The patent gives a 19 step procedure for how to implement the system directly using Adobe Photoshop -- so there need be no confusion as to what they do, although details over the setting variation is the $250 darblet vs the more powerful box, etc. are unclear to me still.
Thanks for posting! I hadn't looked at the patent to discover the Photoshop procedure to simulate the effect. I'll follow the notes in the patent and post the results here this weekend. Cool.
All I know is that watching some of the content they had on that Vango projector at CEDIA was a pretty damn impressive demo. I know they used Avatar for some clips (which is a colorful movie, anyway), but the Darbee box really made it pop.
Joined: 25 Aug 2008 Posts: 2479 Location: Chicago USA & Berlin Germany
Link Posted: Fri May 25, 2012 8:58 pm Post subject:
I would like to get back into this thread but I don't have time between work and travel right now. One thing I did think of that the box could possibly do is some sort of dithering. But this leads me to ask, why should the dithering in a $300 "video processor" be better when added to the dithering inside an $80k+ DP projector or a Sim2 or whatever?
If this really was gold, I would think they would sell the technology directly to the manufacturer to install internally on board the projector. I mean, you are going to buy a $100k projector and then you need a $300 box to make the image better?
The Radiance goes in line with many high end digitals because you can adjust each parameter that is possible and also adjust the magnitude of the parameter. You aren't stuck with one gross setting for HD that applies to all HD TV's, projectors, or whatevers. There is an array of filters that can be added in tiny integers all the way up to really high filtration on the Radiance. And you can change the type of dithering on the HDMI output with the Radiance.
Darn, I was just going to write one or two sentences. I really want to get back to Jeff's Oppo setup thread too because I went through some of the math on that and have som really good answers that will take forever to write down...
Time, time, time...
craigr _________________ *NEW JETI Specbos 1211 Spectroradiometer
Photo Research PR-650 Spectroradiometer
*NEW Klein K10-A Laboratory Grade Colorimeter
Quantum Data QD780 Multimedia Generator
ChromaPure Pro Color Calibration Software
ISFcc ControlCAL Professional Software
Sencore CR7000 CRT Tube Analyzer / Rejuvenater
www.CIR-Engineering.com - craigr@cir-engineering.com
Phone: 865-405-6892
Link Posted: Sun May 27, 2012 1:44 am Post subject:
Craig commented:
> If this really was gold, I would think they would sell the technology directly to the manufacturer to install internally on board the projector. I mean, you are going to buy a $100k projector and then you need a $300 box to make the image better? <
Well, this was originally supposed to ship last October, for $150, but was delayed until this April, at $250. But you are correct, they would like to get manufacturers to incorporate the IP for this directly into their products. Many devices already have the horsepower to utilize their optimized algorithms, and Paul, the inventor, estimates it would add about 18c (that's 18 cents) to the product cost. Or you can buy the box now for 250 clams.
As we all know, digital displays have a terrible problem with sharpness, and need all the help they can get. No, wait. That doesn't sound quite right. As we all know, Blu-ray discs have a terrible problem with sharpness and detail, and need all the help they can get. Hmm, that doesn't seem quite right either. Honestly, with as many real film buffs already concerned about exaggerated sharpness from digital displays, I'm a bit surprised to see that they're clamoring for even more.
If you look at the comparisons here:
you will see that what they're doing is enhancing the contrast, saturation, and sharpness. From my POV, the real accomplishment here is that they're doing so without adding massive amounts of artifacting. However, I'm still not a convert. The way I'd characterize the pairs of images is that the originals are 'natural', while the processed ones are 'dramatic'. Now at first glance, there's something fairly compelling about the ones on the right. They do definitely 'pop', as the company claims. By comparison, the originals seem soft and washed out.
While I do agree with SC that there may be times when you'd want an 'effect' like this, most of the time I wouldn't. Simply because: a) it's NOT natural, and b) it would be visually tiring. I.e., dramatic may look impressive at first, but be less desirable in the long run.
