Recently read through all positive posts about HQ Player, and noticed that many in the Roon community have been clamoring for HQ Player integration for months. This recently caught my eye.
After purchasing it, I quickly discovered these limitations:
1/ HQ Player is a seperate process from Roon, e.g., a network end point. This means audio buffers are needlessly copied between Roon output and HQPlayer input FIFOs adding unecessary latency. HQPlayer should ideally be a shared library (plugin) which is loaded into the Roon process.
2/ HQ Player, and probably Roon as well, does overly conservative buffering to avoid network drop outs. This, in addition to extra out-of-process buffer copying, bumps up the latency (I experience +5 seconds). This latency is so high, that you can only just barely use the transport controls in Roon. There should be a bi-directional throttling/QOS protocol which dynamically adjusts FIFO depth in both apps - it can’t be a static, high value.
3/ HQ Player doesn’t appear to run headless (at least I haven’t figured out how to do this yet). This means you have two Player UI applications up. Let’s just say, generously, that the UI’s don’t look like they were designed in the same century.
4/ Roon looses track of the internall state of HQ Player way too easily. There are probably multiple issues with both apps that need fixing here, but HQ Player should export an API to Roon that let’s it “query” the state of the player dynamically -and/or- there should be an async notification sent from HQ Player to Roon whenever its state is updated. For instance, you can simply bring up the preferences in HQ Player during playback (without saving or changing anything) and Roon will assume the worst and drop the connection.
5/ Roon apparently can get various properties from HQ (SR, filter alg, …), but it would be nice to be able to select the dither + filter alg and other parameters without having to step into HQ.
6/ iOS iPad app doesn’t know about HQ Player properties yet, but I’m sure that is just waiting for Apple app stiore approval.
In short, I really wanted this to work out, but I am disappointed enough to post this here. I really bought into the hype on the forums regarding HQ Player, so Its easy to get swept away by the descriptions from fellow audiophiles claming NS5 dithering makes all the difference in the world. 
Personally, I was hoping I could run convolution IR room correction with Roon using HQ, but I can only speculate on the additional latency impact there. To its credit, Roon Labs responded to a great deal user pressure - that speaks highly of them - thanks for listening - but what we have currently is a very minimal , “let’s get the monkey off our back”, kind of solution which is really not in spirit with the polish in rest of the app. This is an intriguing “proof of concept” implementation.
cheers.
