Spec an Audiophile PC With Me Using CAPS 3.0 as a Starting Point, Adding HDMI for Multi-Channel Hi Rez Audio!!! - Page 22 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #631 of 776 Old 04-09-2014, 12:37 PM
RUR
Innocent Bystander
 
RUR's Avatar
 
Join Date: Feb 2007
Location: California Republic
Posts: 2,212
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 49
Quote:
Originally Posted by djnickuk View Post

Yes, you are remembering the RS20i routings / crossovers page.

I have no idea why Dirac on the pc is causing such issues when it works absolutely seamlessly in the Datasat.

Thanks, Nick! That I/O mapping applies to both calibration measurements and playback?
RUR is offline  
Sponsored Links
Advertisement
 
post #632 of 776 Old 04-09-2014, 12:42 PM
Advanced Member
 
djnickuk's Avatar
 
Join Date: Feb 2011
Posts: 693
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 34
Quote:
Originally Posted by RUR View Post

Thanks, Nick! That I/O mapping applies to both calibration measurements and playback?

Yes, basically one sets up your channel routing, ie, sharing subs on channel 4 as I have done to create a mono entirety to be calibrated as one.

Once you have this setup, you then run Dirac. The first page of Dirac will list 16 channels which pertain to these listed in routings / crossovers. You select what ever channels you want to calibrate, then move onto the next part which is measurements and optimizations

And as you can see, I have channels missed out, channels 11 and 12, so I do not select these in the first page of the Dirac calibration and it's just skips those channels and applies filters as required.
djnickuk is offline  
post #633 of 776 Old 04-09-2014, 03:09 PM
RUR
Innocent Bystander
 
RUR's Avatar
 
Join Date: Feb 2007
Location: California Republic
Posts: 2,212
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 49
Makes perfect sense, Nick. Eric's issue with Dirac's PC product is so fundamental to its use, I'm keen to hear their explanation.
RUR is offline  
post #634 of 776 Old 04-09-2014, 07:03 PM
AVS Special Member
 
Bulldogger's Avatar
 
Join Date: Dec 2000
Location: Clinton,MS
Posts: 6,644
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 20
Quote:
Originally Posted by catonic View Post

Steve, since it is clear that you are both very experienced and capable when it comes to this field of high quality audio, could I ask for your opinion on this matter?

In your experience, is it worthwhile going to the 192kHz /24bit playback audio format or does it provide no audible benefit over 44kHz/16bit playback ?
Thanks.
Steve didn't really read your question. Yes going to 192kHz sounds better than 44kHz playback. As Steve answered the difference between 96k and 192k are not so great. Both 96k and 192k can sound worse than 44kHz if the recording is really bad. Typically that's not the case and you will prefer the higher resolution.

Never become so involved with something that it blinds you.
Bulldogger is offline  
post #635 of 776 Old 04-09-2014, 07:10 PM
AVS Special Member
 
Bulldogger's Avatar
 
Join Date: Dec 2000
Location: Clinton,MS
Posts: 6,644
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 20
Quote:
Originally Posted by edorr View Post

Turns out Dirac does indeed screw up the channel mapping. I asked tech support, about the Dirac runtime engine showing a small 4 before channel 5 and a small 5 before channel six. This means channel 4 input gets mapped to channel 5 output, and 5 input gets mapped to 6 output, and input 6 is just gone. In calibration I specify I have no channel 4 input/output but mappings don't change (i.e. 5_>5 and 6->6), so Dirac clearly screws this up. Good to know and saving others embarking on the Dirac Live MCH journey a lot of aggravation. The workaround I am planning is to temporarily insert a speaker on LFE channel for calibration, do a 5.1 calibration, and take out the channel 4 speaker during playback. I will map my channel 4 to channel 1 and 2 in Jriver during playback. This should work OK.
Cool. Glad you found a solution. I was certain the problem lie with Dirac as they are the newer software. Lynx Studio has been around awhile and something so simple was unlikely on their end. Jriver is just far too tenacious an outfit to let channel mapping go unfixed. I remain the most impressed with Jriver and the effort they put forth compared to any that I have used for anything remotely related to audio.

