997 Video Processing

There sure is a lot of deserved excitement over the upcoming 997 and much of the attention so far has been directed at the Trinnov Optimizer and how it might compare to the Audyssey system. So I'd like to open the discussion on the other major improvement in the 997: video processing.

The 997 will come out at about the same time as Oppo's BDP-83 that will have Blu-ray, decoding of all the new lossless codecs, and all the features of their DV-983H for improving standard DVDs. Cost for this new Oppo will be somewhere around $500 to $700. The new Oppo begs the question how the SO Reon video processing stacks up against Anchor Bay Technologies VRS as implemented in the DV-983H? Is there a consensus regarding the relative strengths and weaknesses of these two video processing approaches?

I've heard mostly good things about both the reon and anchor bay chips. I would assume like most things the quality they produce will be determined by how well they are implemented. The things I would most like to see are upconverting of everything to hdmi and pass through unscathed of 24fps/48hz, deep color etc.

Implementation is important no matter what, of course, and we don't know about how the 972/997's Reon implementation will stack up with others. Doing scaling in a receiver is also more complex than in a DVD player, because the possible array of video input signals increases significantly when you open up the field to any connected video component (from 480i analog to 720p and 1080i broadcast HD). Having said that, what I've seen consistently when following discussions of the 983H is that most people who have used both the 983H's ABT implementation and various Reon implementations is that the 983H tends to appear more film-like. The Reon solutions are often said to be a bit more "plastic" in appearance. It's still a very good hardware platform, and many people are very fond of it.

There are so many things to wonder about at this point regarding the 997, but I'll toss out another that falls into the realm of video processing. One issue that I've seen mentioned regarding some other processors and receivers (I remember it coming up with the Onkyo 905 and the 9.8 and 885 processors that were based on it) is a lack of per-input video controls - particularly a "video bypass" option that would allow really high performance sources (such as the 983H) to pass through untouched while still allowing processing to be employed for other inputs.

You all may have already ready this but here is some interesting info on the Reon chip.

SO Reon video http://www.siliconoptix.com/contentEngin...2e-314dbfaf0518

Thanks Gonk for the insights when comparing ABT and SO video processing. From a video standpoint I wonder if combining the new OPPO BD player with the 997 would be a bit of overkill. If the OPPO takes care of DVD and Blue-ray and HD TV is pretty good as-is, that leaves standard def TV, LD players, VCRs, etc for the 997 to process. As was noted, that would be a tall order for a pre/pro to handle such a variety of inputs. Separate video settings for each input would be very nice in that situation. I recall there was a steady number of requests from 990 users to do just that on the audio side and it makes sense to have separate settings for each audio and video input on the 997.

Am I wrong that most displays will still apply their own processing to an input signal? If so, this makes comparisons between SO and ABT even harder.

Blu-ray shouldn't need much video processing - it's those SD sources that need to be "worked over" to be displayed on an HD display that push the need for good video processing. The BDP-83 should offer a great solution for standard DVD (as the 983H already does today), but it can't help with all of that SD television material you may still be watching or those other SD sources (Wii, VHS, LD, and so forth). Considering how dreadful the scaling and deinterlacing are in every HD cable box I've used (particularly deinterlacing), I find the idea of a good video processor to take proper care of those sources appealing - as long as I can skip that extra processing when my good sources (983H DVD and Blu-ray would be basically it, although I do still have an HD-DVD player sitting around here looking rather forlorn as well), it could be great.

Any display will probably do a little processing, although it's not guaranteed. The signal must reach the display's native resolution (which varies from display to display, although a lot of "720p" flat panels are 768p). If it arrives in some other resolution, the TV must make adjustments. Deinterlacing is particularly tough, so delivering a progressive image to a display can make a big difference. Some display also offer a "per pixel" mode that allows a 720p or 1080p (depending on the native resolution) to be displayed without processing, even if it means sacrificing a few lines of pixels.

I'll have to check, but I don't think my 768p Pioneer does 1:1 pixel mapping and it wouldn't matter if the 997/Reon didn't output at that resolution. Before I bought the Pioneer I was looking at a 768p Panasonic display and the ABT/DVDO vp20 because the vp20 could output at 768p and the Panasonic was 1:1 capable. Cost became a big issue real quick.

I wonder if the 997 could output at 768p? There are a lot of us 768ers out there. I imagine the processing done by the display to take a 720p signal to 768p is relatively minor but it still would be a very nice feature.

