[Marko Zegarac]

Andrew,

Not a question, just wanted to thank you for your work on this website. Although YouTube is flooded with AI content, that doesn’t work for every learner. I learn by reading, comprehending and following (written) instructions. Your presentation of content here is truly amazing. I couldn’t do better myself (and I do teaching for a living).

Looking forward to seeing more of your articles/tutorials on the upcoming technologies!

Cheers,
Marko

(PS As feedback to you, let me add: consider creating a subforum called ‘Feedback’. 🙂

[Andrew]

Thanks for the compliment! Great idea. I will add a feedback section when I got a chance.

[Graeme Nimmo]

I just want to chip in and echo Marko’s comment above to show my massive appreciation for the content you create. I’m an ex-teacher and I absolutely loathe and detest learning from youtube videos and have a massive preference for text and the work you’ve done is great for that, so I’d like to throw you a massive thanks for that.

[Andrew]

Thank you, Graeme! 😄

[Rafael Bravin]

I also wanted to show my gratitude, your site showed me the path to an incredible journey. Since the time I started learning here I’ve grown massively and now I feel like I’m finally able do walk on my own. Even tho I miss some updates and more advanced content (and would like to know when there is new stuff available), for a beginner this was a wounderfull place to start.
If you ever add some newer content with the more advanced stuff that is coming out (flux and newer lcm animate-diff models) I will defnetly take a look, cheers.

[David Rawlins]

Hi Andrew,

I’d also like to add my thanks to you for SDA – it is a fantastic learning resource.

One suggestion is that if you have time, it would be worth refreshing the pages that cite or link to checkpoint models as there are some newer versions and some of them (eg Juggernaut XL_v11) seem very good indeed and a big improvement over the older ones. I can see the move is towards Flux and other new models but some of the SDXL models are still pretty good from what I’ve seen. I’ve mostly stopped using standard SD models but they can also be made to work well as you’ve shown.

[Ulysses Heart]

Thanks for the great content. It has always been useful for me, many times over. I just have one thing to mention, and that is what seems like a big bias toward ComfyUI for all things, which only seems to have increased over time, even as things have changed a lot. I am running a very powerful Mac and have used Auto1111 for over a year, but want to find way of doing AI video locally. My problem is that installing ComfyUI always fails at some point, no matter what, time and time again, and I’ve just given up. I just wonder if you’ve come across any other platforms capable of being locally-installed that seem interesting lately (i.e., like LTXV, Mochi1, etc)?

Thanks very much.

[Andrew]

Video models tend to be much larger than image ones. They are just not made for Mac to run. While Mac’s silicon chip has gone a long way, the models must be rewritten to take full advantage of it. Even if a team manages to do that, there will still be a significant gap from Nvidia cards.

A1111 and Forge’s development is stagnant. Only ComfyUI is still going strong. In the foreseeable future, most new stuff will be on ComfyUI with an NVidia card (PC or Linux). If you don’t want to own one, the next option is using an online service, which is likely cheaper than owning one.

[Author]

Posts