Never become so involved with something that it blinds you.
Bulldogger is offline  
post #636 of 776 Old 04-10-2014, 12:26 AM
Member
 
catonic's Avatar
 
Join Date: Feb 2008
Posts: 157
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 39
Quote:
Originally Posted by Bulldogger View Post

...... Yes going to 192kHz sounds better than 44kHz playback.........

Thanks Bulldogger.
catonic is offline  
post #637 of 776 Old 04-12-2014, 09:51 AM
AVS Special Member
 
edorr's Avatar
 
Join Date: Feb 2009
Posts: 2,446
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked: 24
Quote:
Originally Posted by RUR View Post

Makes perfect sense, Nick. Eric's issue with Dirac's PC product is so fundamental to its use, I'm keen to hear their explanation.

Mapping of non standard configurations has now been acknowledged by Dirac as a software problem, which will be addressed in future version. They will create a feature allow for manual channel mapping in the runtime (DAP).
edorr is offline  
post #638 of 776 Old 04-12-2014, 10:19 AM
RUR
Innocent Bystander
 
RUR's Avatar
 
Join Date: Feb 2007
Location: California Republic
Posts: 2,212
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 49
Thanks, Eric. Now we know it's their problem and not some obscure user setup issue.
RUR is offline  
post #639 of 776 Old 04-12-2014, 12:04 PM
AVS Special Member
 
edorr's Avatar
 
Join Date: Feb 2009
Posts: 2,446
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Liked: 24
Quote:
Originally Posted by RUR View Post

Thanks, Eric. Now we know it's their problem and not some obscure user setup issue.

Well technically it it still MY problem because I have to redo my MCH calibration, but I get your point.
edorr is offline  
post #640 of 776 Old 04-27-2014, 11:42 AM
AVS Special Member
 
Bulldogger's Avatar
 
Join Date: Dec 2000
Location: Clinton,MS
Posts: 6,644
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 20
Quote:
Originally Posted by Chronoptimist View Post

.
If you think that is good, wait until they update to the latest madVR build which is currently in testing. There are some improved scaling algorithms, a new dither option which can significantly reduce the noise floor, and debanding options to improve gradations. (which can also reduce the appearance of compression artifacts)

You won't be able to use these options with low power or passively cooled components though, you need a gaming GPU to use them.
.
MadVR ,"knocks it out of the park," with the new algorithms. I am going to directly compare it to Oppo but don't really see the need. I don't believe the current model Oppo's have a chance at matching this. The performance this offers justifies buying the best GPU that will fit into your HTPC. It's that good.

Never become so involved with something that it blinds you.
Bulldogger is offline  
post #641 of 776 Old 04-27-2014, 11:54 AM
Advanced Member
 
stevekale's Avatar
 
Join Date: Nov 2006
Location: London
Posts: 968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 20
Or just upgrade your key films to a better resolution

Egglestonworks Andra III, Andra III Centre, Rosa (as surround). Rel Stentor II. Theta CB IV. Krell FPB 200 and two KAV 150a amps. Custom-built audio server. Oppo 103EU. Apple TV. Pioneer PDP-LX608D. Synergistic Research "Element Copper" front speaker cable. Cardas Clear Light bal interconnects.
stevekale is offline  
post #642 of 776 Old 04-28-2014, 09:30 AM
 
SBruzonsky's Avatar
 
Join Date: Feb 2014
Location: Gilbert, Az
Posts: 239
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16

The very latest April/May issue of Widescreen Review has an excellent article by Amir Majidimehr "Building High-Performance Audio Servers"! Amir explains the objective specs and performance differences using various methods of transmitting the audio signal out of a computer media server (toslink, coaxial, USB). He discusses and measures performance of two USB to digital converters, the Audiophileo and the Berkeley Alpha. He explains why having a top notch USB to digital converter separate from the DAC is advantageous. From the article, I conclude that my use of the Berkeley Alpha USB digital converter, with my use of USB audio, is indeed the best way to go in regard to sonic performance!

