Where are you getting digital outputs on the Denon Prime 4? I was hoping for Toslink inputs, but will use converters for that.Originally Posted by Reticuli
I'm interested in the performance of the hardware - what happens between the input and the output. A/D > D/A. Processing engine as a separate item, sure.
Last edited by SWS Productions; 03-15-2020 at 11:24 PM.
This was not done on Prime 4, but isn't going to make any difference. Common code.
The digital processing issues of Prime's Engine OS are also present on the analog outputs, as that's what's feeding the DACs on Prime gear. The digital outputs used in the cited testing on the SC5000 and SC5000M, however, rule out the possibility of DA conversion or AD conversion affecting the cited testing, which I believe is what you were previously asking -- whether some audio interface was affecting it. Using the analog outs into an interface you see the same problems, but using the SPDIF removes a possible variable, obviously.
At very loud volumes that are overloading peoples' ears or with a lot of room interference and with no baseline to compare it to (say, comparing Prime with keylock off to Pioneer with keylock off), you might not be bothered by it regardless of the musical content used. With a back to back comparison and with complicated musical material, it's readily apparent.
There is still a slight high-frequency roll-off, though I think that's the least of our concerns. Engine OS obviously shouldn't have it, but it's going be less objectionable than the other stuff.
Never seen DJ gear with Toslink inputs. You mean you wanted coax inputs?
Last edited by Reticuli; 03-18-2020 at 06:37 PM.
^^^ Toslink inputs for CD player interface from rack mounted decks. I'll try some converters at the board and see if it's quiet - that should work ok. Optical interface to the decks would be best.
Why would toslink be better? Unless you've got a serious ground loop issue it's not going to make any difference.
The chance of losing any data on a short s/pdif run is miniscule.
And I have the same skepticism regarding the sound of a digital mixer with more or less "complex" program material.
"Complexity" matters when you are doing digital compression eg. mp3, but a mixer doesn't care how complex a signal is.. a sample is a sample.
On that I was talking about the Engine OS firmware on the players regarding the measured degradation of the sound being more noticeable to me personally on complex material vs minimalist musical content. The weird sound quality we were hearing especially on complex material was why we started doing tests. If you've only got is a beat and a high hat going on, you're not even going to notice frequency range roll-off. I wasn't talking about the mixer nor was I calling this particular aspect of the players measurable and objective. My assessment that the players' measured sound issues are not as noticeable on minimalist stuff is entirely a subjective thing. Think of it as glass half full, but it's also if someone's playing really minimal techno on Pioneer and Denon Prime players and goes, "you're crazy, I don't hear a difference", well, I don't hear a difference in that situation easily, either.
The mixer, in contrast, just sounds a little dry, lacking in low-level resolution, sounding like everything's being pushed through a pinhole sometimes, and has some odd stuff going on up top in the highs in 96khz mode. This is also a subjective statement of mine, though several others have independently said similar stuff. The mixer, does, however, test fine in RMAA; so to find out exactly what might be causing it to sound like that would require better testing methods. And heck, I could be wrong. I'd welcome someone with a big test bench rig to prove me wrong and demonstrate the X1800 is just as accurate or more so than the competition. Like I said, though, the mixer doesn't sound that bad to me, and it does have quite good bass. I also wish the faders were longer, the LEDs were less bright, the FX master LED wasn't just lighting up all the channel FX LEDS, the master meter was the same size as the channels, there was a rotary option, etc, but I digress.
Now back to the playback/player end, considering there is both aggressive low-pass filtering, at least one instance of sample rate conversion and possibly more instances (e.g. massive oversampling then downsampling to 24/96 for the SPDIF), and a band-isolated limiter all currently implemented in the Engine OS player firmware, and sorta badly and in real time at that, the effects of it can be described as being similar to something like lossy data compression, which is what I've described Prime's player digital processing sound as elsewhere. I've previously said that complicated lossless content on the Prime players reminds me of playing the same content with lossy data compression on Pioneers if we're comparing keylock off on both... the degradation is that severe, IMO. Actually, that might be too kind. I might prefer the MP3 version on Pioneer players using the SPDIF if we're talking keylock off comparisons. However, you use these on a sound system in an untreated room at very high volumes and with minimalist productions, and I doubt anyone's going to tell the difference, especially after a few drinks. Start using keylock with extreme negative pitches from zero, and Prime's use of Elastique starts to show past its poor core audio processing sound quality. Good rooms, complicated & dense music productions, and staying near zero pitch with keylock on, and I think it's a lot more noticeable a difference in sound than I wish it was.
