DVD Benchmark - Progressive DVD Player Shootout # 2 - August, 2001
Welcome to Secrets' long awaited Second Progressive DVD Player Shootout! This may very well be the most sought after article in A/V magazines anywhere in the world. We have been working very hard to complete it, and have received e-mails from a large number of our readers, asking for it, so here it is. Several hundred thousand people will be downloading this article, so we are keeping the file as small as possible. - John Johnson, Jr.
For this second shootout, we re-evaluated our tests and looked at whether there were some that weren't giving us enough useful information, or giving us information duplicated by some other test. We also kept watching for new material that tripped up deinterlacers, and found a few new tests. In all, we eliminated two tests and added three new ones.
Unlike the last shootout, where one player was able to pass every test, in this shootout no player came out perfect, though several came close. This isn't because players are getting worse, but rather that some of the new tests are very hard. We think this is as it should be. If we don't push the players to the limits, we won't see improvements made.
Rather than duplicate explanations of the tests we performed, you can go to the explanatory benchmark articles and get the information there. In particular, Part 5, recently updated, will explain progressive scan. If you have not done so, we suggest you read the original Progressive Scan DVD Shootout.
Table of Tests:
WF1 WHQL Film 1. Pass if it stays in film mode, and doesn’t comb. WMM WHQL Mixed Mode. Pass if it returns to film mode in no more than 2 frames, and doesn’t comb. Number in parens is time in frames to return to film mode. WC1 WHQL Chapter Stops 1. Pass if it stays in film mode. Number in parentheses is time in frames to return to film mode. WC2 WHQL Chapter Stops 2. Pass if it stays in film mode. Number in parentheses is time in frames to return to film mode. VZP Video Essentials Snell & Wilcox Zone Plate. Pass if the stationary areas of the screen stay full resolution for the whole sequence. BL Big Lebowski Making-of. Pass if it stays in film mode and doesn’t comb. Number in parentheses is total number of combs in first 30 seconds. GQM Galaxy Quest Menu. Pass if it stays in film mode and doesn’t comb. Yellow circle if it goes to video mode. GQT Galaxy Quest Trailer. Pass if it stays in film mode and doesn’t comb. Yellow circle if it goes to video mode. MT More Tales of the City. Pass if it stays in film mode and doesn’t comb. Yellow circle if it goes to video mode. A13 Apollo 13 Making-of. Pass if it stays in video mode and doesn’t comb. TS Toy Story Chapter 4. Pass if the chroma has no streaky upsampling artifacts. BC Blue's Clues Chapter 19. Pass if no combing. SS Super Speedway Chapter 7. Pass if no moiré effect on bleachers. A Abyss Chapter 4 w/ subtitles on. Pass if no combing on subtitles. Resulting number is number of combs in 60 seconds.
The Test Results
We use a green checkmark to show that a player passed the test, and a red X to show that it failed the test. If a player technically failed a test, but it didn’t comb or show obvious distracting artifacts, this is indicated by a yellow circle in the box. We used the yellow circle very sparingly this time around, as we were able to tighten our criteria based on what we learned from the last shootout.
In a few cases, there is a number in parentheses after the mark, which in most cases tells the number of frames (at 30 per second) it took for the player to recover from glitches and return to film mode. In the case of The Big Lebowski test, it shows the number of times the player combed during the 30-second intro.
Player data table:
Player WF1 WMM WC1 WC2 VZP BL GQM GQT MT A13 TS BC SS A Q50 (10) (8) DVD 1050P (15) (15) (10) RP91 (1) (1) (1) (5) NS700P (20) (30) (15) (1)
DVD-N996 (1) (25)
DVD-2800 (1) (2) DV27 (1) (8) H2000 (1) (1)
WinDVD 3 (1) (20) (3) (5) DVD 50 (1) (25) (3) (5) DV-38A (3) (3) (7)
AD800 (1) (20) (3) (5) XV-S60 (1) (25) (3) (4) 510 (TV) (35)
N/A N/A DVD-RP56 (10) (6) DV-5700 (15) (6)
Notes on individual players:
Philips DVD Q50
This is one of the players that uses the new Sage/Faroudja deinterlacing chip (FLI2200), and it's an impressive chip indeed. The only things that tripped it up were the recovery time on Mixed Mode 1 (we said a player had to recover to film mode within 3 frames to pass), and combing on subtitles, which is a pretty minor thing. All in all, this is a strong performer. With the exception of the Skyworth, all DVD players that contain the Sage/Faroudja chip can be identified by the DCDi logo on the front panel.
