Wondering if anyone compared SSD to spinning hard disk.
Wondering if anyone compared SSD to spinning hard disk.
I've done this on a very limited scale. I have no data, measurements, or scientifically approved testing procedure to back up my opinion either. In my experience an SSD sounds better with my Mac Pro and a Zalman fanless PC running Windows XP. Other people I know have the same opinion and have conducted some extremely basic A/B tests with mastering engineers and product designers. Again, not empirical data. Just sharing my experience with SSD v. HDD.
For me, it is still a struggle to let my trusty Sony SCD-1 transport go because I still think spinning a CD on it and sending it to my PS Audio DAC sounds better than HDD. The difference isn't too far off so I am kind of willing to take that trade off for convenience of HDD. With SSD in the picture, I was hoping it may hit that sweet spot so I can retire the Sony SCD-1.
Is the PS DAC my weakest point?
I'd say your computer setup must be the weakest point. There's no reason a DAC will kill sound by itself, but how it is connected does matter. No further explanations ... just logic (ok, to me :-).
I don't think you should solve this by means of an SSD. To my own experience it (somehow) detailes out everything, and when you already now can hear the PC making things worse, it will get even worse. The exact thing happened to me, while at the time I did not recognize that "fact".
Is it possible that adding an external clock, like the Antelope OCX, would eliminate the difference between SSD and HDD??
Other than timing and jitter, what could it be? The computer is sending a timed stream of numbers to the DAC. The cable between the computer and the DAC could affect the jitter, I guess. But if you re-clock the stream with an external clock before it leaves the computer, I would expect that you would cover up any differences between SSD and HDD.
Antelope OCX looks interesting. I wonder if it will improve things. Apogee Big-Ben is simular right?
Will the Isochrone DA be a more ideal product? Neither one have optical in so how would you connect it to Mac Mini?
I used the Antelope OCX and really liked it. Unfortunately one has to manually change the sample rate on the actual clock to play music with different sample rates.
Chris, I switched to Vista and am very impressed with the new sonics. Have you considered Vista or 7?
So, based on the smattering of feedback on the web, and the latest feedback from the Symposium, it seems that SSD's sound better than spinning disks. Makes sense. Implementations like the Memory player, etc are doing everything they can to load into RAM or flash drives, etc. Questions:
1) At the Synmposium, were the SSD's handling the music, the OS boot, or both?
2) Since SSD's are still somewhat pricey, and if the real advantage (assumption here) is storng music files on them, then is it a legit short term solution to store "the good stuff" on the SSD (say the best 128 GIG I have) and let the mediocre redbook rips continue to reside on the normal spinning drive..........or do you want the processing, the sound drivers and the player software, etc. to be on the SSD if you were forced to pick your most intolerant 128 GB of your 1TB collection and system files? Thx
like cMP/cPlay or XXHighEnd.
Has anyone determined that Amarra is not a memory player? I thought someone said it wasn't, but when I read the description below in quotes from the Sonic Studio website, I wasn't sure. Amarra seems to imply it's doing a lot in memory and I know there is significant latency involved between Amarra and iTunes. And several folks have commented that Amarra sounds better with more memory.
"The first task is that the audio file, represented as digital data, must be read off the media and into memory. This
requires the audio player to guarantee (as best it can) the audio will be read from the disk in time. The player needs
to deliver the audio samples to the Hardware Interface not only on time but also synchronized. Otherwise, we may
hear clicks, dropouts, phase, and balance problems. Sonic addresses this using optimized file handling routines.
We do not look at latency as an issue in a home audio reproduction system unless it's being synchronized with video.
Once the audio is in memory, it needs to be converted to a format the computer can "understand". Current audio
formats for uncompressed audio use a 16 or 24 bit sample and this must be converted to the IEEE floating point
architecture in use by most computers today. This conversion can introduce noise into the audio signal. Should we
truncate, round, scale or perform some combination to achieve the best sounding result? This conversion occurs on
input from the disc and on output to the hardware interface. Based on our experience, we sometimes find that
textbook math does not mean the best sounding math.