One other thing, does the SO Reon support automatic/continuous lip sync? Lip sync has to be one of the most annoying problems and really detracts from HD TV. At times it feels like I'm watching an old Godzilla movie! This feature alone might put me on the 997 waiting list.

The problem with 768p output (as I learned while beta testing the 983H) is that there is no "standard" 768p: the video timings and horizontal resolution vary widely, making it impossible for a scaler to offer a "768p" mode without providing the user with some very complex controls for evaluating and coping with that variation. That is why OPPO omitted such resolution control from the 983H (in addition to a lot of software work, it would have required additional hardware to implement a truly "all purpose" resolution control). It's most unfortunate that 768p has proliferated as widely as it has, because that particular resolution is simply absent from any standards (even the HDTV specs, which include a laundry-list of resolutions beyond the 720p and 1080i that we're used to). On the bright side, converting 720p or even 1080p to 768p is relatively "simple" enough that most TV's can do a respectable job with it. Just get the deinterlacing done first - that's a biggie.

Lip sync controls are a good question. The Reon alone can't be responsible for that, although the implementation would hopefully include provisions for coordinating between the Reon and the DSP section so that any video delays created inside the 997 are matched with corresponding audio delays. There is still the issue of video delays incurred external to the 997 that we'll need provisions to correct for...

What I want to know is if we will be able to bypass any and all processing and simply have the 997 pass along the video signal untouched...

I already have a RadianceXD, thankyouverymuch...

We don't know, of course, but I think it's safe to assume that at a minimum there would be a "pass through" option for the video processing. The bigger question in my mind is if we can bypass processing for specific inputs (which can be desirable for Blu-ray players or DVD players with really good video processing, for example) while still using the processing for other inputs (like our cable boxes that have dreadful scaling/deinterlacing or those less high-performance DVD players).

I'm fairly curious about this as well since I have been debating getting the new DVDO edge and now the 997 comes out to muddy my waters

I'm hoping they implement a large protion of the Reon chip and do it well then I could bypass getting the Edge and just take my current iScan HD out of the system altogether.

I would really like to upgrade to the 997 from my 950 since it's getting a bit long in the tooth and falling behind farther and farther every day and I just can't do both any time soon, especially with the 997 being $1400.

Per Scooby's question, there is potential problems if the Reon processing can't be bypassed.

The Reon implementation in my Integra 9.8 is faulty. There is a color shift. The Reon outputs color space 601 (SD) vice 709 (HD) when outputting 1080i/p. My display (SXRD XBR1) receives the HD signal and uses specs for decoding a 709 color space when it is fact a 601.

Bottom line: Green is undersaturated approximately 20% versus blue and tints are off accordingly.

Onkyo/Integra has never fixed the issue despite the fact that the Reon doesn't have the same problem in other implementations (e.g., my Toshiba XA2 HD DVD player).

I've therefore chosen not to use the Reon with my Integra 9.8. I would fear any implementation that did not allow for a "bypass".

Finally, this is an issue with YCbCr since RGB doesn't use the different color spaces.

I've read a bit about folks really struggling with the way video processing was employed on the Onkyo 905 platform (including the Integra 9.8 and Onkyo 885 processors). My hope is that there will be a per-input bypass option - so that the Reon can be used for those "just average" video sources (cable box, old DVD recorder, and whatever other odds and ends might be floating around such as older game consoles or VCR's) can benefit from a good scaler while those better HDMI-based sources (Blu-ray, well-scaled and deinterlaced DVD) can be set to pass through unaltered. Fingers are crossed...

Can I infer that a 1080p signal from a blue ray disk player via hdmi will be detected as a 108op signal by the 997 and therefore no video processing will occur by the outlaw unit? Additionally if I have a regular DVD player hooked up via component, then will the Outlaw processor upconvert the signal to 1080p and output via HDMI? I was considering getting the new OPPO universal player when it arrives However based on the discussion in this thread if I am satisfied with outlaws upconversion then why bother, just get a basic BD player. Thoughts?

If the Outlaw works the same as the Integra DTC-9.9, then yes, both of your scenarios will work just as you've stated, I believe. 1080 signals will not be upconverted in any way and all other signals will be upconverted.

But I too, like gonk, am hoping that the 997 will offer control over upconversion on a per input basis.

