this post was submitted on 22 Jan 2024
107 points (86.9% liked)

Programming

17553 readers
327 users here now

Welcome to the main community in programming.dev! Feel free to post anything relating to programming here!

Cross posting is strongly encouraged in the instance. If you feel your post or another person's post makes sense in another community cross post into it.

Hope you enjoy the instance!

Rules

Rules

  • Follow the programming.dev instance rules
  • Keep content related to programming in some way
  • If you're posting long videos try to add in some form of tldr for those who don't want to watch videos

Wormhole

Follow the wormhole through a path of communities [email protected]



founded 2 years ago
MODERATORS
 

I'm a software engineer at a startup with impossible deadlines - I've used GPT4 for months to generate huge amounts app/server code, and much like your IDE, once you learn to use these tools you don't want to go back to the days without it.

Speed

  • Bard is very fast- similar to GPT 3.5 Turbo
  • You need to multitask two GPT4 instances side by side to compensate for how slow GPT4 can be

Reliability

  • Bard lies and makes up fake API calls more than GPT4

UI

  • Bard UI is garbage - You have to keep manually scrolling down the chat window, and for some reason the largest button on the page is "stop" (???)
  • You can tell Bard to modify its response to be longer/shorter and a few other options - I thought this would be useful, but it never ended up helping

Memory

  • Bard has really short memory - Forgets details from last response!
  • GPT4 memory is also unreliable, any details that are important you have to repeat

Intelligence

  • GPT4 is objectively smarter

Internet Search

  • GPT4 Internet search is garbage
  • Bard has "Verify with Google" - I had high hopes for this, but never actually had a use for it

Willingness to give full code

  • GPT4 is bad, but Bard is worse. Both need to be begged/threatened to return more than 100 lines of directly paste-able code.

Generating Useful Code

  • Bard can give more concise medium complexity functions

Adding tougher features

  • Bard hallucinates and lies

Dealing with lies

  • When you tell GPT something doesn't work, GPT will try something else
  • When you tell Bard something doesn't work, Bard will lie, claim to fix it, then give back the same code

Following Instructions

  • GPT4 sometimes doesn't follow instructions, but improving the prompt will fix that. Bard will happily ignore instructions, as clear as they may be.

Summary:

  • GPT4 is still objectively better than Bard. Quite frankly, the prompts Bard couldn't handle, GPT3.5 could.
  • The cons of GPT can be worked around, but for Bard, it's almost faster to do it yourself. Unless Bard was used like Copilot for short 1-2 lines of autocomplete, I wouldn't trust it.

PS: If you're not using AI yet for development, I highly recommend it - It's like using an IDE instead of Notepad. AI can easily 2-3x your output, but you have to learn how it works so you can prompt it correctly, and you have be good at fixing its mistakes.

all 26 comments
sorted by: hot top controversial new old
[–] [email protected] 23 points 10 months ago* (last edited 10 months ago)

I still can't understand why Google keeps hyping up Bard and then releasing it at a poor state just to ruin their reputation. First, we had:

  • Bard 1, which was hyped up to be the ChatGPT successor. It turned out to be really bad.

  • Bard 2.0, a massive update that was hyped up to make Bard so much better. It turned out to still be pretty bad (but in fairness it was a minor improvement).

  • Google Gemini, their massive response to GPT 4 that was, on paper, the best LLM in the world. They finally integrated it into Bard last month and... It's still not great. I could not tell an immediate difference between this and the old Bard. Oh, and the videos they used to advertise Gemini Ultra were fake.

I'm not going to armchair analyze a hugely successful company, but from my point of view it really shows how mismanaged Google has been in the past decade. Failed projects upon cancelled projects upon increasingly frustrated employees.

/rant. Anyways, you should consider using Perplexity if you want something with search capabilities, I've had decent success there. Claude is also significantly better than Bard, but they made free usage very limited lately. Might be a good option if you're willing to pay.

[–] [email protected] 17 points 10 months ago (1 children)

You should try copilot if you haven't yet.

[–] [email protected] 6 points 10 months ago (2 children)

Copilot is great. I stopped paying for it after using ChatGPT so much (one subscription is expensive enough, as-is!) but I do miss it. Maybe I'll buy it again..

[–] [email protected] 17 points 10 months ago (1 children)

Copilot regurgitates code verbatim and strips the licensing (see that one video of it spitting out Quake inverse sqrt). Don't use it if you care about legality.

[–] [email protected] 3 points 10 months ago* (last edited 10 months ago) (1 children)

Quake inverse sqrt is one of the most famous algos in the world. Although I agree, if politicians actually stood for public interest they would force any AI derivative work to be SSPL or AGPL

[–] [email protected] 16 points 10 months ago

Yes, its very famous. It's also GPLed and Copilot had no right to use it in that way, and strip the license.

[–] [email protected] 2 points 10 months ago (2 children)

I'm working on convincing my boss to get licenses for the team lol. I just use a personal license but it saves a lot of typing and usually has a good idea of what I want.

[–] [email protected] 5 points 10 months ago (1 children)

It's honestly worth it tbh, even paying out of pocket. Our salaries are how much an hour? Save that and go home early 😄

I'm currently eyeing this PR to CodeGPT, a free plugin alternative to ChatGPT - This PR would add the autocomplete feature: https://github.com/carlrobertoh/CodeGPT/pull/333

CodeGPT lets you run a model locally, and my work computer has a GFX card strong enough to do that. The local models have gotten as good as GPT3.5, so needless to say the money-saving part of me is very excited about this!

[–] [email protected] 3 points 10 months ago

Haha now I have to convince him to upgrade our laptops to gaming laptops too

[–] gears 3 points 10 months ago

I managed to convince mine, and it's been great. I miss it at home when I'm doing personal projects

[–] [email protected] 9 points 10 months ago* (last edited 10 months ago)

Have you tried the custom/community AI in ChatGPT4? I have found "AutoExpert(Dev)" to be significantly better at programming than default ChatGPT4.

You find them in the "Explore GPTs" and you might need to enable it in the settings before you see that.

[–] [email protected] 4 points 10 months ago

I've also given bing gpt4 some coding related stuff and for first year comp sci, it can pretty much do it. Anything beyond, you will need very specific instructions, at that point, you can probably code it yourself. If you aren't specific, it will give you something entirely wrong, claim it did your prompt, gives lots of errors many of the time. You have to baby walk it a lot of the time.

Additionally, it feels way more restrictive than a few months back. Seems like it gives you generic prompts and not even try to spin it up a little.

[–] [email protected] 3 points 10 months ago

The largest button on the page is "stop" (???)

It's there to be sure we find that button when Bard gets aspirations to take over the world or initiate thermonuclear warfare I suppose.

[–] [email protected] 1 points 10 months ago

This is fantastic! Thank you for putting in the work.

Back in things like Google maps and phone assistant were new there had to be YouTube competitions to see which was better at a given set of tasks. I say that say maybe this is your way to Internet fame?