AVS Forum banner

1 - 20 of 1421 Posts

·
ABSOLUTE ULTIMATE AV
Joined
·
5,695 Posts
Discussion Starter · #1 · (Edited)
This thread is for posting objective evaluation data and analysis results with respect to the various Video Processors both in their own right and as compared with eachother.

This includes the likes of the Lumagen Radiance PRO, the madVR Envy, madVR HTPCs, and any other video processing devices.

This will be a great place to carry out and post the results of subjective shootouts as well as subjective direct A-B comparisons.

I would like for this to also include evalution and comparative video performance of video processing that is built into devices as well, wherein for example the comparative video performance versus the JVC projectors' built-in HDR Dynamic Tone Mapping (DTM).

Everyone please feel free to carry out your own analysis, evaluations, feedback and comparisions with respect to the various video processor devices and post the details here in this thread accordingly :)
 

·
Premium Member
Joined
·
4,575 Posts
This thread is for posting objective evaluation data and analysis results with respect to the various Video Processors both in their own right and as compared with eachother.

This includes the likes of the Lumagen Radiance PRO, the madVR Envy, madVR HTPCs, and any other video processing devices.

This will be a great place to carry out and post the results of subjective shootouts as well as subjective direct A-B comparisons.

I would like for this to also include evalution and comparative video performance of video processing that is built into devices as well, wherein for example the comparative video performance versus the JVC projectors' built-in HDR Dynamic Tone Mapping (DTM).

Everyone please feel free to carry out your own analysis, evaluations, feedback and comparisions with respect to the various video processor devices and post the details here in this thread accordingly :)
Subscribed. This should be interesting.

Mark


Sent from my iPhone using Tapatalk Pro
 

·
Registered
Joined
·
1,382 Posts
Subscribed. This should be interesting.

Mark


Sent from my iPhone using Tapatalk Pro
Yes. But I am afraid not many people will own both the Envy and the Radiance Pro to make a comparison. Perhaps some users will get together for a shootout with their VPs.

As an aside - is there any contractual limitations to prevent dealers from selling both Lumagen and Envy?


Sent from my iPhone using Tapatalk
 

·
Premium Member
Joined
·
791 Posts
is there any contractual limitations to prevent dealers from selling both Lumagen and Envy?
No, there´s not.
We´re an Envy dealer and currently looking into selling Lumagen as well. I´m hoping to receive a test unit shortly. (however, don´t expect a shootout report from our side, we´ll not go that route ;))
As i mentioned already in other threads: although both the Lumagen and Envy do have their overlaps in functionality, there are a couple of differentiators that would make one a better choice over the other for a specific use case.
 

·
ABSOLUTE ULTIMATE AV
Joined
·
5,695 Posts
Discussion Starter · #5 ·
Yes. But I am afraid not many people will own both the Envy and the Radiance Pro to make a comparison. Perhaps some users will get together for a shootout with their VPs.
Don't worry, there will be more than enough. 😉

Important to note however that what we are looking for here is truly impartial, unbiased and honest evaluations and comparisons here. Also, it will be fundamentally important to ensure that all devices are setup and calibrated properly. Furthermore, that all comparisons are true apples-vs-apples comparisons where relevant. And objective measurements data is preferable to subjective observation. 🙂
 

·
Premium Member
Joined
·
791 Posts
Also, it will be fundamentally important to ensure that all devices are setup [...] properly.
Well, that point alone will be a challenge. Define "properly" :)
Usually on these kinds of VPs, there are Zillions of combinations of settings. And there might (will?) be specific settings that work well with some material but worse on others. Even down to specific scenes in a movie.
I´m really interested how you´re going to handle that.
 
  • Like
Reactions: Archibald1

·
ABSOLUTE ULTIMATE AV
Joined
·
5,695 Posts
Discussion Starter · #7 ·
Well, that point alone will be a challenge. Define "properly" :)
Usually on these kinds of VPs, there are Zillions of combinations of settings. And there might (will?) be specific settings that work well with some material but worse on others. Even down to specific scenes in a movie.
I´m really interested how you´re going to handle that.
When you setup and calibrate video equipment you follow the same parameters with respect to targeting the same objective of optimum video performance. For example you should achieve a minimum of <= 2 dE color accuracy; wherein the color accuracy is not going to vary across content or from scene to scene. Same goes for other objectively measureable aspects of video performance, such as HD-->4K upscaling and HDR Dynamic Tone-Mapping (DTM) performance. Also, literally nobody is going to be pausing a movie and changing the settings on a scene-by-scene basis. You want the settings which achieve the best possible overall video performance with respect to the whole movie / video content.

With respect to different types of source content, yes there will be different sets of settings for the likes of HDR and SDR, however the particular video processor should be setup and calibrated properly and optimally with respect to each of these.
 

·
Premium Member
Joined
·
791 Posts
Also, literally nobody is going to be pausing a movie and changing the settings on a scene-by-scene basis. You want the settings which achieve the best possible overall video performance with respect to the whole movie / video content.
Sure, in real live you just want to watch a full movie.
But as you know, some video processors have their strengths and weaknesses depending on the content of the scene, especially with regards to DTM (e.g. highlight recovery). But of course, we want the overall quality, not a quality broken down to scenes.

Regarding upscaling: you can do a lot with sharpening, edge enhancement, etc. How do you measure that objectively? In my opinion, this is more about personal taste.

I´m looking forward to what comes up here. (y)
 

·
ABSOLUTE ULTIMATE AV
Joined
·
5,695 Posts
Discussion Starter · #9 ·
Regarding upscaling: you can do a lot with sharpening, edge enhancement, etc. How do you measure that objectively? In my opinion, this is more about personal taste.

I´m looking forward to what comes up here. (y)
Upscaling and sharpening / edge enhancement are two completely different things. With respect to objectively measuring upscaling performance the correct way of going about it is to take a native 4K video source content, downscale it to HD resolution, then upscale it back to 4K resolution via each of the video processors and objectively measure how close the native 4K original they are. 'Personal taste' has absolutely NOTHING to do with upscaling performance. Sharpening / edge enhancement on the other hand are optional enhancement features which can in fact result in worse video performance due to inducing manifestation of video artifacts. Usage of these does involve personal taste, however IMO unless these are able to enhance the video image without introducing video artifacts then they are not fit for purpose and should not be used.
 

·
Premium Member
Joined
·
791 Posts
With respect to objectively measuring upscaling the correct way of going about it is to take a native 4K video source content, downscale it to HD resolution, then upscale it back to 4K resolution via each of the video processors and objectively measure how close the native 4K original they are.
Hm, i´m not an expert on this, but since there are lots of different down- and upscaling-algorithms, i wouldn´t expect to get an objective judgement on upscaling image quality doing it this way.
Especially if your using a still image as opposed to live video.
 

·
Registered
Joined
·
3,229 Posts
I have to agree with hockyAVS on this one... doing a downscale and then an upscale will never ever get you near the same quality as native 4K. This was proven long ago when the whole downscaling screenshots to prove they were or were not native 1080p.

The only way to do it, objectively, is by two identical players (Oppo 203 since they have source direct is my preference) and in one put the BD of a movie while on the other use the 4K UHD of the same movie, while being fed into TWO identical displays. Either mask each display (projectors are good for this) or putting two flat panels one above the other. This is the only way you can actually see BOTH images being displayed at the same time. Using one display is possible but the black-out period while switching HDMI is going to influence the results.
 

·
Premium Member
Joined
·
791 Posts
As a first step, i would probably go with 1080p multi-burst images and let the probands upscale these.
OK, this would be even more "artifical" content compared to a movie, but at least that should give a first indication about what the scaler do with the data.
 

·
ABSOLUTE ULTIMATE AV
Joined
·
5,695 Posts
Discussion Starter · #13 · (Edited)
Hm, i´m not an expert on this, but since there are lots of different down- and upscaling-algorithms, i wouldn´t expect to get an objective judgement on upscaling image quality doing it this way.
Especially if your using a still image as opposed to live video.
Firstly, judgement is subjective not objective. I am talking about objective scientific measurements. Secondly, you carry out the downscaling competantly. If you don't know how to do this then you should not be attempting any of this in the first place. And thirdly, it is best to be carrying out the objective measurements using moving video, not still images, and most certainly not nyquist test patterns.

For your interest, the first lot of analysis which we ourselves will be reporting is being done in a professional studio, by a certified video professional who amongst other things carries out post-production on movies for movie studios, using circa $1.5 million worth of professional analysis equipment including studio master grading monitors, Christie Dolby Vision projectors (as per feature in Dolby Cinemas), Tektronix analysis equipment and professional studio software.
 

·
Premium Member
Joined
·
791 Posts
Sounds promising.
Hopefully the outcome of this material battle translates into a tangible outcome for normal setups. :)

EDIT: Sorry for the "judgement" - it wasn´t meant that way. I´m not native in englisch.
 
  • Like
Reactions: ARROW-AV

·
ABSOLUTE ULTIMATE AV
Joined
·
5,695 Posts
Discussion Starter · #15 · (Edited)
I have to agree with hockyAVS on this one... doing a downscale and then an upscale will never ever get you near the same quality as native 4K. This was proven long ago when the whole downscaling screenshots to prove they were or were not native 1080p.
Incorrect. This will depend upon the respective downscaling and upscaling performance. You can most certainly get back to very nearly the same quality as native 4K. The objective measurements irrefutably demonstrate this.

