Onkyo - DV-SP1000

MPEG Maker: unknown MPEG Model: unknown Deinterlacer Maker: Silicon Image Deinterlacer Model: SIL-504     MSRP: $2000.00 Website: http://www.onkyousa.com

The Onkyo DV-SP1000 is a player I have been trying to get hold of for quite some time now. As most readers of the Benchmark know, I have not been a big fan of Onkyo’s previous DVD players and their reliance on all-in-one MPEG decoder/de-interlacers. The SP-1000 changed that stance and incorporates the popular and very capable Silicon Image SIL504 video processing chip.

This switch-out made a huge difference in performance for Onkyo and puts the SP-1000 in the category of being one of the best DVD players on the market today. The SIL504’s abilities with film based de-interlacing have been heralded time and again at Secrets, and its performance in the DV-SP1000 is no exception. The SP1000 delivered a very clean image devoid of the issues we were seeing with their older players.

The chip passed all of our de-interlacing tests including those for 2-2 material. This chip has a faster recovery time than the Genesis FLI-2310 and doesn’t suffer from any of the macroblocking issues that have been so common on most upscaling DVD players. But that doesn’t mean that this chip doesn’t have its limitations and clear weaknesses.

Currently there are two chipsets on the market that are incorporating advanced diagonal line processing. Those are the FLI series chips from Genesis/Faroudja and the HQV chip from Silicon Optix. Unfortunately, the Silicon Image chip just cannot compete with these solutions in terms of video-based material, and more specifically, diagonal line processing. This is extremely evident with video-based material such as DVDs of TV programs and a lot of Anime or animated features.

A great test for diagonal line processing is found on Silicon Optix’s new Benchmark DVD. Using the three angles test resulted in some pretty horrible jaggies along the edges of all three angles. I also looked at some older Sage-produced test clips that showed obvious jaggies.

Another downside of the SIL504 chip is the 8 bit video processing. Despite the HDMI output of the SP-1000, banding was still evident in gray scale ramps. This is not as much of an issue right now since most digital displays still have 8 bit internal processing, but it will become an issue as these displays start incorporating higher bit processing. The Genesis and Silicon Optix solutions do all of their processing at the 10 bit level. Having the higher resolution processing eliminates artifacts such as banding and gradiations that are pretty common with DVD and digital displays.

Of course there is more to a DVD player than de-interlacing. Great de-interlacing will not make up for a sub-par DVD player in the core performance department. Onkyo has been hit or miss in this department in the past, but this time around they got all their ducks in line.

The SP1000 has absolutely no Y/C delay via component or HDMI. This lends to a cleaner image devoid of the ghosting effect that Y/C delay will bring to an image from misalignment.

The analog frequency response is quite respectable but a tad down in the upper frequencies. This may lead to a slightly softer image than some players, but this is a non-issue if you use the preferred HDMI output. Pixel cropping was a tad excessive, with 4 pixels cropped from the bottom of the image and a combined 6 pixels from the left and right sides.

For our chroma tests, the SP1000 did modestly. This player has no chroma issues with traditional film based 3-2 material and 2-2 material. But flickering and some banding was noticed with 3-2 material with alternating progressive flags. This will cause some artifacts with some of the popular Disney animated DVDs on the market including Monsters INC and Finding Nemo. One of the plus sides of the Silicon Image chip is its full retention of the chroma resolution. This player's vertical chroma resolution was excellent.

The SP-1000 passes a below black pluge via both component and HDMI. It has a selection range for HDMI that will also allow for PC RGB levels if necessary. The default white level for the component output was a very respectable 99 IRE.

Overall, the user interface is excellent, with clean, easy to navigate menus that are very responsive. Access times for the player were quick, and menu navigation and chapter skips were almost instantaneous. Using our full bit rate layer change test, the SP-1000 measured in at a quick one second.

Some of the other highlights of this player include full support of both DVD-Audio and SACD. The player not only features the typical analog multi-channel RCA outputs but also a Firewire output compatible with both high resolution formats. This makes the SP1000 one of the few players offering a complete digital transport solution for both video and audio. Unfortunately it does not support the newest version of HDMI though, i.e., the 1.1 spec. This would allow for transmission of high resolution PCM audio like that found on DVD-Audio discs. The SP1000 is limited to Dolby Digital, DTS, and standard resolution PCM via the HDMI output.

Summary: Not since the Pioneer Elite 59AVi have I seen such a huge turnaround from one of the bigger electronics companies. The DV-SP1000 is a great step in the right direction for Onkyo and I hope a sign of things to come on this front. The SP1000 joins the ranks of my highest recommended players on the market today and is an excellent value.

Video Frequency Response
Onkyo

Panasonic - DVD-S97

MPEG Maker: MEI MPEG Model: 2050004AP Deinterlacer Maker: Genesis Deinterlacer Model: FLI-2310     MSRP: $299.00 Website: http://www.panasonic.com

We had the opportunity to review the S97 last year for our Benchmark, and although this was a hotly anticipated player, it had a few issues working against it. Panasonic has since released two different software updates, and the player I tested this time around had the latest and greatest firmware installed. Panasonic has definitely made some improvements to this player since the last go around, and this now stacks as one of the best players to be found for the price.

