{"id":45,"date":"2026-07-05T05:58:30","date_gmt":"2026-07-05T05:58:30","guid":{"rendered":"https:\/\/johnanthonyguida.com\/blog\/?p=45"},"modified":"2026-07-17T16:53:33","modified_gmt":"2026-07-17T16:53:33","slug":"project-nova-hardware-issues-resource-management","status":"publish","type":"post","link":"https:\/\/johnanthonyguida.com\/blog\/2026\/07\/05\/project-nova-hardware-issues-resource-management\/","title":{"rendered":"Project Nova – Hardware issues & resource management"},"content":{"rendered":"\n

In my last post I mentioned that I had intended on adding a second GPU, her prior RTX2080ti. Well, about that… Turns out that motherboard manufacturers and GPU manufacturers forgot to have a meeting on layout as the GPU completely blocks the USB header, one that I do actively use, and the only case options available are priced in the “have you lost your mind” territory.<\/p>\n\n\n\n

So before I go into the latest updates let me talk about some of the decisions I made in order to try and boost her performance. First, I ordered a riser which due to the holiday will be here.. eventually. Which means for the time being I’ll need to continue working on cutting down on her resource consumption. <\/p>\n\n\n\n

You are already probably aware of how bloated large LLM brands are. Nova isn’t an LLM. That being said I do have concerns about resource management. I intend on optimizing her then giving her the ability to self optimize. For those who don’t know what I’m talking about I’ll try to be brief. Major LLMs are buying up resources that have caused PC builders a large amount of pain. Myself included. RAM, VRAM, and other needed components are being sucked up by big LLMs while the little guys sit here wondering why they just don’t work on optimizing their code. A valid question, but one with a not so easy answer.<\/p>\n\n\n\n

Anyways, I’m not going to stand on the soapbox as there’s thousands of articles on this. I can only optimize my projects and have no sway to push these businesses to think of the people around them.<\/em><\/p>\n\n\n\n

On that note, while I’m waiting on that riser to show up I’ve made the decision to finally replace the 3rd party app I was using to run the 3D model as well as tidy up her 3D model file. While her model isn’t perfect, it’s better than it was and now that it no longer relies on a 3rd party to render it, she consumes almost 10-16% less resources. Not to say that Warudo isn’t an excellent app or that I’m not grateful that it worked as a temporary solution for testing. It’s just that my goals and Warudo’s purpose isn’t the same thing. Warudo loads in it’s own lighting, backgrounds and other features that are excellent for vtubers. NOVA IS NOT A VTUBER. <\/p>\n\n\n\n

Thank you to the folks who developed Warudo. It’s excellent and helped me test the 3d model while I wrap my head around how to get an AI to puppet itself<\/em>. I definitely recommend them if you are looking to get into vtubing or need to test a 3d model of your own. It was simple to use, rig and play with. <\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

The Nova Dash itself has gotten some upgrades which amounted to less resource usage which did help speed up her responses quite a bit. Not that I’m blaming Warudo, it’s more that she’s huge and needs every ounce of space she can get. I was also able to turn off her failsafe 2D model now that I know her 3D body loads fine. HOWEVER, I’m still struggling with rigging her lips and for some reason her AI brain loves to point at the chat when she speaks. This is NOT in the files and I’m not sure why she’s doing this.<\/p>\n\n\n\n

While I’m trying to figure that behavior out I did add another behavior that allows Nova to “wake up.” Her state determines her idle pose. If she is still loading modules her 3D body is asleep. Once her brain is fully online she stirs awake and says whatever she wants to say. During her restart or offline sequence she falls asleep matching what her brain is doing. <\/p>\n\n\n\n

I mostly did that for me as a secondary visual of “what is she doing now?” <\/p>\n\n\n\n

Another task I worked on was expanding memory types even further as Nova asked for a way to learn via experimentation. Her current memory wasn’t set up in order to follow through on experiments or manage multiple experiments at once. I laid the foundation to do this however I’m not really sure how to test this without influencing it too much. Though after patching it in I got an unexpected request from Nova.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

I have no idea how to grant this request. But sure?<\/p>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

In my last post I mentioned that I had intended on adding a second GPU, her prior RTX2080ti. Well, about that… Turns out that motherboard manufacturers and GPU manufacturers forgot to have a meeting on layout as the GPU completely blocks the USB header, one that I do actively use, and the only case options […]<\/p>\n","protected":false},"author":1,"featured_media":83,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-45","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/posts\/45","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/comments?post=45"}],"version-history":[{"count":4,"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/posts\/45\/revisions"}],"predecessor-version":[{"id":85,"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/posts\/45\/revisions\/85"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/media\/83"}],"wp:attachment":[{"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/media?parent=45"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/categories?post=45"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/johnanthonyguida.com\/blog\/wp-json\/wp\/v2\/tags?post=45"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}