For example, in the image of the mountain village, is the dramatic one really better quality? The contrast enhancement has created deeper shadows, and thus more "depth". But in the process, you can't see as much detail in those shadows. I confirmed this by blowing one of the buildings way up, and noted that 4 window sockets quite visible in the original, virtually vanished in the 'enhanced' version. Which proved that real detail is being lost, even though perceptually it may appear more detailed.
In the shot with the blond, both images look good, but are quite different in their impact. The processed one is much more dramatic, with deeper highlights, like it was shot in a different environment. But that doesn't make it better. The 'feel' of the two are completely different. Personally, I don't want every film I watch to look like "Domino", with hyper contrast. Imagine what this would do to a film like Charlies Angels, in a scene on the beach with bright sun, saturated colors, and deep shadows. Just too, too much. Because it is exaggerating the contrast for a dramatic perceptual effect.
Personally, I like to set things up properly, and calibrate them. Then leave them alone, and enjoy the diversity of imagery... some sharper, some softer, more or less saturated, etc. Ya know, like the cinematographer intended. Call me crazy. Rather than try and make everything uniformly dramatic, and loaded with pizazz. Like Ruby Rhod said in TFE, "Everything must POP POP POP!!" Or maybe, not so much. Does "On Golden Pond" need to be hyper-realistic, and (seem) razor-sharp?
Frankly, I would not have expected the folks at the A-forum to be going ga-ga over this. This type of image-processing technology has actually already been available for a long time, for HTPC owners running ffdshow. There are sharpening filters there that can achieve similar results. Some folks love them, some hate them. But they tend to be the most useful in situations where either, a) the source material itself is soft, and could use some sprucing up. Or b) the display device itself is soft, and could use a bit of help. Say, for example, 8" CRTs without super-wide bandwidth, or peaking circuits that already enhance high-frequency detail. A processor like this, used judiciously, might help balance some of that out. _________________ - Tim
Link Posted: Sun May 27, 2012 11:16 am Post subject:
Well Ive got one on order through AVS, by the sound of it There selling out faster than they can get them..
While I use sharpness enhancements sparingly The only thing I dont like about my Radiance is being softer compared to my DVDO VP50Pro both on default or any other setting, others that have actually done side by side comparisons between the to mostly agree..
Apparently this is due to Lumagen performing the enhancements before Scaling while DVDO perform it After Scaling..
Im hoping running the Radiance/ Darblet combo will be A Killer combo, Darbee's algorithm is ment to be much more advanced than whats available in the radiance,DVDOs or any other consumer product to date..
Joined: 25 Aug 2008 Posts: 2479 Location: Chicago USA & Berlin Germany
Link Posted: Sun May 27, 2012 12:44 pm Post subject:
VideoGrabber wrote:
Personally, I like to set things up properly, and calibrate them. Then leave them alone, and enjoy the diversity of imagery... some sharper, some softer, more or less saturated, etc. Ya know, like the cinematographer intended. Call me crazy. Rather than try and make everything uniformly dramatic, and loaded with pizazz. Like Ruby Rhod said in TFE, "Everything must POP POP POP!!" Or maybe, not so much. Does "On Golden Pond" need to be hyper-realistic, and (seem) razor-sharp?
Frankly, I would not have expected the folks at the A-forum to be going ga-ga over this...
I could not agree more Tim. Thanks for your detailed post.
ecrabb wrote:
I'm all for calibration and faithful representation. It's critically important. But, once we have that baseline, why can't we play with the image a bit to make it look a bit "better" if we think that's what it should look like? I liken it to bass and treble controls or EQ. I've never EQ'd my system completely flat because it sounds boring and lifeless. I've always applied a "house curve" because it sounds better. A lot better. Is it faithful reproduction? Not exactly, but I don't care because *I like it*.
I could not disagree more.