SBruzonsky is offline  
post #643 of 776 Old 04-28-2014, 12:35 PM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Quote:
Originally Posted by stevekale View Post

Or just upgrade your key films to a better resolution
That assumes a better version exists. I have hundreds of films which are not available in HD.
And there's very little 4K native content if you have a new display.
Chronoptimist is offline  
post #644 of 776 Old 04-28-2014, 05:55 PM
AVS Special Member
 
Bulldogger's Avatar
 
Join Date: Dec 2000
Location: Clinton,MS
Posts: 6,644
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 20
Quote:
Originally Posted by stevekale View Post

Or just upgrade your key films to a better resolution
I am going to use this to scale 2k to 4K. Also, to my eyes and I believe edorr shares the sentiment, the Oppo looks better than the lower settings for video when using Jriver with blu ray. With the new stuff, that's all changed. Oppo is no match for this. I see a difference with 1080i as well which is HD. ? For others, who will watch cable and Sat TV broadcast, this will offer great improvements. My blu ray concerts in 1080i never looks so good as well. Two buddies use the Oppo 103 and Oppo 93 to increase the quality of Direct TV and cable broadcast. It offers quite an improvement but falls short of this.

There are a lot of different options with this scaling. You can tailor the picture. I have only began to play with this so still have a ways to go before I figure out what I prefer the most. What I have seen so far however, has killed any desire to every buy a video processor/scaler. I love this hobby. This kind of stuff is what makes it so much fun!

Never become so involved with something that it blinds you.
Bulldogger is offline  
post #645 of 776 Old 04-29-2014, 07:02 AM
Advanced Member
 
stevekale's Avatar
 
Join Date: Nov 2006
Location: London
Posts: 968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 20
I have about 360 DVD rips. Unfortunately most of these are NTSC and not even PAL editions. But while getting them up and running via JRiver has made them more accessible, previously they have been available on a hard disk attached to my Oppo. Have I watched a film from this library in the last 2 years? Not a sausage. For a few core favourites I just bought the Blu Ray disc but I find little desire to watch old movies. I also don't see a point for my current display size of "upgrading" to 4K especially when there is little to no native 4k content - even then, it would likely only be worth it if I wanted to substantially increase my display's size as the eye's acuity depends on the relative display size and viewing distance. Adding a 4k display merely presents a problem that doesn't need solving at this point. Direct TV? Having just been in the US for a week I can understand someone's desire to deal with your relatively awful cable/sat service but I am also of the view that there is very little content worth watching and most of what is worth it is HD. Here in the UK, I simply don't bother to watch "TV" and the little I do watch is 1080p. Mostly we would watch TV shows (the latest being The Bridge) via AppleTV in 1080p. So again, unless you impose a problem by installing an unusually large 4K display, there isn't much to upscale except an old DVD library which is barely accessed in my case. How often do you watch again a movie from 5-10 years ago?

It's fun to play around with new technology and Madshi's algorithyms are state of art (although some have been familiar to me for many years as a photography enthusiast that has used s-spline interpolation techniques for upscaling for many years). When you say "the new stuff" what are you running and are you still running MadVR via JRiver? Jinc? I can upscale NTSC DVD rips to 1080p with Lanczos upscaling without a graphics card.

PS Bulldogger, did you visit Eggworks?

PPS I believe all of my Blu Rays are 1080p but I should check this

Egglestonworks Andra III, Andra III Centre, Rosa (as surround). Rel Stentor II. Theta CB IV. Krell FPB 200 and two KAV 150a amps. Custom-built audio server. Oppo 103EU. Apple TV. Pioneer PDP-LX608D. Synergistic Research "Element Copper" front speaker cable. Cardas Clear Light bal interconnects.
stevekale is offline  
post #646 of 776 Old 04-29-2014, 07:52 AM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 7,785
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 163
I have tried many software DVD players over the years. Still I find my old Xcard with SDI out through a Digital Vision DVNR noise reducer/image enhancer, and now through a Teranex scaler is the best I have seen at home. I prefer keeping the DVD in interlaced mode up to the scaler. The DVNR works best in interlaced as that's how it was designed. And the Teranex excels at de-interlacing.

The Xcard however is 2004 technology and I still use Tvedia as the player as it interfaces so well to the Xcard. Would be nice to move to a more state of the art platform and off this complex hardware chain. Perhaps I need to try the latest Jriver stuff but the problem I have found on all these software players is mediocre de-interlacing. Note this is just for DVD, my HD server runs VLC through a Nvidia FX5500 HDSDI card.

