วีดีโอ

When you pick certain Speed options or start with the different batch files or do something like in painting, it downloads the needed files at that time. However once those files are downloaded for what you want it to do, you should be able to just copy the folder and move it to another PC

@@holidayeffects3933 you mean I have to first run and download all the speed and other features that I will use in the 1st computer and then I can copy it to my other pc right? Will it save my time for that 18 gigs?

@@arxkhan6280 Yes that is correct. Whatever feature is working is already downloaded and should work fine when copied

It is open source & free under the GNU Public License

This is Windows only, however this is just a front end. The software it uses behind the scenes is actually all open source and runs on Linux. Check out this video on using the command line version of the tools this program uses. I haven't tried in in Colab but you should be able to create a Colab script by using the steps in the video. Use the Linux commands in the video description (it has Windows and Linux versions) th-cam.com/video/QdYdq3xO7-k/w-d-xo.html&lc=Ugx8-V-wEQA8Z61E_GR4AaABAg

Sorry for the slow reply - I am not sure but you could try turning off the "Reprocess Finished Files" to see if that would skip the frames already done. I am not sure about that, just a guess.

Make sure you have enough space on the C: drive - it downloads many large files

Yes just click the "Input image" checkbox in the lower left corner and then go to the Image Prompt tab to upload up to 4 reference images

@@holidayeffects3933 I thought that is what the checkbox was for. Just wanted to make sure. Getting into coding after working in the data center as a data center tech since 08. Working my way through CCNA, but that's really just to get the cert. Figured I'd jump into a project that would force me to learn code. So I am working on fine tuning a StyleGAN. In the meantime though, I would love to also learn these platforms to keep me motivated. Thanks for the quick response!

I am not familiar with RTX SuperRes since I don't run a card that supports it. However it may well be that the program downloads drivers during it's check period so I would go online, do the Test Compatibility and then you should be able to go offline. However this may be particular feature may have some sort of NVidia cloud thing going on, so can't say 100%. Generally everything runs offline for the other engines, however

​@@holidayeffects3933thanks for the prompt response. Will reinstall it while being connected to the internet.

wish i can help u im still on GTX XD

I suspect your laptop's graphics doesn't support a modern version of Vulkan (which a lot of these AI upscalers use if you don't have an NVidia graphics card). Try updating your graphics driver and give it another go.

@@holidayeffects3933 Thank you. Unfortunately it seems that my graphics card doesn't support it. The solution would be to use the program hosted on google colab and I found some videos talking about it

When downloading from a site like CivitAI, make sure the Lora you are downloading is for base model "SDXL 1.0" and the type is "Lora" and not a checkpoint, vae, etc. Place the downloaded file in to the "models/lora" pre-made subdirectory under the Fooocus directory. Finally check for keywords in the description or sidebar. Many loras won't activate without the keyword. You can see all of that in the sidebar and even filter for just SDXL in Civitai. Then in Fooocus, click off everything under "Style" tab and then in the Model tab select the lora and set the strength typically to somewhere between .7 and 1 and make sure to include at least one keyword in the prompt

Hard to say exactly but it looks like a Vulcan driver issue. See if you can update your graphics drivers and try it again

Yes, the green screen was just to make the animation more reusable so you could use it in multiple videos. Nothing wrong with placing the background in instead, however

I just downloaded the latest version and ran it past Microsoft Defender and Kaspersky Threat Intelligence Portal and both showed no issues. Also this is an open source project and the author has over 1300 patreons who pay to get the advanced version. Do you have basis in fact for your opinion?

Fooocus has added instructions for AMD GPUs and Macs. Scroll down the GitHub page to the section that applies to you like “Windows (AMD GPUs)” for instructions in how to set it up for non Nvidia GPUS

You can use this method for Win. Linux & Mac here: th-cam.com/video/QdYdq3xO7-k/w-d-xo.htmlsi=LxIwx81rTiSyzq1J or Win & Linux here: th-cam.com/video/UF05AbZq-sw/w-d-xo.htmlsi=f2Dah-s6AlBBVBTq or Mac specific th-cam.com/video/juuNQju0xv4/w-d-xo.htmlsi=HemGxI_Mp_Psr7vc

Calm down 😭

You can use out this all GUI method for Mac here: th-cam.com/video/QdYdq3xO7-k/w-d-xo.htmlsi=LxIwx81rTiSyzq1J

This video uses the free & open source version that run on your PC. Of course they would like you to buy their cloud version that uses their servers, but that is totally optional. I have never paid a dime for AI upscaling personally

That PC is running a old Nvidia 1070

THANK YOU SO MUCHHHHH! Bless your heart! I was facing the same error :)

This too also worked for me - unlike the command in the description. Thank you

LOL Totally agree, unfortunately there is a loose wire somewhere between my brain and my mouth that shorts out occasionally

The AI video upscaling part is free, I have never had to use any of the non free features for basic upscaling. Also this is just a front end to various open source projects. You can do the upscaling yourself here: th-cam.com/video/UF05AbZq-sw/w-d-xo.htmlsi=XmWUl7YjooO8tRgN Other video processing stuff can be done with the free software Handbrake too. This software is just convenient in that gives you a nice GUI and does lots of checks when it starts. Totally not necessary, however.

Thanks for your fast answer. I appreciate it. There's a difference between upscaling and upgrading. To simply go from SD to HD in upscaling, as you say you don't need anything more than Handbrake or similar. But now I want more through the use of AI. I've already tried Topaz Video, and the results are spectacular. But you need VRAM memory from a separate video board that you use, which I don't. Only one whole 1 hour long video could I get to finish with Topaz. All my other attempts were to last for several days, and all of them were cut short for reasons I couldn't know, prevent or cure. I still wonder why my 32GB of DIMM memory can't be used for the processing. And 4GB on the external GPU would do a better job. The problem is the life of my GPUs on my computer has been very short on three Nvidia boards I had, and gave up on having one. I wonder if there's a safer and less expensive way to go about that. Can you suggest any?

@@carlose.martinez545 The problem is GPUs are much better suited to AI tasks due to their massive parallel processing which is why most AI tasks depends on the GPU and not the CPU (there is an exception here for Large Language Models (LLMs) which do often run OK on CPUs). This is especially true for image/video processing which is broken down into many small chunks for faster processing. For relatively short tasks you can use the free Google Colab to basically borrow a GPU in the cloud. Because it is free, however, Google won't put up with long running jobs. You can rent CPU/GPUs in the cloud from people like RunPod or Vast ai, but obviously those aren't free. Alternately maybe you could find a friend with a gaming PC that isn't using it over the weekend and bribe them into letting you borrow it for some AI upscaling. I know none of those are great options but AI upscaling (at a reasonable speed) is just a GPU game right now

Do tell me more about exception for Large Language Models (LLMs) which do often run OK on CPUs. What is that? How can I use it for video? How can I find a reliable GPU? There should be some. It did happen with my mobo. After a long run of mobos that lasted about one year, and I'm not talking of cheap mobos, my present Gigabyte GA-Z97X is 9 years old. Same power supply too. Only the GPUs installed on it didn't last.

If that happens you need to describe the pose like "portrait", "full body", "face", etc. as part of the prompt such as "A full body photograph of an Asian woman wearing an ornate, traditional gown"

@@holidayeffects3933 I speak spanish. Maybe is it better to write the prompts in English?

I have only tried English prompts so I do not know how well other languages are supported. However I do know all of the styles internally are in English so I suspect English will give you much better results.

Thanks!@@holidayeffects3933