If one calibrates to standard, than when one watches any material it is presented properly. If one adds something to make the image "better" after calibration, it might be perceived by the viewer as good on one image, but will also be hyper exaggerated on another image which looks terrible. Had one stayed at reference, the image would have looked proper in any circumstance.
I don't really care if something looks "better" on one example or not, it is not correct and will lead to issues with another example. Better is qualitative and is not the same for everyone. This sort of "calibration" approach means that a client / viewer is going to have to change parameters like brightness, contrast, gamma, color, or any other qualitatively adjusted from viewing session to viewing session. This is not my goal for my own theater nor for my client's. I want to know everything is correct and then simply enjoy the presentation as it was intended. I don't want my clients to have to make any changes to their display device after I am finished calibrating it.
Video and audio do not perform at all the same and using one as an example for the other is not a good analogy. One of the reasons I am a video engineer as a pro and an audiophile as a hobby is for just this reason. Audio is actually more difficult because often times what measures well sounds like garbage. However with video if it measures well it is going to look good. The closer video gets to reference the better it looks. Any deviation from reference on video will cause issues with respect to the degree of deviation.
Adding "enhancements" to a proper video signal is something I have fought again my entire career. This is the same argument again with a new form of pixie dust. I would be happy to view Darbee and would be open to it, but I have seen so much of this kind of magic over the years that I know where it is going.
Anyway, you guys have fun with this one...
craigr _________________ *NEW JETI Specbos 1211 Spectroradiometer
Photo Research PR-650 Spectroradiometer
*NEW Klein K10-A Laboratory Grade Colorimeter
Quantum Data QD780 Multimedia Generator
ChromaPure Pro Color Calibration Software
ISFcc ControlCAL Professional Software
Sencore CR7000 CRT Tube Analyzer / Rejuvenater
www.CIR-Engineering.com - craigr@cir-engineering.com
Phone: 865-405-6892
Link Posted: Sun May 27, 2012 2:34 pm Post subject:
ecrabb asked:
> ...why can't we play with the image a bit to make it look a bit "better" if we think that's what it should look like? <
You can, if you want. But should you? If the problem lies in the display device, then you calibrate it to correct that limitation. If the problem lies in a source device (some commented specifically that it seems to bring their DirecTV signal more up to par), then maybe something like the Darblet could be useful on the output of that device. If it's specific-program dependent, then I think you should mostly live with it... especially so for film, though less so for TV shows perhaps.
And this is one area that's getting a lot of commentary... that folks like how it pumps up their sports programming. I.e., they like it to pop and be more dramatic. What they don't realize is that they could create a new memory setting on their display that cranks the contrast, saturation, and sharpness, to achieve a similar effect. Name it "Exaggerated". They may have such a "Dynamic" mode already. But of course they don't want to do that, because it would be tampering with their correct calibration. Luckily, inserting a Darblet will have no effect on that.
cinema mad commented:
> ...by the sound of it There [sic] selling out faster than they can get them. <
I expect to see quite the wave of hype over this new device. Which will eventually taper off, as folks come to their senses. The inventor is correct when he analogizes it to the drop-shadow effect used to create a sense of depth on display text. That's great, but do you want to watch all your films with that "effect" added? Darbee is already promoting a version of the device to add this 'effect' directly to the film masters, in post.
> Im hoping running the Radiance/ Darblet combo will be A Killer combo <
I hope so too. Darbee does appear to have avoided the major concerns with previous enhancement devices (i.e., artifacts they introduce while enhancing detail), such as ringing, edge enhancement, and grain exaggeration (as long as the settings aren't boosted too high). Perhaps used in very judicious doses, it will give you back the bit of perceptual sharpness you feel you have lost with your Radiance. If after using it for enough time to have the novelty wear off, you still prefer to have it in-circuit than out, then that will be a good thing.
Unfortunately there's going to be a lot of marketing-speak with this device, which is nothing more than pure BS. There's actually been some nonsense already, but I expect a lot more. This will be touted as a panacea, the "magic elixir, that cures what ails ya". _________________ - Tim
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum