Does old mf lenses give better image quality compared to new kit lens?

You'd have to test it, or find tests others have made with those specific lenses. "Old MF lens" spans a huge range of image quality; there is not a single answer for all of them.

The test I like to use is to print an ISO 12233 still image test chart onto two pieces of letter-sized paper, which gives about an inch of overlap, so you can choose where to cut or fold one to mate up with the other. (Ideally, you'd use a wide-format printer to print it onto a single sheet, but I don't have access to one.) Print it at the highest resolution your printer supports onto high-quality matte paper. Tape it up to a vertical surface, and light it well.

Once, I did this test by taping the chart to a portable whiteboard and putting it on an easel out in the sun, because I needed to test a small telescope, and couldn't get it to focus closer than the longest free-space dimension inside my house.

Put the camera on a tripod and use either the camera's 2-second self-timer feature or a remote shutter release to avoid camera shake from the button press. Fill the camera frame with the chart.

Minolta MD Rokkor 1.7 50mm (or other similar lens within the same price range)?

Here's an ISO 12233 test of a similar lens, which probably gives image quality a bit better than the lower-spec lens you're looking at, at about 2× the cost, an SMC Pentax-M 50/1.4:

(Click image for full-size version.)

The "M" designator means it was from before Pentax cameras even had auto-aperture features; it came out in 1977.

Note the significant barrel distortion and color fringing. At full-size, you will also see a fair bit of blurriness.

Now, here is a different 50/1.4, Canon's design made 16 years later:

There is more barrel distortion in this lens, but there is no color fringing, and it's sharp clear out to the edge of the frame. This is a much better image, overall, because a bit of barrel distortion won't ruin most pictures, and you can correct for it in post if it's an actual problem. You can't fix blur and CA nearly as easily in post, and doing so loses more image quality than fixing a simple distortion like this one.

Now, simple image quality aside, you also have to decide if you're willing to give up on autofocus and auto-aperture.

As for autofocus, I came up on a manual-focus camera, so I thought I'd have no problems doing this...until I ruined a series of family portraits by focusing just behind the group with the aperture opened wide enough that the subjects fell out of the zone of acceptable sharpness. In hindsight, I could have turned on my camera's 1:1 zoom feature to get a pixel-level view on the camera's rear screen, but I didn't know that at the time. And, knowing it now, I realize that this means I'm going to take even more time to take those shots I still take with my collection of manual-focus Pentax lenses.

Auto-aperture is less of a concern, since rotating an aperture ring is scarcely more difficult than whatever control your camera already uses. The main losses are that you don't get visual feedback of the selected aperture on the camera's screen, and it doesn't get stored in the image's EXIF data. On the plus side, it means you're operating in DoF preview mode full-time. :)