New DVDO iScan Duo [2.0 firmware released] - Page 44

Does it not seem a little unlikely that the EDIDs are wrong with the JVC RS1, HD750, HD950 etc? And then it is also wrong with one of the Panasonic's.

I have discussed this issue with AUTO ending up as 1080i/60 with Ken before, and the real issue seems to be that they don't have any of these units to test with.

I think its more likely the Duo's interpretation of the EDID from the projectors that is wrong, but if Ken doesn't have the kit to test with, I don't see how it will ever get fixed.

I wonder if we should be reciving any new updates soon for this unit.

I dunno, Jon, but I can tell you that I have often had issues whenever I have one "auto" circuit talking to another "auto" circuit. The solution has always been (at least for me) to pick one side to be "hard" set while leaving the other side to "auto", or better yet, hard set both sides when practical. Then that way only one circuit does the adjusting instead of having 2 circuits trying to decide how to adjust to the other. I don't know if my theory is correct or not, but it seems logical to me...

Anyway, Ron's idea worked great!

As an aside, I had written to Anchor Bay customer service just before I read Ron's answer. I am happy to report that I got an identical answer from Ken within a few hours of my writing - a very cordial, helpful reply. Since this is my first experience with AB's customer service, I am very pleased.

One other thing, does anyone know what the difference is between the "1:1 Frame Rate" setting and the "Frame lock" setting? The manual is not very clear.

My BD85 Blu-ray player can use color interpolation to go beyond 8bit (it's setting is ycbcr 4:4:4). Does anyone know if the DVDO can accept a YCbCr 4:4:4 input and benefit from the interpolated color?

It can handle 30 & 36 bit color!

Thanks! I spoke with Ken several times earlier at AB, very nice guy.

Im now trying to find out if my series of Plasma supports 4:4:4. If it does, he said to have the BR player send out native 1080p/24 4:2:2 to the DVDO, then let the DVDO interpolate to 4:4:4 to the display. If my tv does not handle 4:4:4, then 4:2:2 from the player to the DVDO, and 4:2:2 out of the DVDO.

I read that the Duo processes in 4:4:4 so output set to 4:2:2 would mean you are down sampling color. Sending it 4:2:2 is fine but output would probably be best set to 4:4:4 or RGB. Get the Spears and Munsil disc and test every mode to see which mode is best.

Reason he said that is, because it'd be better for the Duo to down sample to 4:2:2 than letting my display do it.

The duo is pretty sweet! Good point!

From what Im getting, is make sure all Inputs are native out, and anything the TV might have to do to the signal after it leaves the Duo, change that and make the Duo do it first

That's true! Running tests with Spears and Munsil disc I couldn't see the difference between 4:4:4 and 4:2:2 from my Pioneer player and my JVC projector processes in 4:4:4 so I stick with all 4:4:4 and calibrate set to 4:4:4 between the Duo and JVC.

I tried all the different settings. Its hard to see any differences with the Duo. Its that good.

It'll be nice to have a central unit that takes in native, knows what your tv needs, and does all the work

Not a technical person, but I think the Duo processes in 4:2:2.

If I am correct, anything sent to it in 4:4:4 must be changed by the Duo for processing, then changed as normal to output 4:4:4.

However, if you see no difference between the inputs then that is the measure of the Duo's capability.

Incidentally, because I have an LCD and not a PJ, I output RGB from the Duo rather than 4:4:4 to prevent my display from doing this necessary change.
I think the Duo does the conversion better.

I've seen guidance for the same reason that the best approach is to output RGB to the panel - i.e. get the Duo to do as much of the work as possible as, in the end, it always needs to be RGB. I don't know if this applies to all screens, but RGB output defo seems the best for my Pana panel.

I read some where it processed in 4:4:4 not 4:2:2. I don't think they would force downsampling and then process. Usually processors process in either RGB or 4:4:4. In my case my JVC converts everything to 4:4:4 before converting to RGB out to the panels. The Duo knows this. It outputs 4:4:4 in auto mode to my projector.

