Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

New Arri Digital vs. Scarlet?

The primary benefit of 192khz is not the bandwidth extension beyond what you can here, but that filter artifacts are minimized and pushed further out of the audible band.

Graeme
 
Steven Caesare said:
If I can only see it as a waveform on adevice, but cannot perceive it as different when it's being played back, that's fine with me.

Out of curiosty Irri, have you done any blind listening tests with your high res audio?

I have, and good down sampling, and/or compression is VERY difficult to hear, if you ar being honest with yourself and methodology.

And that doesn't even take in to account the difficulties even VERY GOOD equipment has in approaching 24 bits of resolution or what "nuances" 192Khz sampling (i.e 4+ times higher than what even the best of human ears can hear) are supposed to impart to recordings.

-sc
Click to expand...

I haven't actually. My old speakers don't match well with the very revealing source or my room (I'm abroad where carpet isn't common) so what time I have has been spent looking into a suitable pair of speakers. I think blind tests are the only ones that count, but once I have bought something I'm happy to let the placebo effect contribute in a positive way. Ignorance is bliss, as they say.
 
Graeme Nattress said:
The primary benefit of 192khz is not the bandwidth extension beyond what you can here, but that filter artifacts are minimized and pushed further out of the audible band.

Graeme
Click to expand...

Indeed. Oversampling assists this in the filter constrction as well.

However, I've never seen an objective double-blind test where somebody could reliably identify recordings where the only variable was the sampling rate.

Strangley enough, many of the high-end equipment vendors don't like to publish such tests... :rolleyes:

-sc
 
Irri said:
I haven't actually. My old speakers don't match well with the very revealing source or my room (I'm abroad where carpet isn't common) so what time I have has been spent looking into a suitable pair of speakers. I think blind tests are the only ones that count, but once I have bought something I'm happy to let the placebo effect contribute in a positive way. Ignorance is bliss, as they say.
Click to expand...

Even if the placebo effect is going to cost you 10x as much?

If so, I have some amazingly transparent Onyx interconnect towers for sale...

-sc
 
Oversampling really helps with output filters, but the high sampling rate helps with input filters.

Objective audio testing is not entirely simple, but indeed, should be done more. What might be a more useful test is to keep the sampling chain at 192/24, but use filters that you'd use for 44.1 to compare with the filters you'd use at the much higher sample rate. Then you'd just be changing the filters and would be able to hear how they alone work. That's one test I'd really like to hear.

Graeme
 
Graeme Nattress said:
Oversampling really helps with output filters, but the high sampling rate helps with input filters.

Objective audio testing is not entirely simple, but indeed, should be done more. What might be a more useful test is to keep the sampling chain at 192/24, but use filters that you'd use for 44.1 to compare with the filters you'd use at the much higher sample rate. Then you'd just be changing the filters and would be able to hear how they alone work. That's one test I'd really like to hear.

Graeme
Click to expand...

I'd like to hear that too.

And would be a test that's contrcutable from a single 192KHz source, rather than the more difficult to obtain pair of recordings at differing sampling rates, taken at the same time, with identical mastering steps, etc...

My bit depth and lossy compression testing has been much eaiser to do...

-sc
 
Graeme Nattress said:
The primary benefit of 192khz is not the bandwidth extension beyond what you can here, but that filter artifacts are minimized and pushed further out of the audible band.
Graeme
Click to expand...
Fine for sampling analogue audio. But why stick to 192 for the final mixed master? It's like sampling 35mm at 16K and not doing a final 4K master but staying at 16K.
 
If you keep the file at 192khz, you don't get the hear the effects of a downsampling filter and because no downsampling filter is perfect....

