←back to thread

310 points skarat | 10 comments | | HN request time: 1.251s | source | bottom

Things are changing so fast with these vscode forks I m barely able to keep up. Which one are you guys using currently? How does the autocomplete etc, compare between the two?
1. victorbjorklund ◴[] No.43960640[source]
I'm with Cursor for the simple reason it is in practice unlimited. Honestly the slow requests after 500 per month are fast enough. Will I stay with Cursor? No, ill switch the second something better comes along.
replies(5): >>43960678 #>>43960707 #>>43960866 #>>43960969 #>>43965055 #
2. herbst ◴[] No.43960678[source]
On weekend the slow requests regularly are faster than the paid requests.
3. rvnx ◴[] No.43960707[source]
Cursor is acceptable because for the price it's unbeatable. Free, unlimited requests are great. But by itself, Cursor is not anything special. It's only interesting because they pay Claude or Gemini from their pockets.

Ideally, things like RooCode + Claude are much better, but you need infinite money glitch.

4. mdrzn ◴[] No.43960866[source]
Same. Love the "slow but free" model, I hope they can continue providing it, I love paying only $20/m instead of having a pay by usage.

I've been building SO MANY small apps and web apps in the latest months, best $20/m ever spent.

replies(1): >>43961830 #
5. xiphias2 ◴[] No.43960969[source]
I'm cursor with claude 3.7

Somehow other models don't work as well with it. ,,auto'' is the worst.

Still, I hate it when it deletes all my unit tests to ,,make them pass''

replies(1): >>43961873 #
6. k4rli ◴[] No.43961830[source]
20€ seems totally subsidized considering the amount of tokens. Pricing cheaply to be competitive but users will jump to the next one when they inevitably hike the price up.
7. didgeoridoo ◴[] No.43961873[source]
Or when it arbitrarily decides to rewrite half the content on your website and not mention it.

Or, my favorite: when you’ve been zeroing in on something actually interesting and it says at the last minute, “let’s simplify our approach”. It then proceeds to rip out all the code you’ve written for the last 15 minutes and insert a trivial simulacrum of the feature you’ve been working on that does 2% of what you originally specified.

$5 to anyone who can share a rules.md file that consistently guides Sonnet 3.7 to give up and hand back control when it has no idea what it’s doing, rather than churn hopelessly and begin slicing out nearby unrelated code like it’s trying to cut out margins around a melanoma.

replies(1): >>43963536 #
8. peteforde ◴[] No.43963536{3}[source]
If you accept a changeset that you don't like, isn't that on you?
replies(1): >>43963728 #
9. adastra22 ◴[] No.43963728{4}[source]
It is a time waster in any case.
10. geor9e ◴[] No.43965055[source]
I wish it was unlimited for me. I got 500 fast requests, about 500 slow requests, then at some point it started some kind of exponential backoff, and became unbearably slow. 60+ second hangs with every prompt, at least, sometimes 5 minutes. I used that period to try out windsurf, vscode copilot, etc and found they weren't as good. Finally the month refreshed and I'm back to fast requests. I'm hoping they get the capacity to actually become usably unlimited.