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

Why does Opus 5 feel worse to work with?

mun-logadan.github.io

Post reply on HN
Print view View original

621–630 of 916 posts

Re: Why does Opus 5 feel worse to work with?

#621

Post by throwaway_7274 »

nonethewiser wrote: post #563
throwaway_7274 wrote: post #284

Earlier quoted context omitted.

As a native speaker, it feels like reading an impression of a literature book by a high school English class’s most overconfident student who’s only ever read LinkedIn-speak. Anyway, you might have more luck just writing to it in your native language. It’ll be equally crummy, but maybe you’ll find it easier to decode.

Yes. “Academic” isnt the right term. Its dense like academic language but its also borderline incoherent.

Even more so than borderline incoherent academic writing like Foucault or Lacan or whatnot, for that matter. It’s less “I don’t understand this and I suspect the author doesn’t either” and more “reading this feels like having a stroke.”

Re: Why does Opus 5 feel worse to work with?

#622

Post by saaaaaam »

galaxyLogic wrote: post #535
saaaaaam wrote: post #303

Earlier quoted context omitted.

Oh, that's interesting - because that's absolutely what's happening in my experience. If I look at the thinking (which seems to have become unavailable in Opus 5 a lot of the time, but was present - and often useful - in 4.8/4.6) you're right - it's having the discussion with itself, and seems unable to distinguish that discussion from discussions with me. BUT it also seems to be related to the length of the chat - t…

Interesting. So if the LLM is having a discussion with itself, am I paying for the tokens it uses for that?

I think so, yes.

When I make API calls, the discussion with itself is part of my token cost, so I assume that is the same in the subscription plans.

Which is why people are surprised when they use their whole allocation in half an hour asking questions Fable about 200 page document.

Re: Why does Opus 5 feel worse to work with?

#623

Post by avereveard »

barrkel wrote: post #12

The single biggest annoyance with Opus 5 is that it writes too elliptically. Sentences that orbit a point, then jump to it like it's a revealed insight. Unnecessarily abstract phraseology. Constantly using inanimate nouns as the subjects in sentences in order to unlock variety in verb choice, especially when it helps construct a sentence where the real action can 'land' like a surprise at the end. It is definitely mo…

[dead]

Re: Why does Opus 5 feel worse to work with?

#624

Post by quintu5 »

vivzkestrel wrote: post #364

- why dont we have a model that would actually ask you 50000 questions everytime you wanted deep work done?

The UX around asking and answering so many questions is awkward.

To work around this, I had claude code build me a questionnaire skill that takes a json file with a flexible schema as input and it then serves up a simple questionnaire web page on a node server where I can read the questions and give my responses either by selecting from preset tags supplied as part of the input or by including a text-based response.

The agent can include references to external images, html files, or mermaid diagrams and the page can render them all inline with the relevant question.

Once I'm done answering, I just save my responses and click a button to kill the server. The agent watching the process sees that it stopped and takes that as a signal to go read the responses from disk.

Works like a charm.

Re: Why does Opus 5 feel worse to work with?

#625

Post by grey-area »

npalli wrote: post #610
Panoramix wrote: post #500

Earlier quoted context omitted.

You're right, and the load-bearing part of the argument is not what you think it is. Two ambiguities worth resolving before moving on: whether what you wrote also applies to ChatGPT, and whether you have custom instructions set up. Failure mode worth flagging explicitly: I didn't read TFA. (I'm becoming allergic to how these things write).

Alas, it writes so much better than the average human that it's what everyone started using. Hence the utter familiarity and now contempt.

No, it does not.

Re: Why does Opus 5 feel worse to work with?

#626

Post by mnicky »

jmuguy wrote: post #258

If anyone knows how to get it to stop adding comments, I'm all ears. Its just getting worse and I'm starting to worry that the comments themselves are poisoning future agents that examine the codebase.

Try using output styles: https://code.claude.com/docs/en/output-styles

Re: Why does Opus 5 feel worse to work with?

#627

Post by rhdunn »

nonethewiser wrote: post #559
herbturbo wrote: post #479

Earlier quoted context omitted.

Perhaps this is related to their new "invisible watermark" concept which would probably require rather contrived language patterns to make possible.

I think opus was released before they included it on model. Its hard to say, but from what Ive read it doesn’t seem like it would have that drastic of an effect. I had the same thought though.

The watermarking is independent of the model. The model itself has the probability weights to determine the next token. The watermark is similar to things like temperature and top_p/top_k in that the watermark adjusts the probabilities in a deterministic way that changes over time to hide tells from word choices.

Re: Why does Opus 5 feel worse to work with?

#628

Post by semiquaver »

grey-area wrote: post #625
npalli wrote: post #610

Earlier quoted context omitted.

Alas, it writes so much better than the average human that it's what everyone started using. Hence the utter familiarity and now contempt.

No, it does not.

You are vastly overestimating the average writer’s ability.

Re: Why does Opus 5 feel worse to work with?

#630

Post by rsalus »

barrkel wrote: post #12

The single biggest annoyance with Opus 5 is that it writes too elliptically. Sentences that orbit a point, then jump to it like it's a revealed insight. Unnecessarily abstract phraseology. Constantly using inanimate nouns as the subjects in sentences in order to unlock variety in verb choice, especially when it helps construct a sentence where the real action can 'land' like a surprise at the end. It is definitely mo…

Yeah, +1 on the comment verbosity. Left alone it's actually insane. I've had to setup enforcement + templates (use ASD-STE100) to keep the cruft down. I'm worried this may impact the quality of outputs though - haven't measured it.
Post reply on HN