The next stage is the gain structure and processing that may take place inside the audio engine. When you adjust the
slider to control the volume a gain process is applied to the audio signal. Even when the gain is at full volume
processing may take place that effects the sound. In the Sonic Studio Engine we perform all calculations using 64 bit
extended floating point math. This allows us a full 56 bit mantissa which allows us to keep the noise levels below the
24th bit found in high resolution audio. When any processing is applied it is important to redither the sounds to mask
the effects of all the rounding and noise that were previously introduced. To address this problem, SSE comes with
two dither algorithms which are optimized for use with Amarra. Perhaps as important as the underlying math is the
efficiency in implementation as each operation has the possibility of increasing the noise."
I was under the impression all (certainly most) DAWs loaded the whole audio file into memory before any editing could begin. Now as Amarra is based on the Soundblade system I've always assumed it does the same.
This could account for better quality audio from SSD even when using "memory" players if the computer is running out of memory or otherwise using virtual memory.
I had never had the opportunity to compare them until the symposium. The difference was clear to me. The SSD was more refined and more "relaxed" sounding, a quality Keith Johnson ascribes to higher resolution. I have no idea why it would be the case. It sounded vaguely like the kind of difference you hear between SACD and red book CD. Although the audiophile difference was clear, it almost sounded polite if that makes any sense. I have heard a number of SACD's that I didn't like as well as their CD counterpart, but I have no idea what other variables were involved. Sometimes the CD sounds punchier and more exciting, even though it has warts.
Some time ago Gordon Rankin reported changing the drive in his laptop to SSD and getting improved sound, even though his music was stored on an external spinning drive.
Here's an interesting link from Pacific Valve concerning their new CyberServer with the addition of a SSD. He presents three videos that explains not only this new item, but the inherent problems with jitter and PC's. Some of his comments will surely spark controversy but it's interesting nonetheless.
What do Amarra, iTunes, hard drives, SSDs, different coaxial digital cables and different toslink optical cables all have in common?
They can each be a part of a bit perfect computer music system, transfer bit perfect data and yet result in different SQ. Imagine that, same bit perfect data, same zeros and ones, but different sounds.
Maybe the Amarra skeptics would disappear if Amarra was only $100 or better yet, free. Then all skeptics that have Mac computers would use Amarra and never look back.
On to the next debate.
I'm not sure I follow some of this thinking.
If the 'bit perfectness' is referring to a transmission of a digital signal(image) from 1 medium (hard drive) to another - input to the DAC then assuming that there is no buffer overflows (data loss) in the path this can be done with 100% accuracy. I have personally done it over thousands of miles (across the atlantic maintaining synchronous connections between 2 continents with ps accuracy) and at speeds in the Gbps range e.g. Hypertransport and FB-DIMM interfaces.
So I have not internalised why SSD's would make any difference. - I did set up a RAM disk and tested unmodified data transfer and there was no difference.
If however the signal is altered e.g. converted to Floating Point, convolved filtered etc.... then the poles/zeros of the filters will of course change the nature of the music. For fun take the output of itunes feed it into any number of the mastering AU's or VST's (which of course is what happens when the music is intially mastered). One can get reasonable phase coherent room equalisation, tube effects, stereo widening etc... Then the bit perfect nature of transmission is mute - then the classic space/time ... FLOPS/memory performance of the computer matters.
Note I am willing to believe that if the DAC is poor the system degrades to a 'streaming' situation then all jitter becomes statistically cummulative then all sources of jitter will have an impact.
On all my systems - Win7. Snow Leopard and Linx. No ifs or buts. Plus it's dead quiet. I do not not know why it sounds better.
And it's getting cheaper by the month....and what is $400 or so in audioland? drop in the bucket.
What he said. Don't know why but I am confident in what I hear, with a variety of competent dacs.
It makes a huge difference. I think it is primarily in timing. Being that everything I play is bit true, it seems that there must be something with timing.
Everycomputer I have now has an SSD.
BUT NOTE! Not all the SSD's are created equal. Some are only 2x the speed of spinning disks. Some like the SuperTalent UltraDrive or the OCZ Vertex are more like 10x the speed.
