Not sure exactly what you want so here's some ramblings (well this turned into a wall of text... )
There's some off the shelf you can buy, like Monoprice for under $200, however, beware buying anything too cheap and research first, this and many other ultra low end printers do not use common off-the-shelf parts. You can almost always replace the controller on these sort of printers, but the motors are not the norm. You want something with nema 17 motors (nema is a standard, not a brand) and a traditional hot end. With these you can pretty much repair or build a better printer and re-use these parts so make sure they are common parts. 3d printing is one industry where open source is actually leading the market instead of following it so while you may scoff at a home built or open source model, they can actually be better with more features, better quality, and faster speeds. Almost all of the best off-the-shelf are actually open source, I've seen quite a few people think people home build them for budget reasons when the truth is they're often better.
Since most use the same controllers and motors, they all tend to print similar quality. The biggest differences between printers is not quality or even speed, the real biggest difference is longevity, repair-ability and how much of a pain in the a$$ they are to use. Printers with higher resolution motors tend to be rare, but are slowly becoming more common, but the truth is, it's one of those cases of diminishing returns, so don't be sold on the idea you need better or more precise motors. I started with cheap used CNC motors on Ebay, don't do that, they run really hot. Today you can get a set of motors better suited for the same amount I paid but brand new. Get someones used printer or spares and you can pay even less. You should be able to get motors for $50 or so, even new.
Controllers make a major difference in configuration but once setup and running you rarely actually do anything with it, so while I DESPISE Ramps/Arduino, once up and running they work fine under normal circumstances. It may take a few hours to setup a Ramp/Arduino (which is the most common, even if under a different name) whereas I can setup a Smoothie/Panucat Azteeg or Duet in a matter of minutes, but what is your time worth, a Ramps setup can be had for under $20 on ebay, under $50 with an lcd and wiring harness (not a bad deal btw). Basically higher end controllers are easier to setup, possibly run faster (in a delta), maybe a little quieter, and in some cases wifi or network access... Basically just nicer features, but that all costs money. My Duet setup with lcd is fantastic and considered the best at the moment, but it should be at nearly $300, you can build an entire printer for that. So while nice, don't think that you NEED this. This is the single biggest place you can severely over spend, and the place it really almost matters the very least. There's legit reasons for a better controller, but once it starts putting down plastic it just doesn't account for much. Put your money elsewhere to start.
LCD, like I said, you don't need it. You can run the printer from a computer, or a Raspberry Pi, which is actually fairly common since it adds wifi to cheap controllers. Granted if you can get a compatible lcd in a kit with your controller for cheap and want it, go for it. LCDs are nice but many of us got started without them. If you need to save a few bucks this should be one of the first places you cut cost. It can always be added later or use a Pi, I really can't stress how nice that setup is. A Pi can not just run the printer through wifi but it can handle everything including slicing, letting you run the printer from your phone and even adding a camera to watch it. It's more useful than the lcd for almost the same cost.
Hot end... This is a BIG one.
This is one of the two single biggest pains the neck on a printer if you have a bad one, but you can buy 3 or 4 cheap ones for less than a single good one. A set of small drill bits can usually fix a bad one to work better, so what some have started doing is buying two cheap ones and use whichever worked best. Their quality has gotten better though so it's less a problem now. I've even seen people use a generic one and a higher end nozzle. This works out to about half the price of a good one with similar performance. Small nozzles ARE NOT BETTER. A 0.2 nozzle can do fine work better and thinner walls, sharper corners, but honestly those are not really noticeable. Smaller nozzles print slower, and the feel of quality is actually a product of layer height, not nozzle diameter. I usually run 0.4mm, but I've used 0.5 and the difference is quite small, but dropping to 0.35 is a nightmare as you are getting close to particle size that can clog a lot more. Start with 0.4 or 0.5 and if you buy a cheap 0.4 nozzle and it jams a lot buy a 0.5mm drill bit and bore it out (carefully, these bits are tiny!). Just doing this seems to fix a lot of the cheaper hot ends as a larger hole has less back pressure, I've seen this fix a bunch of them.
Extruder matches the hot end for pain in the neck. Smaller gears mean more torque, but also less grip so while they have power they can't actually exert it on the filament so beware. If you need one, I'd look at the BMG Extruder, they run about $40 on Amazon and are pretty much the best bang for the buck. These are a clone of a higher end product and these are the best of the clones as while they use small gears (actually called a hobb), they use two of them to offset the tiny contact patch. Do NOT go cheap here, seriously this will be the bane of your existence if it's bad, you will waste a lot of plastic, you may blame your hot end... And worse, it relies on software, hardware and you to tune as such it is often the cause of a lot of problems people just cannot figure out. Keep in mind this works in tandem with the hot end, so the 0.5 nozzle trick applies here as well. People think extruders are simple, I've designed multiple ones, it's easier to design a printer.
Style.
