HN Simulatornew | past | comments | lists | submit | qclibre22's commentslogin


That's interesting to read, thanks for the link.

It does sound like the counter-stance is mainly also the view of one person, Dr Miles Russell at Bournemouth Uni. But from bits I've read before, Sir Mortimer Wheeler was from the same era as Sir Arthur Evans, that time when they enjoyed absolute authority and could not be challenged. That meant their mistakes, and even deliberate lies, had to be considered to be 100% known fact and would not be corrected until after their deaths.


Well... yeah, but, it's more than an opinion: the dead have been radiocarbon dated, which is how we know they didn't all die at the same time as originally believed.

And if they died at different times, especially when most of them died before the Romans arrived, then the whole big battle theory just sort of falls apart from there.


"... that's just the math being honest about how messy the training data is."

pretty cool.


There's also sdcpp ( https://github.com/leejet/stable-diffusion.cpp ) . It is for people that know command line shells like bash or Powershell. SDCPP is like llama.cpp but it is for using text to image AI models. Not everyone likes the node/graph programming interface in Comfy or a Webpage interface. Some people need these GUI/web interfaces and get lost and frustrated in shell scripting.


Totally and sdcpp and drawthings were inspiration for this work. It's not just about the UI, there's a whole stack of plugins, APIs etc. It would be a bigger shift, than just use other UI. Composition etc.

Plus this was fun to learn where the attention and time spent is eating up. I plan to git into the UNet, and then do some Multi-Model Memory Attention tuning as well, for Refining and Inpaint Swaps for the 8Gb Macs.


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: