Can anyone out there tell me what the difference would be in having one of these chips vs the other in your processing?? I see some high end stuff with 96khz and some with 192khz.
Outlaw has 192 where Rotel has 96(one of the differences). Would you want the 192 for any future considerations??
Thanks...I'm learning a lot here.

You'd want 192kHz chips if your processor were processing signals of that resolution... which to my knowlege, the 950 does not. In fact, I don't know of any modes it has that put out a signal with greater resolution than 48kHz.

------------------
Matthew J. Hill
matt@idsi.net

So there's no source material with that resolution?? The 950 says that it has 192 DACs. Some top end products have 96 and some have 192. Am I correct in thinking that 192 chips are not a real important feature to consider in shopping for a pre/pro?
Thanks.

Steve,
I don't have a full understanding of the way that DACs process a signal, but from what I've read a 96Mhz DAC processes at a top end of around 18 bits and a 192 DAC at some thing like 22-24 bits. The 192 DACs also have a designed in lower inherent noise floor shown as the DACs S/N ratio. The audible benefit of this is dependent on the full circuit design as well as the source material.

------------------
Take Care,
merc

There is source material out there that has a 96kHz resolution (DVD-A, DTS 96/24) but the 950 doesn't support any of those encoding formats. DVD-A can also sample at 192kHz in two channel mode.

The 950 does "support" these formats through the use of an external decoder and the analog pass-through, but in that case the 950's DACs are not being used.

------------------
Matthew J. Hill
matt@idsi.net

So we are looking at things other than processing speed??
I am just trying to not have to buy another pre/pro for a few years...don't want to hear that 96khz chip will no longer do the job in a couple of years

It's really not related to processor speed. I don't know what CPU is in the 950 or how fast it is. The numbers we're talking to refer to what resolution digital signal the DACs can convert to analog.

The CPU is fast enough to perform all the processing that it currently supports. It won't support any more. Its feature set is not going to change. Therefore, the CPU will always be fast enough for what it needs to do.

------------------
Matthew J. Hill
matt@idsi.net

MJH - Thanks. Can I assume that the higher resolution the chip can convert...the better the sound dimension, clarity etc.

From glancing at the 950 spec's as listed in the manual, the top digital input is 96khz. The real limitation, however, is finding a source at that level. I understood the limitation on digital frequencies to be created by the recording/movie industry. Thus, for example, the DIGITAL output of an off-the-shelf DVD-Audio or DVD player is limited to cd-quality output (44.1khz). Higher frequencies are down converted for the digital output. Otherwise, you have to use the audio output if you have a DVD-A player.
The exception, at least for my DVD player, appears to be non-copyprotected material recorded at higher bitrates (e.g., 96khz). My DVD player's setup (RP-56) gives the option not to down sample non-copyprotected bitstreams (at I presume a max of 96khz because that's the limit of its DAC). When I have the chance, I will see if I can't record a 96khz sample on my computer (e.g., using SoundForge) and either run it directly to the 950 digitally or burn it on a cd and try it on the DVD player. Except for recording live music, however, I don't see much practical usefulness in this, as most sources of 96khz music will be DVD-A--which per industry standard cannot be sent digitally to the 950 or ANY OTHER external processor. Perhaps there is a way around this limitation by playing a DVD-A disk on a DVD-ROM player in a computer (which will almost always have a digital output).
Not being an engineer, if anyone has reason to believe that these future trials will blow up my DVD player or the 950, please let me know!!
Robairdo

There are 24 bit/96 kHz linear PCM source discs out there (not DTS). I think Chesky (or some other really small but audiophile record label) has put them out. Most DVD players can transfer that digital signal now, but used to be that only Pioneers among a few others could do it.

Would those be HDCD cd's?

No, HDCDs are not 24/96's.

HDCDs are simple linear PCM 16/44.1 discs that have been encoded differently some how... (Filtering, DPS'ed, sound shaped; help... )

I thought HDCD was 20/44.1 ... that it's "claim to fame" was that extra four bits per sample which gave increased dynamic range.

------------------
Matthew J. Hill
matt@idsi.net

Matthew- HDCDs may claim to have 20 bit *resolution*, somehow due to how the signal is encoded (...), but the CD linear PCM red book is strictly 16/44.1.

Maybe I'm confused, but I thought HDCDs were not red book.

------------------
Matthew J. Hill
matt@idsi.net

They have to be red book, otherwise how could a normal (non-HDCD) player play them?

I keep hoping someone will get motivated and look up HDCD on Google or something and fill in the blanks here...

HDCDs are still red book audio CDs. They hide codes in lower order bits. On a regular CD player, this isn't audible -- way below normal sound levels. An HDCD equiped decoder recognizes these flags, and uses them to actively alter the playback sound. This includes dynamic range expansion, and some filtering/dithering effects.

HDCDs developer's claim this offers _equivalent_ quality to higher bit rates, and higher bit depths.

Of course, they also claim that it improves playback quality on non-HDCD devices, too.

Does it really sound better?? Like most audio things, there is no clear answer. I have a HDCD player. Although I _think_ I can tell a difference, I probably couldn't in a blind test.

big mac- Thanks!

Sony also has this thing (in the DAT world) called Super Bit Mapping. Also utilizes simple linear PCM 16/44.1. But what it does, is to move the "noise" from the audio spectrum where humans hearing is more sensitive, to a part of the spectrum where it's less sensitive.

Sort of like a filter , but not really.

The wierd thing? Test measurements say that with SBM, it's worse, but when people listen, it's better. The claim is 20 bit resolution in terms of *apparent* signal to noise ratio.

I thought HDCD's did require a special player or an external decoder fed via the digital out? I do know that XRCD's do not require any special player or decoder and they meet the redbook standard but still mananage to fit more than "normal" data on the discs. I haven't found an XRCD that I already have on standard redbook or SACD, so I can't compare, but the one I picked up to check out the format sounds very good (Count Basie - Basie's Big Band)

------------------
pat----

email: pat@sklenar.info ---===--- home page: Grumpy's Lair

HDCDs will play in any standard player.

I often wonder if the sound quality from these special formats is really due to the formats, or just special care taken during mastering/remastering.

I have a CD from Audiophile, has a piano recording done simultaneously through multiple formats. Analog, various tape speeds, different digital formats, some active filtering/shaping, and HDCD. The change among the various analog, and from analog to digital is pretty easy to pick out. I'd have a hard time being honest about being able to tell the difference between straight 16 bit/44 CD and HDCD.

So, if a disc has been remastered as part of going to HDCD, I would be suspect about any quality differences really resulting from HDCD itself.

I thought I read somewhere that the 192 dacs offer higher bandwidth for signal processing, creating a better sound than chips with less headroom even for lower bandwidth inputs.

Any truth to this, or was I on glue?

Start with more precision, your end result will be more precise. Think of it this way: You're going to do a series of steps on a number, all using floating-point math of very-high detail. Would you rather start with your initial sample only accurate to one decimal point (4.5) or four (4.4511). With a series of steps, the final result could be very different in the two cases. Whether the difference is really audible or not is a totally different question!

[This message has been edited by bigmac (edited June 11, 2002).]

That's an interesting question... at what precision does the 950 process signals internally?

------------------
Matthew J. Hill
matt@idsi.net

Just visited HDCD website. It's an extra 4 bits of resolution. Will make all digital recordings sound better even if you don't have the hdcd chip. Even better with the chip. Evidently most recording studios use this technology for all the stuff they mix - even stuff not designated hdcd and it's supposed to improve the sound. I doubt one could hear it on regular equipment.
The Rotel has the chip...therefore all digital recordings should sound better. It's used even though the light isn't on. On an hdcd cd the light should light up.
BTW - the best recordings I ever heard are from Professor Johnson's recording lab. He uses all tube 2 channel equip and the sound is beautiful even on not so great equipment. This was a few years ago so I don't know where they are available but if you can get your hands on one check it out.