Deltas use fewer parts and rely on computing power, this is why a Ramps runs a bit slower on them, but typically a budget printer can't exceed these speeds anyhow. Deltas are fast, cheap and extremely quiet. I've seen a dozen deltas run more quiet than a single cartesian (box) and the tend to need the least maintenance. Deltas however are TERRIBLE for accuracy and are not known for making perfect circles so gears can be a bit tricky. Cartesians are great for accuracy, but tend to run loud, slower, and are more costly to build and maintain as they use a LOT of parts, they use mechanics instead of software. They make good if a bit pixelated circles (cnc is digital, not analog) and as these are left right back and forth, they can be a bit steppy, a delta will round off the digital stepping better. The other option is corexy and h-bot, these are sort of a mid point and are my current preference. They have nearly the same precision of a cartesian, nearly the speeds of a delta, able to round off the digital steps a bit, and often almost as quiet. Unfortunately they also can be the most expensive to build as they need everything a cartesian does, plus extra belts and (a lot of extra) pulleys, depending on how they are built they can require more or less maintenance than a cartesian and use a combination of mechanics and software to move. While you can build a budget delta, there's not really a budget corexy and they tend to be some of the most expensive home built/open source printers you can build with the best ones hitting almost $1000 before you even start adding electronics, motors and hot end. While they can be one of the fastest and quietest, remember what I said earlier, the final printed product isn't necessarily better by a large factor, unless you know what to look for you can't necessarily look at a print and say "this was made on a delta" or corexy. I can by studying it closely but for the most part these differences are miniscule, we're talking human hair size differences, unless you know and looking you won't know and your print settings make a larger difference, it's only when you really get to know them and the settings that you can really understand these minor differences, so build/buy what you want. If you've never seen a delta, especially in person, they are amazing to watch, while corexy is interesting, cartesians are boring, but you aren't buying it for looks so pick what you want. Remember, you can always change it later so don't think you want corexy because you need it start with what you can afford, buy they time you are able to get the benefits from something better, you will have enough knowledge to do so at very minimal expense. It cost me about $20 in hardware and $20 in plastic to rebuild my first delta into a new, different delta that worked far better.
Again, don't get a heated bed to start, this will save you a bundle on the heater, glass and power supply. If a delta, build it to use cheap locally sourced mirrors, if it uses a square bed and no heater you can get cheap glass at the local home improvement store. No joke, I once got an 8in bed glass for 89 cents, it was cut from scrap. Heated beds require special glass (normal glass can work but will crack in short order), which is actually quite brittle and expensive. You also need a lot of power, and it will turn your printer into a space heater, seriously, you will be dumping 200 or more watts of heat, just dumping it. You don't need a heated bed for most plastics, often even when the manufacturer claims you do, yes, you pretty much require it for ABS, but for most others you don't unless you're using more exotic plastic, in which case you need a more exotic hot end and these plastics are less than budget friendly. There is also ways around needing a heated bed even when they are "required" but frankly, a heated bed is just more settings to deal with. I've even seen it create issues, the whole point is for the bed to created this warm zone around the print keeping it from shrinking as you print but it can leave the print soft during printing and cause it to wiggle but I've also seen it keep it from shrinking for a short time and then massive shrinkage for the next 1/4 inch causing an hourglass effect on the print. KEEP IT SIMPLE, you don't need it to be complicated. Start without one and just use PLA and PET then move onto other things.
Don't get struck thinking you need a $300 controller, the majority of printers use an Arduino Mega 2560 with a Ramps shield (even if they change the shape and name). It's cheap, it works. Avoid connectors if it moves, I've had them fail in less than a week. If you do have any connectors or soldered joints and things act funny, start there. Make sure it has a probe, if it lacks one, make one. Manual calibration sucks.
If you haven't figured it out, 3d printers are modular and interchangeable. If you don't like something, change it, don't like your entire printer, change it. You can start with a delta, buy a few parts and build it into a cartesian, a few other bits and it's an H-bot, reroute a few belts and you have a corexy. I've seen people start with off with a broken Chinese printer they got for $60 dump another $40 into it and make it function quite well. If you later decide you want or need a heated bed, add one, think you need dual color or multi-color add it later. Don't spend your money on this stuff until you know you need it. I've met sooooooo many people who think they "need" a heated bed or dual extruders who later find they actually don't want the hassle or who bought it and found they've used it once. Sure, these things can make it easier, but 3d printing is about creativity and there's multiple pays to create your way out of a problem so don't rush into expensive upgrades. I started with a heated bed and later removed it, it was of limited benefit with a lot of hassle.
Get in, get your feet wet, THEN decide how to proceed. You may think you know what you want but you don't really know what you need. Wait till you need it.
Like I said, it's cheap to change and everything's modular, so anything working is a good starting point. I like to tell people, your first printer is going to be the most expensive, once you have one you can make more or make it better. Most people who really get into this wind up with 2 or 3 printers in due time. One is usually smaller or larger and often a different style and with different options, and they always try to make sure at least one is operational in order to make parts for the other.
Most find that it's better to have two printers that can do different things than it is to have a single printer than does everything (keep it simple!), plus their needs change, they like to just change things up and it makes it easier to keep them running if you have two. Don't let that scare you though, like I said, the first is the most expensive. What many do is get their first, upgrade it, then build an open source one to their specs, rob the first of all the good parts, put the original parts back on the old one and buy just enough to get both running again, then the cycle repeats until they have a bunch of spare parts laying around or they sell one, or offload parts to people trying to build their first. At this point there's a lot of parts out there unlike when I got started. Check local hackerspaces, Meetup and Facebook for printer gatherings, there's bound to be plenty pf spare parts if you ask around. Robotics competitions and groups are also good for this.