Graeme
 
Coming from an audio background and having fought this digital revolution on that front many many years ago there are many elements that go well beyond data rate. It is evident that there are several beneficial elements related to oversampling, far more dynamic range was reached in regards to the 16 bit to 24 bit revolution, but 96kHz/192khz/384kHz/1-bit DSD all became standards without much merit to support such arguments (Sony would disagree as they spent far too much money developing and abandoning 1-bit DSD). After several years and millions of dollars of convertor shoot outs we ended up designing our own with one of the most brilliant minds in A/D conversion & that man was Dan Lavry of Lavry Engineering. This man has forgotten more the most will ever understand about the process of energy conversion, and the question was raised, what's more important the data rate or the quality of the clock and the components that governs the translation? Ultimately there still is much disbelief of going to higher data rates beyond 88.2 (or 96kHz) 24-bit when the right convertors and sync operations are in play. A lot of aliasing and jitter can be caused by the most simplistic of clocks in a cheap 192kHz converter, so does this mean that a cheap 192kHz convertor trunks a high end 96kHz convertor with the right components in the box? Never! Lavry's gold series actually took into account recalibration on wear of it's internal components as it's performance 'changes' over the years due to resistors and transformers wearing and tear over time. Less I digress, this is quite comparable and of interest in this day of the video revolution as it is not always about the data bandwidth. As much as we all wonderfully want to get carried away with 'more is better' ideology that simply isn't relevant any longer... I imagine 4K on Epic is going to far surpass (as photo's are beginning to show) 4K on any of our 'present' RED Ones, but its far more complex and involved/evolved with the processing and data management within these devices rather then "more hard drive space eaten away by more pixels that may or may not contain a perceivable & necessary advantage due to the governing laws that captured and converted it wonderfully into all those 1 and 0's".
This already applies to the quality, look, and style of lenses, and no brand is made alike. It comes down to how do we convert energy of quality and not necessarily energy just in quantity... and this is a skill RED is wonderfully learning and conquering from the ground up and sort of awoken all these other giants now.
Good on you Red Team, don't stop now! :)

Oh yes, and is it not funny how Canon is releasing all these prototype cameras 5Dmk2 and the 7D only to realize there are far too many faults with their present Digic IV processor that are preventing them from overcoming the simplest of problems (any 24p support yet for that 5Dmk2? nope as far as I can tell they have put it on hold as they can not resolve the rolling shutter issue and the downscaling data requirements).
These are fun times & I feel sorry that many of these companies who have done this for far longer are only now stepping up to the plate with the marketing approach... 'this is the next RED killer' ... So to get this straight, camera companies are now intimidated by a passionate & successful business man (and now his team) who started out several years ago making bicycle grips? Again this goes to show that it's the quality/passion of the components (people on Team Red) and not the quantity of them our there (Arri, Sony, Canon etc).
 
Steven Caesare said:
Even if the placebo effect is going to cost you 10x as much?

If so, I have some amazingly transparent Onyx interconnect towers for sale...

-sc
Click to expand...

Note I said "once I have bought something"

How much snake oil is used manufacturing the interconnects? I could be interested :)
 
I got a very thick rug which helped a bit. I think I'll get some new speakers though. Both speakers and source/amp are said to be forward sounding so sound a bit "in your face" combined.

I fear we may have wandered slightly off topic.
 
Irri said:
The only complaint I have heard about REDCODE was from a compositor who said there were compression artifacts. In all honesty, I got the feeling he was really trying to find artifacts rather than having a problem with them.
Click to expand...

Well a lot of FX shots has chosen not to use RED exactly because of this. I've had that problem myself on one that was shot with RED. As the codec gets better it will iron out those problems but right now it can be a problem if you truly have to "dig into" the image.
 
Irri said:
Note I said "once I have bought something"

How much snake oil is used manufacturing the interconnects? I could be interested :)
Click to expand...

On the audio side there is a lot of snake oil in the interconnect market. I have personally taken some medium high end analog RCA interconnects ($70 per 1 meter pair) and compared on a distortion analyzer with 15 cents a foot twisted pair shielded pro audio cable terminated to some reasonably good impedance matched RCA's in the $7.00 each range wired in quasi-balanced configuration with the shield terminated at the input side only and seen less than 1/4 the distortion on my versions, more than 6dB improvement in total noise and distortion on a 1 meter cable in a lab environment. Differences are even greater in custom field installations where there is more EMI and the cable lengths are longer.
 
David Rasberry said:
On the audio side there is a lot of snake oil in the interconnect market. I have personally taken some medium high end analog RCA interconnects ($70 per 1 meter pair) and compared on a distortion analyzer with 15 cents a foot twisted pair shielded pro audio cable terminated to some reasonably good impedance matched RCA's in the $7.00 each range wired in quasi-balanced configuration with the shield terminated at the input side only and seen less than 1/4 the distortion on my versions, more than 6dB improvement in total noise and distortion on a 1 meter cable in a lab environment. Differences are even greater in custom field installations where there is more EMI and the cable lengths are longer.
Click to expand...

Clearly you weren't testing against ultrapure long-grain oxygen free silver interconnects!

-sc
 
You must log in or register to reply here.
Back
Top