Your experience & thoughts on SSDs for MacMini

03-07-11: Perrew
I thought regular HDDs did sequential (what most programs do) loadings faster than SSDs?

That was true in some cases a few years ago, but no longer. As a rough ballpark, I would expect a typical SATA2 SSD today to provide sustained read transfer rates of around 200 MB/sec. I'd be surprised if the stock HDD in the OP's several year old Mac Mini could do more than around 50 MB/sec, and perhaps even less.

And of course in terms of random access times, which are particularly important for an operating system drive, the SSD would probably be more than 100x faster.

Regards,
-- Al

03-07-11: Perrew
I still kind of fail to see what the SSDs do for sonics, are they having lesser electrical noise?

If the music data is being output from the computer directly into a dac, via usb or firewire or s/pdif, it would certainly seem conceivable that replacing an hdd with an ssd might affect (and hopefully improve) jitter levels and noise levels on the signal supplied to the dac. How much of a difference that would make would obviously be dac-dependent.

If the music data is being output from the computer via ethernet or wireless, and assuming that the link works reliably and consistently, I would be surprised if going to an ssd would make any sonic difference. Although I would not totally rule out the possibility that rfi effects could come into play in some setups.

Regards,
-- Al

03-08-11: Perrew
When you optimize the computer for music playback the access speed of the drive might be a non issue so controlled test with SSD vs. HDD should be made

Agreed.

One easy way would be to take once Mac mini with regular internal HDD and connect an external SSD. If the external sounds better we might extrapolate that to an internal SSD sounding better, but if the external SSD sounds worse or equal it might be the interface connecting the drive.

I doubt that any meaningful extrapolation could be made. Regardless of where the music files are stored, the internal drive will be in use supporting the operating system and programs. And its circuitry will still be present and active, generating noise and causing (to at least a small degree) voltage droop, voltage fluctuations, signal crosstalk, ground-bounce, and other effects that could ultimately affect (to at least a small degree) the jitter and noise levels of the output signals of the Mini. All of those effects will differ between an ssd and an hdd, with little predictability.

Regards,
-- Al

03-08-11: Aplhifi
HDD jitter? No such thing, IMO! And let’s not forget that the SSD also has a SDRAM buffer, just like HDD.

It does not matter where your audio file is stored; it is always double buffered (at minimum); first in the storage device own memory buffer, and again to the main system memory. So I do not see a reason for SSD to sound better than HDD.

Alex, no one has mentioned "HDD jitter" in this thread, which I agree is a nonsensical expression. A question was asked as to how changing from an HDD to an SSD could affect sonics. I responded that jitter and noise on output signals of the computer, that will be connected directly into a dac in many setups, could conceivably be influenced by crosstalk, ground-bounce, voltage droop and voltage fluctuations at various circuit points, rfi, and other similar effects that in ANY digital design that involves a large number of signals that have fast edges can result in subtle interactions between circuits that are ostensibly unrelated. Particularly when the destination to which some of those signals are sent (the dac) may be especially sensitive to small amounts of noise and jitter.

I take no position as to the likelihood that using an SSD instead of an HHD will make a sonic difference in any given computer, or as to whether the difference, if any, would be for the better, or if any such effects would be consistent across different computers, because I have no experience upon which to base such a position. What I have said is simply that it is technically conceivable that there could be a difference, in at least some setups with some computers and some dac's. I don't think that is disputable.

Best regards,
-- Al

03-10-11: Perrew
Its the first post by PeterST Im referring too, he is Not advocating what you say thats what Miska says, Peter says to deal with all small spikes.

Yes, that is the post I was referring to also, the one entitled "SSD" and signed "Peter" at the end. As I said, I found it to be somewhat incomprehensible, so it could very well be that I was misinterpreting the point he was trying to make.

All Im trying to say is that the story on SSD having superior sound is suported by some but some others dont support it, so I figure if you really wanna know what your own ears detect youre gonna have to give it a go ....

Agreed 100%.

Regards,
-- Al

03-09-11: Jylee
SSD may sound better than magnetic HD, but it's largely irrelevant today because there isn't any SSD large enough to store the music collection. I have 600GB+ of music from CD, DVD Audio, and HDTracks. It's not practical to build a disk array to store all the music.

Jylee, your statement is all certainly true (a 1TB SSD costs around $3000 these days), except for the part about irrelevance. An aspect of the question about sonics that was addressed in my posts and some of those by Darrell concerned the possibility that sonic improvements might result from replacing the computer's relatively small internal drive with an SSD, while utilizing a large external HHD to store the music files.

As you have seen, Darrell described replacement of the internal HHD with an SSD as a "no brainer." And my comments provided technical rationale for why doing that MIGHT produce a sonic improvement. That rationale applies regardless of whether the music files are stored on the small internal SSD or on a large external HHD.

Regards,
-- Al

03-10-11: Perrew
Here is an interesting take by someone saying SSDs cant improve sound....

After reading it (the article on the lower half of the page) several times, I must say that I have very little idea of what he is saying :-)

I think that one of his main points, though, is essentially that noise produced as a result of activity of the cpu chip, under software control, is the predominant source of internally generated noise in the computer. The reason being that it draws more current than any other device in the computer, and rapid fluctuations in that current (which are primarily responsible for the noise generation) will be greater than for all of the other devices in the computer, including the digital chips in the SSD or HDD (an HDD has significant digital circuitry, in addition to the electromechanical stuff). He appears to conclude, therefore, that all internally generated noise that is not produced by the cpu chip, and all other effects that may be caused by the rapid switching of digital signals elsewhere in the computer, is insignificant.

As someone with extensive experience designing high speed digital circuits (not for audio), that all strikes me as meaningless oversimplification, of things that (due to their complexity) are inherently unpredictable apart from experimentation.

Regards,
-- Al