These new updates were strictly on the software side, so the core hardware elements are the same as before. The S97 uses a new Panasonic/Matsushita MPEG decoder and the Genesis FLI-2310 video processing chip. The 2310 chip is infamous for its macroblocking problem that seems to create digital blocking in the image both on its own and by enhancing issues that are already there. The only company that has made an effort to fix this issue so far is Denon who released some firmware updates to the DVD-5900 that toned the artifact down some, but not enough. Panasonic has now attempted the same thing but with much better results.

I did my standard tests for this artifact via both component and HDMI outputs to my display. Via component video, there were no signs of the macroblocking at all. And I mean no signs. Via HDMI, there was still evidence of the issue, but it was toned down significantly since the last time I looked at this player. While it still has the macroblocking bug (MBB), this is one of the best players I’ve seen with respect to that particular problem.

But Panasonic made some more changes too. It seems that the de-interlacing performance has been tightened up a bit. The S97 offers three distinctive progressive modes: Auto1, Auto 2, and Video. Auto 1 has the typical Faroudja performance that you see in just about every player that uses this chip. It does a great job with both film and video material and is a solid performer all around. Like all implementations, this mode does not support a 2-2 cadence, which is expected. Auto 2 solves this issue. I only recommend using Auto 2 if you know the material you’re watching is using a 2-2 cadence, because performance with other non-2-2 material can vary. Basically, if you are seeing strange artifacts, such as combing, shimmering, or unsteadiness in the image, switch over to Auto 2 and see if it helps. If it does, great! When you’re done watching that material, go back to Auto 1. There is also a Video mode, but there is no real reason to use it as the Auto 1 and 2 modes cover those bases fine.

The core performance of the S97 has also improved with the new firmware. Response times are up, and the layer change time is down. While it is still not down enough to compete with Mediatek players, Panasonic managed to shave about a half second off the layer break time. The MPEG decoder itself still has some issues though. Panasonic did a bit of a back step after the earlier players like the RP-82 and RP-91 in terms of MPEG decoder performance. Our CUE tests revealed some obvious flickering when the material used a 3-2 alternating cadence. While there wasn’t any banding or jaggies, the flickering can be a bit annoying.

The component outputs had some minor pixel cropping, including 7 combined pixels from the sides of the image and 2 from the top and bottom. There is a slight amount of Y/C delay with both the component and HDMI outputs, but it still was within our 5 nanosecond standard. The component outputs are properly set up for a 0 IRE output level, and they pass a below-black pluge with no problems. The measured white level of the player is a tad high at 102 IRE but still within our passing spec.

The HDMI output has some issues with its video levels though. With the brightness and contrast levels in the player set to their default state, this player does not pass below-black or above-white information. I had to set the brightness to +2 and the contrast to -1 to pass the full video signal. This is something that Panasonic should have tweaked in with their new firmware, and I’ve seen the same issue with some Denon players, including their flagship DVD-5910. On the plus side, this player does not incorrectly change the RBG levels when converting HDMI to DVI, which is common with a lot of HDMI players out there. So owners of DVI-based monitors have no worries here.

Summary: I am glad that Panasonic took it upon themselves to further refine the S97 DVD player. It stands as one of the best existing players at this price point.

Video Frequency Response
Panasonic

Pioneer Elite - DV-59AVi

MPEG Maker: Mitsubishi MPEG Model: unknown Deinterlacer Maker: Pure Cinema Progressive Deinterlacer Model:     MSRP: $1600.00 Website: http://www.pioneerelectronics.com

This is the second go around for the Pioneer Elite 59AVi in our benchmark. I tested the player last year, and I liked it so much, it received our Best of Award for 2004 in the DVD player category. A short while after that, rumors started spreading around the net that Pioneer released an updated firmware revision for the player which eliminated some of the issues we reported on. I spoke with Pioneer about it, and they sent me a brand new player to test. We also received a 59TXi receiver and you will be seeing full reviews of both products here at Secrets.

So the big question is, did anything change? Is this version any better than before? Well the answer is no. I ran into the same issues this time around that I did last time. But it still remains one of the top players on the market today.

The 59AVi is based on a Mitsubishi MPEG decoder and uses Pioneer’s proprietary Pure Cinema Progressive de-interlacing solution. We have not been real big fans of this solution in the past, but this new version is a huge step up from past players.

The Elite 59AVi has four different progressive modes to chose from; Auto 1, Auto 2, Pure Cinema On and Pure Cinema Off. Just like the last time we tested this player, we recommend Auto 2. This mode passed all of our tests, including 2-2 and 3-2 Chapter breaks. It also has a very fast recovery time when transitioning between film and video, and vice versa.

Auto 1 does quite a good job as well but had more difficulty with video based material compared to Auto 2. Neither mode does as well as a DCDi based Faroudja solution with video material, and jaggies were quite evident with a few different video clips I had on hand to compare. But Pioneer doesn’t have any issues with macroblocking like Faroudja-based players do.

The core side of the player has some obvious differences between component and HDMI. HDMI is definitely the preferred connection with this player. There is no pixel cropping at all via HDMI, and Y/C delay is perfect. This player also correctly outputs Studio RGB levels via HDMI even when converting to DVI. Another huge plus is support for 480i via HDMI, making this an excellent candidate for future HDMI-based video processors. This is especially so, since the 59AVi also supports iLink for high resolution audio and movie soundtracks.

