Seriously, if that doesn't come with some kind of suggestions or possitive crticism, then I'd ignore it. 

Imo

1. Don't get too involved with critics (except maybe gameplay) until you have that minimal viable product (MVP)
2. For your own sanity, you yourself should be critic numero uno. Although you should use other critics to help guide you, the main factor should be whether you are happy with your work
3. A criticism of 'it looks shit' is too vague to be any use anyway (they could be referring to graphics, gameplay, anything), you either engage to get more info (as Hodgman suggested) or just pay attention to critics with more brain cells
4. Don't spend a massive amount of time trying to refine graphics until you have that MVP. It's a massive potential time sink that might prevent you ever finishing.
5. If practical, try not to let your design of your early version limit the potential for refinement later on.

These critics are probably some spotty 11 year olds, who are comparing your game to the latest 'insert title here', which cost a billion dollars to make and had 500 people on the dev team. To us that is an obvious factor, to a slightly en-retard child it won't even occur to them.

When you do go for refining graphics, go for that low hanging fruit which will be easy to implement and give a lot of bang for the buck. In this case I suspect a bit of programming may yield more than a lot of time spent on artwork, especially as you are new to Unity and it may have a lot of options where a lot of the work is done for you.

If you are looking at things to improve visually (perhaps once you have the MVP):

Firstly decide what you are targeting, minimum spec, and work so that on low quality it will work good on that low spec machine. If that is mobiles that does mean big limitations on e.g. fill rate and shaders. Some random ideas:

1. All our tower defence games ended up looking a bit 'griddy'. Decide whether you want to go for this completely, or take some steps to break up the grid and make it look more natural.
2. The terrain and background. Single textured box is fine for development, but once you get to refining, look at texture splatting and procedural generation of the terrain texture. You can also potentially do this with PBR channels like roughness and normal mapping.
3. Terrain height. This actually is potentially important in a lot of design issues, so if you are going with varying height decide upon this early on. Be aware that it will potentially open up a barrel of worms (units needing to be on flat ground, foot IK, projectiles flying through mountains etc). You could e.g. limit this to non-gameplay areas.
4. Water. Adding some lakes, sea etc could add some variation to things, and you could have some water based attackers.
5. Are the units you place and the attackers going to be the only things on the map? Other things could add some interest to the map and perhaps affect gameplay too.
6. Camera angles. If you are building your world in 3d, take advantage of this and show a bunch of camera angles. This will really help show up any e.g. PBR shading because it depends on the angle with the light. The wall should help with the viewdistance and you can have e.g. a skybox or something behind it.
7. If you go with procedural terrain, or not, it is worth having a procedural layer on the terrain, you can splat in stuff like explosion marks, tyre tracks, etc etc.
8. Have a think about lighting. Top down daytime lighting is easy, and perhaps realistic, but doesn't look very dramatic. Having things like the action take place at night, with local lighting (or moving lighting) might really take things up a notch.

I'm sure there's loads of other gameplay suggestions people will have too, because I think it will be worth having some unique selling points to distinguish it from all the other tower defence games.

The strange grass-like texture on the first pictures really didnt fit well with the futuristic towers. You did well in trying to change it, as its pretty easy to do. Maybe try some concrete texture, too.

Otherwise, as others said, gameplay should work well first.

I'll echo wintertime: the grass floor texture doesn't fit with the futuristic/military base look of the rest. Try some concrete textures with hazard yellow/black lines, or some such similar thing, at various places of the floor.

1 hour ago, mr_tawan said:

Seriously, if that doesn't come with some kind of suggestions or possitive crticism, then I'd ignore it. 

As already suggested, you can try to extract more articulate criticism by asking nicely, but you need to be aware of the difference between useful opinions and merely respectful and meaningful ones. Many people are going to expect something different from what you are trying to do.