With operating systems that swap virtual memory constantly the speed difference can make a huge difference in sound.
Not to mention the load on the power supply which reduces airborne noise and cable noise. SSD drives can eat about 1/10 the juice that a rotating unit will eat.
Hi Paul - I definitely hear you. Your comments are shared by many learned and logical minded readers. Since this thread was started I've had a lot of time to compare HDD v. SSD on very resolving systems. I also demonstrated this to almost 100 people at the Computer Audiophile Symposium back in June.
The results are nearly unanymous every time. There is a clear improvement in sound reproduction when SSDs are used. I can't wait for the cost to keep coming down so people are more willing to try it themselves.
Also, I noticed you are using Amarra. I really like it and won't use my Macs without Amarra. However, don't your comments about bit perfect transmission ring true with software as well as hardware? i.e. If iTunes alone and iTunes with Amarra output bit perfect audio how can there be a sonic difference?
Just an observation :~)
Firstly congratulations on the site; what an exceptional contribution.
At the moment my setup is 3*1Tbyte drives Apple Soft RAID0. This gives me a measured 130Mbytes/sec (using Bonnie+) bandwidth. This Volume stores music - about 600Gbytes AIFF.
I also arranged the music on the volume such that file reads are contiguous i.e. the filesystem does not have to seek to multiple cylinders to recover data. (no fragmentation).
A regular SSD has a continous read b/w of 250Mbytes/sec (it also kills HDD on Random Access - which I avoid by data placement).
So if the B/W to disk is an optimisable parameter the SSD outperforms my softraid by 2:1. Note that it kills single HD especially if the music has been fragmented by the user - I wonder whether this is what people hear?
So now I have on order a Highpoint RAID card. This will give me 750Mbytes sec to my 3 disks - I'll test that and post the results. This is 3x the bandwith of the SSD for around $400 - won't beat the SSD on Random Access but will on contiguous read access.
I also have a couple of my engineers comparing the performance of PCIe SSD solutions from Fusion-IO. These systems give us 100k IOPS. The context is in the world of Supercomputing and Next generation Internet compute and storage architectures (my business) - I'll snag one for a weekend and see if I can get any difference.
My comment about bit transfer applies to s/w and h/w. I don't see any sonic difference between Amarra, Itunes, Logic Pro, Songbird, Play this is what I expect - but I so want to believe Amarra is better.
When Amarra upsamples then I expect to see differences. There is a huge difference in my system when I force the Lynx card to upsample.
My question is a serious one and not just me being a smartass. When you said you don't expect to see differences, were you talking about literally seeing a difference along the lines of spec type performance, or that you don't hear a difference when you use Amarra? Just curious.
I don't hear a difference. I can force a difference by switching on the EQ inside Amarra but out of the box with EQ switched off i cannot hear a difference in the 16/44 recordings.
Where Amarra works well for me is in playing the 24/96 and 24/192.
I've had Amarra for some time. Jon was kind enough to give me an eval copy way back.
I keep going; I feel that as and when Amarra does some signal processing e.g. upsampling promised in rel 1.1 it will start to pull away from the crowd.
I am very open to modifying things and trying new experiments.
Sidenote: I am not into specs much - but the physics/engineering should be understood - this is not rocket science - the big problem is cause and effect. A very typical (Type A) logic error is: I changed (x) and it sounded better hence (x) is good. This doesn't follow at all - by changing (x) one may have affected something completely different which caused the change. Politicians use this argument a lot.
My Macbook Pro runs rather warm, just wondering of SSD would cure or worsen that.
The Netbook that I saw in the store seemed to run like a little oven, and I think they have SSD in them?
See if the link helps to answers your question. You may wish to try for more Google "hits"
How can it sound any different?
A blue ray movie can be ripped to a 5400rpm 1.5tb drive and played back audio and all with no problems, with identical results to an SSD drive when viewed, so explain to me how a faster solid state drive can have any interaction on audio playback with a PC, apart from mechanical silence?
Could this be like fancy cables? Cause it's new and faster it's imagined better? After all the benefits of a SSD are pretty standard fare in benchmarks and PC apps, and no one I know has found SSD to improve their graphics performance?