I am also in the process of archiving my LaserDisk collection, mostly concerts, as UNCOMPRESSED SD video. I have a modified laser disk player that outputs composite digital video (an obscure obsolete broadcast standard) to a reference grade decoder to component SDI. I then put these on a RAID using a Blackmagic capture card and Adobe Premier. It takes about 100GB per hour of 10bit material (70GB for 8bit) - but there is no compression so the archive is as good as it gets. and I can always apply compression later.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is offline  
post #647 of 776 Old 04-29-2014, 01:55 PM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Quote:
Originally Posted by stevekale View Post

I also don't see a point for my current display size of "upgrading" to 4K especially when there is little to no native 4k content - even then, it would likely only be worth it if I wanted to substantially increase my display's size as the eye's acuity depends on the relative display size and viewing distance.
Many people on this forum have been severely under-estimating 4K's benefits. Any of the charts you may have seen posted are based on the wrong information. 4K is effective up to about twice the distance those charts recommend.

But you may still not feel that 4K is worthwhile.
Quote:
Originally Posted by stevekale View Post

It's fun to play around with new technology and Madshi's algorithyms are state of art (although some have been familiar to me for many years as a photography enthusiast that has used s-spline interpolation techniques for upscaling for many years). When you say "the new stuff" what are you running and are you still running MadVR via JRiver? Jinc? I can upscale NTSC DVD rips to 1080p with Lanczos upscaling without a graphics card.
Spline and Lanczos are old news, and have been in madVR since the beginning.
Jinc scaling and the anti-ringing function are relatively new additions (maybe a year old at this point?) but the new stuff is NNEDI3 image doubling/quadrupling. (if your scaling is not 2/4x then it will be combined with another algorithm)

These examples are quite small, because they're taken from a post here, where someone else provided the source image.

You should hopefully be able to see that NNEDI3 scaling produces a nice sharp, detailed image, with very nice smooth edges, rather than the jagged (aliased) or blurry results other scaling algorithms produce.
These improvements are a lot more obvious when you're scaling up a DVD to fill a large screen.

This is a very demanding algorithm though, and can require a reasonably powerful graphics card.
Quote:
Originally Posted by Glimmie View Post

Perhaps I need to try the latest Jriver stuff but the problem I have found on all these software players is mediocre de-interlacing.
It depends what you are watching. Video deinterlacing still sucks on an HTPC compared to serious high-end processors dedicated to the task, and an external processor is probably best for non-DVD sources.

However, if you are watching films on DVD, the opposite is true. It should be perfect on the HTPC, as JRiver/madVR will let you force film-type deinterlacing to be used.
This means that the disc is either played back in progressive if it's soft-telecined, or the two separate fields will be perfectly recombined to produce a progressive frame if necessary.

Another trick madVR has up its sleeve, is that if you are playing back NTSC content, which is stored as 59.94Hz on the disc, it is able to use 3:2 decimation to extract the original 24fps source from the disc, and output that at 24Hz for perfectly smooth playback, as if it were a 24p Blu-ray disc.

A standard DVD player or video processor will perform 3:2 pulldown, but still output 60Hz to the display so there will be judder.
Now there are actually one or two Blu-ray players on the market now which can do this with DVDs, but the feature is still very uncommon as I understand it, and I don't know how well it works with them.
Chronoptimist is offline  
post #648 of 776 Old 04-29-2014, 04:17 PM
Advanced Member
 
Nyal Mellor's Avatar
 
Join Date: Aug 2010
Location: SF Bay Area, California, USA
Posts: 920
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 68
Quote:
Originally Posted by Chronoptimist View Post

Many people on this forum have been severely under-estimating 4K's benefits. Any of the charts you may have seen posted are based on the wrong information. 4K is effective up to about twice the distance those charts recommend.

But you may still not feel that 4K is worthwhile.

Spline and Lanczos are old news, and have been in madVR since the beginning.
Jinc scaling and the anti-ringing function are relatively new additions (maybe a year old at this point?) but the new stuff is NNEDI3 image doubling/quadrupling. (if your scaling is not 2/4x then it will be combined with another algorithm)

These examples are quite small, because they're taken from a post here, where someone else provided the source image.

You should hopefully be able to see that NNEDI3 scaling produces a nice sharp, detailed image, with very nice smooth edges, rather than the jagged (aliased) or blurry results other scaling algorithms produce.
These improvements are a lot more obvious when you're scaling up a DVD to fill a large screen.

This is a very demanding algorithm though, and can require a reasonably powerful graphics card.