The component side doesn’t fair near as well. There is over a full pixel of Y/C delay at 480p. The 59AVi has a chroma delay adjustment in its video setup menu, but it only allows integer pixel increments which is too large to fix the problem. The player also clips a combined 7 pixels from the right and left side of the image via component. The component video white level is at the edge of our acceptable standard and measures in at 102 IRE.

One of the other big drawbacks of Pioneer players in the past has been chroma issues. This player is really no exception, but the main 3-2 and video based CUE errors are gone. But with material that uses an alternating 3-2 cadence (most of the Disney animated DVDs), you will see a flickering problem in solid colors. The Pioneer solution also doesn’t have a chroma filter for 4:2:0 ICP, while the Faroudja, HQV, and DVDO solutions do have this.

In terms of responsiveness, the Pioneer Elite is a bit above average. Menu navigation and chapter skips are pretty fast and never an issue. The layer change is a tad sluggish and measures a bit over a full second. The user interface itself is quite good though and offers plenty of tweakability and lots of set-up modes for video performance.

Summary: I would still rate the Pioneer 59AVi as one of the best DVD players on the market today at any price point. It is also one to consider for use as a total digital transport (universal player) in the months to come. With more and more HDMI-based video processors coming out in the next year, this is a fine player to use with them. I look forward to seeing if Pioneer has a successor for this player at CEDIA-2005.

Video Frequency Response
Pioneer Elite

LG - DVB418

MPEG Maker: Mediatek MPEG Model: MT1389FE Deinterlacer Maker: Genesis Deinterlacer Model: FLI-23xx     MSRP: $249.95 Website: http://us.lge.com/index.do

This is essentially part two of the review for the LG DVB418. I wrote a separate review on the 480p component section of this player as it is almost completely different from the up-converted component or HDMI outputs. While all modes rely on the Mediatek MPEG decoder that I am so fond of, the HDMI and higher component resolutions utilize a Genesis/Faroudja FLI-2300 series video processor. This makes the performance vastly different then 480p component which just uses the MPEG decoder for all of its processing.

In my opinion, the only high point for this player is its responsiveness. The Mediatek chip is extremely fast, making menu and chapter navigation a breeze, and layer changes a thing of the past. This player takes full advantage of that, but, unfortunately, the kudos stop there.

First off, the 418 shows the “macroblocking” issue that has become the Achilles heel of all Faroudja-based players lately. And it is quite bad, especially via component video. The reason is the black level of this player is set to 7.5 IRE which lightens the darker areas, and this is where the problem is most noticeable. Using my regular test scene, Chapter 22 of A Bug’s Life, not only was the artifact clearly visible in all the backgrounds, it extended up into the letterboxing on the top and bottom of the image. This rendered the picture just about unwatchable.

This player outputs the correct RGB levels via HDMI, and passes below-black, unlike the component outputs. It also outputs the correct Studio RGB levels if the HDMI is converted to DVI. This made the macroblocking not as bad with my test scene, but still clearly visible. But the flipside to this coin is the white level which is WAY too high via HDMI. The white level is so hot via HDMI that I had to bring my monitor’s contrast level to almost zero to eliminate the majority of the clipping. Completely unsatisfactory.

There are also several chroma issues with this Faroudja implementation. LG turned the cross color suppressor to ON and left it there. This can result in artifacts with color, including flickering and noise. There is also some extremely strange ghosting that occurred with our chroma test patterns. We use the Microsoft WHQL Test Disc 3 for our chroma tests now. These tests have fish that swim around an aquarium at different speeds and they are completely red. There was no tearing or obvious jaggidness to the fish since the Mediatek doesn’t have CUE issues, but the color appeared to lag with movement, creating a strange ghosting affect. After seeing this, I tried a few different animated DVDs at 720p via component and via HDMI at various resolutions. I used Shark Tale and The Spongebob Squarepants Movie and some other films as well. Thankfully, none of the same chroma delay anomalies were noticed and the image looked stable.

The core side of the up-scaled component video is quite bad. The video frequency response is all over the place and goes up in the middle, causing some artificial ringing in the image. The upper end is also rolled off a bit, making the image seem soft in the fine detail department. 720p and 1080i via component also suffer from over a full pixel of Y/C delay and Pb/Pr delay. It is unusual to see Pb/Pr delay in a player, but the 418 has it quite bad. Pixel cropping via component was also horrible, with 18 pixels clipped from the right side of the image at 720p and 21 at 1080i.

The HDMI output doesn’t improve things much. Pixel cropping is still bad with the right side consistently cropping about 17 pixels regardless of resolution. Thankfully, there is no Y/C delay via HDMI.

Summary: The LG DVB418 is a player with numerous bugs to work out. There is a lot of potential for a great player here if LG would have been a bit more careful in their implementation. Attention to detail can go a long way in the video department, and I really hope LG ups the quality control the next time around. They have a new HDMI-based DVD player due later this summer, and I hope it is a big improvement over this one.