The recovery time issue is something we noticed while watching movie material on this player. When it goes into video mode, it tends to stay there for a while, and it goes into video mode more often than the Silicon Image chipset would. Because of Sage's DCDi video-mode deinterlacing (see Part 5 of the series to read an explanation of DCDi), it's not as noticeable when the chip switches to video mode, but it does sometimes stand out, especially when there is a very detailed background. Still, we'd have to say this is a relatively minor issue, and we would still recommend the player.
An explanation is necessary for the yellow circle on the Toy Story test. This player's MPEG decoder does have the chroma upsampling bug, and in fact it's perhaps the worst we've ever seen. It's clearly visible on any interlaced TV, or running it through an outboard deinterlacer. However, the Sage chip resamples the chroma channel, which almost completely hides the chroma bug. So if you plan to use this player only in progressive mode, the chroma bug is essentially a non-issue. In certain cases, you can see a minor vestige of the bug, but you really have to know what to look for, and even then it's hard to see.
There is no aspect ratio control on this DVD player. It does have a couple of zoom modes, but the on-screen display is present when zoomed. The Q50 produces a very sharp image.
Update: We received a firmware update for the Q50 that resolves the severe chroma upsampling error in film mode. The Q50s that will go on sale in the USA should have this fix already. If you have purchased the Q50 in another country, contact your local service center, and they should be able to take care of you.
DVD Q50 Pixel Cropping Top Bottom Left Right 12 0 6 0
This is the other player in the shootout with the Sage/Faroudja chip, and it's also one of the least expensive players in the shootout, at around $250-300 street, if you can find one. It's an import from China, and has a lot of useful features, including region-free playback, no Macrovision, and PAL playback, though it doesn't convert PAL to NTSC, so you need a PAL-compatible TV to watch PAL discs. We tried some PAL discs on it, and they worked fine.
Unfortunately, the Sage chip wasn't tweaked quite as well on this player as on the Philips, and it failed several of the tests, although by extremely small margins. For example, we failed it on the WHQL Film mode test because it dropped to video mode deinterlacing inexplicably for a second or so when the moving wedge would hit the 9 o'clock position. It did this every time we ran it, consistently. While it's perhaps a technicality, we had to fail it. It does have a working film mode, but it just drops to video mode too often.
It had the same recovery time problem that the Philips did, but with an even longer recovery time, and it dropped to video mode again during the chapter stops 2 test. It dropped into video mode constantly during Super Speedway, which was very noticeable throughout, not just on the bleachers scene we tested for.
However, in practice, most of the time these excursions into video mode are not going to be that noticeable, and its bad edit handling is stellar (0 combs on the Big Lebowski test). The video quality is also quite good, though it has a noticeable chroma delay on the component outputs. Like the Philips, this player has an MPEG decoder with the chroma bug, though not as bad as the Philips, but also like the Philips, the Sage chip hides the bug. However, there is an odd quirk on this player - highly saturated colors, especially reds, flicker. It's a bit like the red flicker on the Toshiba 5109, though not nearly as bad. However, it is there, and is somewhat distracting. We're pretty sure that this is caused by one of the enhancement settings on the Faroudja chip being set too high, so it's possible they could fix it with a firmware update since the Philips does not have this problem.
There is no aspect ratio control on this DVD player. Given the price, it is a great bargain.
DVD-1050P Pixel Cropping Top Bottom Left Right 0 0 1 0
This player has a lot of things going for it. It has many useful features, including auto-scaling of letterboxed movies so they fill a 16x9 display in "full" mode. It's fast, has excellent video quality, and has no chroma bug. Unfortunately, the player uses the inferior (in our opinion) Genesis chipset, and while Panasonic has done an admirable job of squeezing the last ounce of performance out of it, it still can't match up to the Sage or Silicon Image chipsets.