It depends what you are watching. Video deinterlacing still sucks on an HTPC compared to serious high-end processors dedicated to the task, and an external processor is probably best for non-DVD sources.

However, if you are watching films on DVD, the opposite is true. It should be perfect on the HTPC, as JRiver/madVR will let you force film-type deinterlacing to be used.
This means that the disc is either played back in progressive if it's soft-telecined, or the two separate fields will be perfectly recombined to produce a progressive frame if necessary.

Another trick madVR has up its sleeve, is that if you are playing back NTSC content, which is stored as 59.94Hz on the disc, it is able to use 3:2 decimation to extract the original 24fps source from the disc, and output that at 24Hz for perfectly smooth playback, as if it were a 24p Blu-ray disc.

A standard DVD player or video processor will perform 3:2 pulldown, but still output 60Hz to the display so there will be judder.
Now there are actually one or two Blu-ray players on the market now which can do this with DVDs, but the feature is still very uncommon as I understand it, and I don't know how well it works with them.

I agree on the 4K stuff, I'm seeing resolution benefits at viewing distances greater than in those charts also.

However the upscaling in the 4K stuff I've played with (Sony TVs, PJs) is actually very good. Would the MADVR be better?

Master of Minions, Acoustic Frontiers. We specialize in the design and creation of high performance listening rooms, home theaters and project studios for discerning audio/video enthusiasts.
Nyal Mellor is offline  
post #649 of 776 Old 04-29-2014, 04:38 PM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Quote:
Originally Posted by Nyal Mellor View Post

However the upscaling in the 4K stuff I've played with (Sony TVs, PJs) is actually very good. Would the MADVR be better?
madVR's Jinc scaling has been said to produce results of similar quality to Sony's X-Reality Pro engine - one of the highest regarded - but without the forced sharpening component to the upscaling process; and most of us videophiles would want to avoid sharpening "enhancements" at all costs.
Depending on the source material, NNEDI3 image scaling can be a significant improvement over Jinc in my tests, so I would expect madVR to produce better results than most devices scaling to 4K.
Chronoptimist is offline  
post #650 of 776 Old 04-30-2014, 01:51 AM
Advanced Member
 
stevekale's Avatar
 
Join Date: Nov 2006
Location: London
Posts: 968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 20
Quote:
Originally Posted by Chronoptimist View Post

Many people on this forum have been severely under-estimating 4K's benefits. Any of the charts you may have seen posted are based on the wrong information. 4K is effective up to about twice the distance those charts recommend.

But you may still not feel that 4K is worthwhile.

Spline and Lanczos are old news, and have been in madVR since the beginning.
Jinc scaling and the anti-ringing function are relatively new additions (maybe a year old at this point?) but the new stuff is NNEDI3 image doubling/quadrupling. (if your scaling is not 2/4x then it will be combined with another algorithm)

Quote:
Originally Posted by Nyal Mellor View Post

I agree on the 4K stuff, I'm seeing resolution benefits at viewing distances greater than in those charts also.

There's a lot of science, and behavioral study evidence which supports it, in relation to the eye's acuity. (Far more so than for the ear.) As I noted, it depends not just on viewing distance but distance relative to image size. My Kuro plasma is 65 inches (diagonal). Unless I want to go substantially larger, 4K is a waste of space even at my relatively short viewing distance.

But, hey, everyone loves "pixel peeping" as we say in photography.

As I noted above, I use Spline in photography with good results. (Anyone using Photoshop will be familiar with Bicubic.) JRiver Red October is my first encounter with Jinc. NNEDI3 doesn't appear to be available in Red October.

Egglestonworks Andra III, Andra III Centre, Rosa (as surround). Rel Stentor II. Theta CB IV. Krell FPB 200 and two KAV 150a amps. Custom-built audio server. Oppo 103EU. Apple TV. Pioneer PDP-LX608D. Synergistic Research "Element Copper" front speaker cable. Cardas Clear Light bal interconnects.
stevekale is offline  
post #651 of 776 Old 04-30-2014, 04:42 AM
Advanced Member
 
stevekale's Avatar
 
Join Date: Nov 2006
Location: London
Posts: 968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 20
Quote:
Originally Posted by Bulldogger View Post

The other part of the plan was to use the Deep Cool Dracula GPU cooler. That cooler is not widely available but similar coolers, like this one, the one I chose are http://www.prolimatech.com/en/products/detail.asp?id=2444&subid=2458#showtab .