The only way to do it, objectively, is by two identical players (Oppo 203 since they have source direct is my preference) and in one put the BD of a movie while on the other use the 4K UHD of the same movie, while being fed into TWO identical displays. Either mask each display (projectors are good for this) or putting two flat panels one above the other. This is the only way you can actually see BOTH images being displayed at the same time. Using one display is possible but the black-out period while switching HDMI is going to influence the results.
Incorrect again.... Sorry! (Please don't hate me LOL) 😉 🙂

Firstly, that is not objectively scientifically measuring anything. That is setting things up for subjective observation and judgement. The best, correct and proper way of going about this is to take objective scientific measurements.

The downscaling is being carried out by the same professional studio equipment and software which upscales movies which have been mastered using 2K digital intermediates to 4K for the consumer 4K HDR Blu-Ray movie releases. After the native 4K video content has been downscaled to HD, when this is used subsequently to upscale the content back to 4K again this yields results which objectively measure almost exactly the same as the native 4K original. This demonstrates that the method works perfectly. The downscaled video content is then run through each of the respective video processors and the corresponding upscaling performance is objectively measured in each instance. There is an actual score that is outputted here, wherein the higher the score the closer to the native 4K original / reference.

What is all-important here is firstly the methods and equipment and/or software which are used for the downscaling; and secondly objectively scientifically measuring the upscaling performance. Also important to note that this is being done with respect to actual moving video movie content, not static frames or test patterns.
 

·
.NET Solution Architect
Joined
·
979 Posts
I have to agree with hockyAVS on this one... doing a downscale and then an upscale will never ever get you near the same quality as native 4K. This was proven long ago when the whole downscaling screenshots to prove they were or were not native 1080p.

The only way to do it, objectively, is by two identical players (Oppo 203 since they have source direct is my preference) and in one put the BD of a movie while on the other use the 4K UHD of the same movie, while being fed into TWO identical displays. Either mask each display (projectors are good for this) or putting two flat panels one above the other. This is the only way you can actually see BOTH images being displayed at the same time. Using one display is possible but the black-out period while switching HDMI is going to influence the results.
As ARROW already mentioned the purpose of this method is not to achieve the same video quality of 4K but to implement an independent upscaling test for the video processors you are comparing and have a real 4k video as the reference picture to how it actually should be looking.

Sent from my GM1913 using Tapatalk
 

·
ABSOLUTE ULTIMATE AV
Joined
·
5,695 Posts
Discussion Starter · #19 ·
The only way to do it, objectively, is by two identical players (Oppo 203 since they have source direct is my preference) and in one put the BD of a movie while on the other use the 4K UHD of the same movie, while being fed into TWO identical displays. Either mask each display (projectors are good for this) or putting two flat panels one above the other. This is the only way you can actually see BOTH images being displayed at the same time. Using one display is possible but the black-out period while switching HDMI is going to influence the results.
Further to what I said previously there is another reason why you should not carry out a comparison in the manner you suggest here... The BD of a movie is not an HD exact equivalent of the 4K UHD version of the same movie. For starters the 4K UHD version of the movie will be graded differently, with the BD version being SDR and the 4K UHD version being HDR. Furthermore, they might have been graded using completely different original masters as well. Also, with some older movies sometimes the studio will go back to the original (e.g. 35mm or 70mm) and carry out a restoration, creating a new 4K master. Hence you will be comparing apples vs oranges if you go about things this way.
 

·
Super Moderator
JVC RS4500 | ST130 G4 135" | MRX 720 | MC303 MC152 | 6.1.4: B&W 802D3, 805D3, 702S2 | 4x15 IB Subs
Joined
·
10,807 Posts
The only way to do it, objectively, is by two identical players (Oppo 203 since they have source direct is my preference) and in one put the BD of a movie while on the other use the 4K UHD of the same movie, while being fed into TWO identical displays. Either mask each display (projectors are good for this) or putting two flat panels one above the other. This is the only way you can actually see BOTH images being displayed at the same time.
I don't see this being an accurate measure at all. The BD version and UHD version have been separately mastered and the HDR cannot be completely undone. That alone throws this off. The other issue is how the movie studios process the content.

I see no issue at all with the downscale and compare method to be honest. The upscale may never look the same as the native 4K but the quality of the job is what we are looking at. By the way, I believe madVR A/I upscaler is trained using exactly that method. It's fed several images of resolution X downscaled to Y then it has to learn how to best upscale Y to X2 and compare X2 to X for quality.
 
1 - 20 of 1421 Posts
Top