Hacker News: IronWolve

New comment by IronWolve in "Claude Is Not a Compiler"

IronWolve — Tue, 21 Jul 2026 15:42:40 +0000

And claude isnt a replacement for the chain of humans, you dont know what you dont know, thats why we have a person in a role to gatekeep each step.

Anyone using claude can see it misses steps that an engineer wouldnt, it can make bad choices that a sysadmin wouldnt, it can pick the wrong order a project manager wouldnt, it can pick wrong cost models that a bean counter wouldnt, etc etc etc

But, I'm pretending it is and living with those mistakes.

New comment by IronWolve in "AliExpress hit with record $629M fine for counterfeit and illegal goods"

IronWolve — Mon, 20 Jul 2026 18:59:25 +0000

I've known a few people that had their products copied by china and sold on amazon, ignoring patents. Same with copies of books. I've even bought a couple items just to find out they are knock offs. Its a big issue on amazon.

New comment by IronWolve in "FFmpeg gets 1.372x faster RGB24-to-RGBA conversion with AVX-512"

IronWolve — Sun, 19 Jul 2026 20:05:08 +0000

Think AI did this? Finding optimizations seems to be happening everywhere. Even llama.cpp is getting faster using AI, which they dont accept AI patches, so forks everywhere.

New comment by IronWolve in "Tokensave: An MCP Server That Saved Me Tokens While Coding"

IronWolve — Sun, 19 Jul 2026 19:49:07 +0000

Anyone try this yet? reviews? thoughts?

New comment by IronWolve in "Classic Amiga titles, free to download"

IronWolve — Sun, 19 Jul 2026 08:12:25 +0000

When I first got my Amiga I spent all day one weekend copying PD game disks at the local computer store, Think it was like 25 cents to copy one plus price of a floppy. Good times.

New comment by IronWolve in "LM Studio Bionic: the AI agent for open models"

IronWolve — Fri, 17 Jul 2026 04:37:46 +0000

It was fun to try out which local models made better agents. Few thoughts.

* locked to a single dir, so no system wide access.

* no local web search, can be fixed ddg or local mcp.

* no ssh, I want to have it ssh into my server and do the work.

* doesnt show the model being loaded, needs a bar/% counter.

* Can you drag/drop documents in the work dirs, or only + add them?

I love lm-studio, so cant wait to see how this goes. For local I normally use opencode + lmstudio.

New comment by IronWolve in "Microsoft Comic Chat is now open source"

IronWolve — Fri, 17 Jul 2026 01:27:08 +0000

Nice, I just made an IRC app that loads comic chat files a few weeks ago.

I added some updates, ignore bots, new expressions, diff message bubble behavior, more panels, custom avatars, colors, etc.

https://github.com/IronWolve/MaxChat

New comment by IronWolve in "Ask HN: Is GitHub preparing to go behind a login wall?"

IronWolve — Tue, 07 Jul 2026 19:12:24 +0000

With AI and all the new projects/forks/updates, github is one busy place.

I wasn't browsing /trending on github until I started using AI, wanting to see what cool projects people are working on.

New comment by IronWolve in "Claude Played Me for a Fool"

IronWolve — Sun, 05 Jul 2026 19:01:47 +0000

I noticed this, when it was only read a few files from my project, and I had to ask it to read ALL the files.

I then had it make a mistakes file and write every mistake, so it would learn, it kinda worked but it would still make the mistakes. It clearly wasn't reading all of it.

So I made a checklist, and it had verify every item on the checklist, that was my work around to both lazy and short mindedness of the agents. Turn mistakes into items to check for. Traded processing time for better results, ok for me on smaller projects. My run times went from 5-10 minutes from 3 per task, need to start logging tasks effectiveness/efficiency to reduce processing time.

I keep seeing people saying loop engineering is the way to get around these issues, I guess I'm kinda doing that in an adhoc way. Since I'm already looking at adding cost and goals(kinda).

New comment by IronWolve in "Qwen 3.6 27B is the sweet spot for local development"

IronWolve — Mon, 29 Jun 2026 18:43:28 +0000

I think things are moving fast, tested that new vibethink-3B, works on many small tasks/fast, and playing with ornith-35B with a draft vibethinker-3b as a draft gave me some good speed/results.