Just out of interest, why not run a good spec fan-less GPU or are there no 4K capable fan-less cards available at the moment? I would have thought that an entry level gaming card with good memory would be plenty for simply upscaling of video versus the complex rendering required by gaming.

Egglestonworks Andra III, Andra III Centre, Rosa (as surround). Rel Stentor II. Theta CB IV. Krell FPB 200 and two KAV 150a amps. Custom-built audio server. Oppo 103EU. Apple TV. Pioneer PDP-LX608D. Synergistic Research "Element Copper" front speaker cable. Cardas Clear Light bal interconnects.
stevekale is offline  
post #652 of 776 Old 04-30-2014, 05:57 AM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Quote:
Originally Posted by stevekale View Post

There's a lot of science, and behavioral study evidence which supports it, in relation to the eye's acuity. (Far more so than for the ear.) As I noted, it depends not just on viewing distance but distance relative to image size.
Yes, but the information used to generate the charts that are so frequently posted on the forums here is not correct for evaluating a display's resolution.

If you look at the data from recent NHK research, the results are significantly different.
Don't pay attention to that chart at the top - it's essentially making fun of the Carlton Bale chart that is so popularized.

If you actually look at the data, about 100 PPD is where you start to see diminishing returns. If I recall correctly, the Carlton Bale charts are based around 60 PPD. So you would ideally be sitting at twice the distance those charts say.
Quote:
Originally Posted by stevekale View Post

JRiver Red October is my first encounter with Jinc. NNEDI3 doesn't appear to be available in Red October.
It was a very recent addition. Make sure you're on the "latest" update branch. (Help > Update Channels > Latest)
Quote:
Originally Posted by stevekale View Post

Just out of interest, why not run a good spec fan-less GPU or are there no 4K capable fan-less cards available at the moment? I would have thought that an entry level gaming card with good memory would be plenty for simply upscaling of video versus the complex rendering required by gaming.
Something like Spline or Lanczos is not that demanding. Jinc is quite a bit more complex, and NNEDI3 requires significantly more GPU power than anything else - especially if you are going to be scaling content to 4K.
I don't know that a GPU exists which would be able to handle the highest quality settings madVR offers, when scaling to 4K. (though the absolute highest quality options are not necessary for very good image quality)
Chronoptimist is offline  
post #653 of 776 Old 04-30-2014, 06:24 AM
Advanced Member
 
stevekale's Avatar
 
Join Date: Nov 2006
Location: London
Posts: 968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 20
NHK has a very real vested interest in HD. The definition of regular vision (20/20) is, to the best of my knowledge,an eye that can resolve critical detail which subtends to one arc-minute i.e. 60 ppd. Of course it is possible for some to have better than average eyesight. (I believe the physical limit of the eye is something like 20/10 vision.) I will have to dig out what Poynton has said on this issue. Of course, if you want to lift that to 100ppd then, yes, there will be strong arguments for 4K even at sensible viewing distances. (Carlton Bale merely puts the definition of "normal vision" and its relationship to size and viewing distance in pictorial form. He takes the 60 ppd definition of 20/20 vision.)

EDIT: Norman Koren is always good for this sort of stuff. He produced the Imatest software used to test most camera systems. Here's an introduction to MTF a measure of camera and lens resolving power. At the bottom is a piece on visual acuity. http://www.normankoren.com/Tutorials/MTF.html Note the passages on the impact on acuity of pupil dilation.

PS: that top chart at your link is indeed deceptive but classic marketing. Post the a big colourful chart based on 400 ppd and later go on to post in B&W a smaller chart arguing that acuity has an inflection point at 100. But I suspect, particularly having read Koren's overview that the answer is much more complex than the NHK marketing material would have you believe.