Pioneer, by the way, normally does no sample rate conversion on their players. Their playback processing and SPDIF switch to the sample rate of the file being played, so that might be part of why Prime playback is degraded the way it is. With the pitch fader at zero, Pioneers are literally bit-perfect out the SPDIF assuming the player Utility bit depth setting is at or above the spec of the file. SRC in real-time is a tricky and precarious thing. I think the Prime playback support of sampling rates over 96khz is probably also a mistake, too, considering their player SPDIF out is fixed at 96khz. However, I've run tests by post-processing files to very competently convert them up to 24/96 and even 32/192 using iZotope ahead of time, and there's still weird stuff happening to these up-rez'ed files on Prime. So there might be a variety of issues with the Prime playback Engine OS firmware's audio processing that could use improvement. Upon release, the Prime players did, after all, have even worse high-frequency roll-off than now and you couldn't even set a cue or hot cue until the analysis had finished. The M model would randomly change platter (and thereby track) speed, too. Many embarrassing glitches we've been dealing with and helping them track down and resolve. So I wouldn't rule out a variety of other more-subtle audio processing playback glitches, and wouldn't want to just assume it's various properly-implemented and intentional but inferior SRC methods responsible for the current playback sound.
Last edited by Reticuli; 03-18-2020 at 10:14 PM.
For a mobile system with rack mount CD decks (along with other gear in the racks), optical interface is immune from any ground loop hum. Somewhat of a "just because" in the system design. There will be a digital mixer in the rack, lighting DMX controllers & splitters in another rack, amp racks and so on. You can interface 2 or 3 pieces of gear from the same rack into a common point, like a mixer and only one may have hum/noise. Toslink is the best way to interface for that reason. Hum in gear, most times is due to crappy house power and poor power supply design - hence the comment about different gear in the same rack and some will hum and some not under adverse conditions.
Generally, based upon past experience, noise isn't a problem. But there have been a few select times when it was - fortunately those were public dances where the system was turned up and kept turned up for the whole show. It would be annoying during dinner music and it's not a big deal as the decks have Toslink outputs on them.
No discernable difference in the audio quality with SPDIF or RCA/analog for that matter.
What DJ decks have Toslink optical outputs on them?
And how would you know if a coax digital SPDIF connection was being affected by a ground loop in the first place?
Last edited by Reticuli; 03-18-2020 at 10:12 PM.
Yea true, that is fair.. ground loops can be an issue, and given the choice, toslink would be one less ground to worry about.
But that said, I've not had a grounding problem in my system in years.. everything is running off one power distro, everything is balanced connections except for the decks to the mixer which are S/PDIF. I do have a couple of ART DTI's I keep around just in case of problem (for analog isolation, not digital)
What I've been thinking for decades is that the audio industry should come up with a standard for AES/EBU and S/PDIF over ethernet layer 1.. that is, using the electrical standards for ethernet but not the protocols.. because the hardware is cheap and plentiful and it's galvanically isolated. Much better than these old connectors.
But in this day and age.. might as well go all the way.. forget those standards altogether and make everything AES67. Just put an 8 port gigabit switch into the mixer and plug everything into it
OSC (Open Sound Control) has been around for a while, and Dante.
I've used OSC, many commercial audio interfaces available these days seem to use RJ45/Ethernet-based Dante networks since FW400/800 has been largely replaced by them (and USB3.0, many modern PCs/laptops don't have FW ports anymore) but I have no experience with using Dante. Iirc it does both audio and video over Ethernet, as well as DMX and stuff.
Afaik Firewire/USB doesn't support long cable runs whereas Cat6 RJ45 can be used to carry signals as far as ~350m, in practice such cablings are very rare though (100m is very close to maximum inside a building) but networks like that are both rare and expensive, not to mention very technical and complex to set up/install (in case you're wondering I've been digging into the subject lately as I'm designing an intercom)
On a side note in the military we had a digital field phone which I recall was Mil-spec with an RJ connector (similar to an XLR but with a Cat5e/6 plug inside, I've seen them in electronics stores), however we never used it so there is a "standard" but I guess it hasn't made it's way to pro audio yet (this was in 2009)
Some guys on speakerplans were discussing homebrew DMX over Ethernet too with some unusual military-grade connectors but I can already tell from experience that the Mil-spec stuff can be kinky at times, mostly the grooves and the O-ring, not to mention breaking a pin (I heard it once took the army technicians several working hours or days to locate the problem and fix it)
There's also gold-plated and suspended ball-grid type connectors but I don't remember seeing them for sale anywhere, and I'd figure they're *very* expensive although they probably last a lifetime.
A standard RJ45 will most likely break in heavy pro-audio use so it requires a good quality twisted pair Cat5e/6 which can be expensive in the long run along with the reinforced connectors. Then there's the interface hardware which isn't cheap either (and software)
Anyway, sorry for the lengthy off-topic.
Last edited by efinque; 03-19-2020 at 05:18 PM.
Rack Mount CD Decks, not DJ Decks.
TASCAM CD-200 & CD-200BT
How do you know if electrical noise caused by ground loops in peripheral gear? Sometimes it tracks noise level with input channel fader gain setting - sometimes it's just there - sometimes it tracks with master out level, it goes away when interconnecting cable(s) are disconnected or line cords unplugged.
Last edited by SWS Productions; 03-19-2020 at 07:40 PM.
Posting Permissions
Reply With Quote
Bookmarks