![]() ![]() |
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
84.72.46.172
In Reply to: Benchmark DAC1 output stage posted by Champion on November 11, 2006 at 01:31:27:
and it sounds MUCH better, IMO.The Monarhcy M24 is an 8 x oversampling design (BB DF1704 so it handles 24bit 96Khz inputs even if it is not a 24/96 dac) using the Burr Brown PCM63P-K true 20bit DAC chips.
OP-AMPS??? Not a single one. The output filter and gain stage is handled by 1 6922 tube per channel in an SRPP configuration.
Oh yeah, it also has a hell of a good preamp; on a separate PCB even with separate power transformer and supply and accepts one other analog input. This is also a single tube per channel SRPP design.
When this much nicer machine is avaiable for the same price I can't really imagine the success of the Benchmark.
![]()
Follow Ups:
This "Lite-Audio" design is actually very similar to the Lite DAC-60 and DAC-68 designs combined. I am surprised that they are using such an old DAC chip.The PCM1704 is better because it does 24-bit A/D. The DAC-60 uses the same DF1704 upsampler and PCM1704U-K DAC chips. The problem is that in order for the DAC-60 to sound world-class, it needs quite a few mods and tubes costing between $300 and $500. I admit that it is a good value though. I base my new USB DAC on the Lite DAC-60 BTW. I thought it was that good. However, it took a lot of mods and very expensive capacitors and tubes to pass-up the modded Benchmark DAC-1 (DC-coupled) in performance. It's in a different league both performance and price-wise.
![]()
"This 'Lite-Audio' design is actually very similar to the Lite DAC-60 and DAC-68 designs combined. I am surprised that they are using such an old DAC chip."That may actually be a good thing.... I don't think the chips today are collectively better than those fifteen years ago....
Although if I had to choose a modern chip it would be the DF1704, set to "slow rolloff" mode.
"The PCM1704 is better because it does 24-bit A/D. The DAC-60 uses the same DF1704 upsampler and PCM1704U-K DAC chips."
For 16-bit Redbook CD playback, the extra bits are merely for implementing the digital filter. The only thing a 24-bit D/A has over a 20-bit D/A for CD playback is more-precise tracking of (dithered) low-level signals.
The problem with most 20-bit players in the 1990s was not so much output wordlength, but the digital filter function utilized in those days had poor (ringing) time response. The DF1704 set to "sharp rolloff" would exhibit the very same characteristic.
Does your new USB DAC use the DF1704 in "slow rolloff" mode? If so, I'd be interested in trying it.
The Spoiler production version will have a switch to select the roll-off. Availability should be Q1 2007. Some of the capacitors that I use have very long lead-times, so this could delay it. Pricing is $5999.00.
![]()
Two points: 1) The PCM 1704 is a D/A converter not an A/D converter. 2) Also the DF 1704 is an 8 x OVERsampler not an UPsampler (ie. it is synchronous not assynchronous).Where is your justification that this 24 bit converter is automatically better than the PCM63K? Surely not just the number of bits. I agree with you that the 1704 is a great DAC chip but so is the 63.
My understanding is that the major reason they stuck with the older DAC chip is that it has a more robust current output, which makes means they can use passive I/V conversion and a tube output stage without needing a low noise opamp to boost the signal. Apparently, the 1704 has a quite low current output that is not as suitable for passive conversion or for use with tubes. Monarchy actually offers an "upgrade" to the PCM 1704 for their DACs. I inquired into it for my DAC and they advised me not to do it because the analog stage in the M24 is optimized for use with the PCM63. Monarchies older DACs, which use an ultra high speed opamp for I/V conversion will accept the upgrade with no problems.
![]()
"2) Also the DF 1704 is an 8 x OVERsampler not an UPsampler (ie. it is synchronous not assynchronous)."Although commonly used in this context, the terms are often interchanged. Wadia, for example, calls its synchronous oversampling "upsampling".
![]()
![]()
So what I said then is generally correct, is it not?
![]()
Okay, lets split hairs:PCM 1704 has 4 dB more dynamic range. I just has more resolution with 24 bits versus 20. The THD+N is 102 dB for the 1704 and 100 dB for the 63.
The difference in output current is +/- 2ma versus 1.2ma for the PCM1704, so this can be helpful, but it's still a small current IMO. This is the only positive with the 63 IMO.
The reality is that most designs including this one are not resolving enough for you to hear -100 dB anyway. They need a lot of mods to get there IMO.
![]()
"I just has more resolution with 24 bits versus 20."The only condition where wordlength affects resolution is if the input data is of higher wordlength (DVD-A) than the lesser DAC's max wordlength. Although in such cases, it would be more-accurate to say the 20-bit DAC loses resolution. (And most of the losses in 24 bits vs. 20 would be in the form of noise.)
For Redbook CD playback, all else being equal, a 24-bit DAC may sound slightly cleaner than the 20-bit DAC (due to more-precise interpolation), but the resolution would be almost identical. For the extra bits in CD playback are used solely for digital filtering.
![]()
![]()
I listen to a lot of native 24-bit music mastered from tape as well as upsampled 16-bit data using SRC. Seems to me that a 24-bit DAC would take advantage of this.
![]()
I think the M24 is over $1400? That is the list price on their website.
![]()
nope. Direct from Monarchy for $980.
![]()
This post is made possible by the generous support of people like you and our sponsors: