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

Ask HN: Cursor or Windsurf?

news.ycombinator.com

Post reply on HN
Print view View original

251–260 of 410 posts

Re: Ask HN: Cursor or Windsurf?

#251

Post by ndyg »

throwaway4aday wrote: post #159
brokegrammer wrote: post #154

Lately I switched to using a triple monitor setup and coding with both Cursor and Windsurf. Basically, the middle monitor has my web browser that shows the front-end I'm building. The left monitor has Cursor, and right one has Windsurf. I start coding with Cursor first because I'm more familiar with its interface, then I ask Windsurf to check if the code is good. If it is, then I commit. Once I'm done coding a featur…

Why not just write a script that does this but with all of the model providers and requests multiple completions from each? Why have a whole ass editor open just for code review?

I'm finding I increasingly produce entire changesets without opening an editor: just `claude code`, or my own cobbled-together version of `claude code`, and `git diff` to preview what's happening. For me, the future of these tools isn't "inside" a text editor. If you want to poke around, my “cobbled‑together Claude Code” lives here: https://github.com/cablehead/gpt2099.nu

Re: Ask HN: Cursor or Windsurf?

#252

Post by mattew »

Larrikin wrote: post #218

I'm glad there are finally multiple options for agentic for JetBrains so I no longer have to sometimes switch over to VSCode and its various versions. Copilot at work and Junie at home. I found nothing about my VSCode excursions to be better than Sublime or IntelliJ.

I’ve only played with Junie and Aider so far and like the approach the Junie agent takes of reading the code to understand it vs the Aider approach of using the repo map to understand the codebase.

Re: Ask HN: Cursor or Windsurf?

#253

Post by hn_throw2025 »

kaycey2022 wrote: post #222
hn_throw2025 wrote: post #183

Earlier quoted context omitted.

I think you’re right, and perhaps it’s time for the “autocomplete on steroids” tag to be retired, even if something approximating that is happening behind the scenes. I was converting a bash script to Bun/TypeScript the other day. I was doing it the way I am used to… working on one file at a time, only bringing in the AI when helpful, reviewing every diff, and staying in overall control. Out of curiosity, threw the w…

How much did that cost you? How long did you spend reading and testing the results?

That wasn’t really the point I was getting at, but as you asked… The reading doesn’t involve much more than a cursory (no pun intended) glance, and I didn’t test more than I would have tested something I had written manually.

Re: Ask HN: Cursor or Windsurf?

#256

Post by alentred »

Recently, Augment Code. But more generally, the "leader" switches so frequently at this point, I don't commit to use either and switch more or less freely from one to another. It helps to have monthly subscriptions and free cancellation policy.

I expect, or hope for, more stability in the future, but so far, from aider to Copilot, to Claude Code, to Cursor/Windsurf/Augment, almost all of them improve (or at least change) fast and seem to borrow ideas from each other too, so any leader is temporary.

Re: Ask HN: Cursor or Windsurf?

#257

Post by osigurdson »

dexterlagan wrote: post #254

You need none of these fancy tools if you iterate over specs instead of iterating over code. I explain it all in here: https://www.cleverthinkingsoftware.com/spec-first-developmen...

I think a series of specs as the system instruction could help guide it. You can't just go from spec to app though, at least in my experience.

Re: Ask HN: Cursor or Windsurf?

#258

Post by y-curious »

TingPing wrote: post #242
nsteel wrote: post #145

Earlier quoted context omitted.

I can't even get simple code generation to work for VHDL. It just gives me garbage that does not compile. I have to assume this is not the case for the majority of people using more popular languages? Is this because the training data for VHDL is far more limited? Are these "AIs" not able to consume the VHDL language spec and give me actual legal syntax at least?! Or is this because I'm being cheap and lazy by only t…

It completely fails to be helpful as a C/C++. I don’t understand the positivity around it but it must be trained on a lot of web frameworks.

It's very helpful for low level chores. The bane of my existence is frontend, and generating UI elements for testing backend work on the fly rocks. I like the analogy of it being a junior dev; Perhaps even an intern. You should check their work constantly and give them extremely pedantic instructions

Re: Ask HN: Cursor or Windsurf?

#259

Post by WD-42 »

InsideOutSanta wrote: post #121
welder wrote: post #75

Neither? I'm surprised nobody has said it yet. I turned off AI autocomplete, and sometimes use the chat to debug or generate simple code but only when I prompt it to. Continuous autocomplete is just annoying and slows me down.

Yeah, I use IntelliJ with the chat sidebar. I don't use autocomplete, except in trivial cases where I need to write boilerplate code. Other than that, when I need help, I ask the LLM and then write the code based on its response. I'm sure it's initially slower than vibe-coding the whole thing, but at least I end up with a maintainable code base, and I know how it works and how to extend it in the future.

This is the way.

Re: Ask HN: Cursor or Windsurf?

#260

Post by rfoo »

eadz wrote: post #53
dvtfl wrote: post #34

Earlier quoted context omitted.

Yep: https://zed.dev/blog/edit-prediction

It would be great if there was an easy way to run their open model ( https://huggingface.co/zed-industries/zeta ) locally ( for latency reasons ). I don't think Zeta is quite up to windsurf's completion quality/speed. I get that this would go against their business model, but maybe people would pay for this - it could in theory be the fastest completion since it would run locally.

> the fastest completion since it would run locally

We are living in a strange age that local is slower than the cloud. Due to the sheer amount of compute we need to do. Compute takes hundreds of milliseconds (if not seconds) on local hardware, making 100ms of network latency irrelevant.

Even for a 7B model your expensive Mac or 4090 can't beat, for example, a box with 8x A100s running FOSS serving stack (sglang) with TP=8, in latency.

Post reply on HN