Video Frequency Response
LG

Toshiba - SD-5980

MPEG Maker: Zoran MPEG Model: Vaddis 778 Deinterlacer Maker: Zoran Deinterlacer Model: Vaddis 778     MSRP: $149.99 Website: http://www.toshiba.com

Like the Samsung HD-850, the Toshiba SD-5980 is an OEM-based DVD player (meaning that a different company builds it for Samsung) that revolves around the new Zoran video processing. This includes the Zoran Vaddis 778 and the HD Extreme chip for scaling. These players are priced identically, and both provide an HDMI output, and even include an HDMI cable. The Toshiba adds a few more features, including some slots for different media cards (digital camera memory cards) for slide shows.

The Vaddis 778 handles the MPEG decoding and the de-interlacing, and does an average job with both. I was happy to see that Zoran has made strides with their chroma performance since the Vaddis 5 chip. This new chip doesn’t have any signs of the chroma bug and even does surprisingly well with 4:2:0 test material. This test normally requires special chroma filters that are usually found on high end video processors, but Zoran has done a remarkable job with their all-in-one solution.

Toshiba has included three different progressive modes for the end user to choose from. These include Auto, Film, and Video. The Auto and Film modes performed exactly the same, while the Video mode dropped the film cadence processing and didn’t improve on the video processing a bit. Because of this, we only suggest using the Auto mode, as it handles all material equally well. This player is not on par with the better de-interlacers out there, but it will handle the majority of DVDs just fine.

The core side is a bit of a hit or miss for this player. The HDMI output will not output proper RGB levels when converted to DVI which results in a loss of below-black and above-white information. If your display has an HDMI input, you should be fine though. Via component video output, this player’s video frequency response is a bit rolled off in the upper end, but not as much as the Samsung HD-850. It does have a smidgen of Y/C delay, and the player is a bit slower then the Samsung HD-850. The layer change clocked in at 1.25 seconds which isn’t bad, but it is a full second slower then the 850.

Overall, I don’t know which of these two I would pick personally. Neither has the video processing that would fit my needs, but I think either would be fine for most consumers. I wish there were a fusion of the two players out there since both have their highs and lows. If you are considering either of these two players, you should probably make sure you test both of them in the store before making a decision.

Video Frequency Response
Toshiba

Teac Esoteric - UX-1

MPEG Maker: Mitsubishi MPEG Model: M65776CFP Deinterlacer Maker: Genesis Deinterlacer Model: FLI-2310     MSRP: $13000.00 Website: http://www.teac.com/esoteric/

The Teac Esoteric UX-1 Universal DVD player represents the most expensive player we’ve tested to date. I am generally not very fond of boutique players (definition: made in small quantities, sophisticated appearance, very expensive). Some consist of a rebadged mid-line player with a spruced up audio section and a pretty packaging but really nothing special in terms of performance. I was almost afraid of what I might find with this player in those regards, but thankfully, I was pleasantly surprised.

The UX-1 is an original player from the ground up. If I were to compare it to anything else design wise, it has some striking similarities to the rare Samsung HD-1000, but these are loose similarities. The player is a work of art and without a doubt the nicest looking and best built player I’ve ever seen. The entire outer shell is made of brushed aluminum, and even the tray is made of the same material. The connectors on back are enormous and as solid as anything I’ve ever used. Even the buttons on the front face plate and their backlighting would make most players envious.

But for the Benchmark, those things don’t mean much. At the end of the day, it is about video performance, and this is where boutique players very rarely, if ever, fair well. The UX-1 may be the new exception to that rule.

The UX-1 uses a Mitsubishi MPEG decoder coupled with a Genesis FLI-2310 video processing chip. Just to get the bad news out on the table; the player does suffer from the macroblocking problem commonly associated with this chip. In fact, it is probably one of the worst with the issue so far and probably only second to the previously mentioned Samsung HD-1000. Using my reference scene from A Bug’s Life, the artifacts were clearly visible and quite distracting. If you are considering this player be sure to test it out first on your display if possible, as the problem only manifests itself with certain displays and may not affect yours.

The player features component progressive outputs, via BNC connectors, and via DVI with support for 480p, 720p, and 1080i with HDCP compliant displays. It also has full support for PAL playback in these modes and another bonus, full PAL to NTSC conversion. It is very rare that a manufacturer actually tells you that they support this feature but Teac included the option in its setup menus.

Attention to detail does not stop there either. When playing back via DVI there are some nice features for set-up, including selection of RGB mode. Here you can select either PC RGB levels or Studio RGB levels. The problem is, the Studio RGB levels are set incorrectly. The engineers at Teac set the black and white levels to digital 16 and 235 appropriately, but unfortunately they cut out the head and toe room that allow for below-black and above-white content. I am working with their engineers now to resolve this issue, and they have been very responsive on trying to get the problems worked out. This is extremely rare for a high line outfit in my experience, and I applaud them on their wiliness to support their product to the utmost degree.

The Faroudja implementation is extremely good here with only one small caveat. There is no support for 2-2 processing with NTSC material. The player does however support 2-2 processing with a PAL output, which is completely appropriate. But there is a lot of 2-2 based video material authored in the NTSC format so I hope that they will add this with a future firmware.