We spent quite a bit of time evaluating the various deinterlacing modes (called transfer modes in their menu) on the RP91. It has Auto1, Auto2, Auto3, and Video. Video is just as you would think, forced video mode. Auto 1 appears to only recognize clean 3-2 pulldown. Auto 2 adds the ability to recognize 2-2 pulldown, and Auto 3 supposedly adds even more, but we were unable to figure out what. Overall, Auto 2 produced the best results on the majority of tests. In a few cases, we got slightly better results with Auto 3, but more often Auto 2 was better. Auto 1 was always the same or worse in our tests. The results in the table were all with the player set for Auto 2.
The most interesting thing about the RP91's deinterlacing is that it uses a combination of flag-reading and cadence reading. If the material has the progressive_frame flag set, the RP91 trusts it and goes into film mode. If the progressive_frame flag is not set, the RP91 uses the gmAFMC chip to analyze the cadence, and tries to go into film mode if there is a 3-2 pulldown pattern. This does a better job on some things than flag-reading alone, but it's not as good as top-quality cadence reading.
For example, the player fails the Galaxy Quest Menu and Apollo 13 tests, because those are examples of material that have the progressive_frame flag set incorrectly. But, it passes the Galaxy Quest trailer test, because while the material is not marked progressive, the gmAFMC chip recognizes the 3-2 cadence.
In practice, this player tends to do better on more material than other Genesis-based players, but we still wish it could do better. More Tales of the City, for example, is unwatchable because of the many bad edits, and there is lots more material (DVDs) just like it out there. In addition, when the player is in video mode, it's using the video-mode deinterlacing of the Genesis, which is just lousy. In the case of More Tales, you can switch the transfer mode to video, and the artifacts will go away. However, you may find the image too soft to watch.
Like all Panasonic players we've looked at, this one doesn't have the chroma bug. The video quality is excellent. The auto-scaling works well, and letterboxed material scales fine, without any obvious scaling artifacts. The scaling won't turn a non-anamorphic disc into an anamorphic one, but on many televisions, using the player's internal scaling will improve the picture over using the TV zoom, because in "full" mode, there are more scan lines on the screen, closer together, than in "zoom" mode.
All in all, the RP-91 is a good machine, and worth considering for its many great features, but if deinterlacing performance is very important to you, you may want to look elsewhere.
RP-91 Pixel Cropping Top Bottom Left Right 0 0 13 0
This machine was a surprise to us, a mildly pleasant one. We had heard through the grapevine that this player would be using the Genesis chipset, as the earlier S9000ES had. When we got the player, however, we found that Sony was using a chip of their own. It's not quite up to the level of the Silicon Image or Sage chips, but it's several notches better than the Genesis.
The chip uses cadence-reading, so it passes all the "bad flags" tests quite well. As you can see from the chart, the player failed the Mixed Mode and Chapter Stops test, with a pretty significant recovery time. But it never combed on those tests, which is good. It has good bad edit detection - 1 comb on the Big Lebowski test is very good indeed. Only the Sage and Silicon Image chips do better. And it has good quality motion-adaptive video deinterlacing, which is excellent.
Video quality was quite good. Unfortunately, this player has the chroma bug. It's not as bad as some, but it's there and quite noticeable on certain material. One feature that was missing that some people are interested in is aspect ratio control. The Sony DVPS9000ES offered the ability to windowbox 4:3 material, but we were unable to find this feature on the NS700P.
All in all, this is a very fine player, marred only by a tendency to stay in video mode a little long, and the chroma bug. If the chroma bug isn't an issue for you, we'd certainly rank this player well above average for its price point.
NS700P Pixel Cropping Top Bottom Left Right 1 6 0 2
This is a terrible, terrible progressive player (in our opinion). It's a perfect example of a product designed by marketing, trying to snag people who've heard that progressive players are better. In this case, that turns out to be untrue. It has no film mode at all, and it doesn't even have motion adaptive deinterlacing. An interlaced player would be a significant improvement over the DV-434 on almost any TV. Avoid this player!
