(Sometimes several times a day), but I consider it fair to say that I'm not in front of Adobe company. I think with myself Here we have a small company with limited resources but with a creative developer who has been trying to keep the app fresh and alive for years.
My second point is that maybe the waterfall method is more useful for the development of SA:
https://www.seguetech.com/waterfall-vs-agile-methodology/
My third point as a suggestion is that maybe newer technologies can be added to the program faster like plugins:
flying dog Plugin
https://www.flyingdog.de/sd/
thanks
and I wanted to say please see this link for some Ideas:
https://proximacentaurib.notion.site/proximacentaurib/parrot-zone-7...