On the core side the player is good, but not perfect. There are no signs at all of Y/C delay, and the video frequency response is superb. Via the component outputs, the white level is spot on at 100 IRE, but again the player has issues with below-black information. There is also has some pixel cropping on the left and right side of the image that is a little excessive. I am working with their engineers to get this solved.

The player has no chroma issues at all, and the chroma filters and cross color suppressor settings of the Faroudja chip all seem to be set appropriately.

Overall this is by far the best boutique player I’ve had the opportunity to test, but at this price point I wouldn’t expect anything else. It still has its shortcomings which I hope Teac works out, but its performance was far better then I expected. Stay tuned for our review on the audio performance, which is excellent.

Video Frequency Response
Teac Esoteric

Samsung - HD-850

MPEG Maker: Zoran MPEG Model: Vaddis 778 Deinterlacer Maker: Zoran Deinterlacer Model: Vaddis 778     MSRP: $149.99 Website: http://www.samsungusa.com

The Samsung HD-850 is almost identical to the Toshiba SD-5980 that we are also featuring in this edition of the Benchmark. They both rely on the exact same chips but use a vastly different user interface.

The 850 is obviously Samsung’s version of a stripped down 841. It supports HDMI and scaled resolutions of 480p, 720p, and 1080i, but they got rid of DVD Audio and SACD support. For video processing they chose to go with new solutions from Zoran. The Zoran Vaddis 778 is employed for MPEG decoding and de-interlacing duties, and the Vaddis HD Extreme is used for scaling.

Like most Vaddis chips, the 778 is OK with film based material that has good flagging, but it’s not the best with cadence breaks or video material. In fact, it isn’t even motion-adaptive. The image breaks up with any change in cadence, and I noticed some pretty severe jaggies with video-based material.

The HD-850 is different than the Toshiba SD-5980 in that it offers only one progressive mode with no user selectable settings. Some may call this a big deal, but there was really no need for one since even the modes that Toshiba adds don’t bring any performance advantages. This chip has equal performance for both film and video and supports a 3-2 and 2-2 cadence.

The core performance is where you see the most separation between the Toshiba and the Samsung. The Samsung has user adjustments for RBG levels via HDMI including settings for PC and Studio RGB levels. Of course, we always recommend using Studio RGB levels, as this is the format DVD is mastered in. If you have a DVI-based display though, this player incorrectly converts the RGB levels to PC so you will not get above-white or below-black. But the levels are correct if going HDMI to HDMI.

The player has a full pixel of Y/C delay via its component outputs. The Toshiba is a bit better in this regard. The layer change was faster than the Toshiba, coming in under a second,and the menu navigation was also quite speedy. This player is also better in terms of pixel cropping,but its video frequency response via component video is a bit more rolled off than the Toshiba, so if component video is important to you, the image may be a bit softer. Another downside of the 850 is its white level via component video. This player measured at 97 IRE, which is on the low side. The Toshiba was closer to the reference 100 IRE mark and will produce a punchier picture.

The Vaddis 778 did very well in our chroma tests. Even the ICP 4:2:0 test looked terrific without chroma filters. It is nice to see that Zoran has addressed this issue with it newer chips.

Overall, the Samsung is about the same as the majority of the players out there. It has its issues which are mainly trivial, and it is surprising that Samsung doesn’t address these things before they get out. It would be pretty simple. The de-interlacing is average overall and should do fine with most DVDs on the market today. For the discerning videophile, I would have to suggest you look elsewhere. Since the Toshiba and the Samsung have identical de-interlacing performance, you should pay attention to the core performance and figure out what is important to you. Both players have caveats, but some may or may not matter to you.

Video Frequency Response
Samsung

Pioneer - DV-588A

MPEG Maker: Mediatek MPEG Model: MT1389EE Deinterlacer Maker: Mediatek Deinterlacer Model: MT1389EE     MSRP: $129.99 Website: http://www.pioneerelectronics.com

This was a big surprise entry in the April/May DVD Benchmark. The Pioneer DV-588A is the successor to the DV-578 which I reviewed last year. I was quite impressed with the DV-578 and was pretty anxious to see how Pioneer would follow it up. This new offering is almost identical to the 578 but adds a few new features including DivX and Windows Media Audio support.

Like the DV-578, the 588 utilizes a Mediatek MPEG decoder for all of its video processing duties. So again, despite the “Pioneer Pure Progressive” logo on the front, this is not a Pioneer video processing design. The player only has its default de-interlacing mode, but it did have quite a few adjustments for things like contrast, brightness, hue, and gamma.

Overall de-interlacing performance was just a touch above average, and this is definitely one of the better implementations of the Mediatek chipset. Thankfully, the player is motion adaptive, which few Mediatek players are. With all of our 3-2 cadence tests, the Pioneer performed well, but did trip up a bit with some breaks in cadence. This player supports PAL playback and does a pretty great job at PAL to NTSC conversion much like the 578. The 588A also supports a 2-2 cadence and did quite well with our Natural Splendors test scene.

The core side of this player is where I had the most issues. The biggest offenders were Y/C delay and below-black performance. Th 588A would not pass a below-black pluge signal, and Y/C delay was greater than 1 pixel on both the Pb and Pr channels. This is what we saw from the 578 in core performance and is a rather surprising overlook by Pioneer. I am hoping that they will address both of these issues with a firmware update soon.