PPD calculator here http://home.roadrunner.com/~res18h39/calculator.htm

Egglestonworks Andra III, Andra III Centre, Rosa (as surround). Rel Stentor II. Theta CB IV. Krell FPB 200 and two KAV 150a amps. Custom-built audio server. Oppo 103EU. Apple TV. Pioneer PDP-LX608D. Synergistic Research "Element Copper" front speaker cable. Cardas Clear Light bal interconnects.
stevekale is offline  
post #654 of 776 Old 04-30-2014, 12:15 PM
AVS Special Member
 
Chronoptimist's Avatar
 
Join Date: Sep 2009
Posts: 2,559
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 203
Quote:
Originally Posted by stevekale View Post

NHK has a very real vested interest in HD. The definition of regular vision (20/20) is, to the best of my knowledge,an eye that can resolve critical detail which subtends to one arc-minute i.e. 60 ppd. Of course it is possible for some to have better than average eyesight. (I believe the physical limit of the eye is something like 20/10 vision.) I will have to dig out what Poynton has said on this issue. Of course, if you want to lift that to 100ppd then, yes, there will be strong arguments for 4K even at sensible viewing distances. (Carlton Bale merely puts the definition of "normal vision" and its relationship to size and viewing distance in pictorial form. He takes the 60 ppd definition of 20/20 vision.)
Well part of the argument is that you should not be able to resolve a pixel on the display.
So if the Carlton Bale chart is for 20/20 vision being able to resolve a pixel at X distance, then as a general rule you should be about twice that distance away for the screen to look "real". (i.e. no pixels)

But I don't really want to get into this discussion again - there's plenty of that here.
Chronoptimist is offline  
post #655 of 776 Old 04-30-2014, 12:19 PM
Advanced Member
 
stevekale's Avatar
 
Join Date: Nov 2006
Location: London
Posts: 968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 20
Carlton Bale is just a HT installation guy. He isn't driving our understanding of vision.

Egglestonworks Andra III, Andra III Centre, Rosa (as surround). Rel Stentor II. Theta CB IV. Krell FPB 200 and two KAV 150a amps. Custom-built audio server. Oppo 103EU. Apple TV. Pioneer PDP-LX608D. Synergistic Research "Element Copper" front speaker cable. Cardas Clear Light bal interconnects.
stevekale is offline  
post #656 of 776 Old 04-30-2014, 03:02 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 7,785
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 37 Post(s)
Liked: 163
Quote:
Originally Posted by Chronoptimist View Post

...... It depends what you are watching. Video deinterlacing still sucks on an HTPC compared to serious high-end processors dedicated to the task, and an external processor is probably best for non-DVD sources.

However, if you are watching films on DVD, the opposite is true. It should be perfect on the HTPC, as JRiver/madVR will let you force film-type deinterlacing to be used.
This means that the disc is either played back in progressive if it's soft-telecined, or the two separate fields will be perfectly recombined to produce a progressive frame if necessary. ......

Exactly my experience as well. A good hardware de-interlacer always wins for me with video originated material. While most HTPCs these day can do flawless film deinterlacing, so can any hardware scaler. So in the end that's why I still stick to the hardware scaler.

As I have a very large DVHS library now transferred to a RAID, I still run a version of VLC that supports 1080i all the way through. I then go into the Teranex with 1080i HDSDI and out at 1080p/24 (23.98) HDMI. the Teranex does a much better job than the later versions of VLC that do the 1080i to 1080p internally. Unfortunately they have dropped native 1080i output in the newer versions, probably because very few people have any use for it.

For Bluray I just use a Pioneer player with HDSDI and AES 8 channel output.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is offline  
post #657 of 776 Old 04-30-2014, 06:00 PM
AVS Special Member
 
Bulldogger's Avatar
 
Join Date: Dec 2000
Location: Clinton,MS
Posts: 6,644
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 20
Quote:
Originally Posted by stevekale View Post

Just out of interest, why not run a good spec fan-less GPU or are there no 4K capable fan-less cards available at the moment? I would have thought that an entry level gaming card with good memory would be plenty for simply upscaling of video versus the complex rendering required by gaming.
I don't need fanless for my application. I run front projection and all of the projector models I have used have fans with between 19 and 23db noise level. My server will be used for video primarily. The fans I use are not audible to me beyond 3 feet and even then just barely so and still over-shadowed by the noise of the projector fan. They are completely inaudible from my seating position. If I need a second fanless server just for audio, I'll just build a CAPS type server.

Recently I started using a Celerity HDMI fiber optics cable. Xiaolin, at tech support tells me that they recently tested the cable at 18gps, the HDMI 2.0 standard and easily passed at ONE THOUSAND FEET smile.gif http://www.celeritytek.com/product-DFO.html So you could locate the server almost anywhere in your residence, well away.

