It certainly depends on what’s „big“. For audiophiles who care for small but sometimes essential differences when they sum up, tiny can be big 
I recognize significant improvements in top end fine resolution, midrange ease and ambiance reproduction. But only when the recording itself inherits the potential.
Some Hi-rez downloads I bought from HD Tracks do not sound as good as the same CD. Buying HD downloads is hit-and-miss, so I don’t buy anything any more. And I told them so. If they had confidence in their product they should provide a refund if the CD sound better, but they don’t. They don’t care.
But this then is a matter of different mastering. Hires is never anywhere as relevant as mastering. Mastering is the one thing one always has to take care of when interested in sound quality, and the improvement of sound quality by hires is also depending on labels, specific recording successes/failures and genres.
So as you say it’s hit and miss, but I can tell that I personally can care for 95% hit in my case because when taking care of mastering, the worst to happen is, that the recording is not good enough to show improvements by hires, it’s just the same sounding as CD then, which is frustrating, too I admit, as it was money spent in vain.
If you’re into Jazz (especially pre 1970) and realize the huge number of available better masterings done for vinyl than digital, this is the major meaning of listening to vinyl generally imo.
I don’t care. As a buyer of Hi-rez audio I expect it to sound better than the same CD. Simple as that. When it doesn’t, and the merchant doesn’t care, I stop buying.
I can understand this point of view, it’s somehow as if you expect a component with a more expensive casework to sound better than one with a less sophisticated casework presuming you want to avoid the effort of comparing their sound quality.
If you think like that with the expectation to also have best possible sound, I guess you end up being frustrated very often when buying hires or an expensive component with fascinating casework. IMO the only solution would be to buy the least expensive (as you do with avoiding hires) so you’re sure you might just miss something, but you’re also sure you didn’t pay too much. This is fine for most people I guess…as long as they don’t care for the best possible sound quality…and for most a certain sound quality standard is good enough, so they’re fine with that.
In any case, if you want it best, you have to put some effort in it, it’s always like that.
Only buying when you can get some kind of provenance, or only from sites that give some details you recognize ((re)mastering engineer, a company you trust, etc.) helps a lot. Checking your other good disks (or downloads, etc.) for their engineers, etc. might help. I’ve never been disappointed with the quality of Pentatone, Telarc, the RCA rereleases (and Sony for that matter)… Most material that was originally 24/96 (often listed on the label) is good to my ears. I buy a lot from Pentatone, Music Direct, Acoustic music, Blue Coast Music (tho their catalog is smaller and more esoteric), Native DSD, subscribe to http://www.dsd-guide.com/ check http://www.sa-cd.net/ I like the more recent Channel Classics. Opus 3 is always excellent quality… I’m sure I’m missing a lot of others. It’s relatively easy to find a lot of music samples on the web if you want to check the non-sound quality features of a recording, artist, etc. (Amazon samples come to mind, as well as samplers from most labels, sites, etc.) Subscribing to various sites will get you access to a lot of samplers and you may find you agree often with one or more of their reviewers.
I’ve gotten a few duds (including the Paul McCartney releases) but usually that’s when I don’t do a little research.
For me the early SACDs were almost always good: the various companies wanted to show off their better recordings or the better technology. I bought many genres I might not otherwise have sampled and was rarely unhappy with the music or quality.
This is the exact case for Spotify as a tool to pre-screen albums prior to purchasing. If it sounds good in Spotify, the CD is extremely likely to also sound good. That’s how I have been slowly rebuilding my CD collection with better recorded material one album at a time.
Yeah, those sound great. I should have been more specific and mentioned a label such as Deutsche Grammophon. Many of their releases from the 1980’s and 1990’s sound a bit brittle to me.
Deutsche Grammophon recordings often exhibit a drier, precise sound. I long suspected this is a result of the tremendous amount of multi-miking they do.
Last year I discussed this with one of their recording artists, Paul McCreesh (founder and Artistic Director of the Gabrieli Consort & Players, etc.) during a break in a rehearsal of of the St. Matthew Passion. I asked him why his Deutsche Grammophon are warmer than their typical output. He explained this is what he wants and he limited the multi-miking to a few necessary spot mics (e.g., on the timpani).
Bis high resolution recordings are always excellent. 2L also.
Agree. I was introduced to Spotify by my daughter and now it’s burning a hole in my wallet as the pace of building my CD collection has increased.
Would not the CD always sound better given that Spotify is lossy data compressed?
It’s referring to the quality of the recordings. If a recording or mastering is poor, the CD resolution won’t fix it. But Spotify to screen before buying the CD will save one from plunging cash into the unknown.
It is the CD is “likely” to sound good comment which confuses me.
If it sounds good on Spotify, would not the CD always sound good?
INot always. Often times a CD varies in quality for the exact album, the difference being when it was produced. There are CD albums that were produced from the master recordings and others produced later who are copies of other copies and sometimes labeled as “remastered”, keyword for me to stay away. Example: Dark Side of the Moon, Faith by George Michael, I own both first CD releases and compared those to recently manufactured CDs that I gave a friend a gift. Even he could tell the difference when we popped his CD on my system. They look identical in the jewel case and labels, except that mine is a bit “aged” on the paper. His had less dynamics and it sounded as if the top end had been taken off. Just a fullness that we looked at each other and said what the hell? We then listened to the first release of the SACD of Dark Side of the moon, and although it sounded better than the recent CD copy, it did not sound better than my oldie CD. How does Dark Side of the moon sound in Spotify? Horrid. And I can tell what CD may have been ripped to stream it. Yet the Faith album in Spotify sounds good. So screening helps.
Hmm . . . so Spotify is actually of little use for screening as one never knows the source of the streamed file. The CD one purchases may be the same, better, or worse than the streamed version. It is crap shoot.
It appears Spotify would be of use only in determining whether one likes the music itself.
(I believe we all know different CD releases may sound different.)
Still very usable applying my method. I purchase my CDs from Discogs and mostly the old stock new, and older used ones with vastly positive results. The gift to him was from a local shop, very shiny new stock which probably ramped up in production after George Michael’s death.
Yes, as you point out, to discover the music itself before buying. When in doubt, go with the oldest released CD.
Your method will work if one assumes Spotify plays the worst sounding available. Then, if it sounds OK, anything you buy will at least not be worse. 
Some new releases sound better than the old; some, worse. I continue to be surprised at how many sound less good; no one starts remastering project with the goal of producing an inferior product.
But it happens more often than it doesn’t. Giles Martin wanted Sgt. Pepper to sound modern. It sure does, modern bleck with a DR of 7. One chance in a life time to re-mix and remaster a classic lost.