DV-434 Pixel Cropping Top Bottom Left Right 1 0 1 4
This is one of the many players in this shootout that uses the Mediamatics chip. The Mediamatics chip is a very good performer for the price, but where deinterlacing is concerned, it isn't even in the same league as the Sage and Silicon Image. The Mediamatics relies entirely on the flags on the disc to make deinterlacing decisions. And to top it off, it has a very simple video deinterlacing algorithm, simpler even than the Genesis, using a filter that reduces the vertical resolution considerably.
On the plus side, the Mediamatics has excellent scaling, which the Malata makes use of, with controls that allow you to zoom in X and Y separately with 65536 zoom levels. Unfortunately, there are no picture memories, so you can't save several different zoom parameters for, say, letterboxed 1.85:1 movies and 2.35:1 movies. Add multiple picture memories, and the N996 zoom features would be a tweakers delight. As it is, it's a bit of a pain to adjust the zoom every time you want to watch a different kind of movie.
Like the JVC 723GD we reviewed in the last shootout, this player has five different deinterlacing modes, though not the same modes. Frankly, we have no idea what "Still," "Camera 1," and "Camera 2" mode are. We just left the player in Smart mode, which seemed to do the right thing more often than not. Also, the deinterlacing modes are called "Video Filter" modes. Very intuitive.
Overall, the menus and interface are cryptic, and that's being kind. However, we did put in a disc that had a couple of minor read errors, and a message flashed on the screen: "Minor Disc Error: Clean Me!" Now that's user-friendly! We were wondering if you put in a really bad disc, would it say "Major Disc Error: Throw Me Away!"?
The video quality is good on the Malata, and it doesn't have the chroma bug (like all Mediamatics-based players).
Like the JVC we reviewed last time, this player has the bizarre quirk that changing the deinterlacing mode affects the interlaced output. For example, in video mode, the vertical filter gets applied, then the resulting softened progressive picture is converted to interlaced. For that reason, we can't really recommend this as an interlaced player.
The main thing people are buying the Malata for is excellent PAL to NTSC conversion (playing a PAL disc for your NTSC TV), and it does do that well. If that's the primary reason you want the player, it's a good choice.
DVD-N996 Pixel Cropping Top Bottom Left Right 0 0 20 20
This player was eagerly anticipated by many people, as it was the first reasonably-priced player with the Silicon Image chipset in it. Sadly, the early shipments were marred with a variety of production problems and firmware glitches. However, at this point the production problems have been ironed out as far as we can tell, and our final review sample performed quite well. Apparently there are some units in stores with different colored dots on the box to indicate, perhaps, the manufacturing run. We do not know what the codes mean, so you will have to work that out for yourselves.
Like the Camelot we reviewed in the last shootout, the Denon passed all the core tests with flying colors. It failed the Blue's Clues test and the Abyss test, but the Camelot failed those as well, and they don't represent common scenarios.
The basic video quality on the Denon is good, the image is softer than others, which may be a plus for those annoyed by DVDs with excessive edge enhancement. The player is extremely fast at navigation. The layer changes were lightning quick, as were chapter changes and menu selection.
The main Achilles heel of this player is the chroma bug. It has it, and while Denon did adjust the firmware to alleviate the bug somewhat, it's still there and still pretty bad. If this issue doesn't bother you (and it doesn't bother everyone), this player is otherwise a very fine performer and well worth considering. This player also does not offer any type of aspect ratio control.
DVD-2800 Pixel Cropping Top Bottom Left Right 1 2 5 1
This is another player with the Silicon Image chipset, and quite a bit less expensive than the Camelot, though more expensive than the Denon DVD-2800. Its performance was very similar to both of those players, though with more combing on subtitles, perhaps because of differences in firmware settings.
The Arcam has a very stylish look. The remote has a grid of tiny buttons, all alike, which are laid out in a frankly incomprehensible fashion. We found this to be by far the worst remote in the shootout.
The Arcam had some moderate video ringing, though not the worst we've seen, and otherwise had a fine picture. It does have the chroma bug, though, again, not the worst we've seen. It would rank somewhere in the middle of the pack as to visibility of the bug. The Arcam is modular and has been designed from the beginning to be upgraded. The MPEG decoder is located on its own board and can be replaced in the future. They also plan to offer a DVD-A upgrade at some point.
The Arcam is the first progressive DVD player that we have seen that offers the ability to optimize lipsync for interlaced or progressive. This is a feature of the Silicon Image chip, and we are happy to see someone implement it. The Arcam does not offer any type of aspect ratio control.
Overall, the Arcam was a good performer, and worth looking at. If the chroma bug of the Denon bothers you, you might find the Arcam a good choice, as it's not nearly as bad, and the deinterlacing performance is essentially identical. Remember also, we have studied the chroma bug extensively, and can spot a tiny amount of it from across the street. Some consumers have never even noticed it.
DV-27 Pixel Cropping Top Bottom Left Right 0 10 0 0
This player is essentially the same player as the RP-91 from a video standpoint, as far as we could tell. The deinterlacing test results were the same, as it uses the same hybrid flag-reading/cadence-reading Genesis system. The menus were the same, except that the H2000 adds separate R, G, and B gain and bias adjustments. We didn't spend very much time with them. Under normal circumstances, we'd prefer to adjust RGB tracking on the display, and these controls were much too coarse to do any real calibration with. The H2000 includes an additional remote that has direct access to many of the video specific options.
The H2000 also had a YC delay issue that was not present on the RP91.
The biggest difference between these two players is weight. If you're a fan of heavy electronic equipment, the H2000 is among the heaviest we've lifted. Overall, though, we didn't see what additional things you were getting for your money, at least on the video side.
H2000 Pixel Cropping Top Bottom Left Right 0 0 13 0
EAD Theatervision P
This player is often touted as an alternative to the Camelot, with similar performance for somewhat less money. We found that to be essentially true. It is, like the Camelot, based on a Panasonic transport and decoder, so it doesn't have the chroma bug. It uses the Silicon Image SiI503 chip for deinterlacing, and has essentially the same deinterlacing performance as other players using that chip.
One potential advantage over the Camelot is that the Theatervision P is based on a later model Panasonic transport (the RV30) than the Camelot (the A110). We can't really address that, as the Camelot in our test rack has performed incredibly well so far, and we didn't see any problems with either of them.
The EAD has the ability to window-box 4x3 material, but not to scale letterboxed material. However, the 4x3 mode is controlled with a manual switch on the back panel, which is inconvenient to use.
Video quality was excellent. This is a great performer, and well worth considering if it were not for the fact that this player has been discontinued. Unfortunately the replacement model is now using the Genesis solution. However recent reports have said they will offer the Faroudja technology for an additional premium
A cautionary note: several examples of the Theatervision P, including our review sample, have come from the factory with the deinterlacer misconfigured into forced-single-field mode, which looks terrible. You can check whether your player is set properly by looking at the Avia Resolution 200 TVL pattern. If it's solid and detailed, everything is fine. If it flickers wildly, your player is not configured correctly and should be returned to the dealer for service. We also had a sample of a Camelot with essentially the same problem, so it's worth doing the same check with a Camelot as well.
TVP Pixel Cropping Top Bottom Left Right 0 0 5 6
AVS HTPC w/ WinDVD 3.0
In the last shootout, we tested the software DVD decoders WinDVD 2.5 and PowerDVD 3.0, running on some reasonably powerful hardware, but not hardware tweaked and optimized for home theater. The AVS PC is a commercially produced PC designed for home theater from AV Science, so we hoped it would perform better than the PCs we tested for the last shootout. And, in fact, it did. The stutter we saw in the previous shootout was gone, and overall, the picture was stellar.
Alan at AV Science provided us with a PIII 1GHz w/ 256 MB of RAM running Windows ME. It had an ATI Radeon video card with 64 MB of RAM. It came pre-installed with XYX 1.6.0, PowerStrip 3.0 Beta Build 121, and WinDVD 2.5. The ATI video drivers were Version 4.13.7093. It was also equipped with an M Audio Audiophile 2496 soundcard. We ran the HTPC at 1280x72 @ 72 Hz. This was all pre-setup by AV Science. We did upgrade WinDVD to version 3.0.