This is correct only if your display has no color/saturation control for the RGB input.

If it does, then it must transcode RGB to component for processing and back to RGB to drive the panel. This creates two unnecessary conversion steps that may have a negatve impact on the image.


Ken Whitcomb

So if I calibrate with Duo outputting RGB, probably incurring the two steps you mention, this is more demanding of the panel than outputting 4:4:4 and allowing it to convert to RGB?

I ask because I try to give my panel as little to do as possible because I like the way how the Duo processes and did not realise the point you make.

Also, the spec sheet for the abt2010 says the input to the chip is 30 bit 4:4:4!

Hi Jon,
Anchor Bay is well aware of the problem with DUO/JVC RS line of projector. It's only their lack of will and interest that prevent them to fully support these projectors. I've waited months to get the HDCP problem between the DUO and JVC RS15 resolved, so it's unlikely you'll see a EDID fix soon or if it will ever be fixed.

Time is on their side...

Many thanks for the correction.

Have found this spec. sheet but cannot find the processing information other than it uses 10 Bit processing. Perhaps you could provide a link if it is not the one you refer to?

http://www.anchorbaytech.com/product...rs/abt2010.php

That is the spec sheet! Frankly it would be degrading to compress to 4:2:2 to process color. I would be very surprised if they did this.

It can take 4:4:4, 4:2:2 and RGB. A HDDVR is only going to send it 4:2:2, but Blu-ray players can send out 4:4:4. All of this is moot, if your tv can't handle 4:4:4, which many can't.

If you send 4:4:4 from a player to the Duo, and it sends it out that way to a tv that cant handle it, the tv will downsample to 4:2:2. In this, it'd be better to set your Duo to 4:2:2 out, the BR player to 4:2:2 out, so there is no interpolation added, then downsampled, it's just xtra steps that arent needed.

All starts with the display, and whether it can handle 4:4:4. If not, no need to have the BR player and the Duo ADD interpolation, when it just get's downsampled at the end.

I need to correct my last statement, the HDDVR from Directv does output 4:4:4. BUT, once again, if the tv does not support it, would you rather the TV or the Duo downsample to 4:2:2?

I agree with letting the duo downsample if the display doesn't handle 4:4:4. My point was the processes 4:4:4. I just can't find where I read it but it was from someone who should know for sure. If I am wrong I hope someone who knows for sure will chime in.

I know my JVC processes in 4:4:4 before converting to RGB.

Sending 4:2:2 from the player is probably beneficial because Y is in 10 bit but if your player can output 12 bit in 4:4:4 that may look better. I go by what I see testing using the Spears and Munsil disc. Just assuming processing in the Duo is best may not be so. Everyone should view the conversions with the test screens.

Hi Ken

I've had a dig around for the post - I think it was from Chee, but can't find it but defo from someone at ABT. I think the reasoning was that even in the panel can take 4:4:4, it still has to convert to RGB for display. So, is it better to convert to RGB in the Duo or are you saying that a panel will always convert this RGB to YUV even if you're not adjusting color/hue (just by those controls being present)?

pete S

A useful method to determine what works best is to look at the chroma multiburst pattern on the Spears & Munsil disc. See if the finest horizontal and vertical lines (in the boxes on the right) are better resolved in RGB or YCbCr. If there is no visible difference, you should set the Duo to output YCbCr.


Ken Whitcomb

Pete,

There is no color control possible in the RGB domain. In the component domain, the chroma channels (Cb and Cr) can be altered to raise or lower saturation.

Therefore, any set that has a color/saturation control for a RGB input must convert to component to provide that capability.


Ken Whitcomb

Hi Ken

hmmm - sadly, I follow your reasoning - which will mean having to recalibrate the damn thing again as it's bound to have an impact on the setup. grrrrrr

Thanks - I think.....

pete S

There is supposed to be one on the way sometime in Q3 with 3D pass-through support. The EDGE is to be similarly updated.

New toy