I think you are falling into the same trap I fell into and looking at old information. The post you linked is over 2 years old. Yes, afaik Roon still runs on .NET on Apple Silicon but that is now two years’ more mature than it was when those posts were written.
I agree that ROCK is a very elegant minimalist solution (iff you only want to run Roon, I want to be able to run Minim Server as well), but the price/performance of the M4 is pretty compelling. As far as I can see a comparable Gen 14 NUC would cost more than twice as much as the basic Mac Mini M4 and generate something like 10x the amount of heat at a comparable level of performance. Of course that level of performance isn’t necessary just to run Roon but if you want to run HQPlayer and do funky stuff with filters etc. you might have use for it.
So while both approaches have their pros and cons I think the M4 Mini has an awful lot going for it.
Can I just flag up that Roon Core / Roon Server really doesn’t need much in the way of horsepower to run …
I’ve installed ROCK on a number of 8GB RAM / 128 GB m.2 SATA SSD HP T630 Thin Clients (£35-£50 off eBay) for several friends / colleagues and I’ve never been told of anyone that I’ve set them up for having issues.
I’ve even got Roon Server running on an HP T630 of the same spec as a docker container under CasaOS along with MinimServer (also as a docker container) and CasaOS generally doesn’t show more than 3GB of RAM usage and 20% CPU usage with occasional spikes.
I’m sure that it is possible to swamp the T630 with DSP effects or whatever but I’ve had that streaming high res audio to four separate Roon endpoints at home with no audible hiccups or glitches.
I’m sure a little HP Thin Client doesn’t carry much HiFi credibility and maybe something like an HP EliteDesk Mini G1 or G2 (again, can be had from eBay quite cheaply) would be more suitable (I have used those for ROCK installs too and they’ve behaved impeccably) but obviously they also don’t carry any HiFi credibility.
I’ll leave it to those more able than myself to debate the sound quality implications of running ROCK / Roon Server on different platforms…
Well my sonicTransporter i5 (purchased 2016) with 8GB of RAM chokes when playing downloaded DSD128 files and applying my DRC FIR filter. While this is certainly not the simplest use case it is something I want to do, not just a contrived edge case designed to break the system.
Hard to know what the bottleneck is as sonicOrbiter has a simple graphical CPU monitor but no memory monitor. Individual cores can be seen spiking up to 97% or so when it craps out, although average CPU load across all 4 is generally around 50%. I actually managed to catch it a bit higher in this screenshot:
Yep! many of us like to use the same box to do other things. I looked at the price of a NUC and was shocked! As @Phil noted Roon is a trivial workload (unless you are doing complex DSP).
Sure you can roll your own or purchase a vintage PC. But my original point is: at $600, one can have a modern device that is stylish, compact, generates virtually no heat or noise, with a well designed power supply… for less than the cost of a NUC
Even an M1 is light years ahead of my vintage i5 (I’m guessing it is probably a Broadwell/Gen 5 - will have to open it up and take a look). Problem is there is no way of telling what hardware I would actually need for my use case other than trial-and-error.
As I am a Mac user and know my way round MacOS it seems like a good option for me, and the price/performance of a $600 M4 Mini is simply off the charts. The icing on the cake is that it runs cool, the M4 has a TDP of only 22W. It’s really quite a remarkable feat of electronic engineering.
Like you, I have what Apple calls a ‘vintage’ i5 mac mini and it runs Roon without breaking a sweat. Activity monitor uses more resources than Roon most of the time! The Mini idles a less than 10w power consumption. I’m hoping the M4 will be even lower
And I came across someone only this week who compared a fairly “hot ticket” streamer with his incumbent Mac Mini, expecting/hoping to be wowed, and reported that in listening tests the Mac Mini stole the show. Expectation bias override right there!
Thanks Nigel but I hope that you don’t object to my requesting further details so that I can begin to understand what is being compared to what.
I understand that there was a “streamer”. Have we any idea what it was?
2 Was it a streamer or a DAC streamer?
What was the source or sources being used. Are we comparing two UPnP programs or program from a streaming service or services (or both).?
If UPnP, does the streamer have storage or was it fed by an external NAS or other source?
The mac is a general purpose computer. So what music source was used ( or sources)? Or is Mac being used to support Roon core instead ? Is a streaming service’s software installed ? Is a media player software installed? Is the mac being used or NAS storage with the output being sent to an external DAC?
Something else altogether?
Did the comparison involve identical interfaces for both?
am sure that other factors may be pertinent but my main issue is that until we know the answers ( or at least some of them) the comparison has no credence. That is not to say that it may be incorrect just that there is not enough information to warrant further thought.
Apologies @PAR Pete if you would only find this useful if I could answer all those questions but I’m afraid I can’t. I am also conscious that the streamer in question is both relatively new to the market (less than 12 months), has a pretty strong pedigree and is still being evaluated by the purchaser; some of the questions you ask relate to the aspects of the comparison being further examined. I think power supplies are also on the list of variables.
Indeed, he streamer was bought as an obvious upgrade from a “general purpose computer” but turned, initially at least, to not be so.
Re (2): dedicated streamer vs Mac Mini into same DAC.
Re (4): Mac Mini not Mac. No onboard storage in use. No Roon AFAIK.
I do not own and have never used a Mac Mini myself but in the context of the preceding contributions from fellow travellers I thought my (apparently incomplete!) contribution might be helpful to some.
Thanks again Nigel but I am afraid that at the moment it is similar to reporting that I met a guy who had compared apples to bananas, decided apples are better and somehow treating this as some kind of “ proof”. Or at least in support of the concept😟.
Currently I don’t even know how Mac Mini is being used as a music player or in connection with this function.
No it’s not, it’s like someone saying that, like previous correspondents who enjoy apples, they met a guy who similarly enjoyed apples and indeed preferred them to the supposed upgrade to oranges which he’d just bought a bag of. He didn’t say apples are better than oranges and no-one challenged him for proof to back up an assertion he didn’t make…
I don’t think this is a particularly fruitful exchange (see what I did there?). You’re obviously entitled to ignore any post you don’t find interesting or helpful. This would clearly include mine!
Similarly, I can’t remember the context other than it being a business meeting. A participant suggested we were comparing apples and oranges but to me this felt like a gross understatement; I suggested it was more like comparing apples and Wednesdays. As you can tell, I was quite pleased with that.