That said, the deinterlacing performance was, as you would imagine, the same as before. PC DVD software still doesn't have the horsepower to do high-quality MPEG decoding and deinterlacing at the same time, so an HTPC has all the limitations of any flag-reading player, and doesn't have motion-adaptive deinterlacing. The Radeon does offer a fairly primitive motion-adaptive deinterlacer in hardware, but we were unable to get it working with WinDVD. It worked fine using the ATI (Ravisent) DVD player, but we found that DVD player to not be nearly as good as WinDVD. Usability was somewhat better than before, because the AVS HTPC comes with an infrared keyboard. Conceivably, you could set up macros and teach the key sequences to a Pronto or other learning remote, and have a DVD player nearly as easy to use as a standard video component, with the keyboard on your lap in your home theater room.
As before, the picture quality of the HTPC was incredibly good. There was no ringing at all, no chroma delay, just a smooth, detailed, accurate picture. In addition, WinDVD doesn't have the chroma bug. If you don't watch a lot of video-format DVDs or DVDs with odd flags (anime, for example), this HTPC may well fit the bill.
HTPC Pixel Cropping Top Bottom Left Right 0 0 0 0
Harman Kardon DVD 50
This player performed very similarly to the other Mediamatics players in the shootout. Unlike the Malata and JVC players, the DVD 50 doesn't have multiple deinterlacing modes, which makes it perhaps easier to use and less confusing, but it also means you have no options if the player isn't deinterlacing a particular title well. As far as we could tell, the single deinterlacing mode was similar to the "Smart" mode on the JVC and Malata players.
Like other Mediamatics-based players, the DVD 50 had a difficult time with material with bad or nonstandard flags, and its simple vertical filter deinterlacing made all video material (or material marked as video) look soft.
This player is a 5-disc changer, and we found it a little slow changing from disc to disc, and even just spinning up a single disc.
Like the other Mediamatics-based players, the DVD 50 doesn't have the chroma bug, and the video quality was very good. There was very little ringing, and no significant chroma delay. If you don't plan on watching a lot of titles from smaller companies or video-based titles, this is a perfectly fine player.
DVD 50 Pixel Cropping Top Bottom Left Right 0 0 3 3
This is the fancier sibling of the DV-37 we reviewed in the last shootout, and its video performance was essentially identical. The deinterlacing performance was poor, even for a flag-reading player, and its chroma bug is fairly bad. There was more ringing on this player than on most of the others we looked at. This gives the image a very sharp crisp look that so many people are after. This works ok on a CRT, but is distracting on a digital display like the Sony 10HT we used.
On the plus side, the DV-38A (like the DV-37) does have motion-adaptive video deinterlacing, which puts it ahead of most of the pack when watching video material. But while good video deinterlacing is a great thing, and much to be desired, it's the film deinterlacing that you pay good money for a progressive player for. And here the Pioneer does not do well. It has worse bad-edit detection than any of the players in the shootout (7 combs on Big Lebowski, for example), and combed in places that no other player did.
The player has an interesting control that allows you to adjust "Pure Progressive" from "fast" to "slow," which actually seems to be adjusting the balance between temporal and spatial interpolation in the video deinterlacing algorithms. We found this fascinating, but largely worthless to the end user.
Frankly, for this much money, Pioneer could do much, much better. As you'll see when you read the write-up for the Pioneer Elite 510 TV, clearly Pioneer has a good quality deinterlacing chipset. We don't understand why they didn't use it for this player.
DV-38A Pixel Cropping Top Bottom Left Right 1 1 2 3
This is another of the Mediamatics-based players in the shootout, and it had essentially the same deinterlacing performance as the others. However, it had significantly worse video performance than the others. It has a significant rolloff of video high-frequencies, which manifests as a very soft picture, and it has worse video ringing than any of the Mediamatics-based players, which suggests that Apex skimped on the analog components after the DACs.
Like the Malata, the Apex has many options, and a similarly convoluted menu system. The Apex menus are slightly easier to understand, but only slightly.
Like the other Mediamatics-based players, the Apex has no chroma bug. Like the others, the Apex has difficulties with bad or non-standard flags and has no motion-adaptive video deinterlacing.