The player’s video output level was spot on at 100 IRE, and the frequency response was almost ruler flat with only a slight nudge up in the mid and upper frequency range. The sharpness test pattern from AVIA Pro didn’t show any signs of ringing, and the player resolved the full resolution potential of DVD perfectly.

This player does have some pixel cropping issues. The right and left sides of the image combined for 7 pixels cropped, while the bottom clipped off 1.

The best part of the Mediatek implementation here has to be the user interface and responsiveness. The 588 has a completely seamless layer change, and its menu and chapter navigation are extremely fast. I did knock a bit of a point off for its load times though, which I felt were a tad sluggish. The player also did extremely well with our chroma tests, passing all of them easily with the exception of the 4:2:0 ICP test. Surprisingly, this chip didn’t do near as well as it normally does with that test, and streaking was blatantly obvious.

The 588 is a budget universal transport, supporting DVD-Audio and SACD. This support includes full time alignment and bass management, though the time alignment is a bit more limited for the SACD side. Unfortunately, there was no way of telling what crossover point Pioneer uses for the bass management, but based on the available specifications for Burr Brown PCM DACs, I can pretty much conclude that this player does convert DSD to PCM regardless of the setup. While I don’t find this to be a big deal at all, I know there are some people who will frown on this (it is pretty common though).

Summary: Overall, the Pioneer DV-588A is an excellent player for the money and on my short list of recommended players under $150. In reality, I would almost call it a steal. I am quite surprised that some of the same issues that were readily apparent with the DV-578 showed up again with the 588. What is the point of a new model that doesn’t address problems in older models? I really hope that Pioneer takes the time to fix these issues with a future firmware update.

Video Frequency Response
Pioneer

LG - DVB418 - Component 480p

MPEG Maker: Mediatek MPEG Model: MT1389FE Deinterlacer Maker: Mediatek Deinterlacer Model: MT1389FE     MSRP: $249.95 Website: http://us.lge.com/index.do

You can just about chalk this one up as one of the most difficult DVD players I’ve had to test. If this player would have had any larger selection of de-interlacing modes, I would have gone stark raving mad.

The LG DVB418 is the first HDMI or progressive scan player that I’ve seen on the market from LG. This company seems to be going gangbusters, as I’ve seen more and more of their products hitting the shelves this past year. When I saw that they had an up-scaling DVD player for less then $200 street, I figured it would make a nice addition to the benchmark.

Because of the way LG designed this player, I had to break it up into two sections. It uses its MPEG decoder for de-interlacing duties via component video at 480p, but a different solution for 720p and 1080i via component, as well as all resolutions via HDMI. Yes, this player does scale to 720p or 1080i via component but only for un-copyrighted DVDs. The majority of studio DVD releases on the market today won’t play at those resolutions.

The 418 utilizes a Mediatek MPEG decoder for its core processing. Judging from the huge influx of DVD players this year using this chip, it is obviously becoming a very popular solution, which is understandable given its performance attributes. It is extremely fast, handles both PAL and NTSC including PAL to NTSC conversion, and it has excellent chroma performance. It is easily becoming one of my favorite MPEG decoders.

For 480p component video, the de-interlacing duties are handled by the Mediatek chip. Why they did this I have no idea because in the same player is a Genesis FLI-2300 series chip. Which series the Genesis chip is, I am not quite sure, as it was underneath the main video board, and since I planned on returning this player, I didn’t want to risk breaking it.

This makes the overall performance of the 418 very different, depending on the mode. For 480p, the Mediatek did quite well, and this is definitely one of the better implementations I’ve seen. The player is motion adaptive and passes the majority of our de-interlacing tests, including 2-2. It did have some issues with 3-2 Video flags though, as well as the chapter breaks. This is a pretty common problem with most MPEG decoder-based solutions.

The biggest downfall of this player regardless of which mode is used is the core performance. Via component, the video frequency response is excellent, but pixel cropping is hit or miss. At 480p, there are 7 pixels cropped from the sides of the image and 6 from the top and bottom. At 720p, there are 18 from just the right side alone, and that number increases to 21 at 1080i!! The player also doesn’t pass a below-black pluge via component, but at 480p, the black level is set appropriately to 0 IRE. When you engage the Faroudja processing at 720p or 1080i via component, the black level shifts to 7.5 IRE, and there is no way to change it. The white level is also extremely hot at 105 IRE when outputting 480p. At 720p it is toned down to a usable 102 IRE, but that is still on the edge of our acceptable range. The player also has over a full pixel of Y/C delay at 480p.

Summary: Despite what the numbers might suggest, the 480p component output of the LG DVB418 is the best output you’re going to get. You can read the specifics of the up-converted component performance as well as the HDMI performance in the review for the other section for this player. But I warn you, it isn’t pretty.

Video Frequency Response
LG

Oritron - DVD 4119

MPEG Maker: Zoran MPEG Model: Vaddis 6 Deinterlacer Maker: Zoran Deinterlacer Model: Vaddis 6     MSRP: $39.99 Website: www.oritron.com

The Oritron DVD 4119 can be bought for less then $50. It is essentially a drive and a small PC board wrapped in a diminutive shell similar to the Cyberhome players you see at Wal-Mart and Best Buy. Overall performance is a bit below average for a DVD player, but I wasn’t expecting a lot at this price point.

