HDX vs Native

[Cheesehead]

you shouldnt find HDN any different than HDX1

That isn't true in my experience, sound01 will find he can open larger sessions on HDN than on HDX1. That's what I found.
Just today I opened a large film session that was running on HDX3 on my 2019.5 Native rig with no problems. Those extra voices are brilliant!

[JFreak]

Originally Posted by Cheesehead

open larger sessions on HDN than on HDX1

Well for now that is true: with 2019.5 HDN has 384 voices whereas HDX1 still has the original 256 voices until Avid releases the firmware update that should bring extra voices for HDX as well.

But in theory HDX3 session can have 768 voices and opening that on HDN will only open first 50% in case one hasn't bought three extra voice packs now available... (and once HDX gets the voice update, HDN system would need 6 extra voice packs to match maxed out HDX3 session)

[sound01]

Originally Posted by justinhill

Latency is the inherent delay between asking a question and getting a useful answer. In this system, the background noise isn't part of the signal; you have to listen to it first.

Hahahaha. Perfect, thank you.

[sound01]

Originally Posted by Cheesehead

That isn't true in my experience, sound01 will find he can open larger sessions on HDN than on HDX1. That's what I found.
Just today I opened a large film session that was running on HDX3 on my 2019.5 Native rig with no problems. Those extra voices are brilliant!

Thanks, it’s strange that I couldn’t run a full session, which had over 256 tracks, on my hdx1 2019.5 rig without tracks becoming inactive but my colleague could run it on his 2018 HDN rig with no problems, nothing was inactive.

It’s as if his HDN system can allocate voices, as and when they were required, rather than it being set by the actual track layout. Does it allocate busses differently? Am I missing something in the manual.....?

On another note. Anyone want to buy an HDX card? 1 careful driver.

[BScout]

Originally Posted by sound01

Thanks, it’s strange that I couldn’t run a full session, which had over 256 tracks, on my hdx1 2019.5 rig without tracks becoming inactive but my colleague could run it on his 2018 HDN rig with no problems, nothing was inactive.

It’s as if his HDN system can allocate voices, as and when they were required, rather than it being set by the actual track layout. Does it allocate busses differently? Am I missing something in the manual.....?

Yes. You are missing something. HDX uses a DSP mixer. That means voices, bussing, sends, delay, side-chain delay, etc are handled on the card. That's the trade-off for the extremely low latency (and is also why some features are HDX only -- the voices are reserved in the HDX mixer which can double one track's voice count like with punch in and out of cascaded tracks.) Native plugins on Master and Aux tracks consume voices (because the voice use to shuttle from the DSP mixer to the native processor and back.)
So you can use up your voices on an HDX1 system.
On HDN, there is no DSP mixer so everything stays in one place on the native processor. There are some in the post world who would like the ability to define what the HDX card is used for (DSP plugins only, no mixer for instance.)

Quite frankly, other than the low latency tracking and the additional features unlocked, HDX1 is kind of a dumb purchase. HDX2 is where things get interesting.

[LDS]

Originally Posted by JFreak

Yea, so unless I participate in a blind test where I can pick a difference between 2ms and 3ms rtl I still think the difference is irrelevant and as long as you stay safely below 10ms there is no REAL problem for anyone

The problems arise based on the precedence effect, where essentially our brains will mash together the original sound and the delayed sound into one event if the delay between them is small enough, giving preference to the first arriving sound. How small that delay needs to be depends on the type of sound, with faster transient, shorter sounds having a much smaller window for the precedence effect - somewhere between 0 and 5ms. Slower, longer sounds might be somewhere up to 20ms or 30ms. It is when the delay between those two sounds exceeds the period of time that are brain applies the precedence effect that latency ultimately becomes a problem. Because it is perception based, it is probably a quite individual thing. It also makes single sound based latency a different kettle of fish, like playing a VI with midi or a guitar with the amplifier 10 meters across the room. Our brains aren't inflicting what essentially seems to be a psycho-acoustic flight or fight response in those scenarios.

The precise benefit of HDX is that it is a guarantee of latency. It's base RTL will be at least 4 times better than HDN. Native is certainly capable of low latencies, but it isn't guaranteed. The 3ms of RTL for HDN is dependant on the actual session and system that runs it. That is precisely why so many other interface developers still include their own hardware monitoring paths.

[YYR123]

Originally Posted by justinhill

All I know is that when the drummer is waving at the window and calling "bit of a delay in the cans, mate - can you do something about it please...".

Funny story, a few years ago I dropped some cash down and purchased an HDN rig with a HDio 8x8 system used.

I was thrilled b/c I was gonna get closer to having less problems (better functionality) (so I thought)

I got it all setup and had my 8-10 year old daughter come in and overdub a vocal on a session I was working on.....

First thing she tells me “it sounds weird in the headphones, like a delay”

Omg took all the air out of my ballon

I sold it a year or so ago, when i found a killer deal in a HDX card - I won’t be moving anytime soon - no headphone delay

[Eric Lambert]

Originally Posted by YYR123

Funny story, a few years ago I dropped some cash down and purchased an HDN rig with a HDio 8x8 system used.

I was thrilled b/c I was gonna get closer to having less problems (better functionality) (so I thought)

I got it all setup and had my 8-10 year old daughter come in and overdub a vocal on a session I was working on.....

First thing she tells me “it sounds weird in the headphones, like a delay”

Omg took all the air out of my ballon

I sold it a year or so ago, when i found a killer deal in a HDX card - I won’t be moving anytime soon - no headphone delay

Something was wrong with your configuration as you shouldn't have received complaints. I even paid close attention to the existence of this effect yesterday while tracking (after having read this thread and curious to know if I'd merely adjusted to it) and noticed no noticeable delay. With earlier versions of PT and Native hardware, when not properly configured, yes, but not today, and not within the past few years.

At one point we'd even tested this by recording the headphone returns and comparing them to the source. The delay was miniscule, and not unlike what we were accustomed to after years of DSP cards. Again, if talent is complaining about delay in the cans, something needs to be addressed, but not Native as a platform. Thousands of people are doing this without issue.

[glogulus2003]

I recently had to move back to using my HDN temporarily for tracking and I hated it. From my experience, I've not had a great time tracking using HDN even at the lowest buffer settings. This was especially true when sending output to a headphone that might use an aux for reverb and other effects for vocal tracking. So, a few years ago I took the plunge and bought myself an HDX card. Now I've got two. Probably way overkill for my purposes, but I don't have to worry about how it's going to sound when tracking. Works pretty much perfect every time, and then when it's time to mix I can use either native or dsp plugins. I for one am happy I went HDX. Probably will use these cards until they fail because it doesn't seem like Avid will do another hardware platform anytime soon.

[JFreak]

Originally Posted by glogulus2003

it doesn't seem like Avid will do another hardware platform anytime soon.

How so? Historically HDX is ancient and should be replaced with some kind of "accel" version soon..

Though this supposedly coming voice upgrade is going to be a kind of upgrade but still.. history says anytime soon (though don't hold your breath)