Was just trying to see how small I could go and get acceptable results, but yeah, larger Qwen 3.6 with MTP is going to be better. Cant wait to see how AI model (unsloth/local-llm/heretic/reaper/etc communities) are tweaking/engineering quality down into smaller models. Lots of new things coming out.

New comment by IronWolve in "Why AI Is a Bubble"

IronWolve — Mon, 22 Jun 2026 23:27:57 +0000

Somewhat true, the finances might fall apart, might, for some of the biggest companies, but AI is exploding in services industries of all sizes. Also, ai hardware is lasting way longer than 1 year... So a bubble? maybe, too many assumptions yet.

New comment by IronWolve in "GLM 5.2 vs. Opus"

IronWolve — Mon, 22 Jun 2026 07:57:12 +0000

Having issues with coding a render for good looking realistic smoke coming off burning incense, opus 4.8 & gpt-5.5 both have code issues, glm-5.2 did it. Amazing.

The real time 3d fluid dynamics appear to be the tricky part, I wish I still had opus access, would love to see if it can do it.

New comment by IronWolve in "Windows 11 New Media Player Uses 3.5x More RAM, Charges for Popular Video Codecs"

IronWolve — Sat, 20 Jun 2026 16:18:57 +0000

Do people still use the K-Lite Codec Pack so their players have all the codecs installed? Or just use vlc?

New comment by IronWolve in "What was nice about the UI of Windows 2000"

IronWolve — Thu, 18 Jun 2026 16:37:30 +0000

I really liked when XP came out, and you could have both the clean 2000 look with advanced font rendering like cleartype. That was the perfect combo for desktop use.

I had my xp running a blackbox 4win and coLinux/cygwin, then moved onto vmware/virtualbox with windows vm and linux desktop, now I'm doing win11+wsl2 and loving it for cuda/ai/building projects.

New comment by IronWolve in "Claude Desktop spawns 1.8 GB Hyper-V VM on every launch, even for chat-only use"

IronWolve — Sat, 13 Jun 2026 22:30:47 +0000

I wonder. I spin up tons of projects, but I do it differently.

I'm also on windows. But I'm using a WSL linux vm to do the work. Then I have claude ./sync-to-windows, so the src is copied to a windows share via /mnt/drive/work and I just run a windows built.bat.

I have claude take care of the sync/build.bat so it can do things like clear cache, add options, look at logs, etc. This is also how I build multi platform, build on linux, src works on linux/osx/win

New comment by IronWolve in "I replaced Spotify with a homemade FM radio station"

IronWolve — Mon, 08 Jun 2026 19:16:31 +0000

I'm a big fan of streaming radio, and been using apps for decades, so decided to make a modern version. Also, wanted to try AI coding out, and it worked out really well. Been having a blast.

https://github.com/IronWolve/StreamTuner-ng

Tidy PSU – PD-64 C64 PSU Brings USB PD to Commodore 64

IronWolve — Mon, 25 May 2026 20:23:03 +0000

Article URL: https://theoasisbbs.com/pd-64-c64-psu-brings-usb-pd-to-commodore-64/

Comments URL: https://news.ycombinator.com/item?id=48271257

Points: 17

# Comments: 5

New comment by IronWolve in "Vibe-coding video games with Claude"

IronWolve — Fri, 24 Apr 2026 14:55:34 +0000

I made my first vibe coded game, made casino style slots with lines, bets, free plays, auto play, in a linux shell, with color graphics, amazed how well it came out. Amazed how well it came out, with fixing graphics and adding features.

New comment by IronWolve in "Nvidia's DLSS 5 uses generative AI to boost photorealism in video games"

IronWolve — Tue, 17 Mar 2026 22:36:03 +0000

Might be nice on older games that dont have 4k HD texture packs, or even games like Skyrim or replaying older Witcher 1&2. Its upto the user right? Let people play older games with some better looking gfx. Seems like an easy win, use it or dont use it.

New comment by IronWolve in "Ask HN: Battery life for graphical Linux VMs (or Asahi) on Apple Silicon laptops"

IronWolve — Sun, 13 Jul 2025 22:51:25 +0000

About 50-75%, also depends on cpu workload and brightness.

Running a VM will probably give you better life.