The Apex is inexpensive, and it's not a terrible player, but these days there are probably better choices for a similar amount of money. The Malata, for example, has better video quality, and is region-free to boot.
AD800 Pixel Cropping Top Bottom Left Right 0 0 3 3
We are getting to the point where reviewing another Mediamatics-based player gives us a certain feeling of deja-vu (all over again). This JVC is essentially identical in every key way to the JVC XV-D723GD we reviewed last time, minus the DVD-A support. The menus and displays are the same, the deinterlacing is the same, and the firmware appears to be exactly the same. All the quirks we outlined in that report are still there, including the annoying way that the interlaced output is derived from the progressive output, and is thus affected by the progressive deinterlacing mode. We informed Mediamatics and JVC about this problem on the 723GD, but apparently the news never made it to the right people.
Rather than go over all the quirky features of the JVC, it's easier to just read the previous review.
Video quality was very good. As we said in the previous review, if you can live with the quirks, the basic video quality is excellent, and it has some nifty features. Given that this new player is even cheaper, it might be worth considering.
XV-S60 Pixel Cropping Top Bottom Left Right 0 0 3 3
Pioneer Elite Pro-510 TV
This is a TV rather than a DVD player, but we included it in our deinterlacing tests because it's often touted as having an excellent deinterlacer, so good in fact that there's no need to use a progressive DVD player with it. We wanted to put that assertion to the test. We did all our tests in the factory default deinterlacing mode, labeled "advanced". We found that the "standard" mode was significantly worse, and we can't imagine why they included it.
What we found is that essentially, the reports are true. The deinterlacing on the Elite is excellent, in fact better than most progressive players. Its bad-edit detection is excellent, and it has a fine motion-adaptive video deinterlacer. The one area it didn't do as well in was recovery time from video to film mode. It tends to drop into video mode fairly quickly, and takes a relatively long time to get back to film mode. And because it doesn't have anything like Sage's DCDi, the change to video mode is often fairly noticeable.
As to whether you don't need a progressive player with this TV, that's hard to say. The Sage and Silicon Image based players were better at deinterlacing, though not by a huge margin. Perhaps more importantly, we did notice the resolution loss when using the internal deinterlacer, because the signal must be converted an extra time from analog to digital, deinterlaced, then converted back to analog. With an external player, the deinterlacing is done before the signal is ever converted to analog, so there is no resolution loss.
It's probably not worth getting a lower-end progressive player like a Genesis or Mediamatics-based player to mate with this TV. But if you want the best quality possible, adding a progressive player will make a small but worthwhile difference.
While we are not actually reviewing the TV at this point, it is worth noting that this TV has some of the most pronounced ringing we have ever seen in a consumer display device. We did disconnect SVM (Scan Velocity Modulation) and we pulled down the service mode sharpness adjustment from 0 to -128.
(Note: we didn't do the Abyss test on the Elite TV because that test wasn't yet on the list when we did the testing of the Elite, and we were unable to get the unit back to do that test.)
This player was a surprise. Coming so soon on the heels of the RP91 (reviewed above), we expected it to be similar, and use the hybrid flag-reading/cadence-reading Genesis chipset of that player. Instead, we found that it uses the Sage/Faroudja chip, which is amazing in a major brand player at this price (around $230 street). The box and front panel don't say anything about Faroudja or DCDi, but it's in there. And except for a few minor hiccups, it's all configured correctly.
Like the other players with Sage deinterlacing, the RP56 had a long recovery time when it switched to video mode from film mode -- too slow to pass our test. And it combed a few times on the subtitles on "The Abyss", which again was pretty normal for this chip. Otherwise, it passed all the tests with flying colors. Unlike the Skyworth, it passed the Super Speedway test, and it doesn't have the chroma upsampling error.
The video quality was very good, especially considering the price point of this player. There was very minor ringing, nothing unusual. There was no significant Y/C delay on the Avia pattern. There was some apparent roll-off of the highest frequencies, but not outrageous. We haven't yet seen the player on the Tektronix analyzer, but what we saw on the test patterns looked quite nice. It's not reference quality video, but much better than most players at this price.
