SDI vs HDMI 480i - Page 2  

thanks Mike, do we need anymore evidence?, like I and others in this thread have said, SDI has proven better than HDMI time and time again, even on the same players and this is on a pure picture quality level, not considering all the other benefits

-Gary

If you look in the 59avi owners thread in the DVD forum (May-ish 2006?) you will see discussion that even with proper player settings the 480i from HDMI is not "correct" - what is on the DVD. There's a couple aberrations when moving up through the grayscale. Plus I can't recall if the 59avi outputs YCbCr. If you are picky, SDI is the best way.

larry

I agree Larry. It is a mystery why that is the case and why Pioneer did it this way.

Dave,
Carl did all his testing with Avia Pro test patterns and a Lumagen HDP (some function in it can display a value that can be converted to black level). Don't you have access to these? It would be cool if somebody could do a similar evaluation of the 970HD. Hint, hint.

larry

The 59AVi has one of the very best HDMI interfaces. It's one of few players that can actually output native digital interlaced component video over HDMI. The Pio and Arcam DV-79s go one better and can output YCbCr 4:2:2 video - supposedly just the same as SDI. But people invariably say an SDI interfce is better still. I've never heard anyone say there is no difference.

Nick

I own 3 or 4 HDMI players that are SDI'ed. SDI is better in each case, sometimes a small margin, other times it is larger.

The Oppo 970 is a lovely cheap transport though.

I think we all now agree SDI in practice is better, question is why or better can we expect a perfect HDMI player. The reason for asking is simple we need these kinds of tests in the new secrets tests for example. First there was a demand for 480i over hdmi now that we have a good load of them (about 10?) its time to have ways to compare them to SDI (as we all agree that is the accepted reference).

I really hope we are moving that way since who of us would not love a 'perfect' hdmi in the form of a cheap $150 player if only as a extra or to help out friends who are not willing to go the sdi route (and its costs and troubles).

Daniel.

A bit off topic but I'd love to hear SD-SDI/DVD comparison with HD-DVD. And will we be seeing HD-SDI modded players and VP's? Has anyone seen HD-SDI?

I wish santa could give me a HD-SDI modified HD-DVD and Blueray player 4 christmas. i`ve been realy good this year.

Larry,

I don't have access to the Oppo 970, but if you want to buy me one I'll gladly take it. I can probably get manufacturer accommodation pricing too

I also don't have a Lumagen anymore, so if you want to ship that along with the Oppo you're more than welcome to!!!

Well ideally no one should live in poverty in first world countries. Alas, the reality of situations is not always so rosey.

The moment you do one thing to the video between the MPEG decoder and the HDMI board, the picture stops being the identical.


Well Denon doesn't have the greatest history of doing things "right", so I wouldn't pin too much hope on them.

It's certainly not impossible to have a bit identical HDMI out, but by the time the money men have picked over the machine, and any design failings have been factored in, the DIY approach will usually make more sense.


Being the addition of a digital interface, it's different from a parts upgrade. The audio analogue of SDI is I2S, something I'm intending to add to my DVD player sometime when I find the time. The same rules apply with I2S (and unlike SDI and HDMI, I2S is genuinely and demonstratably much superior to SPDIF).


Point to point wiring can in fact be superior to a PCB.


SDI is connected up by paralleled single ended cabling, so grounding isn't an issue assuming it's soldered properly, and I've yet to see a player without sufficient internal current headroom to support the mod.

The more expensive version of the Crystalio II comes with 2 HD-SDI inputs. The Vantage-HD has an optional dual HD-SDI input expansion board.

Reports say that both Toshiba's and Samsung's HDVD players have already been successfully HD-SDI modded.

Very damn cool. I can't wait to see this.

Question - With the path of HD-SDI being:

HD-SDI Source -> HD-SDI Video Processor -> HDMI Display (1080p)

Will HDMI negatively impact the signal?

With HDMI 1.0 - 1.2 yes! Why? Because a good VP internally calculates in 10bit 4:4:4 and HDMI 1.0 - 1.2 can not transport that. So the VP has to dither the video data down to a format HDMI 1.0 - 1.2 can understand/transport. HDMI 1.3 with its DeepColor feature should be able to transport the video data in the same format the VP is ending up with, so there should be no loss in image quality (provided there are no transmission problems like jitter, dropouts or whatever).

However, the big question is whether there's a *noticable* impact with HDMI 1.0 - 1.2. Maybe yes, maybe no. I can't really answer that question.