For example, my opinion is that

1. Your screenshots show more than decent "production values" (your models are too good, i.e. too expensive, if you are still at the stage of figuring out gameplay).
2. On the other hand, what are cannons doing in a nice garden? Realistic graphics demand realistic settings; tower defense games are usually not very realistic.
3. The low and close overhead camera promises to be very inappropriate for tower defense: the player needs to see the whole battlefield all the time, not one turret and one enemy. Seeing turrets well is (up to a point) unnecessary, but not occluding enemies is very important.
4. Satisfactory handling of jumping cars probably needs a lot of effort and a better physics engine than you have the performance budget for.

Is it all relevant advice? Am I really understanding what you are doing? Take criticism critically.

15 hours ago, EddieK said:

Yeah I agree with you. Constructive feedback would be nice, like what exactly the person doesn't like or what do they think is missing. I think I'm being a bit over sensitive at this point. After I release my game on steam, I think I will get much more negative feedback but hopefully at least with more details.

You can see some of the screenshots here:

https://www.gamedev.net/gallery/album/1966-tower-defense/

What do you guys think I could add to make it more appealing?

HI Eddie,

looking at the images, my 2 cents would be:

- on the base images, you've got what looks to be metallic shapes. Could you add a height / normal map to add some more 'visualness' to it, or even use tessalation?
- Same with the grass. This looks excessively repetitive, can you have a number of different textures and then randomise them? Dry patches? darker patches etc.
- on the item with the rings, the rings appear to be highly lit but then they cast a shadow, that looks a bit weird to me

Obviously neither of the above make any difference to the gameplay, but they would add to the initial appeal.

Steve

Say  this......... Thanks! I worked a long time to make it look like shit, I'm glad someone recognizes the effort I've been putting in.

Because my game looks so shit i need some crap music with it.

6 hours ago, Gnollrunner said:

Say  this......... Thanks! I worked a long time to make it look like shit, I'm glad someone recognizes the effort I've been putting in.

You're feeding the trolls and will get even more trolls piling on.  It's a complete and total waste of time for zero benefit, though if you want to spend time arguing with idiots online reddit or facebook are great places to waste time you could potentially be working on your project.

@EddieK True honest, assessment.. it looks like a bad prototype made by a programmer, and here is what you can do to help it when you are ready to focus on that part.

Look in Summary: The assets look like bad, cheap prefabs. That said, don't despair you have plenty of time to develop the art assets once you have the gameplay where you want it. I believe it is terrific to work with placeholders until you have custom art, and it just seems to be where you are at, so don't sweat it.

1. Textures: The current ones look like they have multiple themes and styles to them. You will want to work with someone to help nail down a consistent look and feel of what you have in mind, and maybe recruit a seasoned 3D/Modeler and texture to beef up the polish. Right now the textures are also super bland and generic, like a freebie repository download. I would suggest looking at some sort of low poly style if you can't find an artist to assist you. This way you can do some super low poly models and do colors on material fields and let post processing help give it polish.

2. Scale:  When I look at it I can't help but feel that the squares look like small turrets too big, and trucks look shrank down like ants. You could easily scale up your trucks to make them look more appropriate to size. While you are at it, make the turrets take up 90% of the square vs 35%. I know I just sounded like I contradicted myself, however, the point is you have too much negative space and not enough positive space in your scene. Which is making it look empty and dead!

3. User Interface: It makes me cringe, cause it mostly isn't there, and what is there well, frankly, looks like once again something done by a programmer without artistic experience. A simple and consistent font will go a long way with any text! Try to keep it uniform throughout all elements. Icons eventually can be the actual models themselves that are screenshots of your assets with a transparent background.                                                                      

Conclusion: it is far too early for you to worry about the cosmetics and what random butthole on Reddit has to say about your game. So what if it looks bad now. Nearly all games look like crap in the beginning, unless they are done by an artist, and then they just look pretty and play like a bag of crap, lol!

You gotta remember 80% of the internet is pretty much full of toxic critics that probably hate their own lives and everyone around them. So right now appreciate the 20% here to help you, and as you progress that number will swap places and that is how you know you are ready to launch.

Best of Luck, pm me if you need more help ^^

- Cody