There were a couple of minor issues: first, when we switch the player to forced Video deinterlacing, it turns off DCDi. Since DCDi is only useful on video material, this is clearly a mistake. However, since the Sage chip does such a good job of automatically selecting Video or Film mode, there's essentially no reason to ever manually switch modes. Also, like the Skyworth, this player has a small flicker in saturated colors because of the Sage's cross-color suppression filter being set too high. Panasonic could fix both of these issues with a firmware update; whether they will do so is anybody's guess.
The RP56 has no aspect ratio control at all for scaling non-anamorphic DVDs, and doesn't have an interlaced/progressive switch on the remote (it's on the front panel). If you have a TV that locks into Full mode with 480p signals, this may not be the player for you.
Overall, this is an excellent player at an excellent price. We're very happy to see deinterlacing of this quality being put into an entry-level player like this. Let's hope we see more such players.
DVD-RP56 Pixel Cropping Top Bottom Left Right 0 0 6 7
Kenwood Sovereign DV-5700
The Sovereign line is Kenwood's new high-end line of DVD players. They are apparently not going to do a single-disc model, at least not at first, and are concentrating on the changer market, which up to now has not had a model with a top-quality deinterlacing chip. The 5700 is a 5-disc changer, and Kenwood will also be releasing a megachanger, the 5900.
This is another player based on the Sage/Faroudja deinterlacing solution, and it performed very similarly to the others on our deinterlacing tests. It seemed to recover just a hair faster from video mode to film mode on the Mixed Mode test. Otherwise, it performed as expected, which is to say very well indeed.
The Kenwood, unlike the other players with the Sage that we looked at, also includes the video enhancer chip that Sage offers as a companion chip. It adds selective edge enhancement on the luma and chroma channels, which can make the picture look sharper, especially on a display with a soft picture, like many CRTs. In the menus on the Kenwood are settings to control the enhancement amount, and we preferred it in animation mode, off, or set fairly low, like 1 or 2 (the default is 8). However, different people have different preferences for edge enhancement, so the appropriate setting depends a lot on whether you prefer a sharpened picture or a smoother picture or something in between.
There is also a switch to turn off DCDi, which seems not very useful to us, and an "animation" switch, which turns off some of the enhancement settings that produce odd artifacts on patches of saturated color. We'd normally recommend leaving it in that mode all the time, but unfortunately turning it on adds a significant Y/C delay. (We have a pre-production model, and we have been told they made a couple of changes to the shipping product.) So we'd recommend leaving it off, unless you see odd artifacts in the saturated colors that look something like little bites have been taken out of the edges of objects. If you see it, you'll know what we mean. If you don't see it, then don't worry about it.
There are a few minor quirks of note with this player. First, it's very slow. It takes a long time to spin up a disc, and a long time to change discs. Menu navigation is slower than average as well. We assume that some or all of this is because of the added complexity of the changer, as other DVD changers we've looked at have also been slow. In addition, the remote has a very strange design that we weren't fond of at all. It has very few buttons, and each button has two or three different functions. You select which function group you want by adjusting a three-position switch on the side of the remote. For standard playing, pausing, and scanning, you can leave it in the top position, but if you want to do anything more complex, you find yourself switching back and forth between the different modes, and it's not at all intuitive.
The Kenwood doesn't have the chroma bug, as it apparently uses a Matsushita (Panasonic) MPEG decoder. The video quality was good, with mild to moderate ringing on the Avia pattern (the ringing is much worse if the enhancer is set to the default position), but no noticeable roll-off of the high frequencies.
This is a good player, with top-notch deinterlacing. We looked only at the 5-disc changer, but we expect the megachanger to be similar. If you're looking for a progressive changer with great deinterlacing performance, this is a good choice.
DVD-RP56 Pixel Cropping Top Bottom Left Right 0 0 6 7
- Don Munsil -
Last Updated: 9/29/01
DVD Benchmark - Article Index
Part 1 - Video Part 2 - Audio Part 3 - Functionality Part 4 - Usability Part 5 - Progressive Scan Part 6 - DVD Audio
© Copyright 2001 Secrets of Home Theater & High Fidelity
Return to Table of Contents for this Issue.