What Madshi has done with the latest version of madVR is revolutionary in my opinion. It may appear that we are having the same debate but we are NOT. He has "laid down the gauntlet," with video procession on a level that has certainly NEVER has been done for such low cost before. I submit that the quality of the processing is so good that it will truly test your resolve to have a completely fanless system. It is not often that in my time in the hobby, I have seen such a development. My hat is off to this gentleman. Men have been made very wealthy and contributed far less than what he has given the home theater community. I can not heap enough praise on him. I am that impressed.

I will however concede there is no "one size fits all." Lumagen is fanless and excellent processing. That is a solution for those unwilling to relocate the server or to have even inaudible fans in the room.

Never become so involved with something that it blinds you.
Bulldogger is offline  
post #658 of 776 Old 05-07-2014, 05:18 AM
Advanced Member
 
stevekale's Avatar
 
Join Date: Nov 2006
Location: London
Posts: 968
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 15 Post(s)
Liked: 20
Quote:
Originally Posted by SBruzonsky View Post

The very latest April/May issue of Widescreen Review has an excellent article by Amir Majidimehr "Building High-Performance Audio Servers"! Amir explains the objective specs and performance differences using various methods of transmitting the audio signal out of a computer media server (toslink, coaxial, USB). He discusses and measures performance of two USB to digital converters, the Audiophileo and the Berkeley Alpha. He explains why having a top notch USB to digital converter separate from the DAC is advantageous. From the article, I conclude that my use of the Berkeley Alpha USB digital converter, with my use of USB audio, is indeed the best way to go in regard to sonic performance!

"My preference is for the bridge, as that gives me many more choices of AV products than looking for one with the right USB device."

Interestingly he doesn't cover the ease with which SPDIF can be galvanically isolated (Such as with the Juli@ XTe) and then, of course, buffered and reclocked in a good DAC like one from Theta with Jitter Jail II! The other thing that's forgotten in these tests is that you are rather unlikely to hear noise down 80-90dB anyway - especially in a normal room with much less audible bandwidth. That's not to knock the true quality of the Berkeley product.

Egglestonworks Andra III, Andra III Centre, Rosa (as surround). Rel Stentor II. Theta CB IV. Krell FPB 200 and two KAV 150a amps. Custom-built audio server. Oppo 103EU. Apple TV. Pioneer PDP-LX608D. Synergistic Research "Element Copper" front speaker cable. Cardas Clear Light bal interconnects.
stevekale is offline  
post #659 of 776 Old 05-07-2014, 09:42 AM
Advanced Member
 
Nyal Mellor's Avatar
 
Join Date: Aug 2010
Location: SF Bay Area, California, USA
Posts: 920
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 68
Quote:
Originally Posted by stevekale View Post

"My preference is for the bridge, as that gives me many more choices of AV products than looking for one with the right USB device."

Interestingly he doesn't cover the ease with which SPDIF can be galvanically isolated (Such as with the Juli@ XTe) and then, of course, buffered and reclocked in a good DAC like one from Theta with Jitter Jail II! The other thing that's forgotten in these tests is that you are rather unlikely to hear noise down 80-90dB anyway - especially in a normal room with much less audible bandwidth. That's not to knock the true quality of the Berkeley product.

I didn't read the article but I would assume that a DAC with an internal async USB receiver would be the best solution, since that way you avoid the clock recovery that happens when you use an external USB/SPDIF converter.

Master of Minions, Acoustic Frontiers. We specialize in the design and creation of high performance listening rooms, home theaters and project studios for discerning audio/video enthusiasts.
Nyal Mellor is offline  
post #660 of 776 Old 05-07-2014, 04:46 PM
 
SBruzonsky's Avatar
 
Join Date: Feb 2014
Location: Gilbert, Az
Posts: 239
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 16
Quote:
Originally Posted by Nyal Mellor View Post


I didn't read the article but I would assume that a DAC with an internal async USB receiver would be the best solution, since that way you avoid the clock recovery that happens when you use an external USB/SPDIF converter.

 

Spend a few bucks, buy the mag in print or online, and then read the article. Then tell us what you think!!!@@@

SBruzonsky is offline  
Reply Ultra Hi-End HT Gear ($20,000+)

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off