The 4119 is based entirely on the Zoran Vaddis 6 MPEG decoder. This chip does all of the video and audio processing for the player, including de-interlacing duties. This is one of the cheaper Zoran chips and doesn’t perform at the level of the Vaddis 5 or 778 which seem to be a bit more popular amongst manufacturers.

This player is not motion adaptive, which I consider a deal breaker with any progressive scan player. At this point in the game, all DVD players should pass the test, even at $50. It did reasonably well with film-based cadences, but any break in cadence or bad flag resulted in obvious combing. The player does recovery pretty quickly though, so this shouldn’t be a huge distraction.

Like most flag-based players, the 4119 handles a 2-2 cadence pretty well. This is great for video material. But when I watched some video-based material, jaggies were clearly evident with horizontal lines. This was also extremely evident using the 3 line test on the Silicon Image test disc. Those of you who like to watch DVDs of TV shows may want to look around some more if you don’t like a jagged image.

The core performance of the player was decent but not great. Pixel cropping was a bit extreme on the right side of the image with 8 pixels cropped, but no other sides had any cropping. The player has no signs of Y/C delay, and its video frequency response was only a bit down in the higher frequencies. This contributed to a slightly softer image than with higher line players. It does pass a below-black pluge pattern, and its white level is at a respectable 101 IRE.

On the interface side, this player did about average in terms of speed. Menu navigation and chapter skips were quick, but not blazing, and the layer change clocked in at around 1.5 seconds.

The player did quite well in our chroma tests, and this was probably the most impressive aspect of its performance. Even with our 4:2:0 ICP test, it did not show any signs of the infamous CUE problem. There was no banding or jagged edges at all on our fish tests using the Microsoft WHQL 3.0 test disc. Zoran has made great strides in this department with both the Vaddis 6 and Vaddis 778 chips.

Summary: In the realm of under $50 players, the Oritron DVD 4119 fits in with the rest of the pack. Its video performance is average at best, and generally not too bad. If you are on a tight budget, you might prefer a Mediatek-based budget player, but this one would still be OK.

Video Frequency Response
Oritron

Sony - DVP-NS775V

MPEG Maker: Sony MPEG Model: CXD9804R Deinterlacer Maker: Sony Deinterlacer Model: CXD9804R     MSRP: $129.99 Website: http://www.sony.com

First off we’d like to give a huge thanks to Derek Smith for giving us the opportunity to test this player. Derek graciously volunteered his player and even came over when I did the tests.

The Sony DVP-NS775V is a mid to lower line entry from Sony that doesn’t have many bells and whistles other than support for SACD playback. The 775 is built on a Sony platform all in one video chip that handles both MPEG decoding and de-interlacing for progressive playback. Like the last few Sony players I’ve tested, this one shows that Sony has come quite a way over the last few years in terms of video performance. I look forward to trying out their new flagship ES players that are due out later this year.

The Sony video processing is in line with most standard players on the market right now. It handles film cadences quite well but will occasionally trip up when that cadence changes. So with the majority of film DVDs out there, this player should do fine in the de-interlacing department. It does lack a mode for DVDs using a 2-2 cadence though, so video performance may vary.

Equally important to video processing is the player’s core performance. This is the building block to a great picture. Keep in mind that most displays nowadays contain some pretty advanced video processing, including scaling and de-interlacing. Sometimes it is more beneficial to rely on that processing instead of the player’s own processing. But a player’s core component performance is critical so that the display’s processing has the best signal possible to work with.

The 775 fared okay in these tests but could be improved upon. The most glaring issue was Y/C delay. Using a bowtie pattern on Video Essentials and our digital oscilloscope, we measured greater than a full pixel of Y/C delay on both the Pb and Pr channels. Y/C delay will cause a “ghosting” effect on the image that almost looks like a halo on the side of the delay. It is essential that the luminance and chroma channels are perfectly in sync with each other to provide the best video image.

The video frequency response of the player was excellent and just about ruler flat. I did spot some pixel cropping on the left and right side of the image with a total of 7 pixels being cropped. The player did extremely well with our chroma tests. The only one it didn’t pass was the 4:2:0 ICP test which requires a chroma filter that is normally only found in high end video processors.

The player did pass the below-black pluge pattern test, but its white level was a bit high at 103 IRE. This will exaggerate white levels and produce an artificially high contrast level.

The user interface was simple to navigate, and response times were quite good. I thought the initial loading times were on the sluggish side, but menu navigation and chapter skips were quite speedy. Using a full bit rate layer change, we measured the delay at about 0.75 second, which is excellent.

Overall, the Sony is a decent entry level player, but a bit more attention to detail would have gone a long way. It surprises me that so many companies seem to drop the ball on the easy stuff this late in the game. I hope that the next offerings from Sony that are due out later this year are tweaked in a bit better.