Mark,
I have both AVIA and DVE, and for both patterns, the SDI interface does pass BTB.

BTW, just watched UltraViolet last night via the DV-59AVi. The story natch, sucked big time, and after the 1st 40 minutes or so, I started to lose interest. But the movie is serious eye-candy, and is about the best mastered SD title to date. Compared to my Toshiba HD-DVD, the only major issue is the lack of resolution, otherwise, the colors and contrast are very close to HD-DVD standards. There is still a lot of life in SD-DVD, which still makes my investment in SDI a good choice that will serve for some time to come.

This is highly unlikely if the HD players are built to HDCP licensing requirements. Those requirements state that no unencrypted video data buss can be exposed. That means either the unencrypted video must remain inside a chip or must be routed on internal layers of the PC board using blind vias. In layman's terms that says there is no place to solder to.

Glimmie, but DVD players also have to implement HDCP. And lots of them are SDI moddable. Why should things change with HDVD?

In theory the HD players aren't supposed to exposed the signals requied to hook up the HD-SDI interface, however I think the early kit has been shown to expose the signals required, but its not gauranteed in the future.

John.

From reading the mega DV-59 owners thread on the DVD forum it sounds like most of the gurus that did a lot of testing there don't use "direct" mode and enable some processing to get around the BTB problems and get a better picture. This has basically convinced me to SDI mod the 59. I'd like to have at least one pure source to feed into a quality scaler and then let the scaler do it's thing.

Has any one use the Marantz DV9600 as a digital transport and compare the SDI vs HDMI?

I wasn't aware that HDCP requirements extended to SD DVD. The do have CSS and Macrovision on the NTSC & Y/C outputs. I have even seen some with Macrovision on the Y component output. But even so, the players that have been modified are all older models where HDCP was not required even if it is now. HDCP implementation is rather new even though it has been discussed for the past three years.

There is however the issue of manufacturing cost. If you look through the "Poor Man's SDI" thread, you will see there are a few new players that have combined MPEG decoder chips which include the analog DACS and NTSC encoder. Here, the digital video buss is not accessable. This is simply a result of large mass production where the engineering costs of a model specific single ASIC pay off. You can't sell DVD players at Wall-Mart for $39.99 with third party chip sets and multilayer PC boards to interconnect them. Be assured this will happen to HD DVD palyers as well.

FWIW, all HDMI equipped DVD players have HDCP activated. Doesn't that mean that the HDCP requirements do apply to SD DVD players? Why else should they activate HDCP on the HDMI output? They could differentiate themselves from each other if they did NOT activate HDCP. But still every DVD player has HDCP activated, and I believe to have read multiple times that this is legally a must.


That's not true. There are several HDMI equipped DVD players (all of which have HDCP activated on the HDMI output) which have been successfully SDI modded. E.g. Pioneer's flagship player, all the Arcam DVD players and also an Oppo, just to name a few examples.


That's quite possible/probable. But still only some DVD players are built this way, not all. And it's only done to reduce cost and not to make SDI mods impossible.

the only players without HDCP are those that somehow bypassed the test by avoiding having to sign the HDMI/HDCP/DVD license agreements (E.g. Oppo, Bravo).

are there any HDMI(not DVI) based DVD players that don't have HDCP?

-Gary

Not that I have come across but you 'can get' some discs where HDCP isn't applied as no copy protection is needed and the HDMI output does not have HDCP. Superbits are lovely

The easy way around HDCP is to have a DVI out player and them HDCP is not mandatory, as others have hinted at already.

I've used one for a while running 480I 4:2:2 10bit to a Pioneer XDI Plasma. Picture was excellent, I might be able to borrow the player to compare my SDI modded DV-868 if anyone is interested.

Right now I'm more interested in a HD player with I-Link for Audio, all these players with one HDMI output cause problems for poeple like myself who use an audio only preamp with a video processor.

Gary,

HDMI implies HDCP

Yes, I know that direct mode shouldn't be used, for that very reason. That was the point of the comment I made about it and why I said should when talking about direct mode output. You'd think that since they called it DIRECT, that it would be just that without clipping of the head and toe room

I wonder if it's possible. The Sigma chip in it has the scaler intergrated into it, which would suggest it's not possible, but the Crystalio II uses one, which would suggest it is.

Have you tried asking over at Pixel Magic's forum? Good luck, I'd like to know the results, too.