The Outlaw 950 definitely will accept 24bit/96kHz PCM. My Pioneer DV-47A outputs two channel 24/96 through its SPDIF on DVD-Audio discs and the 950 sounds quite good on this material. This is what I use when I don't want to listen to DVD-A's in multichannel.

True to a point, but what you're talking about is sample size (bits) not sample frequency. IIRC the 950 processes everything at 24/96 or below so it seems like the 192kHz capability would be mostly just bragging rights.

Most pre/pros that divulge any info at all claim to do all math (processing) as at least 24 bit quantities internally to prevent rounding errors from being audible.


Charlie

>>> IIRC the 950 processes everything at 24/96 or below

Just asking, but do we know this? I suppose we could look up the part number of the Cirrus/Crystal DSP chip that's used, and maybe they'd say.

I don't know enough, but even if the
[Added:] Sort of like oversampling in the old days.

But now that I think about it, maybe more than 24/96 processing capability doesn't get you much anyway. The max rate digital signal you can input is 24/96. And if your inputting an analog signal, you're stuck with either the 24/48 ADCs the Outlaw has, or just keep it straight analog. Anyways...



[This message has been edited by Kevin C Brown (edited June 11, 2002).]

Nope... still, who the hell cares??? Those that may care have already been afixed on that blue light that emanates from that other pre/pro faceplate.

Just admit it and move on to paying me for my Naïveté...




------------------
Take Care,
merc

Just a few theories about how it could be better. If they sample at 192KHz, even on a 96KHz signal, you are still sampling the same level twice. This can give some error detection to reduce jitter, ie if you get a 1 and a 0 for the same level, it can probably be corrected based on the bits around it. If the manufacturer was bold enough, they could interpolate between the two samples to create a more smooth transition. This would allow the use of cheaper and lower order output filters which tend to mangle the phase less than higher output filters required to smooth more jagged or stepped signals. I would think the 192KHz DAC COULD sound better if the manufacturer actually used them and didn't just throw 192KHz DACs on to put it on the spec sheet.

Ah - that would be the ADC, not the DAC.

IIRC the ADC on the Outlaw 950 tops out at either 48 or 96 kHz (which is plenty - Nyquist and all) by 24 bits deep.

I suppose one could use the DAC bandwidth for some sort of noise shaping or dithering or ..., but if this was the case I would imagine marketing would have told us about it. I suspect the difference in cost was tiny and the device fit the same socket.


Charlie

The 950's ADCs are 24/48.


We are still missing something. If you only needed a 16/44.1 DAC to properly decode a CD, then why did "quadruple oversampling" sound better and measure better when they hit the scene?

*I* don't know what it is, but there still must be some benefit to the 24/192's for the DACs on the Outlaw...

Well, for one, the 24/192 DACs have a lower noise and better dynamic range than the 24/96 DACs on the Rotel. If I remember correctly, the 950's DACs are THD+Noise spec'd at -100 versus -91 and the A Weighted Dynamic Range is spec'd at 120 for Outlaw and 101 for the Rotel.

------------------
Take Care,
merc

Check the latest Sound & Vision as to what they measured s/n in the Outlaw.

Steve: Do some research into HOW S&V tests and what their S/N figure really means. It is not the same as the standard A weighted S/N spec that manufacturer's use and should and can not be compared directly to each other. S&V's S/N test is top ended at the theoretical perfect(absolutely no noise at all, the component is totally invisible from input to output)and the spec'd S/N figure that is used by manufacturers has no top end, theoretically.

Finally, it can be argued that a S&V S/N rating of -72 cannot be audibly discerned from one of -76.

What most people don't understand is that the S&V tests show quite good noise levels for the Outlaw 950. Based on each manufacturers specs, I'd guess that the Rotel might S&V test at just under the Outlaw, yet, that difference would probably not be audibly discernable. Why not call Brian at S&V and ask him for yourself about their testing procedure and how the Outlaw 950 sounded?

------------------
Take Care,
merc

[This message has been edited by merc (edited June 13, 2002).]