Video Frequency Response
Sony

Harman Kardon - DVD-31

MPEG Maker: Panasonic MPEG Model: MN103547JRB Deinterlacer Maker: unknown Deinterlacer Model: unknown     MSRP: $299.99 Website: http://www.harmankardon.com

A big thanks goes out to Jon Oliva who volunteered to send this player to us. I had been on the lookout in my home town but kept coming up short.

The Harman Kardon DVD 31 is their newest progressive scan player and looks to be the follow-up to the DVD 22. Not only does it feature progressive scan playback, but it also adds full DVD-Audio support. The player uses a Panasonic MPEG decoder, but we’re not sure what de-interlacing solution is employed. There was another chipset on the player’s video board, but it was covered by a massive heat sink that I didn’t feel comfortable removing. Based on its processing performance though, I would guess that they relied on the Panasonic chip for de-interlacing duties, as most stand alone de-interlacers perform better than this.

The DVD 31 has three progressive modes that can be accessed in the set-up menus. The modes are Auto, Movie, and Video. Each one is a bit different from the other, but overall, I would have to recommend using the Movie mode. It performed about the same as the Auto mode with film-based content but recovered faster from a cadence change. It also supports 2-2 based cadences which the Auto mode surprisingly didn’t. The video mode does not recognize film flags at all and is for the most part worthless.

The 31 did about average to below average with video processing. It has a tendency to trip up with bad flags or breaks in cadence. It also showed a few video glitches, including some weird shimmering with vertical scrolling titles. The player is motion adaptive thankfully, but it drops out of this mode when it is in a looping state. Since most test patterns just continually loop until you press skip, this artifact was a bit annoying.

The core performance of the player was a mixed bag. We used to be huge fans of Panasonic’s MPEG decoders back in the days of the popular RP-82 and XP-30. Based on the tests we’ve done of the 31 and the new S97 players, it seems that Panasonic is letting their quality control slip a bit with their chips. This was most evident with our chroma tests. The player passed all of them except the 4:2:0 ICP test, which is expected. But, what wasn’t expected was how bad ICP material looked. This is exactly the opposite of the results we saw with the older Panasonic chips.

The layer change was also pretty bad, measuring in at around 2 seconds. The image also broke up consistently every time we did the test because of the pegged out bitrate. This would make layer changes painfully obvious for higher bitrate transfers.

Another downside is the player’s inability to pass a below-black pluge signal. This is another one of those deal breakers for me, since the player is now clipping information in the video signal that was intended to be seen. The white output of the DVD 31 is also right on the edge of our acceptable standard and measured in at 98 IRE.

The frequency response tends to drop down a bit in the upper frequency range, contributing to a somewhat softer look than higher end players. But, when I watched some footage from some of my DVD collection, this was not overly apparent. The image looked pleasing with most of the titles I managed to put on.

Summary: Overall, I wasn’t very impressed with the DVD 31. It definitely could use some tweaking in, especially in its core performance which is critical to achieving a great image. If you are looking for a player in this price range, I would have to recommend looking elsewhere.

Video Frequency Response
Harman Kardon

McCormack - UDP-1

MPEG Maker: Mitsubishi MPEG Model: unknown Deinterlacer Maker: Pioneer Deinterlacer Model: Pure Cinema Progressive     MSRP: $3495.00 Website: http://www.mccormackaudio.com/

The UDP-1 uses a Pioneer (re-badged Mitsubishi) MPEG decoder and Pioneer’s Pure Progressive de-interlacing technology. This seems to be based on the older video processing that was around with the DV-45 or 47 days and not the newly improved DV-59AVi performance.

For de-interlacing, the player offers three modes; Auto, Pure Cinema On, and Pure Cinema Off. We recommend just using Auto, as Pure Cinema 1 had the same results, and Off didn’t seem to do anything right. Even in the Auto mode though, this player’s performance was far from perfect and didn’t perform near as well as most other players we have tested. This de-interlacing solution is flag-based and had the tendency to trip up with almost every break in cadence we threw at it. It is also not motion adaptive, which should be a given this late in the game.

The player failed the majority of our de-interlacing tests including our high detail and bad edit tests. The UDP-1 does offer 2-2 processing though, but its performance with diagonal lines in video-based material was very poor. Jaggies were quite evident using the popular Faroudja flag test and the clock test from the Silicon Image test disc.

On the core side, the player did about average. The video frequency response has a bump in the mid band that may cause some exaggeration in detail and some slight ringing in the image. Unlike the Pioneer players it’s based upon, the UDP-1 did a great job with our chroma tests. They seem to have employed a chroma filter that Pioneer didn’t use in their releases. Y/C delay was evident with about a full pixel of delay showing on our oscilloscope. There was also some pixel cropping, with a combined 7 pixels missing from the left and right side of the image.

McCormack has said before that the video side of this player was not the focus when they built this machine, but they felt it was decent enough for most viewers. I don’t completely agree with this since there are plently of machines that cost a fraction of this player that do a far better job in the video department. The audio side is quite good though, and we will be following up these tests with a full review of the audio side of this player. I guess it really depends on what your main concern with a player is: the video or the audio. The UDP-1 seems to focus on the audio performance.

Overall, this is not a player I can recommend for video performance. Like the older Pioneer Elite players, there are just far too many issues to be worked out for the discerning video enthusiast.

Video Frequency Response
McCormack