DVD 480p vs 720p vs 1080i

[junior77blue]

I'm just curious how much 'better' the picture quality gets when using DVD players with scaling built in to them.

I just purchased the Denon 2900 and I'm wondering how its output via component @ 480p compares to say a Denon 3910 with scaling upto 1080i via HDMi or DVI.

I could see this being beneficial with HD video. But since DVD is 'currently' not HD what are the true benefits?

Albeit, I understand with DVI and HDMI its all kept digital.

[Andrew Pratt]

That depends on your monitor's abilty to handle those other formats. For example with my 1/4 HD projector I saw little to no improvement upsampleing the image to 1080i while others claim to see significant improvements.

[junior77blue]

Well, I have a Hitachi LCD RPTV, 60". That is 1080i capable.

Not sure what size the LCD chip is inside.

[Kevin P]

Since you have the upscaling DVD player and a RPTV that (I presume) has the HDMI or DVI inputs, why not hook it up and see for yourself?

Since LCDs already have native-res scalers built in, you may or may not see an improvement depending on which scaler is better. Hence my suggestion to try it and see for yourself.

[Shane Martin]

Saw no improvement here. Buy the best 480P dvd you can is my opinion.

[JonMarsh]

1080i just means it will accept a signal and display something on the screen.

Most of the LCD and DLP RPTV systems are 1280X720 or 1368X768. Some will do 1:1 pixel mapping with a native rate input signal- some will scale anything coming in, and actually overscan the image. In general, before considering an external upscaling player, it's a good idea to get an idea of how well your RPTV is handling different resolutions. Overall, I don't think there's much to be gained resampling a signal several times. I have seen the situation where some chipsets in digital projectors work better downsamping than upsampling.

I think it's dangerous to generalize about what will work well or not work well- only testing the individual model seems reliable. For example, the Pixelworks chip in a Sony 10HT/11HT/12HT isn't all that great at upscaling (maybe not very many taps in the filter?), and if a higher res signal is available, it may look better downscaling from 1080i.

In my own experience, it's easy to see qualitative differences in upscaling players on a CRT RPTV that works native at 480P and 1080i without processing, or on a CRT FPTV. The players are not created equal; other factors such as Y/C delay, Chroma bug, etc, deserve just as much scrutiny.

Then, there's the matter that there have been problems with DVI levels on many of the DVD players- incl. black crush (Samsung comes to mind), white crush (Zenith), and inability to pass blacker than black.

Really, there's no substitute for veiwing any player you're considering on your own set. A good quick test that isn't open to subjective interpretation is to see how well the player does reproducing the HF luma signals in test patters from the AVIA DVD or VE when upscaling on your display, versus running straight through at 480i or 480P on your display.

No upscaling process creates detail that isn't there to begin with. And if the sampling rates are too different, and not close to integer related, you run the risk of actually lossing HF detail. Again, the test patterns will help establishing that.

But in general, don't lose sight of all the other factors affecting DVD playback quality. Check out the Benchmark reviews at Secrets of Home Theater and High Fidelity for some comprehensive tests. I use an upscaling player, but critical to that player isn't just the upscaling feature, but the DCDi deinterlacing of video, the freedom from white or black crush, and ability to pass blacker than black, as well as adjustability and a responsive transport. Also, it was cheap.

~Jon

[junior77blue]

This is just a theoretical question at this point.

I don't have a Denon 3910 or other DVD player to 'try' out on my display to see if there is a difference between 480p or 1080i.

I'm not sure what method 'Secrets' uses. I need to review their test method.

[JonMarsh]

A well designed upscaling player on the right display may be a good solution- I use one at 1080i on a CRT projector, and it produces a very film like image with a natural looking reproduction of detail. Done correctly, it can upsample the pixel samples into a more natural looking image. It doesn't add any detail not present in the original data, but may render that detail more pleasingly and naturally.



This is a small section from one frame in the Fifth Element. The lower portion is representing pixel data points from the native picture (480p). The upper part of the picture is the same data resampled to 1080p. Bandwidth limiting of a CRT projector can so some of the horizontal interpolation, but only resamping and filtering vertically as well as horizontally can render the upper section. Some people prefer the upper section over the lower. The display must be capable of significantly higher resolution than 480P to do this, of course.

Secrets tests quite a few aspects of player performance, including de-interlacing behavior, behavior with mis-flagged material, chroma upsampling error, etc, as well as video bandwidth, transport layer switching, etc. They probably have the most comprehensive DVD player tests on the planet.

~Jon

[junior77blue]

Do you know which output 'Secrets' use? HDMI or Component for their tests.

[Shane Martin]

They measure both.

[GATheater]

Yes, there can be a difference!

Using my Denon DVD-2910 (which has both 480p and 720/1080 upconversion), along with my SharpVision XV-12000 projector, with a native resolution of 720p, I do see a significant difference in image quality. This would imply that the scaler in the Denon is superior to the scaler in the projector. But, it's the image that matters, so I don't care!

Technically, the bottleneck s/b the resolution of the DVD itself. But, I believe the difference to be more one of matching the resolution of the projector to that of the source....

Now, I'm running a screen size over 160" diagonal, so it is very likely that I'd see more issues with picture quality than on a 70" screen. So, this should all be within the context of your own setup and viewing pleasure...

I don't see a reason to pay the extra bucks for the Denon 3910, which is about twice the price as the 2910, though.

enjoy!

[junior77blue]

Well, supposedly better sound qaulity is one of the major differences....but I personally have not listened to either one.

[GATheater]

3910

Yes. You are right. The 3910 does have superior sound than the 2910. But, my understanding is that most of the differences are in the Music space and not the HT space, as HT is thru the digital outs, anyway. I am an HT purist, in that I listen to my music DVD's outside of my theater and don't like to mix the two. Never could stand to mix my mashed potatoes and peas, either...

The 2910 does have excellent sound capabilities, though. And, it's a deal in the 500 to 700 bucks range, even at the local E-mart. But, that's for another forum...

[junior77blue]

Some of us are unfortunate enough to have to mix and mash. And yes, if your doing all the decoding at the preamp, then there's no reason to opt for the 3910.

I feel more comfortable purchasing the 2900 at the discounted price.