Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude: Elevated errors across many models [resolved]

status.claude.com

Post reply on HN
Print view View original

151–160 of 173 posts

Re: Claude: Elevated errors across many models [resolved]

#151

Post by pixard »

hotfixguru wrote: post #85
evilturnip wrote: post #13

I do feel like for all their dogfooding of AI coding, their own software/APIs are quite buggy and work against their message. Claude Code is especially buggy in windows terminal. The rendering is quite slow, choppy and lines frequently get garbled. In contrast, using antigravity cli is the exact opposite: fast, smooth and very responsive.

Claude code + tmux is SO buggy. Things rendering all over the place.

Turn off vim mode if you use it. There is some kind of bug with it I found it basically unusable. With vim mode off it sort of works ok. Only breaks occasionally.

But hey coding is a solved problem.

Re: Claude: Elevated errors across many models [resolved]

#152

Post by black_knight »

orphea wrote: post #65
sunaookami wrote: post #47

Earlier quoted context omitted.

That's the first time I saw someone prefering GPT-styled output over Claude ;) It's the complete opposite for me, GPT is way too verbose (even after telling it to STFU), overwhelms the user with thousands of options and doesn't just answer a question without shitting out thousands of paragraphs. Also the overall tone is way too enthusiastic.

Dunno, I prefer GPT 5.5 too for the same reasons as the parent. Extremely subjective but had better results with it too. Maybe I just got unlucky with Claude a few times, but even the latest Opus was dumb .

Fascinating how people have such complete diametrically opposed experiences. I guess both models have it in them to behave very differently in different circumstances and we have very little idea what pushes them in this or that direction. I guess it does boil down to luck!

Personally, Claude Opus (and in the few interactions I had with it, Fable) has been the far the superior experience. GPT-5.5 seems dumber and more certain about presenting me bullshit. Opus has better humor, and is less pretentious in its presentation. But this may all boil down to how the models react to my prompting.

What is without a doubt is that I wish they both were more intelligent – or maybe it is their wisdom I find lacking!

Re: Claude: Elevated errors across many models [resolved]

#154

Post by slashdave »

crystal_revenge wrote: post #86
evilturnip wrote: post #13

I do feel like for all their dogfooding of AI coding, their own software/APIs are quite buggy and work against their message. Claude Code is especially buggy in windows terminal. The rendering is quite slow, choppy and lines frequently get garbled. In contrast, using antigravity cli is the exact opposite: fast, smooth and very responsive.

> in windows terminal This is an aside, but I'm really struck by how many people on HN use Windows (based on repeated mentions I've seen in comments). I've worked for a pretty wide range of companies over the last decade and only one, maybe two companies even had any people that worked on Windows machines. I haven't worked at a company where devs used Windows in 15 years (and even that company eventually switched to…

Windows is the de facto standard in most companies.

That said, I bet you will be really hard pressed to find a single Windows machine at Anthropic.

Re: Claude: Elevated errors across many models [resolved]

#156

Post by sterlind »

vidarh wrote: post #95
crystal_revenge wrote: post #86

Earlier quoted context omitted.

> in windows terminal This is an aside, but I'm really struck by how many people on HN use Windows (based on repeated mentions I've seen in comments). I've worked for a pretty wide range of companies over the last decade and only one, maybe two companies even had any people that worked on Windows machines. I haven't worked at a company where devs used Windows in 15 years (and even that company eventually switched to…

My biggest client right now is about 2/3 Windows 1/3 OS X in the dev team. It was very surprising to me, but I think I freak then out with my maximised tiled iTerms on multiple screens...

[dead]

Re: Claude: Elevated errors across many models [resolved]

#157

Post by Alifatisk »

throwaw12 wrote: post #111
wxw wrote: post #21

I’ve been using Codex w GPT 5.5 more than Claude Code recently. I think Anthropic won the marketing game because Codex is quite good, even better IME.

One very annoying thing about Codex: You can't do anything until all MCP servers are loaded and connected (or failed). Can you move it to background connection?

Unless the MCPs is connected through STDIO, I see no reason for them to block Codex from accepting user prompt. I know other harnesses doesn’t wait for all MCPs to connect. They can establish their connection during a session, they’re dynamic after all.

Re: Claude: Elevated errors across many models [resolved]

#158

Post by recursive »

ls612 wrote: post #125
crystal_revenge wrote: post #86

Earlier quoted context omitted.

> in windows terminal This is an aside, but I'm really struck by how many people on HN use Windows (based on repeated mentions I've seen in comments). I've worked for a pretty wide range of companies over the last decade and only one, maybe two companies even had any people that worked on Windows machines. I haven't worked at a company where devs used Windows in 15 years (and even that company eventually switched to…

Windows laptops suck ass compared to an MBP. Windows desktops are pretty nice, you just need to do a lot of first time setup to remove all of the cruft and make sure you have a local user account etc. But for a typical dev building a desktop every 5 years or so that isn't a big deal.

For work, I'd always be using a domain account anyway. Never used a local user account for work.

Re: Claude: Elevated errors across many models [resolved]

#159

Post by smcleod »

pixelpoet wrote: post #102
crystal_revenge wrote: post #86

Earlier quoted context omitted.

> in windows terminal This is an aside, but I'm really struck by how many people on HN use Windows (based on repeated mentions I've seen in comments). I've worked for a pretty wide range of companies over the last decade and only one, maybe two companies even had any people that worked on Windows machines. I haven't worked at a company where devs used Windows in 15 years (and even that company eventually switched to…

While we're doing anecdotes: I essentially never meet people who own an iPhone, and I've lived (3 months+) in a huge number of countries and cities. The handful people who do are American; bubbles exist :) For a long time, high end graphics and games was mainly done on Windows and Visual Studio. I'm from that world, and only made the Linux transition last year November after Microsoft forced everyone's hand.

Shows how different countries are. In Australia the vast majority of people seem to have an iPhone, with Android being the outlier (at least in the major cities, I can't speak to rural areas). In the workplace it's far more common to see devs running Mac's than Windows, obviously there's pockets where you still see Microsoft but it seems pretty rare amongst engineers these days here. People will say I'm living in a bubble but I'm a consultant that hops around all sorts of clients so if I am it's probably a bubble of being in a city (Melbourne in my case).

Re: Claude: Elevated errors across many models [resolved]

#160

Post by smcleod »

evilturnip wrote: post #13

I do feel like for all their dogfooding of AI coding, their own software/APIs are quite buggy and work against their message. Claude Code is especially buggy in windows terminal. The rendering is quite slow, choppy and lines frequently get garbled. In contrast, using antigravity cli is the exact opposite: fast, smooth and very responsive.

I've never seen any developers really use a windows terminal before, if you're on windows wouldn't you be using the WSL or whatever if you're doing development?
Post reply on HN