I'm a programmer, not an artist, and I've always followed the advice of using crappy programmer art to get the game functional and playable before looking to find artists to make it look appealing.

However, lately I'm seeing more and more people doing the opposite, people who are artists first or know someone, and they get the look of the game settled before making the game playable. And I can see why: they immediately have something to post and show people, starting the marketing machine early.

Have the times changed? Do we need to be locking down the visuals earlier to generate a following while the gameplay gets worked out? Or does the old advice of getting the game fun and playable first still hold? Does that delay marketing and generating interest too late in the development process now?

I may be biased as I'm a sound person, but I'm personally a fan of getting the game looking nice as soon as possible, especially when it comes to a game that's story-driven. Like you mentioned, it helps with marketing really early in, but I also think it helps keep people motivated as it feels like they're working on a more polished project relatively early in.

I think of course, it can depend on the game as well and what it's priorities are for the player.

Nairou said:
… using crappy programmer art to get the game functional and playable before looking to find artists to make it look appealing … people doing the opposite, people who are artists first or know someone, and they get the look of the game settled before making the game playable. And I can see why: they immediately have something to post and show people, starting the marketing machine early.

They do both, and it depends on many specifics.

Games in environments that need a pitch need something to pitch. For some games that means piles of concept art, lots of beautiful drawings that convey what is needed for the pitch. For some games that means a tech demo, enough of a single concept implemented to demonstrate the mechanic even if it isn't beautiful or polished or complete. For some games, you need more than the single concept in a tech demo, you need a bigger proof of concept for the entire game created, with a bigger exploration of the concept than a tech demo. For some games that means a vertical slice, a complete playable level including somewhat polished graphics and animations and sound and menus and gameplay and functionality. Once the pitch is made you will need to iterate repeatedly to reach a finished state.

Similarly for games that need a marketing direction, you need something to market. Repeat all the potential different things you may need, you may need to have previously done market research or allow marketing to do it and bring in results. You will need numbers for cost estimates, numbers for business plans, numbers for sales estimates. You will need to iterate repeatedly to reach a finished state.

For games where you are trying to reach a key demo, or publishers, or early access, or whatever else, they all have different needs.

If you're trying to experiment with game design ideas, you need enough to explore the design. If you're trying to experiment with different emotional styles, you need enough to do that instead.

There is no universal answer. Figure out what you need, figure out the order you need to do it, which is unique to your project, and follow your own route.

Well, there is a lot of new “plug and play” engines that can be easily learned like Renpy, or have a lot of documentations like Blender and Unity. Where it's easier to jump into some types of games genres like visual novels, Cursed runes, er, I mean open world survival crafting, and point and click adventures it doesn't quiet help make things where a bit of coding knowledge will help like Classic Castlevania platforming, racing games, Loaded esque top down shooters.

But honestly don't focus on visual art, you can write text adventures that lean on flavor like humor or action.

If you're a single individual working alone, then you can only work on one part of the game at a time. That also means that you can't meaningfully market and work on your game at the same time. Early marketing is therefore a trap for you, as you'll eventually either end up neglecting the marketing or neglecting the game.

If you've got a marketing guy, let them do the marketing. If you've got an artist, let them do the art. But if it's just you, focus on getting the gameplay right first and worry about graphics and marketing later.

Actually the answer to this question is simple. To entice a user you have maybe at max 5-10 seconds. Here your visuals need to appeal to him/her. Visuals will bring in the interest. However to keep the interest and make him/her keep coming back you need the game play to be interesting. Which comes first does not matter, however getting the game play out of the way by getting it done makes it much easier to focus on the art quality.

In our game Funtabulator release currently on google play , we finished the logic and game play first. Then we concentrated on the visuals and tried to make the game cute.

The two things from your post are not mutually exclusive.

Before you spend time (and/or money) on high quality art, make sure it will not go to waste, make sure the game is worth finishing. That did not change.

If you want to show your game to someone, it better look good. That did not change either.

What did change a little is that some games are shown to the public earlier in the development process ("early access" and things like that), and that pushes the need for good looking art earlier. But all of the above is still accurate.