As I understand it, the DTC-9.9 will only pass 1080p through unaltered if it's 1080p/24 or if a number of settings are specifically disabled (all of the Reon picture controls and the OSD pop-up). If a 1080p/60 signal hits the Onkyo or Integra units and there are picture control settings (noise reduction, brightness/contrast, etc.) or the OSD pop-up is enabled, the Reon chip would have influence on the signal - even though the resolution remains 1080p. This is why I am hopeful that the 997 will offer a per-input video bypass option that would allow specific inputs to be flagged as "pass-through" while still making full and optimal use of the video processor for other inputs that would benefit from it.

I thought that the DHC 9.9 was assignable by input & that this was an issue with the DTC 9.8? I could be wrong, but that is what I understood.

Did they fix it in the new generation? I hadn't heard that, but it'd be good if they did.

I don't understand the need for all of this video processing in pre/pros and receivers. Most DVD/BR/HD DVD players as well as displays have sufficient and in most cases very good processing capabilities. Switching is a different story, but I don't really think we need more video processing.

The rationale as I see it is that you can purchase a single good scaling and deinterlacing solution and use it for all of your sources - which is why standalone video processors came into being in the first place. What's happening is we have started seeing those processors get shoe-horned into surround processors. The difference between average and exceptional in this field is potentially huge (especially for deinterlacing, although scaling and other features such as noise reduction also contribute).

Personally, I go back and forth a bit these days. On the one hand, a good standalone video processor like the DVDO Edge that was able to get instructions from a receiver or processor (to automate the video switching) would be a great solution that would also fit with the traditional "separates" model (replace one piece of the system without paying to replace others). On the other hand, we do end up with a lot of components, remotes, and cabling to manage in that scenario that could be simplified by integrating the video processing into the receiver or processor. As always, there's no one perfect answer...

The processing capabilities in most TVs suck compared to a good video processor. The margins on most TVs are so thin these days and consumers aren't willing to pay more (in most cases) to get a TV/projector with a good processor.

Agree with Gonk. The Reon in my Onkyo 885 is going completely to waste because I don't want it mucking with my blu-ray picture, and it is global-or-nothing.

Which is a shame, because it did make a pretty worthwhile improvement in my Dish Network pic. Not worth the mucking with the blu-ray picture though.

The 9.9 may well have rectified this, Gonk, as it was a major peeve among the 9.8/885 crowd, and I assume Onkyo paid some attention to that demographic, since the unit sold far beyond their expectations.

ndksurfer wrote:

[QUOTE] I thought that the DHC 9.9 was assignable by input & that this was an issue with the DTC 9.8? I could be wrong, but that is what I understood. [QUOTE]

...and I ask him and others: Is it true that it is NOT assignable in the Integra 9.8?

Does anyone know whether it will be assignable in the 997? It think PeterT and Scott know. Wanna tell us?

It is true that it is not assignable on the 9.8, Onkyo 885, or Onkyo 905 receiver.

gonk,
Thanks for clearing that up.
Here's hoping.

Still (off topic) I would like to read a good shootout between Audyssey and Trinnov.

I think we'll see at least one (probably more than one) once some Trinnov gear lands. It will definitely be interesting.

According to the manual for the DHC-9.9, the control for upconversion remains unchanged.

However, it should be noted that HDMI sources should NOT be upconverted in any case based on what the manual says (only S-video, composite and component sources are upconvered).

http://63.148.251.135/redirect_service.cfm?type=own_manuals&file=SN_29344765A_DHC-9.9_En_web.pdf

This is what the manual states for the DTC-9.8 as well.

http://63.148.251.135/redirect_service.cfm?type=own_manuals&file=DTC-9.8_En_A.pdf

Of course, this doesn't mean this is how it works in real life.

But a video processor like the Reon does more than merely deinterlace and scale. One of the concerns that I saw from 9.8, 885, and 905 owners (and that I found pretty reasonable) was that it was not possible to make use of the other Reon features (noise reduction, for example, or picture adjustments) with sources that needed them (cable boxes with really bad scaling and deinterlacing, VCR's, cheap DVD players, etc) without also applying those features to all inputs. That's where a video processing bypass option for each input would be valuable.

The manual finished downloading, and I think there has been a change in the 9.9 that would fix this issue. It's not a bypass control - instead, they added an adjustment called "Picture Adjust" for each video input that provides separate controls for the Reon (including a bypass setting).

Nice...well, hopefully, the 997 offers more granular control. More flexibility is always a good thing.